Psych 120A Final

Representativeness heuristic

-"Steve is very shy and withdrawn, invariably helpful, but with little interest in people, or in the world of reality. A meek and tidy soul, he has a need for order and structure and a passion for detail" -Is Steve more likely to be a: (A) salesperson (B) librarian -in fact, steve is 83 times more likely to be salesman than a librarian. there are more than 15 million salespeople in the US and only 180,000 librarians -Another example: Kahneman & Tversky (1973) •Showed subjects personality descriptions allegedly sampled at random from a group of 100 professionals (lawyers & engineers) •Subjects told: 70% lawyers, 30% engineers (BASE RATE) -Description of an individual: Dick is a 30-year old married man with no children. A man of high ability and high motivation, he promises to be quite successful in his field. He is well-liked by his colleagues -Lawyer or engineer? -subjects predict: 50/50 -Base-rate neglect! -People judge probabilities based on the degree that the situation is similar to, or representative of, their stereotypes or knowledge -They do this even when there is other information that a rational person would use to make the best possible decision -Base-rate neglect: tendency to ignore the "prior probability" of an event -People make incorrect judgments even when they are explicitly informed about the base rates •Heres an example that researchers did their first experiment on representative heuristic and people were presented with these little vignettes just a paragraph about this person and their qualities •then participants in this experiment were asked is steve more likely to be a •salesperson or •a librarian •most people say librarian •but actually steve is 83 times more likely to be a salesman than a librarian just if you base this on the overall probability of someone in the world being a salesman vs. a librarian •in this case people are neglecting those base rates and going with the representativeness of how much a typical librarian fits this description vs. a typical salesperson •this description doesn't seem very representative of a salespersons qualities •but they did a later experiment where they tried to overcome some of these limitations •in this one they showed subjects personality descriptions that were sampled from a group of 100 professionals •and they told subjects that 70% of the personality descriptions are of lawyers and 30% were of engineers so they told them the base rate •and then they pulled a card and read it to the participant •"dick... •this is a pretty neutral personality description there aren't a lot of qualities that are specific to lawyers or engineers and then participants were asked lawyers and engineers •and most people said 50/50 •because its ambiguous as to how these qualities weigh in some people guessed lawyer some guessed engineer they didn't have a lot to go off of •so on average about a 50/50 split but really if they didn't have much to go on they should have guessed lawyer because lawyer is a much more likely guess in this deck of index cards •and the same thing applies to salesperson and librarian that if you have to pick the likelihood that someone is one or the other the baserate should matter - the rate at which these things occur in the world and people tend to ignore that •the definition of representativeness heuristic is that "people judge probabilities based on the degree that the situation is similar to, or representative of, their stereotypes or knowledge •and people do this even when there is other information available to them like a 70% 30% detail that a rational person a person that is using proper logic to come up with their judgement would use to make best possible information •so people ignore base rates and tend to base things on the representativeness of something rather than the prior probability that it would be true •and this leads people to make incorrect judgements in many contexts even when they are explicitly informed about base rate

Mental Sets Examples

-15% of the people in Topeka have unlisted numbers. You select 200 names at random from the Topeka phone book. How many of these people will have unlisted numbers? -many people come to this problem with the wrong mental set - they think of it as 15% of 200 is 30 - they are thinking if you have a random sample of 200 names and you know the base rate here of 15% so 30 would be the answer •but the correct answer is 0 because an unlisted number wouldn't appear in the phonebook - so the people who are listed are already the 85% who chose to be listed the 15% who were unlisted would not appear in the phonebook so in order to think about this problem you have to break free of the mental set of trying to treat it as a mathematical equation •heres another insight problem that illustrates this mental set •type 1 problems are like this so in each case you get a false statement - so 3 equals 9 minus 1 so that's not true so you get to move one match here and try to achieve a correct equation •so how can you do that here? •You slide the X over and make it a V so 3 = 4 - 1 •So participants did a lot of these kinds of problems in this experiment •Type 1 insight problems - its an insight problem in that you have to think about it and realize oh its not about changing the operation here its about here I can change this X here from having a value of 10 to a value of 5 •And then participants got other problems like this called Type 2 and type 3 problems so now we have 6 = 6 + 6 so that's not going to work •Move the positive sign into an equal sign so 6 = 6 = 6 •So people have a lot of trouble with this one because they are stuck in this mental set of trying to change the numbers here and they are not realizing here that changing the operation here achieves this •Having two sets of equals is not typical but it does work •Here is another one so 9 = 6 - 3 •Again type 2 and 3 problems involve changing the operations so people didn't do as well with these •Heres another problem so this is the 9 dots problem - your goal is to connect all nine dots... •Lines need to be connected that's one of the constraints and you need to connect all 9 dots and you get no more than 4 lines •Initially people think of this and they try things like this.. •Or they try something diagonal... •It seems unfair because your mental set made you think you needed to stay within the box but the key here is to think outside the box •Another solution people sometimes have is this one but it only works if the dots have some degree of space so that you can touch these ones - if the dots are just points with infinitely small area than it doesn't work but if you think of the dots as circles then some people will do this •Heres another problem like that - you have a candle, matches, and tacks and your goal is to get the candle onto the wall •So this is what you are given - the initial state- candles, matches, and tacks •Your goal is to get the candle onto the wall •In this solution people try doing various things - they try to tack the candle onto the wall that is a common suggestion •Or use the melted wax in some way to glue it •The solution is tacking the empty match box onto the wall in some way you can put the candle on it like this or tilt it sideways and put candle inside of it and this will be a candle holder •So the mental set you have to break free of is trying to think of the matches as being something that will be useful to you for just lighting the candles and here you are sort of thinking the match box has some usefulness

Prospect Theory

-A bigger impact of losses than of gains -In general, we are more loss adverse... -Losing $50 compared to winning $50! •Kahneman and Tversky developed a theory called prospect theory that operationalizes this •If you plot value on this axis and outcome from gains to losses on this axis you see this asymmetry in the curve •So in both cases there is diminishing returns so as you gain more and more it begins to be of limited value •If you could have an opportunity to win 1 buck or 2 bucks that's a big difference or 50 vs. 100 that's a big difference psychologically •But if it's a 1000 dollars vs. 1001 or even 1000 vs. 2000 it starts to dimish in both cases it's a lot and the added value (psychological value) diminishes •And with losses you see this as well •But the point here is the asymmetry so if I plotted an equivalent point •So lets say this is plus 50 dollars and minus 50 dollars - the positive impact for your well being and happiness of gaining $50 is not mirrored with the negative impact of losing 50 dollars •The prospect of losing 50 dollars looms larger and is more negative and more imparitive that you avoid that than the prospect of gaining $50 here •So when you are at the same point plus 50 or minus 50 there is this asymmetry in the curves •So losing compared to winning 50 •And the thing to remember is that losses loom larger than corresponding gains so to get the equal behavior here you would need to be at a much lower loss - maybe like 20 dollars - a loss of 20 dollars has the equivalent psychologically in terms of bringing negative feelings as the positive feelings of gaining 50 so its not a 1 to 1 in losses to gains •And that's what prospect theory shows •And this is true for a wide range of experiments

Using logic to make deductions (conditional statement)

-A conditional statement has the format "If X, then Y." -The first part (antecedent) provides a condition under which the second part (consequent) is guaranteed to be true. Modus ponens -Premise: -If P, then Q -P -Conclusion: Therefore Q -Arguments of this form are valid Affirmation of the consequent -Premise: -If P, then Q -Q -Conclusion: Therefore P -Arguments of this form are invalid •now we are going to talk about conditional so another form of logical reasoning - these are if/then statements •so if X then Y •so the first part is called the antecedent and the second part is called the consequent •so the if X is the antecedent and the then Y is the consequent •and how these work is that if the first part is satisfied then the second is guaranteed to be true that is the set up of a conditional •so there are different ways of setting these up •one is called modus ponens: and these are logical statements in the form of if P, then Q so that's the set up here •so if P, then Q and then you are saying P is true and then you are asked to evaluate this conclusion •therefore Q •arguments of this form are valid •so if P is true, then Q •and then Im telling you P is true - therefore Q •so this is valid •its easier when put into concrete terms: if farm has peacocks, then they will have quails •the farm has peacocks •therefore, they have quails •affirmation of the consequent is another kind: •premise: if P, then Q •Q •Conclusion: •Therefore P •This is invalid •If farm has peacocks, then they will have quails •The farm has quails •Therefore they have peacocks •Invalid •I never said that because I have quails I have peacocks I said if I have peacocks then I have quails •People are affirming the consequent and it's a logical reasoning error •The consequent the second part here - the Q

The Bronze Coin Problem (Insight)

-A stranger approached a museum curator and offered him an ancient bronze coin. The coin had an authentic appearance and was marked with the date 544 B.C. The curator had happily made acquisitions from suspicious sources before, but this time he promptly called the police and had the stranger arrested. Why? -Metcalfe & Wiebe (1987): gave subjects the Bronze Coin problem and complex algebra problems -Subjects made "feeling of warmth" ratings (how close to the solution) every 15 sec before they arrived at the solution. -For insight problems, there is no linear ramp-up of "feeling of warmth" ¤ no subjective sense that one is getting closer to solving the problem ¤ Feelings of warmth do not predict accuracy of solution ¤ "Aha!" moment may only mean that you've discovered a new approach to the problem, not necessarily the solution -A period of "incubation" can often help ¤ walk away from the problem for a while ¤ The solution may come in a flash •when you know the answer to this it will come to you like - youll be 100% sure that you have figured out because its one of these insight problems - it feels like all or none in a sense - like you aren't close and then suddenly you have this insight •there isn't a lot of information here •BC which means before Christ is something that could never have been marked on a coin •We have decided as society and a world that we would demarcate the labels of dates based on relative to the age of one person and when we count backwards we are counting backwards from 0 so 544 BC was not the year that people would put on their calenders in 544 BC - its only something in retrospect we have applied with anything that happened before the year 0 we are going to call it 1 BC, 2 BC and so forth •So there couldn't have ever been a coin with the date BC its not how the dating system works - this is all relative to an event that happened 544 years in the future - so that was the moment of insight you needed for this problem •In work by Metcalfe and Wiebe in the 80's they gave subjects the bronze coin problem and other complex problems that were not insight problems but algebra problems that people had to think about and work through •And subjects were asked to rate what they called the feeling of warmth - how close are you to the solution - so as you are thinking about this one you are gaging am I getting closer? And they had to turn a dial and say how close they felt on this scale of 1 to 7 •So 0 is cold and 7 is you've solved it •And for algebra problems you could see that people were starting out with a little higher feelings of warmth but that the increase of the feeling of warmth was gradual - so this was the time before the solution - so the solution happened here, this was 2 seconds after the solution •But for the algebra problems there is this gradual ramp up - it starts out here kind of mid way getting warmer getting warmer •So the subjects in this experiment knew I feel like im getting closer to the solution •For the insight problems it was more sudden - so theres this slight bump up her from 2 to 3 but its really this non linear jump •This sudden jump that you don't have the solution you don't have the solution and then when you look at the moment they solved this •So here when they solved it they know they solved it and when you go here 15 seconds before they solved it they were very cold so something happened within those 15 seconds before they solved it where they had this sudden insight • It turns out its not related to accuracy - so when you look at warmth on average for an incorrect answer and warmth for a correct answer if anything people felt they were warmer on the trial •so this is sorting the trial after the fact •so people get some problem some they ultimately get a correct answer some they give an incorrect answer •everytime they give an answer they feel they are super warm cuz they got to their answer so their warmth ratings are matched here •but going back if you look at their warmth ratings people on incorrect answers felt they were warmer than those who had correct answers •there is really no relationship or no positive relationship between feelings of warmth and being accurate •so for insight problems there is no linear ramp up of feelings of warmth - saw that on the graph two slides ago •so no subjective sense that one is getting closer to solving the problem just sort of comes to you in this sudden aha moment •and feelings of warmth do not predict accuracy of solution - feeling like you are close is not a useful gage for how close you are for insight problems it is for algebra problems and other kinds of well-defined problems that involve incremental steps and not this moment of insight •and another point here is about incubation - this is still a bit of a mystery why incubation works it seems to be something about what happens when the mind is at rest or wandering and having random associations or spreading of activation from one set of ideas to another set of semantically related ideas - but something about letting your mind calm down and letting the problem incubate or sometimes people call it marinating •you don't even have to be actively thinking about it and in some moment it can pop into your head - it can happen while you are lying down in a field or for some people in their dreams it could happen while you are walking down the street and suddenly a solution comes in it could be minutes later or days later and the solution may come in a flash •in general this is a phenomenon people describe in solving various complex things like physics or architecture etc

Using logic to make deductions (syllogism)

-A syllogism consists of 2 premises followed by a conclusion -Is this syllogism valid? -All P are M. -All S are M. -Therefore, all S are P. In concrete terms: -All Plumbers are Mortal. -All Serial killers are Mortal. -Therefore, all Serial killers are Plumbers. -The Atmosphere Effect: being influenced by the global impression or "feel" of the premises -All dogs are animals. Some animals are pets. Some dogs are pets. -"Sure, I have a pet dog, so that logic is true" -All sharks are animals. Some animals are pets. Some sharks are pets. -"A pet shark? No way!" -Belief bias involves a tendency to: -Accept invalid conclusions if they are believable -Reject valid conclusions when they are unbelievable •We will start with deductive •So one type of logical reasoning that is useful in demonstrating peoples abilities to make deductions in different kinds of scenarios •Syllogistic reasoning •So a syllogism is 2 premises followed by a conclusion •So it's a good example of deductive reasoning •You have two principles about the world which im telling you in the set up are true •So this being a premise you have to assume these are true •So with deductive reasoning premises are known to be true •So all P are M •All S are M •And this is what you are trying to evaluate, the conclusion: Therefore, all S are P •And the question is is this a valid deduction given these premises •Class voted 50/50 •If I put this in concrete terms meaning im going to replace letters with words making it a little bit less abstract in your mind •So All Plumbers are Mortal •All serial killers are mortal •Therefore, all serial killers are plumbers •So in this case if I polled you on it many of you would say that's not valid reasoning because plumbers can be mortal and serial killers can be mortal and UCLA students can be mortal - many different groups can be mortal but it doesn't have any barring on how the groups relate to each other •And if you think of it in a venn diagram mortals is this large class and the premises are that all P or M meaning there is no P outside of M - P is going to be intirely within M it could be the same size as M or could be a subset but it cant be this outside •So if you were going to think of this and make a venn diagram P is a subset of M and S is a subset of M •That is all im telling you in the premise - I didn't make any statement about the relationship of S and P so its possible that some P's or S's could overlap according to these premises but they don't have to overlap •And its certainly not the case that all S's have to be P's its perfectly reasonable that these two things could be true and no S's could be P •People have a much easier time when you put this into concrete terms •If you are trying to do a logical problem on a test it might help to go from abstract letter representation to actual words like this because it helps you to think about it and realize that its possible that some serial killers are plumbers but not necessarily the case that all serial killers are plumbers or that any serial killers are plumbers •So this is the phenomenon called the atmosphere effect: being influenced by the global impression or feel of the premises •In this case the idea that if all P's are M's and all S's are M's you are just kind of thinking well P's are M's and S's are M's so S and P must be related because they are both part of M and you are falling victim to this atmosphere effect - just being influenced by the global impression and not using proper logical reasoning •Heres another kind of example - so instead all we are just going to use some •All dogs are animals •Some animals are pets •And then the line means the premises are done •This is the conclusion: •Some dogs are pets •Is this a valid or invalid conclusion? •¾ of you are saying yes •and the logic here is sure you might think I have a pet dog I know plenty of people that have pet dogs the logic makes sense - yes all dogs are animals and yes some animals are pets and you know also that some dogs are pets •but does these follow from these premises •so here is another example just with sharks •all sharks are animals •some animals are pets •some sharks are pets •just replaced dogs with sharks •and if you give people this as an experiment people are less likely to judge this as valid and in fact if you draw a venn diagram or think about it logically - I mean for the 23% of you that got this right •although it seems to be a true one its not true - it doesn't logically follow that some dogs have to be pets - dogs are an example of the class of animals •some of them are dogs - so dogs are a subset of animals •and of the class of animals some are pets and again it doesn't mean dogs have to be pets •just for the same reason as the previous one •so this shows you another phenomenon called belief bias •this is a tendency to accept invalid conclusions like the ones you just saw here if they are believable •so some dogs are pets is believable - it makes sense as an conclusion even though its not something that logically follows from those premises •so people are using their beliefs - your beliefs about the world to reason - in this case to reason incorrectly and accept an invalid conclusion •and people will also reject valid conclusions when they are unbelievable - if there are some valid conclusion that logically follows from the premises and yet its not believable in the real world people will tend to reject it even though they should accept it •you can see here this is the percentage of conclusions accepted - so for valid conclusions perfect performance would be 100% and for believable premises valid conclusions were accepted basically by all participants •so if the conclusions believable and the reasoning is valid people got that right that was the easiest •when the conclusion is unbelievable difficult to believe and the reasoning is invalid people rejected it almost all the time •so people are doing well in these extreme cases but when the conclusion is believable and yet it is invalid people should be rejecting it •and yet 30% of the conclusions in this experiment were accepted so this is showing you - its not the case that they always accept it sometimes invalid conclusions they do reject but they show an increased tendency to rate something as valid when its believable •and here they should be rating these as 100% if they were using proper logic this should be up here at a 100% and the fact that its diminished means the unbelievability of the conclusion is leading people to rate it as invalid even when it should be valid

Utility theory

-According to Utility Theory, a rational person should try to calculate the expected utility (or "expected value" as it is referred to in your textbook) of each option and choose the option that maximizes this -Expected value = (probability of a particular outcome) x (value of the outcome) -e.g., If you are playing roulette and there is a 1/20 chance of winning $100, the expected value of the gamble is $5 -Do our preferences for particular outcomes map directly onto their expected value? •so this is the utility - what will it do for me if I buy this product - what will it do for me if I chose to go to this college •its really how important and how valuable it is •like what will it do for your happiness - and your well being and your future •and then how likely is each outcome - sometimes your making decisions and things are uncertain - trying to decide should I pick this insurance policy or this and the thing you are trying to evaluate is how likely it is you are going to get sick for example •so probabilities are important •so according to utilities theory a rational person should try to calculate... •expected value is the probability... •its very concrete to calculate this in gambling type situations so these are good to use as an example because you know what the expected value is other things like when the value is expressed in terms of your happiness or well being its harder to quantify it - you have to approximate it with some number •but here with roullete - lets say table had 20 options and there was a 1/20 chance of winning 20 dollars so the expected value of this gamble is $5 if the casino charged you $5 to spin a wheel that had 20 numbers the house would break even because 95% of the time youll lose and 5% of the time youll win but everytime you win you get a $100 •so to play this game is $5 and on average if you win enough you should win 1/20th of the time •so that would be break even •if they charged you $6 per spin they would be making money and if they charged $4 per spin you should just take all your money and keep playing that table because in the long run you are going to come out ahead •so in some cases its very clear •but do our preferences in most situations actually map on to their expected value •so heres some examples that illustrate this may not be the case •would you choose: •a certain $50 •a 50/50 chance to win $100 or nothing •most of you want the certain $50 •both options have the exact same expected value and yet in this case most subjects will chose the certain $50 that's very typical results •it illustrates that human beings tend to be risk averse when it comes to gains •so when you have the chance to do double or nothing here that for some people is scary you don't want to risk getting nothing when you could have $50 for certain •so even though these two gambles have the exact same expected value people behave differently •and heres another one •would you choose: •a certain loss of $50 •a 50/50 chance of a $100 loss or no loss •its almost mirror reverse of prior results •so both options have the same expected value •this one is -$50 - a probability of 1 times an outcome of -$50 •this is a probability of .5 times a probability of -100 so the expcted value is -$50 •so they are the same expected value and yet most subjects will chose the latter option and this shows that people are risk seeking when it comes to losses •interesting phenomenon that people in these two scenarios behave so differently •the option should be 50/50 according to expected value if that's all we went on - these are basically equivelant but that's not how people behave •so a loss of a given size looms larger psychologically - meaning its more psychologically salient •the opportunity to take a risk to avoid loss is worth it for them •in the gains domain when you have the chance to gain an extra $50 with the coin toss people were risk averse •whether coming out of class with 50 or 100 dollars it doesn't matter ot people that much vs. whether you lose 50 or 100 dollars people are willing to try to avoid losing the 50 dollars

Why do some subjects need hints?

-Analogies depend on similarities in structure ¤ People tend to focus on surface features, which can be superficial in solving the problem -Analogy use depends on expertise within the relevant domain ¤ Experts think of problems in terms of deep structure ¤ Understanding the structure of a problem is far more effective for solving future problems than simply memorizing the solution •Want to make the point of why people need hints: so analogies depend on similarities in structure •People often focus on surface features or superficial details which are not going to be that helpful - if you are thinking about soldiers being soldiers and rays being rays and don't realize there is some deeper connection there you are going to miss that point •In order to achieve a solution with an analogy you need to think of the problem in terms of deep structure - represented more abstractly and then understand how different schemes might relate to your problem •And this helps in all field - like as a scientist doing psych or neuroscientific experiments sometimes I might talk to a phycicist or chemist who is dealing with a totally different kind of problem but if I can make analogies between what they are trying to accomplish and what I am maybe ill have some new incite into my research •So by thinking about the deep structure we gain insight

Kohler (1927) Insight

-Bananas hanging from top of cage out of reach of chimps -Some empty boxes where present in the corner of the cage -After sitting quietly for a few minutes, they suddenly produced a solution... -Heres an example of a problem he did - put bananas in cage and hung them on top and they were out of reach of the chimps •And then he had some empty boxes sitting in corner of cage available but just there and the chimp would sit quietly and try to realize what do I need to do to get these bananas •At first they might reach to test if they are tall enough - and then you get discouraged and then at some moment this insight comes to you and you realize oh I can use the crates •So at first you are reaching and its hopeless then you put one crate up and get in top of that crate but you still aren't tall enough and then stack two crates and then you realize this chimp has had this moment of insight where they realize the crates would be their way to get to that solution •So here is examples of it - sometimes even stalking up two to 4 crates to get the bananas •Heres one with a pigeon trying to get something banana like or a seed - trying to do just like a chimp or child would do - am I tall enough - now it realizes well look im taller still not tall enough there - now it had its insight

Inductions and confirmation bias

-Confirmation bias -More responsive to evidence that confirms one's beliefs -Essentially we ignore disconfirming data -Can lead to perpetuation of unfounded stereotypes: - "All [insert group of people here] are bad drivers" -Superstitions: -"If I velcro, un-velcro, and re-velcro my batting gloves, I'm more likely to get a hit." -Or conspiracy theories: -"The 9/11 attacks were an inside job" -Notice examples that fit this pattern more readily (biased attention) -Will recall examples that fit the pattern more readily (biased memory) •so inductions are very linked to a phenomenon called confirmation bias - this is another common reasoning error where people seek out evidence to confirm their beliefs - we saw it a little with the card selection task that the people who wanted to flip over that 3 to see if it was a G that would be evidence that confirmed the premise •so if G then 3 so you really wanted to know whats on the back of that 3 because you thought if it's a G then it will confirm my idea •this figure of this crook here is showing that people who believe more crimes are committed durng a full moon will notice everytime it's a full moon they will notice or read the news or look on the television and see there is a crime - and everytime something in the world confirms their hypothesis they notice it and remember that and all the times it's a full moon and its not a crime - or its not a full moon aand there is a crime they aren't noticing that - so they are just noticing examples in the world that seek fit their theory •and this happens a lot in political discussions also that people find evidence that confirms their existing beliefs they watch the news channel that echos their own belief and just keep confirming and bolstering their beliefs and ignoring or discounting all the examples that disconfirm your belief •another phenomenon discussed in the book related to this is illusory covariation •this illusion that from your sampling of the world - like oh theres a full moon and heres crime and heres another full moon and theres crime and unless you look at how much crime there is when there isn't a full moon you are going to not have all of the data - you might think there is a correlation here when there is not because you aren't sampling all the cases •this phenomenon of confirmation bias to define it is you are more responsive to evidence that confirms one's beliefs •and try not to confuse it with belief bias - deductive reasoning - the tendency to accept invalid conclusions if they are believable •so confirmation is about your beliefs but it is related to inductive reasoning •and we ignore disconfirming data •this is the basis of any stereotypes that when people have these stereotypes of groups of people that are completely unfounded but are derived from people finding evidence - so if you think all _____ are bad drivers •you have this stereotype and then everytime you see something that confirms this stereotype you notice it and increase the strength of that stereotype •and everytime you see someone in that group who is a good driver you don't even notice it •this happens with superstitions - like lets say one time your at bat and you do something with your glove and you get a great hit then you just keep doing that because you feel like oh everytime I do this I hit well and the times I tried not doing it I didn't hit well so then you develop this superstition or you need to do this ritual before the game or before your test because the time you did it it worked and the time you didn't do it it didn't work •you don't really test it experimentally and say ok for the next 5 bats im going to do it and then the other 5 tests I wont do it and ill compare the data to see if this actually had an effect •conspiracy theories also - if you believe that the 9/11 attacks were in inside job you could watch 100's of youtube videos of other people confirming your beliefs and getting all this information that confirms this conspiracy and yet there is so much information out there and reports and investigations to the contrary that people who believe one thing wont necessarily look at or seek disconfirming evidence about what actually happened •so with confirmation bias there are two forms of the bias it can bias your attention - these happen simultaneously so it makes you more attentive to things in the world that fit the pattern that you believe to be true •and they make them more readily recalled from memory - so things if you are trying to think of examples like all USC students are like this and you have some bias against them you just think of all the examples that confirmed your belief about what USC students are like and youll notice those qualities and so it will perpetuate your stereotype •and people tend to not notice the disconfirming evidence

The Hill-climbing strategy

-Find some measure of the distance between your present state and the end state. -Take a step in the direction that most reduces that distance. -A potential local maximum problem. -The hill-climbing strategy is the heuristic to choose an option that moves you in the direction of the goal. -However, many problems require you to move away from the goal state. -E.g., walking two blocks east to a bus stop, in order to catch a westbound bus •one common strategy for solving problems is known as the hill climbing strategy and this is just generally you are starting down here metaphorically and you are tryinig to get to here (the solution) and you are just going to work whatever you can do that will get you closer to the solution •so you are just going to climb this hill to get to this solution •you take a step in a direction that reduces the distance •and this is just a metaphor for solving any kind of problem - you have some goal and you are going to take steps that are going to get you closer and closer to goal •one limitation of this is there is a potential local maximum problem •what this means is that if you are walking up this hill and trying to achieve this goal here you can get to here and now you are closer to the goal then when you started but in order to get even closer and reach the goal you actually have to come back down this hill nad now you are a little further away then you were here •you are closer to the goal on the top of the hill then you are at the bottom of the hill - you are moving away from the goal temporarily •if you're strategy or algorithm - this set of rules that you are going to do to achieve your problem - ill do whatever the next step is thatll get me closer •if this is your algorithm - hill climbing - you are going to get to the top of the hill and stop and not going to realize that maybe moving a little further away from your goal temporarily will ultimately lead you on a course that will help you achieve your goal •now here is a concrete example of it •the dog is trying to get to this bowl of dog food and the dog if its using the hill climbing strategy is going to be stuck at this local maximum •so its right here very close to its goal - physically close •and any step to the left or right will get it further from its goal and in order to get to its goal its going to need to get a little further away come to the gate and go around •so hill climbing strategy works for many problems but often it can be a suboptimal strategy •heres another example - trying to get to westbound bus and your trying to head west - lets say to the beach from here but maybe the closest bus stop is two blocks east so your goal is to get west but you might have to go two blocks away from your goal to achieve a step that is going to move you closer •so in this metaphorical hill climbing you are at the top of the mountain you need to go a little further from the goal by going to the bottom of the mountain and then youll have a path available to you that will bring you closer

Another example

-Five Orcs and five hobbits are on the east bank of the Muddy river. They need to cross to the west bank and have located a boat. In each crossing, at least one creature must be in the boat, but no more than three creatures will fit in the boat. -and, of course, if the orcs ever outnumbered the Hobbits on either side of the river, they will eat the Hobbits! therefore, in designing the crossing we must make certain that the hobbits are never outnumbered, either on the east bank of the river or on the west. -how can the creatures get across without any hobbits being eaten? -now lets try another problem •its useful because it will help us to understand the steps of problem solving a little bit more •here you have five orcs... •"at least one creature must be in the boat" these are the constraints •boat is limited in size and you cant send the boat empty it wouldn't have anyone to stear it •of course you cant leave hobbits and orcs together or else they will eat the hobbits - so that's another constraint •it requires 11 moves •this diagram shows you what we call the problem space •so the problem space is the total number of paths available to you •so the problem space here is actually very vast this is just a subset of it •you have 5 orcs and 5 hobbits on one side and 0 on the other and then you can start sending them across •so you could send three across - so then you would have 5 orcs and 2 hobbits •or 0 orcs and 3 hobbits •or you could send 2 across or you could send 1 of each •so in each one of these they would lead to multiple choices - and this would lead to multiple choices •and so forth so the initial state is the resources you have •the 5 orcs and 5 hobbits and the boats •the goal state is what you are trying to achieve here •the goal state is stated as getting all creatures across without having any hobbits eaten •the operators are these actions that you can take - so sending boats back and forth •and the path constraints are that you cant have the orcs ever out number the hobbits on the shores and there are certain limitations to how many creatures can go on the boat •so this problem I don't have the full solution

Functional Fixedness

-Functional fixedness: The inability to realize that something that has a certain use might also be used for performing other functions -Participants fixate on the box as having only one function: keep matches. -When the match-box was shown empty, with matches on the table, subjects were more likely to solve the problem -The failure to break free of these mental sets has a term you need to know - functional fixedness - the inability or the difficulty in realizing that something that has a certain use in most contexts might also be used for performing other functions •So participants in this study and many of you when thinking about this is you fixate on the boxes as having one function of keeping the matches - the box is just there to hold the matches but the matches are the things that will need to light the candle or doing something else with the matches but we aren't thinking of the function of the box •So when the matchbox was shown to participants empty it helps participants break free of the functional fixedness so if they show them the matches and there is a matchbox and the box is empty people were more likely to solve it •So about 80% of participants solved it getting the correct solution of putting the candle into or on the box when the box was shown as empty otherwise only 40-45% solved it •Here is another example of functional fixedness •In the 2 strings problem which is another famous problem you enter a room in which two strings are hanging.. •Your goal is to get the two strings together that's all you have to achieve •The correct solution is tie one of the strings to the pliers and use it as a pendulum and swing it and then walk over to the other and when it swings back you grab it •The problem is when you are in this state you cant reach them both at the same time but if you could be over here and swing this one then you can grab them •In the 1930's it was found that people had a hard time with this one in general because the insight of tying the pliers to the string didn't come naturally •The functional fixedness was thinking about the pliers as something for pinching or something you could use maybe to give you that extra few inches of extension if you are holding the rope with the pliers but not thinking of it as making the string function as a pendulum to use it as dead weight •But they facilitated insight when you took a group of subjects and had the experimenter accidently brush up against one of the strings and then the subject could see it swaying a little bit that helped people solve the problem even though it was unconscious they rarely reported noticing that happen but these unconscious cues led people to restructure the problem and break free of this functional fixedness and start to think about other things that might be useful here •Now here is an example from a movie you might have seen of overcoming functional fixedness - there is a canon here, he wanted to get across from one ship to another and most people would view a canon as a weapon to shoot a cannonball at another ship or target but overcoming the functional fixedness of thinking of a cannon as only having one affordance or only one thing it can do he thought of the kickback a cannon could give after it fired and being able to use that as a way to catapult himself across - so that's another example of breaking free of this set

Finding solutions to everyday problems

-How to get to class on time -How to stay awake for this entire lecture - ...how to make an informative and entertaining lecture? ̈ How to get a good grade in this course -How to complete all of the requirements for your major -How best to spend your winter break -How to get a great summer job -How to end a war/conflict -want to set the stage with a range of different kinds of problems that one might encounter -from basic problems like how to get to class on time - now most days this wont be a problem but some days there might be a traffic jam - there could be something wrong you could have a flat tire in your car or bike - you have an emergency that comes up and then you have to trouble shoot the issues surrounding the problem •how to stay awake for this entire lecture? •How to make an informative and entertaining lecture? •How to get a good grade in this course? •How to complete all of the requirements for your major? •How best to spend your winter break? Maybe you have plans already made maybe there are still decisions •How to get a great summer job? •Or really large problems like how to end a war/conflict? How to provide good health care to the country in an affordable way? •Things like this these are massive problems •And all of these problems have various constraints so certain things you can do - steps you could take and steps that might not be permissible •And some of them have multiple paths that would get you to the same outcome and some of them might only have a single path or a small number of paths and this is what we are going to explore from a cognitive psychological framework

More representativeness heuristic examples (misconceptions of chance)

-If you flip a coin 6 times, which of the following sequences would be a more likely outcome? A. H-T-H-T-T-H B. H-H-H-H-H-H C. Equally likely -People expect that a sequence of events generated by a random process will be representative of a longer random sequence -In flipping a coin, people think H-T-H-T-T-H is more likely than H-H-H-T-T-T (which seems nonrandom) or H-H-H-H-H-H (which seems like an unfair coin) The Gambler's Fallacy -After a long run of red on the roulette wheel, people think that black is now "due to happen" -Chance is viewed as a self-correcting process Casinos and roulette -Casinos often post the last 20 numbers that occurred. -People try to use this information to make their bets -They might expect that a number is "due" to come up -But there are no patterns in roulette •Coin tosses •This is another example of the representativeness heuristic •If you flip a coin 6 times, which of the following sequences would be a more likely outcome? •H-T-H-T-T-H •H-H-H-H-H-H •Equally likely •Equally likely is right because everytime you flip a coin its just a 50% probability of all of the potential six coin sequences - this is one potential sequence and this is one potential sequence and there are many others - you could write out every possible sequence and each of these is a unique sequence •People are judging this second one - for those that said this is more likely than this it seems like a coin flipped should behave like this - that it is representative of randomness to have some of option A and some of option B a coin that keeps landing on heads isn't very representative of randomness •This is something that happens a lot and you see it in las vegas gambling type games •And its known as the gambler's fallacy this manifestation •So its most commonly seen when people play roulette - ball spins around and some numbers are red and some are black - you can bet on all different kinds of things but a popular bet is to bet on red and black •So after a long run of red on the roulette wheel, people think that black is now due to happen •There is no reason to look at what it landed on in the past - if this is a fair table everytime you place the ball give the thing a spin wait for 30 seconds for it to settle and it should be completely random •So it really should matter at all what happened in the past - not like the ball has a tendency to do one thing or another if you think this is a fair table this should be irrelevant and yet people don't feel this way they think if its been red so many times either there is some reason to bet on red because it's the hot color or that black is just due to come up - it would be too unlikely to keep being on red so it has to be on black the next time •And there really are no patterns - if you just analyze this over enough samples it should be random every time •And the same thing happens in sports that you have a baseball player who has a 333 batting average you expect 1/3 of the time they are going to get a hit 2/3 of the time not and so if they have 10 in a row at bat with no hits you think ok now they are due because you view chance as this self correcting process but yet really with probability theory it just means each time they are at bat they still have a 1/3 chance of getting a hit it doesn't matter statistically speaking what happens in the past Gambler fallacy - one more example of representativeness heuristic

Inductive reasoning

-In many cases we cannot rely upon deductive inferences -Instead we rely upon induction -Induction does not guarantee a correct answer! -However, some inferences are more likely to be right than others •in many cases in the world to make conclusions we cant rely on deductive inferences and science is like this we do experiments we are trying to generate data and pieces of information to test some bigger theories - so we want to understand how something in the world works and we are going to make observations and after making a lot of observations we will make an inference or an induction •so induction doesn't guarantee a correct answer but we can accumulate a lot of information and have some degree of confidence and probability that we are right •and some inferences are more likely to be right than others •so again induction is going from specific instances to general conclusions and ill give you some examples

The problem with induction

-Induction is only guaranteed if we have experienced all possible instances -e.g., there are black swans in Australia •is this guaranteed to be true? Well no because we have only sampled a subset •unless you have sampled every known swan on the planet you cant know that you are certain you can just say out of all the data ive gotten this seems to be the case •so this is inductive inference - its probably correct if you've have 3265 observations that have confirmed it but it could be incorrect •there could be someone later that finds one that's not white there are black swans in austrailia •so maybe you've never encountered one in your country or at the zoos you've visited but there are potentially examples that are unsampled that would violate your hypothesis and this is the case with a lot of scientific theories youll think something based on your data •ex. the capacity of working memory is 7 items and you do a lot of experiments that confirm that and then years later someone will say its actually 4 items you did it wrong •theories end up changing when theres disconfirming evidence

Weighing the options...

-Is it always better to have more options to choose from, and does this make us feel better about our eventual choice? -You want to buy a new stereo, and have yet to decide which model to buy. You pass a store that is having a one-day clearance sale. They offer a popular Sony stereo for just $99, well below list price. Do you a) Buy the Sony stereo b) Wait for other options -You want to buy a new stereo, and have yet to decide which model to buy. You pass a store that is having a one-day clearance sale. They offer a popular Sony stereo for just $99, and a top-of-the-line Aiwa for $149; both well below list price. Do you: a) Buy the Sony stereo b) Buy the Aiwa c) Wait for other options -with more options, people opt to wait! •Going to talk about decision fatigue - or the paradox of choice so people have many options and with amazon shopping and online things there are so many options available •This guy wants to do the right thing for his health but there are just so many kinds of mixed greens and field greens and spinach etc •In this experiment subjects were asked you want to buy... •If you poll people in this experiment 2/3 would say yeah sure ill buy it its on sale •Some people would wait for other options •If you just tweak this ever so slightly "you... •Now 46% will wait for other options •So the fact that there are two choices here are causing more people to wait because they are both on sale and it makes you wonder well there isn't one compelling thing that's drawing me in here - there are many choices they are all on sale maybe ill wait and something even better will come along - this happens with many different decisions when there are many different choices people get overwhelmed and sometimes they don't want to make the decision at all

An empirical study of the representativeness heuristic

-Kahneman and Tversky (1983) presented participants with the following description: Linda is 31 years old; she's single, outspoken, and very bright. She majored in philosophy. As a student, she was deeply consumed with issues of discrimination and social justice, and participated in anti-nuclear demonstrations. - How likely are the following statements? -Linda works in a bookstore and takes yoga classes. -Linda is a bank teller. -Linda is a member of the League of Women Voters. -Linda is an insurance salesperson. -Linda is a bank teller who is active in the feminist movement. -Results: The conjunction fallacy -Participants tended to estimate "Linda is a bank teller who is active in the feminist movement" as more likely than "Linda is a bank teller." ♦Description represents one's stereotype of a feminist ♦ People ignore basic probability principles ♦ Two events in combination cannot be more likely than just one of the constituent events •heres an empirical study of the representativeness heuristic done by kahneman and tversky and this is a famous one that really introduced this heuristic to the world •in this experiment subjects read a little description of a person •this one is about linda •linda is... •so that's what they read •and then they do a series of ratings like this •how likely are the following statements? •Linda works in a bookstore and takes yoga classes •Linda is a bank teller •Linda is a member of the League of Women Voters •Linda is an insurance salesperson •Linda is a bank teller who is active in the feminist movement •Only two important questions in this experiment and one was linda is a bank teller •Many of you thought this was unlikely •The other important question was that she was a bank teller who is active in the feminist movement so few of you thought it was unlikely but many of you thought it was likely •Participants tended to estimate that linda is a bank teller who is active in the feminist movement as more likely than linda is a bank teller •And this is a logical reasoning error - and it shows you that people are succumbing to the representativeness heuristic - that description of lindas personality had elements that made you think that she was representative of someone that might be a feminist and that's true that's what the description was trying to convey •But people are ignoring basic probability principles - if every single bank teller was active in the feminist movement which is not the case but if it was at best they would be the same the probability of being a bank teller and the probability of being a feminist bank teller could be equal but if there are some bank tellers who aren't feminist its going to be a subset •So its impossible that being a bank teller active in the feminist movement is more likely of linda than just being a bank teller •And here are the ratings •If you did it on a 7 point scale with 7 being least likely and 1 being most likely you can see the peoples average rating overall •The important point here is that bank teller who is active in the feminist movement is ranked as more likely than bank teller and you replicated that results •This is called the conjunction fallacy - that A and B should always be more unlikely or at least no more likely than just A alone or B alone •Its impossible that being a bank teller and active in the feminist movement could ever be more likely than just being a bank teller because this has a second condition that needs to be satisfied •This is an example of the conjunction fallacy but also evidence for the representativeness heuristic that the reason people judge this bank teller and feminist to be more probable is because the description just seems to be more representative of a feminist and people aren't using logical reasoning in the appropriate way to realize that as they are making these responses their rating this conjunctive one as being more likely than the constiguent element here

Mental Sets

-Mental set: a frame of mind involving a certain representation of a problem, its context or a procedure for solving it ¤ May cause you to adopt an ineffective strategy and prevents problem-solving ¤ You may make wrong assumptions without realizing it ¤ You may find it hard to approach the problem in a new way -Some problems require a change in mental set; i.e., re-representing a problem •another phenomenon in problem solving is mental sets •this is a frame of mind... (you are creating this schema or mental set a framework for solving a problem) •sometimes this leads people to get the right solution sometimes having a mental set can allow you to not only solve one problem but solve a lot of problems like it but there are many situations where this can be ineffective where you can sort of have one set in mind have an idea of how you are going to solve it and then you cant break free of that and you have trouble thinking about problems in different ways you are making the wrong assumptions without realizing it •so ill illustrate this with a few problems that show how powerful mental sets are •so you find it hard to approach the problem in a new way •and some problems require re-framing the problem in a different way in order to break free of your mental set

Phoneme segmentation

-Normal speaking rate is 150 words per minute - ~15 phonemes per second - A major challenge for speech recognition is phoneme segmentation •With respect to phonemes one thing that is amazing is our ability to listen to someone speak •Normal speaking rate is 150 words per minute •This equates on average to 15 phonemes on average - so that's really a lot of information being transmitted quickly and what your brain needs to do is take those sound inputs and send the auditory signals to your auditory cortex and start to figure out what phonemes were said and in what order •But they are coming so fast so you need to segment them so you can figure out where one word starts and where another word ends and the speech signal is really this continuous barrage of sound •This is just a diagram - this on the Y axis is the frequency •So low frequency information to high frequency and then here is time so this is the word cant •So you can see it has a lot of low frequency energy •The colors here indicate the power - or the amount of information or energy present at each frequency •And it starts with the c (the initial sound) has this big sweep across all frequencies and then it becomes more low frequency for the "an" •And then at the end the "t" has some high frequency so that's what this diagram is •So heres the word cant again but its in the context of a sentence •What you see is that this utterance here when isolated kind of matches this but its going to be changed a little bit •You saying "you cant do that" would be very different for whats happening with this t sound than "you cant handle" •We sort of combine the cant handle •Its almost like we put it together •So the word you say next influences the current word •The point is that this "cant" even though it looks similar to the isolated cant is changed by the insueing word and all speech is sort of like this that when we are planning to say more than one word the word that we are going to say next already influences the way we articulate the current word -and thats a phenomenon called coarticulation: the pronunciation of a phoneme is changed by the following phoneme -This creates a real challenge - it allows us to speak in a way that flows well - so we are more efficient by not having to finish each word, take a pause, and then begin the articulation of the next word - the speech has this continuous stream and we do it in a way that is easier on our vocal apparatus •But when you think of it from a speech perception point of view it's a real challenge because we cant look for this template its going to be altered by what comes next and somehow the system needs to solve it - so that's one of the big challenges its one of the things that AI algorithms (like SIRI and Alexa) need to figure out because people don't speak each word one word at a time - it needs to know how the sounds will be changed in general in your language •This even happens within a word so its not just about words together but the word truth vs. tooth - the way the intermediate sounds here the "oo" sound will be modified by the r here where as following the t itll just sound different •And between phonemes and words: the can't handle example we just saw -Between phonemes in a word: truth vs. tooth -Between phonemes in different words: "can't handle"

Insight

-Occurs when the solution to a problem seems to come suddenly -First identified by Wolfgang Köhler in his study of chimps -Now we are going to talk a little about insight because that problem involved a moment of insight for those of you who solved it without having seen it before - wow this is daunting - you start realizing how am I going to get up to 90 and then you might take a pause and have that moment of insight like oh I get the lake is going to be full on this last day well when will it be half full well the day before because each day it doubles •So insight occurs suddenly and its an interesting phenomenon that's been studied for a long time •Wolfgang kohler studied this in chimps and brought this into the attention of cog psych in the 20's

More options = better decisions?

-Often leads to NOT making a decision -Also in applied areas: medical decisions -Drug A or surgery -Many doctors thought drug treatment might help -Drug A, Drug B, or surgery -With 2 drug options and difficulty justifying one drug over another, surgery option was chosen more often (avoiding the need to make a drug choice) •This applies also to medical decisions - often there is some drug available that can help or a surgical option •If the doctors are suggesting surgery or drug •But lets say there are three options, drug A, drug B, or surgery •And so this is the percentage of doctors referring patient to surgery -doctors saying surgery is the better option here •In both cases many get referred to surgery but when it was surgery vs. one medication the doctors were more conservative and said lets try the drug medication - there is one drug we will give it to you and try to avoid surgery •If there were two drugs and surgery - doctors were more likely to suggest surgery •I think it is because the same phenomenon - there were two choices you don't really know which one to chose they might both be pretty reasonable and surgery is a different kind of choice and seems like well now the doctor isn't forced to pick one of the drugs so you avoid that choice and its not a huge effect around 10% more likely but its still a consequential example of this phenomenon the paradox of choice its been called or sometimes decision fatigue - that after making decisions your brain feels really tired but when you have more options it often overwhelms people and you kind ofshut down becaue you don't want to decide and there are too many factors and not enough data •Another reason is that with more options people are more likely to succumb to regret so if you pick one thing but there are so many other options available you start to notice all the other people have - like I chose to get this phone it seemed cool but now I wish I had this feature or this feature - so many other things out there that you could have done •This happens even with dating - now with dating apps theres so many people available that you could meet and some people think it was easier when there were just a few choices and you could pick one - now its like not good enough, not good enough, etc because you always wonder that there is going to be someone better out there •This is a modern phenomenon where you have many choices available to us - if you live in a small village and there were a few foods to eat and only a few people to know people didn't have all these decisions and weren't burdened by the gravity of these decisions even small decisions like chosing lettuce can be taxing on peoples minds

Russian Blues

-Russian speakers divide what the English language regard as 'blue' into two separate colors, called 'goluboy' (light blue) and 'siniy' (dark blue). •These would be fundamental color terms that they use - instead of having one color term for blue they use goluboy for light blue and siniy for dark blue •In English when you put these up and say what color is it people would say blue, blue blue •We have the voccab to say dark blue light blue we know about navy - we have a richer color vocab but if you had to pick one term we would say blue and in Russia if you had to pick one term you had to commit to either goluboy or siniy •So in an experience like this its now taking out the memory delay part which might have been a confound in the previous experiment and now just saying which of the bottom squares are identical to the top? •So you have these two choices - is this one identical? Is this one? •This is actually an exaggerated example - the experiment was actually harder then this - and they expect subjects to be accurate but the question is how quick will they respond - the reaction time •So they are just deciding left button or right button •And there are two different types of trials - so within category trial both of these would have the same term - so these would both be goluboy •And so a Russian speaker there is no distinction between the verbal label •Here you have guloboy guloboy and then siniy - and this one will be presumably easier for the Russian speaker if the color label influences their perception •And so the stimuli were chosen from this spectrum here •So each trial was composed of a sample and then probes and the question was whether the two probes fell across the category boundary -Russian speakers performed the task more quickly when the two shades straddled their boundary between goluboy and siniy than when all shades fell into one camp. -English speakers showed no such distinction. •English speakers were equally fast with this trial type and this trial type -A verbal interference task (rehearsing an 8-digit number) abolished between-category advantage in Russians, highlighting the role of language in color perception. •But there was kind of a wrinkle here so they did another experiment where subjects did a verbal interference task - they tried to say well how much does this depend on the verbal labeling we do in our minds even if we aren't speaking it that when we see these we label them and give them a word and they said well how can we make people not do that? •Well one way is to tie up their phonological loop or verbal working memory system with a verbal task •Like rehearsing an 8 digit number - so if I give you an 8 digit number which is taxing for WM and ask you to rehearse it and make you responsible for reporting it at the end of the trial •Now you are going to be rehearsing the number and now you aren't going to have the verbal system available to you to kind of covertly label each of these as siniy or guloboy and when they did that it abolished the advantage of russias •Russians performed just like non-russian speakers once you remove their ability to silently name the colors •Showing this isn't a fundamental difference in the way the Russians perceive the colors and make these judgements - but rather when your language is available to you in the normal state - when you aren't rehearsing an 8 digit number they will show a facilitation because they are vocabulary for these two blues allows them to perceive the world in a way that is a little different form a nonrussian speaker •So when their language is available to them they have facilitation in this trial type - so language is influencing their performance in a significant way but when you remove that ability to label them it shows that the fundamentals of visual perception - of the way the retina is working - are the same its not changing their ability to distinguish blues per say its just allowing them to better at this task by virtue of having a richer vocab for blues

The Means-End Heuristic

-The best known problem-solving method -2 major parts: 1. Divide the problem into smaller subproblems 2. Solve each of the smaller subproblems -How should you get from UCLA to the Empire State Building? ¤Fly from LA to New York -Takes care of the biggest difference. ¤ That creates new sub-problems -Getting from UCLA to the airport -Getting from a New York airport to the Empire State Building ¤Each of these new sub-problems needs to be solved. •now a more common and probably the best known problem solving method that people apply to problems is called the means-end heuristic •it works better then hill climbing in that it doesn't have this local maximum problem •and means-end has two major parts so first you are going to divide your problem into smaller problems (subproblems) •and then you just solve each of the sub problems •so heres an example - you are trying to get from LA to NY specifically from UCLA to empire state building •so there are many ways you can do this •the biggest problem here is getting from LA to NY that is the biggest step in the problem •so you want to try to solve this first - maybe plane ticket or train •now you have subproblems so how am I going to get from the UCLA campus to the airport - flyaway shuttle, uber different ways to solve that •and then getting from NY airport to empire state building maybe you are going to take the subway •so there are all these little problems but first you are going to deal with the biggest difference •now you have solved the bulk of it then you'll deal with these other things •and the sub problems can be solved - and they might have sub problems of their own

Decision Making

-The goal of most decision making is to get the most/best stuff as often as possible -This involves two kinds of information -How important is each outcome? -Utility -How likely is each outcome? -Probability •decision making: •the goal of most decision making is to get the best and most stuff as often as possible •usually that's what we are trying to maximize to get good outcomes - get lots of money - get good grades - get happiness and stuff •there are two kinds of information that are important - there are various outcomes we are trying to decide between and we need to know how important or valuable each outcome is

Representativeness heuristic

-The representativeness heuristic may lead us to believe that smoking must be okay for your health based on one example (anecdotal evidence or "man who" stories). -We often assume that what is true of one instance of the category must be true of the category as a whole. -Flyer from the American Smokers Party, using this one woman to mock anti-smoking efforts and illustrate that smoking won't necessarily kill you -But in reality, smoking causes 1 out of every 5 deaths! And smokers die on average 15 years sooner than nonsmokers! -Despite ample evidence, some people don't believe in global warming... -But you're succumbing to the representativeness heuristic if you mock the idea of global warming just because of a few days of unusually cold winter weather... •this slide is meant to illustrate another manifestation of the representativeness heuristic in everyday life •this is the fact that many people anchor on specific examples like I know a man who has smoked cigarettes everyday of his life like maybe two packs a day and now he is 95 years old and is still doing just fine he is the happiest and healthiest guy I know and these kinds of stories of someone who did something that might statistically be bad and yet sort of did fine lead some people to believe oh smoking isn't hazardous to your health because of this one example •this is a common reasoning error where you anchor on one example being representative of a whole class of people - you focus on a man or women who did something in particular doesn't have to be related to a health behavior and therefore this is a characteristic that you ascribe to the whole group - it could be drinking sodas everyday and therefore its not going to be unhealthy •heres an example of this this is a flier from the smokers party and they chose this picture of a women lighting her 100 bday cake with her cigarette as an illustration to try to make this point like if smoking is so dangerous why is this 100 year old woman doing just fine •heres another example: tweet "my grandfather smoked from the age of 13 and didn't die until he was 90. STFU smoking doesn't kill" •this is a clear example of falling victim to the representativeness heuristic you have a sample size of 1 or sometimes 2 or 3 people and therefore you assume that those cases should be representative of the whole •just in reality smoking is extremely dangerous and if you look at data from WHO and other epidemiological studies its causing millions of deaths each year and shaving off an average of 15 years of peoples lives and causing 20% of all preventable deaths in the world •same thing goes for global warming you see this same reasoning error - so some people persistently refuse to believe in the body of scientific evidence and all of the data that the planet is warming and that's because the data is complex to understand and it's a thing that happens over long periods of time •but you would be succumbing to the representativeness heuristic if you mock the idea of global warming just because there is a day in the winter or maybe even in the fall or spring where it is unusally cold so these kinds of cartoons illustrate people falling victim to the representativeness heuristic •the creator of this cartoon where this guy in the car is saying you want me to call and tell them you'll be late for your lecture on global warming? Its mocking the idea that just because they are going through unusually snowing weather that oh yeah the planet is warming well how does that explain the snow storm that we have here in march •heres another illustration - conservative cartoons that are playing into the representativeness heuristic by saying just because its particularly cold right now therefore the planet cant be warming when really this is a phenomenon you cant measure by any one instance you need to look at data overtime and measured at all different ways like satelittes, ground temparatures, ocean temperatures so this is a common reasoning error •and no one exhibited this more than this senator from Oklahoma - he brought a snowball to the senate floor saying the snow falling in DC is evidence against global warming and again this is taking this example of oh wow we have a snowy day it doesn't usually snow here this time of year global warming is a hoax •these are just some practical or real world examples of how people fail to see the bigger picture and anchor on these individual instances

Example of induction

-We would like to form universal generalizations about the world -All X's are Y -e.g., All swans are white -These are based on our previous experience - Swan #1 was white -... -Swan #3265 was white -Therefore, all swans are white •so in abstract forms all X's are Y's •here is a concrete example all swans are white •we want to make some generalization - we want to conclude that swans are white •so we sample the world - we see 1 swan that is white we see another that is white •we see 3265 swans and they have all been white •and now we have sampled 3,265 swans and everytime its come up to confirm this idea that swans are white •so we make a conclusion that all swans are white

Critical Period for Language Development

-There seems to be a critical period during which language develops readily and after which language acquisition is more difficult and less successful. -The extraordinary cases of Victor, the Wild Child of Aveyron (France, 1800), and Genie (United States, 1970) seem to support the critical period hypothesis -Other evidence for critical period comes from: -Studying the effects of damage to language areas in the brain (children recover more readily than adults). -Studying the ages at which a second language is acquired •But there is really this critical period and if language isn't learned by a certain age it becomes very difficult if not impossible •So there are these rare cases of human beings - one case in 1800 this guy named victor who is sometimes referred to as the wild child of Aveyron in france •And later a more famous case in the US of Genie - not going to go into detail •But in these cases and about 30 other that have been documented of children who were essentially raised in the wild - and they are just not exposed to any language input •So they go through the early stage of their life without having any language exposure and then later they are discovered and people try to teach them language and it just becomes in each of these cases nearly impossible with all of the instructions and the attempts to teach them what words go with different things and how to put words together without having been exposed to language at this young age - being sort of exposed to it after the critical period had expired - the system of language learning is no longer receptive to take the inputs of language and map it on to this mental structure so you can learn the syntax and begin to use language in a generative way •So theres other evidence for the critical period - some comes from patients who have damage to language related areas - when you see these kinds of brain lesions that might be caused by a stroke or in a child more often by a head injury or a neurosurgical procedure to remove a tumor that also affects an area of cortex devoted to language - in all of these cases the children that have parts of their brains removed do much better than adults - in some cases you can even remove huge portions of the brain and the child will be able to adapt to that and overcome and learn language •Whereas if you remove language related cortex in an adult its very difficult to regain language you need a lot of speech therapy and even with all of that sometimes it never comes back •Another source of evidence is studying the age in which a second language is acquired - so for people who are bilingual you can look at how well they will do with a new language they learn later in life as a function of when they learn it - so if you learn two languages natively (these data are people's scores on tests of English for participants who English was either their native language and learned it at the same time as they learned their other primary language or English was acquired between the ages of 3-7, 8-10, etc •So the later you acquire language the poorer your score even if you spend quite a lot of time studying and trying to master it and obtain fluency your often still going to have lower ultimate attainment •So that's other evidence for the critical period 3-7 years old is still good if you want to expose your children to a new language but if you wait longer - you wait till high school or even middle school and start teaching them a new language they might never be as good as if they were exposed early

The Tumor Problem

-You are a doctor faced with a patient with a malignant tumor. It is impossible to operate, but unless the tumor is destroyed the patient will die. There is a kind of ray that can be used to destroy the tumor. If the rays reach the tumor all at once at a sufficiently high intensity, the tumor will be destroyed. Unfortunately, at this intensity the healthy tissue that the rays pass through on the way to the tumor will also be destroyed. At lower intensities, the rays are harmless to healthy tissue, but they will not affect the tumor, either. What type of procedure might be used to destroy the tumor with the rays and at the same time avoid destroying the healthy tissue? -Typical solving rate: 10% -Some subjects were given other stories to read; one included the story of a general attacking a fortress •This is another classic one from the 40's: tumor problem •You are a doctor... •Only 10% of people typically solve this problem so it's a hard one to solve •Then they did another condition where some subjects... •And this story involved the general wanted to attack this really strong fortress or castle and to do it he positioned his armies all around and they all attacked at once taking the fortress by all sides so instead of having all the troops enter at once and break through the main door and get fired at by surrounding it and having weaker batilions come in at once they succeeded •So subjects who read this story about a completely different thing generals attacking fortress then just had the insight you just had •If you think this the fortress is the tumor inside the body and you use a lot of rays that are weaker intensity that are all triangulated or fixated on the same spot that each one in and of itself wont damage the healthy tissue that surrounds the tumor but when they all reach that point together the integrated signal will be enough when it accumulates at that central point to destroy the tumor •So that is the solution - with no hint only 10% of the participants get it •If you provide an analgous problem like the fortress but don't suggest that helps well it does more than triple their accuracy so now over 30% are getting it right so just by thinking about the fortress you have primed them to think about multiple weak forces that are converging to create a strong intense attack and if you just give people the suggestion that oh that story you read maybe that could help you that helps almost everyone around 80% solve it

Typicality effects

-criticism of classical theory -some members of a category are more "typical" than others -they are verified more quickly (sentence verification task) -but these differences are related to non-necessary features (e.g. ability to fly), which are not included in the classical theory •This brings us to this other important concept of typicality •In the classical theory I said no individual member of a category is considered the best or most typical •A member of a category is either in that category or out - so you either meet the rules or you don't but that's not the way it seems to work in real life •When you ask people questions like this - so is the following typical example of a vehicle you'll see there are different ratings •If you do these kinds of ratings youll see that people are able to look at individual instances of a category and assess of all the different objects they know to be members of that category of ex. vehicles some are more typical than others •And this is just a property of most categories that sometimes challenges these classical views •Another test to assess typicality is sentence verification - looking at how long it takes you to respond •In each of the ones that are true the question is how long it takes you •And some of these will take much shorter than others - a robin is a bird? Do you think that would be faster than an ostrich is a bird? Yes •The sentence verification test is another indirect way to assess typicality - you can ask people to rate typicality - this is another way of probing it seeing how long it takes to verify these statements •If all members of a category were equally good it shouldn't take you longer to say a penguin is a bird rather than a bluejay is a bird but the fact that it takes you longer means its less typical and when you are assessing your mental representation of birds it takes more mental steps to verify that •So those are called sentence verification tasks •And these differences are related to non-necessary features so the fact that the penguin takes longer - you are being slowed down by the fact that many things about the bird are non bird like - it doesn't fly it doesn't look like most of the things you label a bird and doesn't have the same qualities •This refutes the classical theory •In this experiment done in the 80's they had subjects rate pictures like this for typicality - these were the ratings you got •High typicality for robin (On 7 point scale) •You get this range •Interestingly here subjects should all say 1 its not typical of bird at all because its not a bird but the fact that it wasn't averaged as a 1 either means some of the subjects didn't realize this wasn't a bird or thought the bat had a mild degree of bird-ness •So that is interesting •And the same thing applies to reaction time in sentence verification tasks •So this is a figure from a paper that shows why typicality ratings end up the way they do and it has to do with features that are not necessary for category membership but are typical for category membership •So something like a robin gets a high typicality rating because it meets all of these characteristics we associate as being characteristics of typical birds - so the robin satisfies all of these •The cardinal satisfies all of these •The eagle satisfies most of these - doesn't sing or eat words and insects so less typical •The penguin only satisfies two of these •A bat has three of the 6 features so the reason people view a bat as being bird like and it takes a little while for them to reject it in reaction time tasks and they don't always give it a 1 rating on the typicality scale is it satisfies several bird like features •This explains other kinds of errors - like the fact that people would consider whales or dolphins as being typical of fish •So another type of task is a production task: in this type of task you might be asked something like how many examples of fruit can you name in the next 30 seconds •If you were to do this task most people would go through typical fruit first, apples bananas oranges etc •And lower on the list will be things that are atypical so kiwis, tomatos, olives, persimmons etc •The idea that a category could have certain members that are better examples - like apple, banana is a better example of fruit than others indicates that category membership likely isn't this absolute you are in or you are out - it is graded - so each instance - each type of fruit has a certain degree of fruitness and a tomato even thought technically a fruit is less fruity than an apple or an orange •Just has to do with how typical or how well strongly you associate that with your representation of this category •So this is an example where it shows for all these different fruits peoples typicality rating on a 7 point scale •You can see the most typical and the least typical - and the same for birds •These were the results of the experiment in the 80's •So this actually extends to other types of categories like numbers •For even and odd numbers the rules are simple - classical categorization should handle this just find •Certain numbers or even certain numbers are odd - there is not ambiguity in math as to which numbers are even or odd •And yet people will rate certain numbers as being more typical examples of odd numbers than others •People will rate 3 as a better example of an odd number than 7 or 71 •4 is more even than 18 •and this is a figure from your book of different stimuli and these are typicality ratings on a 7 point scale •so 4 is going to have a higher typicality rating for even numbers •3 has a high typicality rating for odd numbers •so the larger the number the less typical generally in these studies •it also affects generalization so if you give subjects these logical statements - penguins can catch disease X (give them name of disease penguins can catch) then you say how likely is it that all birds can catch disease X and they make a rating of that (so they are using one piece of information and then making this inference and you measure their percentage here on questions like this) •and another class of trials - robins can catch disease X how likely is it that all birds can catch disease X •and these are similar trials - people will rate this statement as more likely (robin) •because robins are such a typical exemplar of birds they are more willing to make this generalization to the whole category •whereas if penguins can catch the disease you might think other more typical birds might not get it • Some members (or even non-member!) of a category are more "typical" than others • (Bats more like birds than penguin) TYPICALITY EFFECT (issues with classical theory) -some items in a category is more "typical" than others • They are.. - Rated more typical when asked participants out right. - Shorter RT when asked if it goes in the category - More likely to be named in production task - More likely to agree with generalisation statements. - Considered more attractive • Family resemblance accounts for typicality effects, but classical theory does not.

Why the difference in behavior? (the wason task and evolutionary psychology)

-do we have a "cheater detection" module? -Leda Cosmides and John Tooby have argued that people are good at reasoning in particular situations that have been present in evolution -Evolution has selected for the ability to reason in social situations, such as detecting cheaters -It has not selected for the ability to solve abstract logic problems •it suggests that when you frame this in more concrete terms - sort of a real world scenario •it helps people apply the reasoning and particularly its been argued that people do well in this kind of reasoning because we might have some cheater detection module that through evolution and the way that social pressures have shaped our society and interactions with others that it is very useful to detect other individuals in your community who violate our social contracts •so cheating whether on a partner or test or breaking any kind of law or rule is something that we want to be good at noticeing and reasoning through and so Cosmides and Tooby have argued that the reason people do so well in the Beer and age cards but not the letter and number cards is because they are good at reasoning in these situations that involve social contracts - and detecting cheaters in this case •and that we are not good at solving logical problems its just not what our brain evolved to do •that we aren't designed to think about if P's and Q's or if G than 3 scenarios but we are good at applying logic in other kinds of contexts •this is controversial and the book gives example of other data that came out that suggests the social contracts and cheater detection idea needs to be expanded to accommodate more cases - but I just wanted to put that out there

Probabilistic theories of categorization

-failure of classical theory led to proposal that category representation may be probabilistic rather than deterministic (rule-based) Two approaches: -prototype theory -exemplar theory -both are based on the idea of similarity •moving on to the other types of theories - the classical theory cant handle typicality very well •prototype and exemplar theory can both accommodate it •they can accommodate it because they are based on the idea of similarity - every member of the category has a certain degree of similartiy with other members of the category but they differ in the way they handle this

Another example: the lily pads problem

-water lilies are growing on Blue Lake. The water lilies grow rapidly, so that the amount of water surface covered by lilies doubles every 24 hours -on the 1st day of summer, there was just one water lily. On the 90th day of the summer, the lake was entirely covered. On what day was the lake half covered? -so here is another well-defined problem -about half of you guessed 89 but the rest of you are distributed almost evenly across the other options •So there are a few approaches •Some of you may have taken a forward solution where you start with one then you double it •And then you take that result and you double it and you continue on and on •Now this would take a long time to go through - to basically try to work forward •So on the first day theres 1 and you are trying to go through and think how long would it take to cover half the lake and you are going to go back and figure it out •So to do it this way it would look like this •You would realize if you did this and looked at all these numbers that the one where you have exactly half is 89 so 89 was the correct answer •But the backward solution is much better - you start on day 90 you don't even need to know how many lilies there are •If lilies double each day going forwards that means they halve each day going backwards •So if you think of the problem in a different framework here the day at which they will achieve half their final magnitude is the second to last day •On the last day they will have some number doesn't matter what and because it doubles each day that means the day before that it will be half that •So sometimes it helps in many problems including mazes for ex to work backward from the end •This is the means-end analysis - essentially applied in reverse •So you are trying to set it up - here we are at the end sort of how am I going to get backward and satisfy the goal

Two types of reasoning

Deductive reasoning -The process of reasoning from one or more general statements (premises) to reach a logically certain conclusion -Conclusions follow directly from premises using rules of logic -Guaranteed to be correct (If you follow the rules!!) Inductive reasoning -Reasoning that constructs or evaluates general propositions that are derived from specific examples -Probable guesses made on the basis of prior evidence -Not guaranteed to be correct •Going to talk about reasoning and decision making •Two types of reasoning - deductive and inductive •So deductive reasoning is the process of reasoning from these general statements about the world - we call these premises to reach a logically certain conclusion so if you apply the rules of deductive reasoning the conclusions follow directly from the premises using the rules of logic •And people aren't always good at using the rules of logic but if you do follow the rules of logic you are guaranteed to be correct •So in deductive reasoning you have general principles about the world you are using them to reach a conclusion about a specific instance •Inductive reasoning is the flip side •So you are going to derive some general statement about whats likely to be the case broadly in the world based on specific examples and this is always going to be a probabilistic assessment •That you have sampled the world - sampled various specific instances and now you are going to derive a general conclusion •These are not guaranteed to be correct •So this figure illustrates the difference between deductive and inductive reasoning - so deductive reasoning you go from these general principles and it allows you to make statements about specific instances - inductive reasoning you sample the world you gather information from specific instances and then you derive general principles

Problem Solving

Four Components: ¤Initial state: Resources that you have at the outset ¤Goal state: The desired end product ¤Operators: A set of operations or actions that can be taken to reach the goal state ¤Path constraints: Rules that cannot be violated •So problem solving has 4 components as psychologists break it down •So every problem as an initial state - these are the resources you have at the outset •So just the initial setup of the problem gives you some resources available if you are trying to get to class on time maybe some of the resources are your bike or your car •The goal state is the desired end product - what you are trying to achieve what you are trying to solve •Operators is a term for what kinds of operations or actions one could take to reach the goal state •And then there are path constraints so rules that cannot be violated - there might be certain things that are just intrinsic to the setup that you have to adhere to •So you might want to get to class on time and there might be traffic so you might just decide im just going to drive the whole way on the shoulder but that violates a path constraint that that is not allowed - you cant drive on the shoulder or sidewalk •you cant run the red lights or you could make those decisions and assume some degree of risk

Example: Tower of Hanoi

GOAL: Getting all the disks from the first to last peg. RULES: Must move disc from top of pile -Peg moving to must be empty or have bigger disc on it -Many sub-goals ¤ e.g., getting largest disk to the far peg first -To solve, must be willing to temporarily move away from the goal •so heres an example of a psychological task called the tower of Hanoi •so the goal here is to get all the disks from the first peg to the last peg •and the rules are that you must move a disc from the top of the pile •and the peg you are moving it to must be empty or have bigger disc on it and here is an example of one with three pegs •so here is how it starts •you have a small disc, medium disc, and big disc and your goal is to get them like this •so this is your starting state and this is your end goal and these are your constraints: can move one disc at a time, cant move a disc from the bottom of a pile only from the top, and you can only move a disc to an empty peg and if you move A to here you cant put B on top of A the peg must have a bigger disc on it so you have to keep it always from big to little on each peg •so with the means end heuristic you can break this into sub goals - one subgoal that everyone should have is that I need to get the C disc here - that's a big part of what im trying to achieve •im going to get this one out from under here and over here •getting the largest disc to the farthest peg and then once I do that I can work on other goals •and to solve this you must be willing to avoid the hill climbing heuristic which has a pitfall here •you have to temporarily move yourself away from the goal •and here is just an example of the solution - you move the small disc here, then the medium disc in the middle, then you move the small disc back to the middle •now that that has happened you can move the big disc to this empty peg so you have achieved your first goal of getting the big disc to the empty peg •so these are even more subgoals here - freeing up the big disc is one subgoal, in order to move the large disc you have to free it up another subgoal is freeing up the third peg - this one needs to be empty, and the third one is moving the disc over and then after this you can move this one back and then this one here and then this one there •so here is the problem space for the tower of Hanoi problem - you see the initial state and each one of these has different branches and this isn't even illustrating the full problem space - each one of these intermediate ones would have more branches as well so you can see this is one potential path and one potential path and you are trying to pick which one of these paths leads you to the solution so it ultimately is this set of actions

Solving Problems by analogy

Gick and Holyoak (1980) -If you provide an analgous problem like the fortress but don't suggest that helps well it does more than triple their accuracy so now over 30% are getting it right so just by thinking about the fortress you have primed them to think about multiple weak forces that are converging to create a strong intense attack and if you just give people the suggestion that oh that story you read maybe that could help you that helps almost everyone around 80% solve it •UCLA's own Holyoak in the 80's •Trying to solve a problem by analogy you are trying to map different things on •The surface features of the problem are different - one is a tumor and one is a fortress •In one case you have armies and in one case you have gamma rays •In one case it's the number of troops in the other case it's the intensity •So those superficial details don't matter because the underlying solution here is sort of the same - you want to get at the deep structure of the problem and analogies help you get at the deep structure •A bunch of different rays converging to create a powerful signal •So this is the convergence schema here (look at lecture)

Using logic to make deductions (modus tollens, denial of the antecedent)

Modus tollens -Premise: -If it is raining, then Alicia gets wet (if R then W) -Alicia does not get wet (not W) -Conclusion -It is not raining (not R) -Arguments of this form are valid Denial of the antecedent -Premise: -If it is raining, then Alicia gets wet -It is not raining -Conclusion: -Therefore, Alicia does not get wet -Arguments of this form are invalid •So another type of reasoning is called modus Tollens •Premise: If it is raining, then Alicia gets wet (if R then W) •Alicia does not get wet (not W) •Conclusion: •It is not raining (not R) •Arguments of this form are valid •So if it is raining then Alicia gets wet - there is never going to be a case in the world where it would be raining and she doesn't get wet because I told you if its raining shes going to get wet •You have to accept the premise at face value - so if its raining Alicia will get wet - Alicia does not get wet so if it was raining she would have gotten wet •The fact that she didn't get wet means that its not raining - it has to mean that •People don't do very well on this one •Motus Tollens is harder for people •And the last one is called the denial of the antecedent •If it is raining, then Alicia gets wet (if R then W) •It is not raining (not R) •Conclusion: •Therefore, Alicia does not get wet (not W) •This is invalid •All im saying is if its raining Alicia would get wet •It doesn't mean that if its not raining she couldn't also get wet - she could jump in a pool she could go in the shower etc •This conclusion that she does not get wet does not follow from these premises •People have a hard time with this •The fact that some of you struggled and got this wrong is normal and that's what im trying to illustrate - that people are not very good at many forms of logic •This is a summary slide form your textbook reading - showing that most people accept these arguments - so affirming the consequent and denying the antecedent are both invalid •And people accept them as valid more than they should •It will hlp you when doing this to think of them concretely so even though I put these in a concrete case to help you its still hard •So making it concrete can only help so much •Its really challenging •And these are data from a study done by marcus and rips in the 70's where they had people make inferences or assesments of different logical problems •Modus ponens - the if P, then Q - P therefore Q •People do really well - 100% of people got that right •Modus Tollens which is also valid fewer people - 60% got that right and the rest got them wrong and thought it was valid that's the if P, then Q - not Q, therefore not P •And then affirmation of the consequent and denial of the antecedent the majority of their subjects rated them as being invalid but you could see the fact that this isn't 0 where it should be if people were perfect means that many people are endorsing these arguments even though they aren't valid logic

What is the "format" of human thought

Plato -"The soul when thinking appears to me to be just talking. . ." Kant -"Thinking is speaking to ourselves" Watson (influential behaviorist) -Thought should be equated with the production of (subvocal) speech -But this can't be the whole story...? -People can have thoughts that are difficult to express. -If people thought entirely in words, words expressing new concepts could never be coined because there would be no way of imagining their meanings. -Infants and nonhuman primates are capable of relatively sophisticated forms of thinking, even in the absence of language. -There have been a few adults who grew up without language, but surely these people had thoughts -Thus, there must be aspects of thought that are independent of language -But can one's language nonetheless influence the way one perceives, understands, and thinks about the world? -"The limits of my language are the limits of my world." -- Ludwig Wittgenstein (1922) •we are going to explore today what the format of human thought is •and we can go back to the ancient greeks - they had something to say about this •so plato remarked "the soul.. •he was making the observation that when he was thinking through things - trying to ponder how the universe worked and all the things that plato and his fellow greek intellectuals were trying to understand about the world is that it was in his mind - just like talking it was happening with language - that language was organized in his thoughts •Kant: many centuries later echoed this in his remarks that "thinking.. •That when you are thinking you are thinking in language •And Watson who was an influential behaviorist actually took it further and thought that "thought.. •Subvocal speech meaning speech that you are not necessarily saying out loud but inner speech when you are thinking - when you are sitting there waiting for the bus on the bench that those are happening in language that your just not vocalizing them •When you wake up in the morning and are planning your day - that's language •These aren't just thoughts that are just floating around without words attached to them but its speaking to yourself intimately intertwined •But it seems to many people that this cant be the whole story - there must be something else •And these are just intuitions about why that view of thoughts being equivalent to subvocal language might be inadiquate •So im sure all of you at one point or many points have thoughts that are difficult to express - where you have some feeling or experience and when you try to put it into words it just seems wholly inadequate like you didn't capture the way that you experienced that •This is certainly the case when speaking about emotion - when you are having certain feelings - like love, loss, regret and its hard to put them into words •But even just other kinds of profound experiences like religious experiences something seems to get lost when we try to nail it down with sentences •This is a cartoon that kind of illustrates that •If you sometimes know what you want to say but you cant find the words than is there some part of you that deep down knows what you want to say and if so how did you have that thought without language •So this is one kind of intuition pump about this •Also if people thought entirely with words its been argued that words expressing new concepts or ideas could never be coined or could be difficult to coin them because there is no way of imagining their meaning - so how do you come up with something that is completely or totally new - new sort of ideas without having the ability to think beyond the constraints of our language •Also infants as well as nonhuman primates (chimps, gorillas) are capable of relatively sophisticated forms of thinking, even in the absence of language •So we can assess them and know that even though their communication seems to us to lack a lot of the language like qualities of human communication which goes for the infants and apes - it seems like there is quite a lot that they can understand so the baby that is born figuring out how the world works they are not narrating in their head with language what is going on and yet they are gaining understand about all kinds of things about the world - and this is certainly the case for many animals as well - sort of what is it like to think without having a structured language •And as I mentioned a few minutes ago in the case of those wild children that are discovered in the woods and grew up without language - in these horrible cases the children have trouble learning language related to the critical period but another thing more philosophically is sort of what were they experiencing during that whole time and they had no language and were interacting with the world - surely they had thoughts and experiences and memories and it just wasn't being organized with the aid of words to scaffold those thoughts •And so one could argue that there has to be some aspects of thought that are independent of language •Where we can have semantics without mapping them onto specific words •And the question that is can ones language nonetheless influence the way that we perceive and understand and think about the world •And Wittgenstein - says "the limits... •That once we run out words it limits what we are able to understand •And one could argue that modern humans as weve come up with the scientific vocabulary for describing how things work, for describing things about physics, things about biology and so forth that having that vocab then allows us to understand the world in a different way •So its kind of related

Another test of deductive reasoning (the wagon 4-card selection task)

The Wason 4-card selection task -Each card has a number on one side and a letter on the other -Proposed rule: -"If there is an G on one side of the card, then there is a 3 on the other side" -Which cards should you turn to test this rule? -Another proposed rule: -"If someone drinks beer, then he/she must be 21 or over" -each card has age on one side and drink on the other -Which would you turn over to test this rule? •We will do another kind of test - this is called the Wason 4-card selection task •This is also a logical task that people struggle with •Each card has a number on one side and a letter on the other •And heres a rule •The proposed rule again is a premise - im laying this out and I want you to evaluate this •If there is a G on one side of the card, then there is a 3 on the other side •You are going to try to test this rule •So in this case you are actually trying to test this whole statement - you aren't evaluating a conclusion as you were in the previous slides •You are saying is this rule a valid rule? - is this a fair thing to say about this deck of cards •You want to test this rule which cards should you turn to test this rule? •Obviously you would want to turn the G - if there is a G on one side then there should be a 3 on the other - you want to know wants on the back of this card •But there is one other card here you should want to turn over to test this rule •So the majority of you wanted to know whats on the back of the 3 •Picking 3 is the one most people pick which makes it an interesting demo so we replicated the results of hundreds of experiments that have been done with this task and people want to know whats on the back of the 3 it seems appealing because you know if there is a G on one side there should be a 3 on the other so know you are curious if there was a 3 maybe there would be a G but there could be many cards - I didn't say A's couldn't have 3's on the other side and B cant and C cant etc •It could be the case that almost every card has a 3 on the back and 3 is just a common one to have on the back •The fact that this has a 3 gives you no information - you don't learn anything by flipping this over if there was a G or any other letter it would have no barring on this rule •If you flip this over and it was G youd think oh yeah that confirmed my hypothesis but it doesn't actually give you any added information •There is only one card that does and the right answer here was the 7 •The only card that could have value for you in this mini experiment of trying to test this rule and theory is the 7 because the 7 has the potential to disconfirm this if you flipped over the 7 and there was a G on the back that would give you information that was actually useful •That would tell you something - you would know this rule is no good and this is a phenomenon that you see in all forms of human reasoning but in sciences people fall victim to this as well •That you are trying to seek evidence that confirms your theory rather than trying to seek out instances that disconfirm it •If you have a theory and you really want to test it you should try to disconfirm it and everytime you fail to disconfirm it - that builds your evidence that its true •But just seeking out cases that confirm it here like flipping this one over - it could make you think well that's interesting this one happened to have a G on the back also but it actually doesn't give you information that tests the rule •And here youll see it in this example •It's the same thing but instead of letters and numbers its drinks and ages •If someone drinks beer then she or he must be 21 or over and each card has age on one side and drink on the other •Which would you turn over to test this rule? •Beer obviously because if someone drinks beer you want to know their age is at least 21 •But there is one other card you would want to know •Its C you want to know if the 19 year old is drinking beer or not that's the thing that would give you information •For the same reason that flipping the 7 would be •the person drinking diet coke you don't care what age they are that has nothing to do with this premise - so the rule is that if someone drinks beer they have to be 21 so doesn't matter what age the diet coke drinker is •the 23 it doesn't really matter what they are drinking people at that age can drink whatever they want they can drink water - nothing, diet coke •if the 23 year old was drinking beer that would be fine but doenst help you test this rule •knowing what the 19 year old was drinking is useful - and if I started with this problem instead of the previous one even though not a 100% of you would get it right many more would get this one right than the previous one and the fact that people are better at this situation than this situation tells us something

Language and thought

The Sapir-Whorf hypothesis -Strong interpretation (linguistic determinism) -Thoughts and behavior are determined by language -The language you speak determines the concepts and categories that you use, and as a result, shapes what you can think about. -No solid evidence that certain languages forbid a speaker from thinking about certain concepts! -Milder interpretation (linguistic relativity) -Thoughts and behavior are influenced by language -Language influences what we pay attention to, and this shapes experience, which influences how we think •So whorf proposed along with his mentor and college sapir whats become known as the sapir-whorf hypothesis •And this hypothesis has two different flavors •So the strong interpretation which whorf advanced is sometimes called linguistic determinism •And this is the idea that our thoughts and by extension our behavior our actions are going to be determined by our language •So language really constrains us - the language we speak determines the concepts and categories that we use, so we think we are just going about and exploring and creating some kind of structure and category of the world but really its going to be very constrained by our language and someone with different languages - with a different language and different concepts might understand the world differently •And so language - not just influences what you think about but shapes what you can think about •By not having certain words there might be certain concepts that are just hard for you to ponder or comprehend •And with certain words there might be things that you see that other people don't experience in the world •So that's the strong interpretation •Decades later it has become clear to most people who have studied this that there actually isn't good evidence that certain languages forbid a speaker from thinking about certain concepts •So we aren't even going to go down the road of trying to make the case for the strong interpretation •So a milder interpretation which is sometimes called linguistic relativity and I will say that in the book both of these are referred to as linguistic relativity •Others make this distinction and use this other term of linguistic determinism for whorf's idea •And save the term linguistic relativity for a milder interpretation which has more evidence behind it •That our thoughts and behavior are going to be influenced by the language we speak •So language influences us in a number of ways - it's the words that we have and the concepts that those words convey influence the things we pay attention to which in turn shapes our experience and the things we notice and care about and ultimately how we think but it's a matter of influencing not constraining us or forbidding us from having certain thoughts or understandings

Types of Problems

Well-defined problems ¤All four aspects of the problem are clearly specified -Water jug problem -Hobbits and Orcs problem -Figuring out how to get from LA to New York n Solving a maze Ill-defined problems ¤ One or more of the aspects of the problem are not well specified -"Achieving world peace" -Goal is no more war, but what else is required to realize this goal? How will disputes be settled? Resources allocated? -"Having a great time on vacation"; "Writing a perfect grad school admissions essay" •so there are two different types of problems generally •there are well-defined problems and ill-defined problems •in a well-defined problem all aspects of problem are going to be clearly specified •so even though the water jug problem is not a typical problem youll face in day to day life nor hobbits and orbs problem - they are well defined •it means that the set up of the problem •that what I tell you about what you have available to you and what you need to achieve is clearly specified •figuring out how to get from LA to NY is another well defined problem - doesn't mean there is just one solution there is many solutions - you could hitchhike book plane tickets, take buses, drive with friend - many ways to do this •solving a maze is another well defined problem - there is a start and finish •usually there is only one path that gets you there sometimes there might be multiple options that work •but your goal is clear you need to get here to here - you aren't allowed to go through the lines you have to keep pencil within the lines and so forth •and then there are ill-defined problems that are problems one could try to solve but one or more aspects of the problem are not going to be well-specified •and this makes it a little harder to apply the same problem solving structure •so achieving world peace is an example of an ill defined problem •so you have this goal that is kind of abstract or maybe the goal is concrete - you want no more war and youll know you achieved it when there is no war •but what can you do to get there •how will the disputes be settled once you have achieved it how will resources be allocated •there are many different factors that are at play here and its hard to nail down what it is you are trying to solve with a problem so vast and complex that it is ill defined •other things like having a great time on vacation are also ill defined - you know you want to have fun that that is your goal but there are many different paths you could take to get there and there is uncertainty - you don't know that booking this particular thing will lead to fun and things could go wrong and so forth •this is kind of a continuum problems aren't completely ill-defined or completely well-defined this problem could be somewhere in between - like some parts of it you know there will be things you can plan that will generally lead to fun but the problem doesn't have all the elements of a well defined problem •writing a perfect grad school admissions essay - you can try to do a good job •you can try to follow advice of an admissions coach or websites that tell you how to write a good essay but you don't know who will be reading it and what they will be looking for and there are so many aspects of it that are not really specified

Example 2: The Timeless Hopi

Whorf's claim: "Hopi may be called a timeless language. [...] It does not distinguish [with grammatical structures] between present, past, and future of the event itself." -the problem with this example... -Although Hopi may not have grammatical markers for time, they do express time in other ways: -Translated from Hopi: "Then indeed, the following day, quite early in the morning at the hour when people pray to the sun, around that time then he woke up the girl again." •The next example that whorf pushed was that about the hopi language •He was interested in this language because it seemed like their grammatical structure didn't contain the affixes that we use - the suffixes that illustrate the passage of time •So whorf quote, "hopi may be called a timeless language. [...] It does not distinguish [with grammatical structures] between present, past, and future of the event itself." •So the way that the words are used they are not modified as we do in English and many other languages to indicate whether we are talking of a verb that is indicating a past occurrence or a future occurrence or something that is happening right now •But this just to cut to the chase also seemed to be untrue •So although Hopi may not have grammatical markers for time, they do express time in other ways: •When you look at their language they don't modify words to indicate past tense or future tense but they do express time in other ways •So whorf was drawn to this because he thought if the hopi language doesn't contain these grammatical structures to illustrate the passage of time then the people wouldn't have the same understanding of time the way that speakers of other languages would - that everything is timeless and the connection of people in the present with their ancestors and sort of feeling their connection with the future generations •That everything from the past and present is woven together and whorf thought well this is amazing - by not having the grammar for time they just don't see time in the same way its not this linear continuum going from things that are before to things that are after but really kind of this interwoven fabric where everthing exists in this timelessness •But the hopi do have words for time - if you really try to translate something that is spoken by hopi speaker and look at the English words you would need to do that translation you see all of these indications that there is this appreciation of time and the passage of time •So then indeed, the following day, quite early in the morning at the hour when people pray to the sun, around that time then he woke up the girl again." •Now when you translate it to English you have to use terms like woke vs wake vs. waking •You have to commit to something and that's hard in the translation - but you can see here the following day quite early in the morning at the hour - so there are all of these vocab terms that seem to be present in Hopi that allow the description of time •So I think whorf turned out to be wrong about that one also as a strong example of linguistic determinism

Broca's vs. Wernicke's aphasia patients: a modern comparison

•And this is a more modern view - so these cases show you what might happen in one individual's brain but if you look at 36 patients with Broca's aphasia and 11 patients with Wernicke's aphasia and you draw on each brain where the lesion - so what region is damaged in patient 1 and 2 and then you just look at the overlap •So the red shows you here 100% of the patients who had this diagnosis when you mapped out their lesion with an MRI had damage •You can see in general the area of Broca's aphasia is this left frontal region and actually the area that is most in common with these 36 patients is this deeper region a little more medial in the frontal lobe - the lateral part is also affected but less consistently •In Wernicke's you can see it is here sort of at the post - these are showing the three slices and you can see this here would be at the posterior temporal lobe pretty consistent •And this is just a table with a lot of stuff showing you a side by side comparison •With Broca's aphasia speech is halting - sometimes called telegraphic it comes out in spurts single word or two words and it stops •Wernicke's aphasia the speech is fluent - it flows even though it might be nonsensical •In broca's aphasia there is a tendency to keep repeating words or phrases (this is a form of perseveration - so perseveration is a general term you saw it in the Wisconsin card sorting task and that's the tendency to keep sorting on a dimension that is no longer relevant •In broca's aphasia a patient will get stuck and then just keep using the same word even though they know its wrong - they cant break free and come up with the right word •Generally in Wernicke's aphasia there is little spontaneous repetition •Broca's - syntax is disordered •Broca's area is really important for taking all the words and ideas you want to say and organizing them in a syntactically correct way and without that all the patient can do is produce these single word utterances or short phrases -Wernicke's aphasia syntax adequate •Wernicke's aphasia has normal or mostly normal grammar •broca's aphasia disordered structure of individual words and comprehension intact •wernicke's sometimes inappropriate or contrived made up words (jargons), comprehension not intact

Background knowledge aids sentence parsing

•How do we resolve this? With many of the ambiguities there are other cues in our environment that help us which is known as extra linguistic context (our background knowledge) •In this case the detectives examined by the reporter revealed the truth about the robbery vs. the evidence examined by reporter revealed the truth about the robbery •Knowing some information about the scene like seeing who is talking to who can help you resolve some ambiguity by this phrase - the detectives are being examined by the reporter that doesn't really make sense here -A. the detectives examined by the reporter revealed the truth about the robbery -B. The evidence examined by the reporter revealed the truth about the robbery -readers are momentarily confused when they reach the "by the reporter" phrase in Sentence A. -that is because they had interpreted "examined" as the sentence's main verb -reader's aren't confused by Sentence B, because their background knowledge told them that "Examined" couldn't be the main verb (because evidence is not capable of examining anything) -The extralinguistic context refers to factors outside of language itself. -The sentence "Put the apple on the towel into the box" is a garden-path sentence, unless the sentence is uttered with the appropriate visual context •Extralinguistic context refers to factors outside the language itself •Things in your environment •If you wanted to look at a table like this and convey something to another person you could say put the apply on the towel into the box - that could be a garden path sentence because you are thinking put the apple on the towel - so they may think you want them to put the apple on the towel but really you are saying put the apple on the towel on the box •By looking at this its telling you there are two apples - one of them is the one the speaker is referring to - if there was just one apple you wouldn't need to say that - but because there are two you are resolving the ambiguity by giving a little more information and its allowing you to resolve the garden path sentence

Example

•In these algebraic problems - so problem 1 and problem 2 about force and angle and friction coefficient and so forth •So novices will say that these two are similar - they are more similar in terms of their superficial features they both look the same they are both incline planes they both have objects and angles •This problem looks different •But experts will say these two problems are both more similar to each other - they are realizing the deep structure - so conservation of energy is the work energy theorem •So these two problems are actually more similar in terms of their deep structure where as these two are more similar in their superficial structure so the point here is that when you are solving problems by analogy you are really benefited by thinking about the deep connections and not just trying to map the superficial features

Attribute Substitution

•This is just a figure from your textbook I put up to show you different types of attribute substitution •So the top is about the availability heuristics - if you are trying to judge the frequency of occurrence in the world like how many words are there that end in -_n_ vs. -ing your really trying to judge something that is unknowable that you don't have direct access to unless you work on dictionaries and just have all that information in your head about the English lexicon you are going to have to base this on something else and the thing people base it on is the availability heuristic - so how easy it is to generate cases to mind so you are substituting this availability attribute for the real thing you are trying to measure •For the probabality of event being in a category like linda being in the feminist movement people are using the representativeness heuristic the resemblance of that event and other events in the categories is what they are using to make the probability judgement but this doesn't always work Availability Heuristic -you want to judge...frequency of occurrence in the world -instead you rely on...availibility in memory: how easily can you think of cases -this usually works because... events that are frequent in the world are likely to be more available in memory -but this strategy can lead to error because...many factors other than frequency in the world can influence availability from memory Representativeness Heuristic -you want to judge...probability of an event being in a category or having certain properties -instead you rely on...resemblance between that even and other events that are in the category -this usually works because...many categories are homogenous enough so that the category members do resemble one another -but this strategy can lead to error because..many categories are not homogenous

Framing effects

•This last bit im going to talk about framing - depending on how you frame something or set it up people will see it one way or another •With this glass you could see it as half full or half empty there are just two different perspectives on the same thing •These are real data from organ donations - so the percentage of drivers donating organs •These are different European countries •Drivers meaning people go to the DMV they check whether they want to donate or not •And some countries have high donation rates like near 100% and other European countries have lower donation rates •And countries that are similar in their culture and background or geography like Austria and germany have very disperate ratings here •It cant be about peoples cultures per say •The form asks you to either opt in or opt out •so these countries it said you will automaticllay be enrolled in an organ donor registry please check this box if you would like your name to be removed from that •so these were opt out countries where the default was that you are in and these were opt in countries saying please check this box if you would like to be added to our donor registry •and just the way you frame the form - people tend not to check the box but if you don't check the box here you are in - the default is that you are in •here the default is that you are out you have to opt in •and it just dramatically changed peoples decisions the way the form was set up

Phrase structure organization aids the reader

•phrase structure is very useful and the organization of a sentence will aid the reader •in this way of parsing - just by the way the lines are •the •large tomato made •a satisfying •splat when it •hit the •floor •that's not the way we would normally parse this - its much easier to convey if its written like this •the large tomato •made •a satisfying splat •when •it hit •the floor •so parsing it like this can help you read the sentence and interpret it - if its delivered like the first one its more confusing

Finding solutions to everyday problems (Die Hard example)

•so going to start with a typical everyday problem made famous in movie die hard: this is similar to one of the problems in your textbook chapter •need to fill jug to exactly 4 gallons •the things you have in your initial state are two empty jugs one is 3 gallons and one is 5 gallons and you have water available to you - so that's sort of understanding the problem •you cant fill the 3 gallon jug with more than 3 gallons of water that is one of your constraints and the 5 gallon jug you could fill it 80% full and that would achieve your goal except you aren't going to get it exact and you need to get it exact •so that's not going to work either •so the first step he is going to take is fill the 3 gallon jug full so now he has 3 gallons of liquid in that jug •so pour the 3 gallons into the 5 gallon •that is the next step •so now the 5 gallon jug has 3 gallons in it so what good does that do? Its good because you only need one gallon of water to make it 4 •so the 5 gallon one is 1 gallon short •so now the 5 gallon jug has 3 gallons of water in it •so it had to do with filling the 3 gallon jug again and now you top off the 5 gallon jug so not how much is left in the 3 gallon jug? 1 gallon •so here is a depiction of the steps that were actually taken to solve this •so you have the 3 gallen jug and the 5 gallen jug and initially you are going to fill the 3 gallon jug full so it has 3 gallons in it •then you are going to pour it into the 5 gallon jug so now this one has 3 and this one is empty •now you are going to fill the 3 gallon one again so now they both have 3 gallons and you have a total of 6 gallons •but you pour this one into here that is when you get this new affordance •you have one gallon of water you have preciously measured so this is one of the moments of insight pouring the 3 gallon jug into the 5 gallon and having 1 left because 1 is something youll need •so now you have 5 here - so now youll dump this one out completely (5 gallons) •and now you have 1 and 0 •you transfer the 1 into here and then you can fill this one (3 gallon) again with 3 and then add the 3 to the 1 (5 gallon) •and that's what you need to get the 4 •so these are the steps to solve that problem

The basic level of categorization

-just as certain category members seem to be privileged, so are certain types of category -for ex. what is this object? -Eleanor Rosch argued that there is a basic level of categorization -neither too general nor too specific -tends to be used in speaking and reasoning about categories -here, "chair" is the basic-level category as opposed to "furniture" (more general, or superordinate) or "wooden desk chair" (more specific, or subordinate) -it is easier to explain what features are common to members of basic-level categories than for other levels -what do all tools have in common? -what do all hammers have in common? •One important point that was recognized as cognitive psychologists began to study categories and how people carve up the world and use category terms for things is there seems to be this basic level of categorization where some members of categories are priveliged in a way •What I mean is if I ask you to name this object what would you say? Chair - almost everyone would call this chair and that's one particular way of categorizing it - it's a level of specificity that conveys a certain amount of information about it •But you aren't categorizing it as a wooden chair or a particular furniture style of wooden chair •You aren't categorizing it in the broader way of furniture •And Eleanor rosch highlighted that there is this basic level - its neither too general or too specific and it's the way that we tend to speak about categories in a natural way •So here chair is the basic level category - if this is the object you are looking at and you wanted to know what is the way people would commonly refer to it its just whatever most intuitively comes to mind will usually be what rosch and other cognitive psychs would consider to be the basic level category •So furniture is going to be a supper ordinate (so one or more levels above) and wooden desk chair would be more specific (that's subordinate) •So basic level here is chair and then above it is furniture or even above furniture could be man made things and so forth and you aren't going to refer to it like that - that wouldn't convey very much information •If I put this up and you said furniture there is a lot of other types of furniture it could be •And if you said wooden desk chair that's more specificity than you typically need that would only be necessary if you were at a store in the chair section and you were looking for a particular kind of chair and you need to convey that level of specificity •So the basic level is where you usually end up at for just naming an object in the world •So whats this? Airplane (that's the basic level) •So you wouldn't call it an aircraft because that would open up too many possibilities - it could be a blimp, helicopter or any number of things so airplane is a much better description for what this is and you wouldn't call it unless you weretalking to some pilot friend or there was some reason to give this level of specificity a Cirrus SR22 G3 - that's the subordinate level - that's more information •So heres just a summary of this - if you called this an elephant that would be its basic level •If you called it a mammal that would be its superordinate level if you called this monkey that's basic level if you called this macaque monkey that's subordinate level (specific type of monkey) •And this curious George is just one level below subordinate level - it's the individual level the specific exemplar of monkey here is curious George just like snoopy is an individual specific beagle •Heres just one more illustration: these are all basic levels - I showed picture of chair you would call it chair •If I showed you picture of table you would call it table you wouldn't call it furniture and you wouldn't call it kitchen table or dining room table unless I asked you could you be more specific? •If I showed you bed on screen you would all call it a bed not a twin bed or furniture •One more tip: its easier to explain what features members of a basic category have in common than for other levels •So what do all tools have in common? Well theres not really a lot to say - like all tools can be used to construct or deconstruct things - all tools might be something you could hold potentially but its harder to say what they all have in common - some have blades, some you turn, some you twist •But what do all hammers have in common? All hammers have a lot of features in common - hammers are a basic level category •Hammers are something you hold - they usually have at least one end that has a hard mallet on it that is made of metal •The other end typically can be used to pry things open or pull out a nail and so forth •Same with apples all apples have the same sort of shape - at the top a little stem - at the bottom the base - if you cut them open they have seeds •All fruits have less in common - fruits all have certain qualities that make them fruits but this is another way to think about it -Neither too general nor too specific • Tends to be used in speaking and reasoning about categories (Eleanor Rosch) • It is easier to explain what features are common to members of basic-level categories than for other levels - Tools vs hammers

A tale of two systems

System #1 -Intuitive -Automatic -Immediate -Most heuristics are produced by this system System #2 Analytical -Controlled -Consciously monitored -Rule-governed -Serial -More cognitively demanding -Flexible -We generally make little or no use of this system -System #1 rapidly generates intuitive answers, which can be monitored/evaluated by System #2, though this latter stage may be rare •So two individuals among many who have studied this really moved the field forward in a significant way so I just want to highlight them here •Tversky and Kahneman really developed this framework for heuristics and produced a lot of experimental evidence and changed the field of economics (made economists think about human behavior) •One of the things they are particularly well known for is this idea of two systems - it's a simplistic notion to think of mind as having these two systems •System 1 and system 2 and there are obviously more complexity - this is just a framework that is a useful starting point •Just as a general framework they outline these two systems and its pretty useful •So system 1 is this fast operating intuitive system for making decisions - gut decisions that we can do almost automatically so you don't need a lot of thought or deliberation you can come up with judgements using the system almost immediately and most of the heuristics that we are going to discuss today are produced by system 1 •System 1 is what allows us to apply these mental shortcuts and make very quick judgements so we aren't paralyzed by every decision that we have to make in the world •But system 2 is this other system that is available to us and even though we don't use it as much it allows us to analyze all the information available we can use cognitive control to inhibit things that might not be relevant that might be details that seem like they could be important but they are red-herring so they really aren't relevant to the current situation •This is consciously monitored - meaning we are aware of the thinking that's going on if you are making a decision about where to go to college you are likely more in system 2 •Its rule governed - its cereal meaning one step at a time you sort of have a lot of different pieces of information you have to shift your attention from thought to thought as you are working through a particular decision - it is more cognitively demanding but that affords greater flexibility •For better or worse we make very little use of this system - for most of the judgements we make when they are inconsequential we can do them mostly on autopilot in system 1 •Heres just a cartoon illustration with purchasing behavior you can see system 1 is going with your gut its more emotional - like you walk past a shoe store and you have to have those •System 2 might say no this isn't something I budgeted for im going to wait these aren't practical etc •Also with food choices - system 1 might be about your hunger is driving you towards something and the aeroma of fast food vs. system 2 is focused on a healthier choice that will ultimately fill your stomach and ease your hunger but maybe not be as appealing (another ex of how it plays out in everyday life) •And kahneman wrote a book about this several years back called thinking, fast and slow - the fast is the system 1 and the slow is system 2

Processed meats rank alongside smoking as cancer causes example

•this is a recent study that came out •the WHO classified processed meats as being dangerous as a carcinegon and this lead to a lot of cartoons being published •this figure shows you the approximate amounts of foods that will raise your health risks by 18% so if you have 6 thin slices of ham per day or 1 ¼ hot dogs a day this would increase your risk of having colorectal cancer by 18% •that's what the experiment showed that lead the WHO to issue this guideline •so that's true from the study they did •but the important thing here is the base rate of colorectal cancer is very low - only 5% of people will get this in their lifetime so the chance of getting colorectal cancer regardless of what you eat is low •and so the 18% increase looks like this if you break it down •so 5% of people would get it overall and eating these meats would increase it to 5.9% •you are really going from a 5% chance to an almost 6% chance even though its an 18% increase - you are 18% more likely to get it then you otherwise would be •so if you became vegetarian you could reduce your risk by 18% but really instead of being 5.9% likely you would go to 5% likely •the base rate really matters •things that are infrequent if something is a 1% risk and doubles to a 2% risk that's a 15% increase in risk and people tend to ignore the base rate in thinking about these •compare it to something like smoking •smoking increase your risk dramatically - your base rate for getting lung cancer is extremely low as a non smoker or just as a general person .2/.4% and if you smoke its still less than half the chance that youll get it but its going from .2% to 24.4% •that's a dramatic increase it increases your risk by about 2500% •this study in wired clarifies it and shows that the base rate matters but people make this mistake a lot

Cognitive Control/Executive Function

-processes that flexibly shape and constrain our thoughts, memories, and actions to allow us to accomplish our behavioral goals Putative cognitive control functions include: -inhibition of prepotent responses -selection of target representations/responses -retrieval of information from long-term memory -shifting of attention/task-switching -coordination and monitoring of working memory •These are processes (a set of mental operations that you have available to you that allow your cognition to operate in a flexible fashion - it allows you to allocate your attention in a way that is consist with your goals (what you are trying to accomplish) and it helps shape and constrain your thoughts (what it is that you are currently thinking about - are you paying attention and trying to learn whats on the slide, are you realizing your mind is wandering and trying to bring your attention back - these are all cognitive control type processes as well as to shape your memory (is this something important that I want to remember am I trying to on a test retrieve a particular detail from the past and im going to use cognitive control processes to guide that search and filter out any irrelevant details that come back to mind to try to get back to that one memory im searching for •Cognitive control is very important for your actions (motor behaviors - is something im doing right now the appropriate thing to do in this context or do I want to inhibit this action - do I want to not swing at a bad pitch - or not swinging at a tennis ball heading right for my racket but about to go out •You are using cognitive control in many contexts in sports to accomplish our goals •So these are just some putative cognitive control functions: there are many things that would be under this list •One is inhibition of prepotent responses: prepotent responses are just a response that are very automatic - when the phone rings your prepotent response would be to pick it up but its only appropriate in certain contexts like if its your phone - but if its not your phone or you are at someone elses house you don't want to pick it up •selection of target representations and responses: this illustration here is a reference to the Robert frost poem "the road not taken" in many situations we have multiple potential actions available to us and how do we select which one is going to be the most valuable for us in this situation or the one we want to pursue out of many possible options •retrieval of information from long term memory - when we are trying to remember something (this could be semantic memory - remembering a fact that we are trying to get on a crossword puzzle or retrieval of information from episodic memory trying to recall details of a particular experience) oh yeah that trip was so fun what was that thing we all did together and then you are sort of searching your memory and trying to guide that search using executive control to come up with strategies to cue yourself so you can find the detail you are looking for •shifting attention and task switching - we talked about this a little bit with using cellphone when driving - executive processes are very intimately tied with the control of attention and when you are multitasking you are constantly balancing how your attention is allocated to different elements of your environment and different trains of thought - you are trying to monitor the road - engage in the motor processes needed to control the vehicle but you might also be in this case on the phone and keeping up a conversation and paying attention to that - thinking about what you are going to say next and in this case maybe glancing in the mirror to get your lipstick right so a lot of quick shifts of attention and trying to regulate that and keeping everything sufficiently active so you don't fail at any one task •and then coordination and monitoring of working memory: spelling bee you get the word and you have to think about how to spell it you cant write it down before you start speaking you have to model it all out in your head - and keep track of sometimes they will start spelling it and then stop and you have to remember what you have already said you cant go back to the beginning - this happens all the time you need to keep track of what you have already done and what still needs to be done Definition: - Processes that flexibly shape and constrain our thoughts, memories, and actions to allow us to accomplish our behavioral goals. • Include... - Inhibition (of prepotent responses) - Selection (of target representation/responses) - Retrieval (of info from LTM) - Shifting (of attention, switching tasks) - Coordinating, monitoring (of working memory)

Many possible categories

Natural vs. artifact -Natural: groupings that occur naturally (birds, trees) -Artifact: designed/invented (computers, sports cars) Stable vs. ad-hoc -Stable: natural/artifact are typically stable -i.e. people generally agree on what goes into them, and what are the criteria of inclusion -Ad-hoc: unstable categories defined for a special purpose or within a specific context (e.g. "things to write on") •There are natural categories - things that exist in the natural world - like birds and trees that have certain properties - they are natural things that have well defined properties in the world •Vs. artifact categories which are categories of things that are created by humans that are designed or invented - ex. sports cars, computers •And then there is stable vs ad-hoc categories •So stable categories are categories that are going to be generally consistent over time there isn't a lot of disagreement about what a member of the category is so people would generally agree (this is a tree - this is not a tree) •And there are criteria that people can explain for what would constitute a member of that category •Where as ad-hoc categories are categories you can make up on the fly - things to grab with you if your house was on fire and you don't necessarily have like a list or you haven't thought about that much but in the moment you could create that category and then say well I have a few minutes and this is how much I can carry - these are the things that are important - I want my photo album, computer, pet •Ex. things to write on - you could write on paper, your hand, sidewalk etc. there are many things and you don't have in your mind a stable category of things to write on and different circumstances you can come up with different solutions (ex. you could use a stick to write on sand) •These are all things you can develop as a category as needed -Many possible types of Categories - Categories of categories: stable vs ad-hoc - Stable: generally consistent over time • Natural vs Artifact - Ad-Hoc: unstable categories defined for a special purpose or within a specific context

Specific way we can classify different phonemes

Voicing - Whether vocal folds vibrate ([z], [d], [b], [v]) -Or not ([s], [t], [p], [f]) •its easy to determine whether a phoneme is voiced or un-voiced depending on whether your vocal cords vibrate or not •another way that we distinguish between phoneme is manner of production Manner of production -Whether air is fully stopped ([b], [p], [d], [t]) -Or merely restricted ([z], [s], [v], [f]) Place of articulation -Where in the mouth the air is restricted: - Closing of lips ([b], [p]) - Top teeth against bottom lip ([v], [f]) - Tongue behind upper teeth ([d], [t], [z], [s])

Recognition Memory Task

-Subjects were given a specific Maunsell color chip to remember. -After a 30 second delay, they were given two target chips (the old one and a new one) and had to recognize the original chip. -Results: Both English and Berinmo speakers showed better performance when the two test colors were associated with different color words in their respective languages -Berinmo speakers do best on wor-nol trials -English speakers do best on blue-green trials •So heres an example of one kind of memory task that can be done to test this •The subjects see a color - so they study a color sample like this - it comes on the screen for a few seconds and this is going to be a visual working memory test and then it goes away •In this case you had to wait 30 seconds so kind of a long delay period to hold onto a color •You saw it you wait, wait, wait and eventually even longer than this you would see two chips and you would have to decide which was the one you studied •So was it this one? Or this one? •So in berinmo both of these chips would be called nol •So this would be called nol and this would be nol •So to get this right they cant use their labels they have to remember that the original one was more this kind of a nol than this kind of a nol •Now here is another trial - they study this one then it goes away they wait, wait, wait 30 seconds later you get this •Was it this one? Was it this one? •Now in berinmo this one would get the label wor for things that are yellowish and most would call this nol •So in this case if they encoded this as nol and then waited here this one would match the color name and this one wouldn't •So these are the two different trial types so colors that share the same labels as the probes or colors that share different labels from the probes •Someone doing the task can try to remember the hue as a visual memory - just trying to keep this color in mind and visualize it but we also can label it with the words •So if you are trying to remember this and you encode it as nol or we might encode it as green and then these two come on and now this one is green and this one is blue - and so in English oh I encoded green and this probe is green and if it matches in your language that is going to be easy its going to be easier when this one is one we would call green and this one is blue •For berinmo speaker its going to be hard here because this is a version of nol and now they get two different choices of nol and which one - was it this kind of nol or this kind of nol •That's going to be hard for the berinmo speaker •The opposite is true here - the encode one as nol and then in the probe here they get a nol and a wor so these are two different labels so the one that matches clearly will be the nol - its easy for them in this trial because they knew they encoded some variance of a nol and here in the probe they get a wor - oh I know it wasn't a wor •Whereas in English these are just different shades of green •So the correct answer berinmo speakers will do better when tested with wor-nol than nol-nol •English speakers will do better in the opposite so in this case these don't strattle our color boundary these are both green in this example •English speakers would do better in this trial type because this is sort of green and this is teal or blue -And here is an illustration of that here - so for berinmo speakers if you give a chip that is over here and a chip that is over here for the probe they are both within the category that is going to be hard - for an English speaker a chip here and a chip here would be across category so that would be easy •The opposite is true over here so a chip chosen here and a chip chosen here are both variances of green so its going to be hard - you have to remember the specific green to get it right •Here all you have to remember is does it match my label - so itll be facilitated by their language •So that's one example of how your language influences your performance on this kind of test - but it does seem to have to do a lot with the verbal labeling - you are waiting 30 seconds you might not keep that color vividly in mental imagery for the full 30 seconds -so you might just kind of wait and the probes come on and if one of them is the same label as the one you studied then you are going to get it right

How are categories stored in memory?

Classical theory: -rules Probabilistic theories: -prototype theory (prototypes) -exemplar theory (individual instances) •Going to talk about three different types of theories: •First classical theory: in which we use rules to define whether a given object or entity in the world is a member of a category or not •Then there is probabilistic theories: instead of having these rigid rules argue that certain properties will define or increase the likelihood that a given instance will be a member of a category •There are two different classes of these theories •There is prototype theory which argues that we use prototypes - mental representation of the average of a given category •And then exemplar theory argues that we have a wealth of experience with individual instances of a category - so for category of dog we have encountered many categories of dogs - those are stored in our brain and if we want to know if a given animal is a cat or a dog or something else we can compare it not to the template (which would be the prototype theory) but to these instances and if its more similar to all the things we have labeled as dog than that would be how we categorize it

Summary: Approaches to categorization

Classical theory -concepts have definitions (necessary and sufficient conditions) Prototype theory -theres a summary representation for each category Exemplar theory -no summary representation - a concept is a collection of individual instances Theory-based categorization -categories include causal explanations •No individual one here is the right answer - there are many ways that we use both prototypes and exemplars for making different kinds of judgements so we surely store exemplars and exemplars are useful to us in certain circumstances it does seem like there is a lot of evidence that we have prototypes and for certain kinds of categories prototypes or exemplars don't really serve us that well and these theory based explanations allow us to make judgements •Even with example of blackbird in bat suit - is it more similar to blackbird or bat it depends on what your theory is and what would define its category •If you base it off genetic makeup that isn't going to be based to visual similarity its going to be based on a theory of what it means to be a blackbird or a bat

Types of phrase structure rules

Descriptive rules -characterize the language as it is ordinarily used by fluent speakers -Linguistics aims to provide a descriptive grammar of language Prescriptive rules -standards for how language "ought" to be used For example: -Don't start a sentence with And or Because. -Don't end a sentence with a preposition -Don't split an infinitive We should keep a healthy dose of skepticism about prescriptive rules -This type of English I just can't put up with. -Up with this type of English I just cannot put. •there are different types of phrase structure rules •there is descriptive rules and prescriptive rules •so descriptive rules characterize language as its ordinarily used by fluent speakers and so the goal of linguistics is to study the language and provide descriptive grammar of the language - so how if you just listen to people speak English - if you read written English you can figure out what are the phrase structure rules of English and how are they similar or different to those of other languages •so its language as its ordinarily used but prescriptive rules are advice for how the language ought to be used or should be used •these are standards that are conveyed - in for ex. the elements of style one guy that has strong opinions about what grammatical constructions you might see commonly in peoples writing are good ones that people should use and which ones are bad and should be avoided at all cost •so these are just general principles or prescriptions or advice - don't start a sentence with and or because •don't end a sentence with a preposition •don't split an infinitive •these rules have had their time and place but many of them have been eroded in modern writing for various reasons - I don't think its all just because of texting and the degradation of our grammatical standards - I just think its evolving and certain things like starting a sentence with and or because I sometimes find it useful to start a sentence that way •we should keep a healthy dose of skepticisim about these prescriptive rules •if you adhere to them too closely and try to always follow them you would come up with sentences like this... •this type of English I just cant put up with •but if you don't want to end a sentence with a preposition you would end up with up with this type of English I just cannot put •and that sounds ridiculous

Fluency guides judgments...

-We assume that if we encode/process something quickly, we will remember it well. -Ease of processing influences judgments of learning -But we often mistake our sense of fluency for a sense that we have learned something well. -Stock performance data from the New York and American Stock Exchanges -Alter & Oppenheimer (2006) showed that fluently named stocks (e.g., KAR) outperformed disfluently named stocks (e.g., RDO) -"...all else being equal, a simpler name outperforms a complex name. Cleverness, uniqueness, and novelty are assets in some settings, but in the financial world, they can be liabilities. If you're trying to choose a ticker symbol for a new venture, there's nothing wrong with asking a monkey to throw darts at a board filled with possibilities—as long as the board is filled with symbols like WUBA, OCIP, and LEAF rather than BRX, SRLP, and MVNR." -Adam Alter, The New Yorker (Nov 13th 2013) -Evaluation of student essays, consumer products, and advertising materials influenced by: ¤ font ¤ printer quality ¤ use of long words •fluency guides our judgement in many ways •fluency is the ease of processing (how easily we can encode something into memory or process it) •we assume that if its easy we will remember it well and also it might be higher quality •this influences if you ask people in memory experiments to judge how well they are learning something if they feel like they are processing the information readily and its not taking a lot of effort they will tend to judge - yeah I am learning this well I am very likely to remember this •but people often mistake their sense of fluency for a sense that they learned something well - you could have a professor or lecture where everything seems fun and interesting and then you feel like this is easy to pay attention to its kind of interesting and exciting and then you might mistake that for actually learning and not take effort to go through and study the slides and challenge yourself and test yourself and read the book and try to do the real learning because you mistook your ease of processing it while you were sitting in class for doing the real work of learning and memorizing the information and understanding the concepts •here is an example of a experiment Oppenheimer did with stock exchange data •they looked at just stocks that were listed on the stock exchange that had either fluent names that you could pronounce easily like KAR or disfluently named stocks like RDO •stocks that had names that don't really come into a pronouncable acronym •so they divided this into pronounceable and unpronounceable and this is just the duration from the initial offering •these stocks were all coming out onto the market with an IPO and you looked at the percentage change in the share price on the first day from the initial listing many stocks go up on the first day - if they had a pronounceable name they tended to go up much more than the unpronounceable name and by a week it diminished a little but it was still a big effect •by 6 months it was a small effect and by 1 year it was non-existent •so in the long run people start to learn about these companies and they look at the balance sheet and they make stock decisions based on financial advising and system 2 reasoning •but initially people are very influenced surprisingly so by the pronounceability of the stock name •so heres a quote from an article that one of the authors of the study wrote in a piece in the New Yorker •so the ticker name shouldn't make a difference but people are unconsciously influenced by this in their judgements •evaluation of students essays, consumer products, and advertising materials influenced by: font, printer quality, use of long words •"Consequences of erudite vernacular utilized irrespective of necessity: problems with using long words needlessly" •in this study he did a series of experiments that presented participants with essays and some used the thesaurus heavily and used long words and tried to sound smart by enriching the vocab and some just used more common words •and in almost every case the essays that used the long words were actually judged to be a lower quality - because the reader had to put so much effort in to understand them and they did this attribute substitution where they took the disfluency in reading this and feeling stuck on the words and readability to finish their appraisal of the cogency of the argument the author was making •and for this work Oppenheimer won the Ig Nobel Prize

Other Cog conflict assessments

To test response override: - Stroop test: "Report ink colour." - Flanker test: "Which direction is the middle arrow pointing?" - Simon test: "Push left arrow if see something red. Push right arrow if something green." • Undetermined responding - Stem completion: e.g., s t a _ _ ? - Verbal generation: e.g., wheel ? • Error commission - ERP evoked whenever subjects realize they have made an error (presumed from the ACC)

The PFC and cognitive control

•going to get back to what the PFC does for cognitive control and this is generally focused on the DLPFC from here on out •so I gave the example that the phone is ringing and do you pick it up? •That is the question we are going to explore and go over a computational model that explains a framework for how the PFC might regulate behavior •so how do you know what to do when the phone rings •this is from an example - an influential review that was put out - this is the general model •the phone rings and there is a set of relay processes (these are nodes in the information processing network) and these are the behavioral outcomes you can either answer the phone or don't answer the phone •so when the phone rings the prepotent response is to answer the phone and one of these pathways is going to be active and one is going to be inactive •so in this case the phone rings and action answer is activated •now what we are going to add is these two potential contexts - so you are at home or at a guest house •in real life there can be more than two contexts but these can be thought of as something that the prefrontal cortex is representing •or information that is available to you from your senses - processing the environment and knowing where you are (at home or at a guest house) •and in this case if you are at home this context is going to be activated •so this part of the model is actually happeing outside the PFC •its just visual recognition that you are at home - recognizing the environment •and its activating a node within the PFC and all of these other nodes here might also send signals to the PFC •and what the PFC is thought to do is to link these up •so this is similar to what we talked about in terms of how memory formation and retrieval works •with the hippocampus is binding together different elements of your experience and indexing them •here your PFC is taking information about your current context and the situation (you know the phone is ringing so you are getting information about that happening - you know the context is that you are at home and the PFC also knows that an action plan that is good when the phone is ringing is to answer it - and its sort of linking thse up and it knows its learned through experience with phones that when you are at home and the phone is ringing you should probably pick it up •and here im adding this extra arrow - reward signals •the reason the prefrontal cortex has learned this is because picking up the phone in this at home context has been something throughout your life that has been rewarding •it's the right action to take - if you do it in time you take the call and you find out what the person was calling about - you communicate - if you don't do it in time you miss the call you might miss out on something •and so its something you've learned from a young age •and heres just showing the VTA neurons (that means this area here the Ventratagmental area - its an area of the mid brain that has neurons that are rich in dopamine that are sending dopamine to the PFC via the nucleus accumbens •just know that the reward value is going to be a modulator of how the prefrontal cortex aquires these behavioral schemes •now if you are a guest at someone elses house you want to activate a different PFC schema here - now the guest context is activating a different node and the phone is ringing which is activating the same node but in the context of being a guest you want to chose an action plan •so in this case not to answer - you want to hold or override that prepotent response and this is also potentially sculpted by reward signals or times that you answered the phone and it wasn't appropriate and someone told you oh don't answer that •so you've learned that there are certain contexts where you want to take one action and certain contexts where you learn that's not the appropriate action •ultimately over time as you've experienced the phone ringing many times in your life the PFC becomes less and less necessary •here the fading you see in these diagrams of the lines going from the PFC to these different behavioral plans is weakening - its involved but less important - you don't need as much cognitive control or regulation because its something that you've really learned over time •its evolved and is going to weaken and this will end up just becoming a dominant plan in the at home context and in the guest context you are overriding this and you might need a little bit more PFC input because this is less dominant - that's why these lines are less thick •this is just a schematic model - none of these nodes are supposed to represent specifically defined regions of the brain •in the end answering or not answering might be motor action and these ones might be premotor plans •this might be the actual processing of the sensory stimulus and this might be recognizing what the stimulus is •this is just a schematic to show you the role of the PFC as this switch board that is taking into account all that is happening in your environment - receiving all this information and deciding which action plans might be useful in different contexts •now if you damage the PFC or remove it entirely - now the dominant plan will be carried out in context even when not appropriate •so if you are at a guests house and the phone rings and you don't have any PFC input trying to regulate this and say this dominant prepotent reaction of answering a phone that is ringing is not going to be what I want to do here - the person might go ahead and do that •a person with PFC damage might answer the phone at a guests house and not realize that that is inappropriate in that context

Language in the Brain

•then I just wanted to relatively quickly summarize a few gross descriptions of how language is organized in the brain •gross level meaning sort of talking about whats happening here in the left inferior frontal cortex (Broca's area) back here in the posterior temporal cortex (wernicke's area) and the connection between them but there are many areas in the brain beyond this figure that contribute to language in different ways

Utilization/Imitation Behavior

•now you see this kind of thing in patients with frontal damage - these are photos taken from this famous French neurologist - he highlighted two different types of behavioral problems you see in patients with frontal lobe damage •one is called utilization the other is called imitation •utilization behavior - illustrated by these drawings here - he just put a stethoscope or tongue depressor on the table and here the patient is taking the stethoscope and trying to listen to his pulse and do his tongue depressions •in this case the patients career was a nurse so this isn't something anyone of you with forntal damage would do but she wasn't asked to do this she just is seeing the object and engaging in this dominant stimulus response behavior - that in her past she knows that when there was a stethoscope or tongue depressor there was something she used to do with that •and now she is taking that object and utilizing it with its proper function but just doing it in the wrong context •the doctor did not ask her to do this and is flaborgasted she is taking it to this extreme •heres another patient some nails and a hammer were placed on the table and there was a picture - and the patient without being prompted in the doctors office is hammering a nail into the wall and hanging a picture •and the patient might think that this was what the doctor wanted - why else would the hammer and nails be there but really its supposed to be an illustration that the patient cant help themselves but recognizing what actions are implied by the object and just doing this dominant behavior that when there is nails and a hammer the thing you would do is nail them into wall and hang the picture •imitation behavior is just a form of copying •in this case the neurologist is sitting here and puts his glasses on and here the patient has a set of glasses in front of him and he puts them on - he will even imitate the doctor putting glasses on if he is even wearing glasses already •so you cant see it well on this slide but he already had glasses on and he put a second pair of glasses on •in this pic the doctor starts praying and this patient starts praying •these are just examples of the kind of stimulus driven or automatic behaviors that fail to get overridden or regulated when the PVC is damaged •you see this in children - this is a child engaging in what might be utilization or imitation behavior •children have a very underdeveloped PVC •here is a container that is useful to put things in and he cant help but put himself into it •imitation you also see - its very important for children to imitate •and without PVC you do a lot more imitation •you can see that the underdeveloped brain in children is a lot like what you see in many frontal lobe patients -frontal lobe patients: No longer able to inhibit pre-potent responses.

Innateness hypothesis (Chomsky)

-"Something has to account for the fact that my granddaughter picked out part of her environment as language related, and almost reflexively developed a language while her pet kitten, a chimp or songbird, exposed to exactly the same data, didn't take the first step and couldn't conceivably take the second."

The critical distinction between syntax and semantics

-"Colorless green ideas sleep furiously" (Noam Chomsky) -compare with "Furiously sleep ideas green colorless."* -"Twas brillig and the shlithy toves did gyre and gimble in the wabe" (L. Carroll 'Jabberwocky') -Illustrates that sentences can be syntactically correct, even when meaningless. •and another important point about syntax is that it operates independently in the way of semantics •noam Chomsky pointed this out with the sentence he put out there, "colorless... •and this sentence uses real English words and puts them together that seems grammatically allowable -you don't read that sentence and think this isn't something one could say •but you may just think I don't understand at all what that means •and so each of these words has a meaning but when you combine them in this way its syntactically correct it adheres to the rules of syntax -it has a viable phrase structure in English and yet it doesn't really mean anything that is easy to ascertain •but if you rearrange the words you can have a sentence that is equally non-sensical and in this case most people would look at that and think this is not syntactically correct •so the fact that we can look at this and say yes that is syntactically correct and say no this one is not •means that the meaning of the words isn't important necessarily for the syntactic parsing - the meaning can help us sometimes understand it better and parse it more easily - in this case we are automatically taking these words and automatically building it into this phrase structure tree that automatically lets us know yes that is a reasonable word order in the English language •there is nothing that syntactically bothers us about that •heres another ex from louis carrol - from just made up words •its syntactically allowable - it works - it doesn't seem to challenge our notion of how English words should be organized •and these examples indicate that even when a sentence is syntactically correct it could still be meaningless

Availability Heuristic

-Are there more words that start with t or more words that start with k? -Are there more words that start with R or more with R in the third position? -Are there more words that end with -ing or more words that end with -_n_? -Most people pick the first answer in each case -More words starting with t come to mind than words starting with k. Correct! -More words that start with R come to mind, but actually more with R in third position! Wrong! -More words ending in -ing come to mind than ending in -_n_. Wrong! -In fact, the latter is the superset of the former -The availability heuristic may lead us to believe that we always do the housework ourselves. -The judged probability of an event is related to how easily that event can be brought to mind. -Which is more likely? Death by: all kinds of accidents, or strokes? strokes -suicide, or homicide? suicide -cancer, or heart disease? heart disease Overestimated: -All Accidents -Tornados, Floods -Cancer -Fire -Venomous bite/sting -Homicide Underestimated: -Diabetes -Stomach Cancer -Stroke -TB -Asthma -Emphysema -Judgments are clearly influenced by how often each factor appears in the media -While homicide and fire are often reported, stomach cancer doesn't get as much press -What about shark attacks!? •So we are going to sort of go through some of the heuristics that these researchers developed and described over their career •The two main ones we are going to go over today are the availability heuristic and the representative heuristic •So we will start with availability and we will do it with a demo •Are there more words that start with t or more words that start with k? T •Are there more words that start with R or more with R in the third position? •Are there more words that end with -ing or more words that end with -_n_? •Words that end in -ing its easy to think of them I mean you could just take verbs and add ing •Blank n blank seems harder to think of it especially if you have framed it in your mind like how many words have n as the second to last position and you might think of words like end, and, one - but words that end in _n_ are all the words that end in -ing plus words like -ang (sang, rang) etc...there are many words •There are many times more words in this condition and it's a superset of the former meaning every word that ends in -ing satisfies this but then there are so many more •This is also wrong - there are many many words that have r in the third position •The fact that thinking of them is hard is the point here - the availability heuristic is how easily the information that you are looking for how readily it comes to mind •This is a question of which has more paths - a structure or line that connects one of the X's in the top row to an X in the next row to an X in the final row •The answer is both the same •But the fact that most of you said A is what I was hoping to illustrate •This is the intuitive option because of the availability heuristic because just a glance at this shows that there are many potential paths that come to mind - you try a few in your mind - your like I could start here and go to any of these options so there are many more choices at the first level and then there are all these choices at the second level and it just seems like there are so many different things •Where as here at each level you only have two choices so it seems very limited - it could go straight down, zig zag, or skip by twos •But if you quickly simulate it for like 5 or 10 seconds it doesn't seem like there is as many paths and you would need system 2 to solve it more analytically either by representing it as a mathematical equation or by realizing the number of combinations here is quite large •So another example of the availability heuristic that plays out in everyday life is for ex people who live with roommates or partners you might feel like you are the one always doing housework - that you are always tidying up doing the dishes •We tend to inflate our estimates of how much we do and the reason is its easy to think of all the times you did the dishes - I did the dishes yesterday and the weekend its always me because those are the things you did so you remember them and they come to mind but you don't necessarily notice all the times that the dishes or things sitting on couch were tidying up by someone else - unless you are very observant - those wont come to us as readily so it certainly seems to us that we are doing more things than we might actually do •And so a definition of the availability heuristic which applies to all of these demos and some more coming up is that the judged probability of an event or estimation of how frequent something is is related to how easily the event can be brought to mind •In essence we are being asked to come up with a true probability like what percentage of the time do I do the dishes relative to my roommate but we don't have an answer to that so we do this thing called attribute substitution •We take one attribute that we are trying to estimate (the frequency of an event) and we are substituting in another attribute which is how readily they come to mind or how easy it is to quickly simulate or recall a few examples and that leads us to make a judgement based on something that is slightly different and usually they are correlated and usually they help you get the right answer but they can cause you to get the wrong answer in some of these demos we have done •So here is another one: •Which is more likely? •Death by: all kinds of accidents, or strokes? Strokes •Suicide, or homicide? Suicide •Cancer or heart disease? Heart disease •If you had said cancer or homicide or those who said all kinds of accidents in each case you might have been using the availability heuristic because we hear a lot about accidents (you might think more there are many kinds of ways you could die by accident - car accidents, falling off ladder etc) and strokes you might know a family member or friend or famous person who has had one but they are not necessarily as commonly discussed •Just like homocides hear about them on the news or on tv shows - all kinds of killings but suicides are less covered •Same with cancer - cancer gets a lot of attention and for awhile back in the day heart disease got less •This is a clip from the LA times from awhile ago: sunbathing more dangerous than terrorism •There are many things we devote a lot of thought to and change our behavior to avoid them and terrorism when it happens is tragic and devastating but the exposure people give themselves to UV by being on the beach without sunscreen is very bad •There is a lot of evidence that that is going to be more dangerous for you and more likely to ultimately end your life then being a victim of a terrorist attack •This is a cartoon - this guy is afraid of flying in planes because they can crash but he is texting and driving •Illustrates the double standard - texting and driving is much more likely to kill or permanently injure you than terrorist attack •Its just the numbers on distracted driving and death is staggering •This slide is just an infographic that shows that the probability of death from various terrorists which are extremely infrequent events compared to these other causes of death •1 in 7 people die of heart disease in life and cancer •so certain things people tend to overestimate •heres an example - when these researchers did a study and asked people to estimate the frequency that you would die of different things •so for each of these they made their ratings - and people tended to overestimate the things on the left (fire, homicide, cancer) •and underestimated things like diabetes, stomach cancer etc.. •and in all of these examples the availability heuristic plays out in that just the ease at which these come to mind - where you can quickly come up with examples of someone you know who has died of a particular thing - if its easy to think of people or you have read it a lot in the news or saw it on tv its going to be judged more frequently •so judgements are influenced by how often the factor appears in the media •homicide and fire are frequently reported - stomach cancer doesn't tend to get as much news •shark attacks is another classic example - some people are deathly afraid of going in the ocean - shark attacks do happen and they get a lot of coverage when they do - its good for people to be aware of which beaches have had recent shark attacks but the chance of dying is so low and so this daily show video shows us about the availability heuristic •those are all examples of judging the frequency of events and how people are influenced very much by the ease of recalling examples of these

The Whorfian question

-Benjamin Lee Whorf (1897-1941): "Are our own concepts of time, space, and matter given in substantially the same form by experience to all men, or are they in part conditioned by the structure of particular languages?" -Whorf's bold idea: "We dissect nature along lines laid down by our native languages. The categories and types that we isolate from the world of phenomena we do not find there because they stare every observer in the face; on the contrary, the world is presented in a kaleidoscopic flux of impressions which has to be organized by our minds—and this means largely by the linguistic systems in our minds."

Neuroscientific data on categorization

-Certain kinds of categories are represented in different brain areas •Anomia patients with lesion at different region showed deficits in naming different categories.

Ventromedial PFC (VMPFC)

-selection and inhibition in social/emotional contexts -NOTE: includes orbitofrontal cortex region -he would act more instinctually and wasn't regulating his social and emotional behavior and that's largerly because the rod that passed through his head damaged this area of the prefrontal cortex that we now know is very important for regulating our social and emotional behavior •in different contexts certain actions are appropriate and others are inappropriate and this part of the brain right here the ventromedial PFC which includes the orbitofrontal cortex region (the area behind your eyeballs) controls that -emotional or social regulation part

ACC and Cognitive Conflict

-Conflict is when you have multiple competing representations active - so in the case of this the conflict is that the word itself is leading you to want to say blue but the ink color being red is your goal here - so you have to overcome the conflict between those •When you saw the ink colors and the letters were all X's you didn't have that conflict •The thought here is that this region of the brain that is kind of situated in the medial surface called ACC (anteriorcingulate cortex??) •This is a different area from the VMPFC an orbitalfrontal cortex component its more of the dorsal medial PFC •And this area right above the corpus collusum called the anterior cingulate is one area proposed to play an important role in detecting the presence of conflict •You need to know when there are multiple competing options available to you that this is a good time to engage in cognitive control - that something is happening now that I need to be aware of and perhaps take myself off of autopilot and do something a little different because the context demands it •Heres another example of a task that would require - we gave the stroop test example - congruent means that the prepotent response (reading the word) is consistent or congruent with the word here •So this is going to be easy you can do it on autopilot essentially - this one is going to be the hard one cause its incongruent - there is conflict and this is a neutral one the word apple shouldn't affect your ability to say this is red •So these two (congruent and neutral) will have similar response times - you might get some facilitation here because the response you want green is consistent to the word •This one will be an intermediate response time and this one will be the slowest •The flanker test your goal is to say what direction the middle arrow is pointing - and these arrows to the left and right of it are just distractors •In the congruent case they are consistent with the response - so you want to say left and in this case you want to put right but everything else is pointing left so there is this conflict •And these 0's have no barring they are neither congruent or incongruent •And the simon test is just another example of that - if you see something to the left you are supposed to respond to the left but the incongruent test you are supposed to do the opposite if you see something to the left you respond to the right and that's hard for people to get the hang of •Another situation where you would want cognitive control is when there is undetermined responses - so in a stem completion task if you see the word sta___ and you are supposed to complete it with a word - there are multiple words that could go here (star, stamp, stand, stars etc.) •And if I give you another task such as verb generation and you get the word wheel - you can do roll, spin, drive if you had to come up with a verb that went with wheel •When you have multiple choices available to you and you are supposed to chose which is the best or the right answer in a certain context that can also be a situation where there is conflict •Some nouns don't have as much conflict - like if the noun was scissors most of you would say cut - there aren't a lot of choices there •Another example here: is error commission - the anteriorcingular cortex has neural activity that can be detected with EEG and this ERP (event related potential) and anytime subjects realize that they have made an error - if you are doing a task and it's a difficult task and they make mistakes occasionally everytime they make an error - an electrode positioned right above the ACC shows this characteristic wave form of response that is timelocked to the error and that's more evidence that this region is involved in detecting conflict because whenever you get something wrong you are going to want to realize I just messed up what do I need to do for the next trial and all the ensuing trials after that to make sure im a little more careful and that I engage in cognitive control so I don't engage in another error -Conflict is detected in ACC • ACC signals to the DLPFC to upregulate attentional control • To inhibit word reading, and announce font colour

Language Acquisition

-Humans pass through several stages while learning language: 1. Cooing stage. Begin to utter a wide range of sounds. 2. Babbling stage. Utter a smaller set of phonemic sounds. 3. One-word stage. Speak out words and morphemes. 4. Two-word stage. Production of two-word sentences. •Babies go through different stages where they start cooing, and then babbling, and then they eventually start learning words for things, and then start putting words together •They might use a verb "want cup" or "blue cup" and then they become sentences •So you don't really need to know much about the ages of the different stages - there is a lot of variability there but I just put this here as a lead in for what is a more covered topic in the book which is the critical period idea •That language acquisition is something that we take for granted because any child raised in an environment with language will start to develop understanding of language - and then eventually speaking of language - and then it will increase in complexity - and then the syntax will become more and more correct

Example 1: Eskimo Snow

-Eskimos have many different words for "snow" •These examples were initially introduced by whorf as evidence for linguistic determinism idea •And in each case youll see they have some weaknesses too •So one of the things whorf was interested in in his assessment of linguistics and cross-cultural comparisons is that eskimo people in Alaska had this really rich vocabulary for snow with many words •Heres just an example of words for snow •If you look at all these different words and try to come up with a translation of what they mean you can see there is a rich vocab for snow •And whorf felt like this is amazing this is exactly what I was looking for as an example of this phenomenon •"That we, as English speakers have the same word for falling snow, snow on the ground, snow packed hard like ice, slushy snow, wind- driven flying snow - whatever the situation may be." (we call it snow) •"To an eskimo, this all-inclusive word would be almost unthinkable; he would say that falling snow, slushy snow, and so on, are sensuously and operationally different, different things to contend with; he uses different words for them and for other kinds of snow" •Now this idea kind of captivated people - like wow if they have all these words for snow other languages and other parts of the world where people have different things important to their lives like what kind of vocabulary did they have - does this really limit an English speaker to notice these differences in the weather or different kinds of snow by not having this vocab but as people studied this more it became clear that whorf wasn't really being fair to English - and other languages that actually do have a fairly rich vocabulary for snow -the problem with this example... •So even though we don't use these terms all the time and they might be more important in the life of a person whos whole existence depends on knowing different kinds of snow - we do have a pretty rich vocab •This is just an ex of English terms for snow or snow like things •And so this ended up going down as a big myth and these two books - their titles illustrate that this was once a big idea that people thought was good evidence for linguistic determinism but turned out in this case word myths: debunking linguistic urban legends •Or in this case the great eskimo vocab hoax •These books make the point that this is not as true or accurate as whorf said •And this cartoon does a nice job illustrating the absurdity of whorf's claim •But this idea - this romantic element of whorf's idea lives on - even in modern advertising this ad here in Fiji, theres.. •Is trying to convey that in the language if there is no word for something that means they couldn't even conceive - like they didn't need a word for it because everything is so pure and clean in Fiji - and if you don't have a word for it it would be hard for them to even comprehend it •So this is just another version of this ad - the land that time and airborne toxins forgot •But this is not actually true •In Fiji , their language does have words for different kinds of pollutants or litter or trash or toxic waste and they also have a problem with pollution just like everything else

Displacement

-Language allows us to think of, and communicate about, 'things' beyond what is immediately 'sensed' -"On Tuesday, we will have our next in-class quiz." -"Dark matter is a type of matter which neither emits nor scatters light or other electromagnetic radiation and is estimated to constitute 83% of matter in the universe" -"Imagine no possessions. I wonder if you can. No need for greed or hunger. A brotherhood of man. Imagine all the people. Sharing all the world..." -"Despite the negative press covfefe" •this idea that language allows us to think of and communicate about things - about stuff that extends beyond the scope of our immediate moment - our experience in the present - so we aren't limited by that •next Tuesday we will have our next quiz - talking about the future (simple ex. of displacement) •this is more abstract "dark matter..." (none of us have seen dark matter to observe it its only indirect through different measurements - all of this is abstract - but this sentence allows us to convey something that humans will never be able to directly observe) •john lennon quote (he is conveying a vision of the future and humanity that could be potentially, if these were true - its something you can imagine but not something you can see all the time)

Chomsky: Universal Grammar

-Language faculty is a genetically transmitted algorithm (set of procedures) for developing a grammar -Influential idea, but highly controversial! -Children can in principle acquire any natural language as their native language. -How can the innate hypothesis (language faculty) account for this? -The set of principles it incorporates must be applicable to all languages: -Universal Grammar: a set of genetically endowed grammatical principles which determine the structure and range of possible grammatical operations in natural language

Example 3: Color terms

-Most languages seem to take their basic color terms from a set of just 11 color names -When only some of the color names are used, naming falls into a hierarchy of five levels -Berlin and Kay (1969): Color hierarchy -In 2 color term languages the terms correspond to Black & -Dani tribe of New Guinea use only 2 color names: 'mili' (dark) and mola' (light) In 3 color term languages they correspond to Black, & Red -Languages with additional terms items are added as follows: -yellow, green, blue then brown, then purple, pink, orange, and gray -Commonalities across cultures/languages suggest that there must be some universal properties that underlie our interpretations of colors •One thing we know about color is that when you look across languages most languages have a basic set of color terms and there seem to be 11 fundamental color names that languages with rich vocab for color usually have these 11 basic terms and then you can go from there and expand your color vocab •But some languages actually have fewer color terms so when a language has fewere than 11 naming seems to be in this hierarchy of five levels •So this was work that was done by berlin and kay in the 60's and they found that some languages only had 2 color terms •So the language didn't have any discernable vocab items to refer to the colors of the world except for 2 so they could say in the new guinea language that something was dark or light •They didn't have words for red, blue, green, purple there was nothing that could be found in studying the language that would convey a label for those other colors •Some languages add a third term and in every language that had three color terms it was for white, black and red •So if you are going to have one color term that is not white or black researchers found that it was almost always or always red •And then as languages added additional color terms - so there are some languages that have 6 color terms and it would generally be black white red •And then yellow green blue •Yellow green blue kind of generally come together in a cluster - not always •But if you were going to add more color vocab those would be the next and then if you were going to add another term brown usually comes in then purple, pink, orange, and gray •You don't need to know the order of these •Just trying to indicate that across cultures these languaes that develop in different parts of the world often isolated from each other there is something fundamental that we all tend to have the same way of referring to color and when we have impoverished vocab for color the fact that it happens in this hierarchical structure - so some languages only refer to dark white and red and others would add these terms - seems there might be some universal properties that underly our interpretation of colors •It also raises the question that languages that lack certain color terms might have differences in their perceptions •I want to start with this ex. here •Which of these two look the most similar to each other? •The correct answer is 2 and 3 in terms of their wavelength of light •The light energy that is hitting your retina when you are looking at this vs this are much closer to each other in terms of the wavelength that is reflecting back than this and this - these ones are further away and yet most of you picked that 1 and 2 were more similar because English speakers were refer to these two both as green •They would say that this (1) is a deeper lime green and this one is a seafoam green (2) and this one (3) is more like a teal or turquoise •So this color of 3 is more of a blue like color and these are two are more like green •So if we had to use basic color terms like one of our color terms we would say this is a shade of green this is a shade of green and this is a shade of blue •So this line here in black is generally where most people would place their boundary for referring to things that are green vs. things that are blue •If you just have a set of choices and you are getting color squares like this and you are just deciding is it green, is it blue, is it yellow using basic terms you have to draw the boundary somewhere and for most people these two strattle that boundary

Sequences of phonemes

-Only some are acceptable in a language. - For example, the sequence [tl] is not acceptable in English -Adjustments for certain phoneme sequences. - For example, the [s] sound becomes a [z] in words like "bags" •Another thing about phonemes is that some are acceptable in a given language •Like English you cant put together the tl - there aren't words that allow that combination it becomes hard to know if a word did have a tl how you would say it without adding a guess of what vowel would be in between •And we also make adjustments in certain sequences - when the s comes at the end of a word like bags we give it the z sound - bagz •With clocks you give it that s finish but with bags you give it that z sound •And these are just things you come to know through learning to speak the language you modify the phonemes - so if you are trying a computer algorithm to recognize phonemes you cant hard code the spectral structure of each one because they are going to be changed in different contexts •You would need to take into account these kinds of modifications •And you can also build in things that aren't allowed so that its not making guesses that are very unlikely and that helps restrict the search space

How do we get to know language?

-Plato's Problem: How can children with different linguistic environments arrive at an accurate grammar relatively rapidly, and with finite input? -"Man has an instinctive tendency to speak, as we see in the babble of our young children; whilst no child has an instinctive tendency to brew, bake, and write." [Charles Darwin, The Descent of Man]

Another example

-Speakers of different languages may be obliged to think about the world differently -Gender-neutral languages (like English): -I spent yesterday evening with a neighbor. -The fork said, "Wash your hands." -Gendered languages (like German): -I spent yesterday evening with a male neighbor. -Speaker compelled to choose between: Nachbar or Nachbarin -The masculine fork said, "Wash your feminine hands." •Another example: speakers of different languages may be obliged to think about the world differently •So in a gender neutral language like English I could say I spent yesterday... •I coulds say that to you and you wouldn't know if the neighbor was male or female and English doesn't command me to disclose that •But in a gendered language like german you have to say I spent yesterday.. •There is no neutral way to refer to it so you have to convey more information and this changes the way that we communicate with each other because certain things in English that you can do like using this neutral term you cant do in other language •These languages force you to commit •A fork said wash your hands •If a fork is a cartoon character here the fork doesn't have to be masculine or feminine but in a language like german fork is going to be masculine so if you draw a fork as a cartoon character its probably going to be more masculine and hands are feminine so its kind of forcing the speaker to convey this information •These are all examples of language and thought •You are what you speak???

Categorical perceptions of phonemes

-Voice onset time (VOT): The time between the beginning of the pronunciation of the word and the onset of the vibration of the vocal chords - "ba" your vocal chords vibrate right from the start - "pa" your vocal chords do not vibrate until after a short delay -Another property that is covered in the textbook: categorical perceptions •So when we here phonemes - each of the phonemes are usually interpreted as being one of the 44 phonemes we know in English •We don't usually interpret it as being ambiguous and so the way that different phonemes like ba and pa are distinguished are by voice onset time •Voice onset time is just the time between the beginning of pronouncing the word and the onset of vibration of the vocal cords so when you say ba your vocal cords vibrate right from the start there isn't a long voice onset time - and pa your vocal cords don't vibrate till after a short delay •So how do you distinguish whether I say ba or pa well you do it based on computing in your auditory system the voice onset time and that allows you to tell if its ba or pa •They differ on that dimension •Now you can acoustically manipulate sounds - like artificially change the voice onset time to create a range - a continuous set of possibilities - now these illustrations here are confusing - they are spectral sweeps but what they are showing here is just different examples of manipulating the voice onset time systematically •And what im going to play for you is what each one of these sounds like •What did you all here for these ones? Ba (first row) •And for these ones? Da (second row) •Last row? Ga •Now some of you may have had a different cutoff •Some of you may have started hearing da half way through - and some of you may have continued to hear Da till it switched to Ga •So there are individual differences as to where it switches - the idea of categorical perception is that its discrete •That people don't usually report a blending or ambiguity - that each one is usually interpreted as being one viable phoneme •So either ba or pa or ga and we don't show this blending -Our categorization of phonemes shows abrupt boundaries, even when there is no corresponding abrupt change in the stimuli themselves. -This phenomenon is referred to as categorical perception. •So this figure here illustrates sort of two possible ways that a data could have turned out from an experience like this •If you alter the voice onset time - adding milliseconds of voice onset time or subtracting milliseconds of voice onset time hypothetically you would have seen a continuous function •So this is the percentage of subjects identifying the sounds as ba and that would start at 100% here and decrease to 0% here •And it would just be gradual that as you change it peoples perception changes •And this is the percentage of people identifying the sounds as pa in this case and you would see the opposite curve •But the actual data are much more discrete - so people are at 100% everyone agrees and then it drops off precipitively - so there are different points where it changes but it is the case that here its 100% - here its 80% - but it drops very steeply and then settles here at 100% •So even though you continue to increment here and here instead of it being this graded representation where you perceive certain phonemes as being like this is a good example of ba - and this is like a pa-ish ba and a ba-ish pa - its not really how it works people interpret it as one or the other •That's called categorical perception - so your brain is assuming that whats spoken must be one of your viable phonemes and given the information its settling on one interpretation and when it jumps it does so discretely

Syntax

-The systematic way in which [categories of] words can be combined and sequenced to generate meaningful phrases and sentences - Rules apply to grammatical categories (nouns, verbs, prepositions, adjectives, adverbs) •Syntax is the next level - so we have phonemes that make up morphemes and the morphemes make up words and the words are combined with syntax •syntax is the systematic way in which words or more accurately categories of words or classes of words (by categories we mean nouns, verbs, adjectives) can be combined in sequence to generate meaningful phrases and phrases are combined into sentences •so the rules that syntax gives us apply not to individual words but to categories or classes of words •so you have rules that tell you what happened with nouns and verbs and prepositions and adjective - and there are some modifications for individual words especially in English there are a lot of violations to these rules or exceptions but in general syntax tells us what is or isn't allowable with certain categories of words -One kind of syntactic rule is a phrase-structure rule, a constraint that governs the pattern of branching in a phrase-structure tree. -One such rule specifies that a sentence must contain a noun phrase (NP) and a verb phrase (VP). •and one ex, of a syntactic rule is a phrase structure rule - this is just a constraint that governs the pattern of branching •so s here is a sentence and the sentence is composed of a noun phrase and a verb phrase - and the noun phrase here happens to consist of a determinant (the) and an adjective (ferocious) and a noun (dragon) •and the verb phrase consists of a verb and another noun phrase - so chased is the verb, and then the object of the verb its another set of another determinant, another adjective, another noun •you don't have to know how to take a sentence and break it down like this but you do need to know that this is a phrase structure role and in English we have a rule that a sentence must contain a noun phrase and verb phrase

The Wisconsin Card Sorting Task

-another test to assess PFC function -subject is given cards with shapes that vary on three dimensions -color, shape, and number of shapes -subject sorts cards into two piles -experimenter provides feedback after each response -the sorting rule will occasionally change and subjects must adapt Cognitive Processes Engaged during WCST: 1. Must internally maintain goal 2. Selection demands are high 3. flexibility required to shift current goal 4. Need to inhibit previously relevant information -patients with PFC lesions are impaired at the WCST -they exhibit perseveration -keep responding based on a previous rule despite evidence against this rule -may continue for as many as 100 trials • another test that is used by neuropsychologists to evaluate how well patients are able to engage in goal directed regulation of behavior and patients with frontal lobe damage will do badly at this • so in this task subject is given cards and they have to sort them but they don't know how they are supposed to sort them they just have to figure it out through trial and error • each card has colors shapes and numbers • you can sort them based on three different dimensions, shape color, or number • and the experimenter just provides feedback on each response • so you set up some piles and they get a card here and if they are going to try to sort by color and they put a card here they could be told oh that's right - update their hypothesis that they are supposed to organize by color • if they put it here and its wrong - well maybe on the next time they will think numbers are relevant maybe I should have put it here • the sorting rule is stable for a few dozen trials and then it changes unexpectedly and the subjects are sorting and suddenly they are told no thats wrong and they need to figure out well has the rule changed? Whats my current goal now and through trial and error discover the right rule •and most people - after you were told you were wrong quickly people adapt to that •you need to engage a number of cognitive processes •you need to maintain your goal and your goal can be one you internally generate - like im going to try to sort this into the pile the experimenter tells me is correct •the selection demands are high meaning you have three different choices for each card and you need to select what the best choice is •you need to be flexible - when the experimenter tells you its wrong you don't want to keep doing the wrong behavior you need to shift yourgoal and your hypothesis on which sorting dimension is currently relevant •and you need to inhibit the previously relevant dimension - you might have sorted the prior 20 or 30 cards by number and then you are told that's not the correct dimension anymore that's not going to get you the right answer and then you need to adjust •so patients with PFC damage do badly on this task •it's a good assessment of whether someones frontal lobe has been compromised •they exhibit whats called perserveration: they continue to respond based on a previous rule - whatever was working for them previously despite evidence against the rule •kids do this too - they figure out what the dimension is initially that gives them the right answer and then they have a hard time when you tell them to change to make that switch because it requires inhibitory control and this goal directed regulation of behavior to say that rule which was working for me a few seconds ago is no longer right -they get stuck in a rut doing the same thing over and over again even though its wrong

Problems with exemplar theory

-assumes that many individual exemplars are stored in memory without "blending" -has trouble accounting for people's ability to extract general properties of categories so as to allow classification of new instances •problems with exemplar theory are that it assumes that all these individual exemplars are stored in memory without blending but really our memory is not perfect •there is often a lot of blending - some individual exemplars are stored pretty well and we can picture them •other things might get blended with similar experiences we don't really need to remember every dog we have ever encountered or every cat we have ever encountered we can usually just extract the concept usually in life you want to move away from the exemplars you don't want to say oh that's a chair im going to store that in memory as chair 1 and that's a chair and now that's chair 4939012 and you are storing them each •we don't need to remember every chair we have encountered we need to learn what in general people will call a chair - so I can use that word effectively in communication •so exemplar theory has trouble accounting for peoples ability to extract general properties of categories to allow classification of new instances also to know whats in general common about these •so if I show you something that you haven't ever seen before how are you going to categorize it •you can do it based on visual similarities but you might not have thought about the general properties • Assumes that many individual exemplars are stored in memory without "blending" • Has trouble accounting for people's ability to extract general properties of categories so as to allow classification of new instances

Problems with prototype theory

-assumes that information about individual instances is not stored, or at least is not used to guide categorization -however, people do seem to store information about individual exemplars, and can sometimes be influenced by these specific exemplars -also, prototype theory doesn't have a way of taking into account the variance of a given category • protype theory has some problems - it assumes that information about individual instances is not stored, or at least is not used to guide categorization (focus on the latter part - if you think that prototypes are all we have then you would just encounter things in the world and use them to update your prototype - oh that's a dog now im going to update my prototype - so no one believes that - but those that subscribe to prototype theory would say this isn't how we do categorization by comparing to individual exemplars •but people do seem to store information about individual exemplars - and our judgements are often influenced by very specific exemplars •so you might have a particular dog that you know that you would view as your prototypical dog •whats a prototypical democrat or republican you might think of a particular person and have that be your prototype as opposed to doing some mental averaging •so prototype theory doesn't really have a way of taking in the variance of categories •some categories have very little variance and others have more - so for ex. bananas is a category - there can be different degrees of ripeness but there isn't a lot of other variance compared to other categories like cakes which can take on many different shapes, sizes and forms • Biased based on individual experiences - Prototypes are developed based on which category members one has been exposed to. • Extreme form of prototype theory assume that information of individual exemplars are not stored at all. • Cannot account for failure within category. • "Average" doesn't always work. ISSUES WITH PROTOTYPE THEORY: EVIDENCE FOR EXEMPLAR THEORY • Category variability information is not modelled in the prototype theory, but people can do it based on distance from closest exemplar and category variability.

Prototype theory

-categories are represented by the average of all members of the category -e.g. the concept "bird" is represented by a prototype that is very similar to a robin and different from an ostrich -the prototype need not exist in the real world -the category representation is abstract -does not store information about specific exemplars •prototype theory: imagine two birds perched on a tree branch •when we think about a category of birds, we would generally be visualizing something like this on the left •so if I asked you to draw that most would be likely to draw something like this •the right is not the prototype •so categories according to prototype theory are represented by this averaging of all members of a category •the concept of a bird is represented by a prototype that is more similar to robin/bluejay/sparrow than to emu/ostrich/penguin •and the prototype doesn't need to exist in the real world - you can have a prototypical bird representation that's not like any individual species •doesn't have to be that your prototype is the robin or sparrow your prototype could be some kind of averaging that doesn't really exist •the category representation is abstract - you could draw a prototypical car and you don't have to pick whatever the most common - doesn't have to be a chevy, Toyota or some common sedan car but it could be something like the average car that you see •and it could be different each time - you could draw prototypical car one day and a different one the next day •when you are asked to draw it - you are asked to make something specific but the representation in your mind is abstract and may not have - its not the case that in your brain you have this stored prototype of a car that looks this specific way •but according to prototype theory we don't need to store information about specific exemplars - as we navigate the world we obviously learn many exemplars - there are certain animals we encounter a lot, pets and they have names and their individual exemplars, people that you are friends and family with and they have names •we store all of these exemplars but we also have prototypes that are the generalization that goes beyond the specific exemplar - so this doesn't deny that we store information about our experiences and unique things (this is what my bicycle looks like and this is a prototypical bike but when people are asked to make category judgements prototype theory says they are doing it based on comparison to a prototype) - there's a summary representation for each category - Issues: relies on similarity, can't account ad-hoc categories, or for variance within a category PROBABILISTIC THEORY - there's a summary representation for each category Abstract • People don't naturally make definition of categories. • Categories are represented by prototype • Category judgments are made by comparing new exemplar to the prototype (via familiar resemblance). • Statistical mean • Graded categorization • No necessary and sufficient features for categories • Instead, categories are defined in terms of a set of typical or likely features. • Category members share a family resemblance. • Category membership is determined by comparing the instance to a category Prototype The prototype is an ideal category member or average of the various category members. • To determine membership, ask "How much does this resemble my bird prototype? EVIDENCE: TYPICALITY EFFECTS • Which of the following is a "better" bird? • Which of the following is more like a bird? • TRUE/FALSE: A ______ is a bird. • Name a bird.

What is a game?

-criticism of classical theory -Ludwig Wittgenstein's famous critique: -what is the necessary feature of the concept of a game? -competition between people or groups? (solitaire) -has a winner? (jumping rope) -provides amusement or diversion (are professional athletes amused or diverted?) -for many categories there are no clear defining features -various members share various features, but there is no single feature that is necessary •and Ludwig Wittgenstein had this very famous example in the 50's critique of classical categorization about what is a game? •So he asked the question what are the defining features if you were going to come up with a classical definition of a game - what would you say? •So you might say well it involves competition between people or groups - but in each case there are examples of games that violate this - so solitaire is something most people would consider as a game you could play and there are others like this that are just done individually - so maybe that's not one of the rules •You might say a game needs to have a winner - but jumping rope might be considered a game and there are no winners - there are many games kids play that are fun but don't have a clear winner •Provides amusement or diversion - maybe that's the defining feature that all games provide that but some games can be very intense (athletes have fun if you win or have a great play but its not always amusing or diversing - sometimes its very intense like boxing or martial arts match which is game like but not necessarily fun) •So the point of this - for many categories there are no clear defining features •So various members will share certain features but it is hard to come up with rules that work for every member •Heres the figure from the book so we talked about possible definitions - what about games being played by children well gambling isn't played by children •Engaged in for fun - professional sports (fun for audience but maybe not so fun for players) •Has rules - well playing with legos doesn't have rules • Many categories have no clear defined features that works for every members ("games"). • Proposed instead "family resemblance"

Criticisms of the classical theory

-defining features often can't be found -you can often remove any particular feature and some object will still be a category member -non-necessary features affect categorization -e.g. which of these shapes is a parallelogram? •one thing is that defining features often cant be found - for categories like these shapes - people can agree what the rules are but in many cases in the real world certain features might be common in a category but aren't necessarily going to be present in every instance •and another problem is when you ask people to categorize things sometimes non-necessary features affect their judgements - so when you say which of these shapes is a parallelogram most people will say this one is the parallelogram •that's true this one is a parallelogram but actually all of these are parallelograms and many people will get that wrong or fail to realize that even though this one is a typical parallelogram and we will talk about typicality in a few slides because that is a very important feature that happens in categorization - that some members of a category are considered more typical •so this is a typical parallelogram - or the parallelogram you learn when you learn about parallelograms but this one is also a parallelogram it just happens to also be a square which is a certain type of parallelogram - and this one is a rectangle and this one is a rhombus but they all meet the criteria for parallelograms perfectly •and if we represent categories classically we shouldn't be affected by this - so this is one reason why the classical theory is a little weak to explain human categorization and the other being that the defining features sometimes cant be found or are too ambiguous •so for ex. of that - what is a bird - well birds have different attributes that most people would agree on generally to define birds •so if you are trying to look at an animal and identify if it was a bird or not you might think well if it has wings that's something birds would do - do they fly, does it have feathers, sings, builds nest in trees and so forth •but each of these features may not be necessary •a penguin is considered a bird and should be categorized as a bird if you are categorizing correctly but it has flippers instead of wings - it doesn't fly •it does have feathers but they are different then the feathers of other birds •sings - it makes noises - doesn't sing like other kinds of birds but still •builds nests - they don't really build nests -it doesn't eat worms or insects • so its hard most of the time when you put up a category and ask people to write down classical type definitions like what is a tree - what is a bird what is a mammal •youll write certain things down which are your rules for the category but there are often examples that are considered members of the category that violate one or more of these and it makes it very difficult to come up with classical rule based definitions for most of the things we categorize readily in our everyday lives •so a variation of any one defining feature does not seem to affect our categorization very much here

What is a drunk?

-its late at night, you are walking home when you see someone jump into a swimming pool fully clothed. -you think "that guy must be drunk!" -why do you think that? -classical categorization: stinky breath, wobbling walk, impaired speech, maybe one entry is "jumps in pools at night"? -theory-based categorization: the concept of drunk involves theory of impaired judgement, which explains the man's behavior, so you induce he must be drunk -knowledge-based causal theory •heres an ex. of when you might need theory based categorization - something like a drunk •and the same kind of thing could apply when thinking about abstract categories such as what is a game? •its late at night.... •Classical categorization would say there are certain features of drunkenness you look for and if present this could satisfy category of drunk (stinky breath...) •And you have this other entry for jumps in pools at night that this is one of your definitions for drunk behavior and probably not •The theory based categorization would say you have this concept of drunk and what drunkenness is and what it does to you and how it affects your behavior •Its this causal explanation you have a theory of drunkenness you've learned and you can apply this theory to explain the mans behavior •So your category of drunkenesss is consistent with this description and this views categorization as a knowledge based causal theory

Theory-based categorization

-we know much more about categories than a list of their features or their values in dimensional space -categories provide explanations for how things work in the world -the same way that theories provide explanations for scientific phenomena -they center on causal relations between entities in the world -theories guide perception by leading us to believe that particular features are interesting and others are not •and we need some way of knowing which features are being compared - so this brings us to the final framework theory based categorization •we know categories are more than just a list of features or rules - the classical theory is going to be inadequate or even their values in some dimensional space (bananas they have a certain shape property and so forth that make bananas similar to each other) but many categories in the world are hard to accommodate in this framework so the theory based categorization model says categories provide explanation for how things tend to work in the world and this is the same way scientists explain scientific phenomenon •so there is a set of causal relationships or properties that link things together •they all have certain attributes - the categories we use are based on causal relationships •theories will guide our perception by leading us to believe that particular features are interesting and others are not - categories include causal explanations • Centre on causal relationship between things - Categories provide explanation of how things work • Knowledge-based - Theories guide perception by leading us to believe that particular features are interesting and others are not Categories include causal explanations!

Some Brain Regions

• PFC - Late maturing • VMPFC: Ventromedial PFC (red) - Selection and inhibition in social/emotional contexts • DLPFC: Dorsolateral PFC (blue) - Regulation of non-social cog. functions - Executive control of working memory • ACC: Anterior Cingulate Cortex - Register conflicts - Upregulate other PFC processes

Upregulating Cognitive Control

How do you alter the degree of control? -you pay closer attention to the raid if you're driving while its raining as compared to while it's sunny -you pay even closer attention if it's raining and dark out, than if it's raining but light out -need a mechanism that monitors or detects cognitive conflict -Hears just another everyday scenario - if you are driving and its rainy or foggy weather you are going to want to pay closer attention to the road - you need to be more careful •When it is sunny and clear and the conditions you are clear you can go on more autopilot but when it is rainy or snowy you are going to want to slow down and make sure you have more control •You are altering the degree of control based on your context and you need a mechanism for this and for the stroop test that is going to monitor the degree of conflict

Other types of aphasia

Global aphasia: -Nearly complete loss of comprehension and production of speech. Some single stereotyped words might still be retained Conduction aphasia: -Preserved comprehension -Spontaneous speech has proper syntax and semantics -Impaired repetition & paraphasic errors (phonemes and syllables will be dropped or misplaced) •global aphasia: sometimes patient will have stroke that doesn't just affect inferior frontal lobe or posterior temporal lobe but a huge area of the brain - this would be a particularly devastating case of brain damage where you see nearly the entire left temporal lobe is obliterated •and this patient might lose both comprehension and production so this is a cripling form of aphasia where you lose all language abilities entirely •and then conduction aphasia which is a bit more nuianced •if instead of damaging broca's area or Wernicke's area you damage the white matter pathways - the arcuate fasciculus is this white matter bundle - it's a large bundle or tract of axons that connect neural ensembles in the posterior temporal lobe to neurons in the frontal lobe - so its like the 405 connecting these two areas •so it's a major information transition highway in the brain •and if this is damaged or severed you can see a condition called conduction aphasia: a conduction because these fibers allow information to conduct like an electrical wire from here to here by breaking that wire - patients usually have reasonably good spontaneous speech with adequate or proper syntax and semantics (meaning the grammar is fine, the meaning is clear) •and their comprehension is usually pretty good but they have trouble repeating so if you give them a few words to repeat like: rose, ball, key •and they just have to say it back: rose, ball, key •their going to have a difficulty doing that •you are basically disrupting the phonological loop their not able to take information that their auditory cortex sends to their posterior temporal lobe and get that signal to the frontal lobe to speak - because this is the speaking part of the brain - this is the part that is going to be selecting what phonemes you want and how to organize this into words and how to organize the words into phrases and the phrases into sentences - that's all happening in the left inferior frontal lobe and this posterior region is taking the auditory input and determining what it is •sort of which words were spoken - analyzing the meaning of those words and all of that •so connecting these two up is very important in most forms of language - and when this is severed patients will have impaired repetition and paraphasic errors (this is when you substitute phonemes or syllables that are supposed to be in the world - like instead of the toothbrush calling it a toobey - you got part of it right and then other syllables were incorrectly added or subtracted •other examples would be luliphone for telephone - or desks for decks and so forth

Cog Conflicts: STROOP test

Read words = prepotent response, almost automatic • Almost never need to announce word colours • Incongruent ink colour and colour words --> conflict - Which region detects these conflicts? ACC • Behavioural results: Slowed RT

We use heuristics!

-A "Rule of thumb" or mental shortcut •Often based on past experience •Does a good job most of the time •Save us time and energy •Not guaranteed to be correct...and errors tell us important things... •And so with most judgements because we cant know everything we need to use a set of strategies or mental shortcuts and those are called heuristics •So these are rules of thumb that tend to help us in many situations - we base them on our past experience there are certain heuristics that seem universal to human cognition that most people have these similar strategies for simple kinds of judgements and these are useful because most of the time they serve us well they allow us to come up with an answer that is going to be a decent guess or good enough saving time and energy not forcing us to look up probabilities and do math and figure out conditional probabilities of potential different outcomes and how they interact •So we are not usually doing all that effort we are just trying to come up with judgements that are good enough - and these are not guaranteed to be correct and when they fail they show us important things just like illusions do for visual perception

Principles and Parameters

-A person's syntactic knowledge can be modeled with two mechanisms: -A finite set of fundamental principles that are common to all languages -e.g., a sentence must always have a subject -A finite set of parameters that determine syntactic variability amongst languages -Binary parameter that determines whether a given principle is ON or OFF -e.g., whether or not the subject of a sentence must be overtly pronounced -The human mind is endowed with a biological mechanisms in which universal principles are embedded -The child's learning task is thus 2-fold: 1. Structural learning: Determining the appropriate value for each parameter (parameter setting) 2. Lexical learning: Learning the language specific symbolic association between meaning and sound

How does this work?

-Are we aware of this? -Often, not really! -Highly available cues guide our judgments •we are usually not aware of this •usually these kinds of things are operating behind the scenes so fluency and availability cues in the environment are guiding and influencing our judgement in ways that we don't realize

The availability heuristic in the classroom

-How soliciting more criticism can actually boost your course ratings -PS...but only if your class is good! -Fox (2006) experiment -List 2 ways in which the course could be improved (easy task!) OR list 10 ways (relatively more difficult...I hope!) -Then ask for an overall rating of the class -Which results in higher ratings? -People who had to list 10 ways gave overall higher ratings! •how bout soliciting more criticism in a course - this is another experiment done by Fox •he asked participants to list either 2 ways in which a course could be improved or 10 ways and then give an overall rating for this course - like how much do you like psych 120A with Rissman •people who list 10 gave it a higher rating •listing 10 the fact that it is hard to come up with 10 - it's the same reasoning to think about why this experiment turned out the way it did - if its hard to come up with 10 criticisms of the course 10 specific things that could be better then you might think as you go on like maybe getting to the first 5 or 6 wasn't too hard but as you are straining yourself - ok whats one more thing that could be different or change - and you start to realize well its hard to think of these maybe it was actually a good course whereas the people who listed 2 rated it lower

Family Resemblance

-Ludwig Wittgenstein proposed that members of a category have a family resemblance to each other -dark hair, glasses, a mustache, and a big nose are typical for this family but do not define the family -Wittgenstein proposed that instead of having classical rules we should think of categories as having this family resemblances and its illustrated her by a figure in your book with an actual set of faces that might be members of a family •But family resemblance can apply to things that aren't people like so any kind of category •Chairs can have a family resemblance structure but what he meant by this is that in trying to come up with a way of grouping these together because they all have a certain degree of similarity what you need to think about is holistically that there are certain features that they all tend to share but no individual feature is required so you would look at all these faces and think yeah these can be grouped in the same category there is certainly something that they all have in common but when you try to isolate any individual feature - like having glasses (that's not required) - having the dark vs. light hair (its not that they all have dark hair - two have light hair) •The nose shape are different - some have big noses some have small •Some have moustaches some have no moustaches •They all have beards of a certain shape - so maybe that's one of the features and the hairstyle is a similar shape but in this example there are certain features that are typical of them but don't define the family •So heres a quote from Wittgenstein " I can think of no better expression to characterize these similarities than family resemblances, for the various resemblances between members of a family: build, features, color of eyes, gait, temperament, etc overlap and criss-cross in the same way" •Heres an illustration of chairs - so no individual chair here meets all the criteria so this one might be considered a chair because you can sit on it but it doesn't have legs •This one doesn't have legs it has rockers •All chairs go on the floor accept this one is a hanging chair •Some of them swivel - some have backs some have no backs •So they all have general properties that are similar but no individual features necessary or sufficient to be considered a chair per-say -There might be many "characteristic features" but it is not necessary to have all of them to be a "Jones."

Morphemes

-Morpheme: smallest unit of meaning within a language -Morphemes can be divided into: root words and affixes (i.e., prefix, suffix) - "dog" single morpheme (dog) - "dogs" 2 morphemes (dog & -s [number]) - "studied" 2 morphemes (study & -ed [tense]) -"restudied" 3 morphemes - "related" only 2 morphemes •They can be divided into different components: so a word like noodles has a root word noodle and an affix - an affix is something that can be at the front of the word or at the end of the word •If its at the front of the word its called a prefix if its at the end its called a suffix •So dog is a single morpheme - its just a root word it has no affixes •Dogs is 2 morphemes so you are using the affix -s (which is technically a suffix - to indicate number/quantity - more than one) •So you have changed the meaning - now there are two pieces of meaning conveyed - dog is the general entity (the thing in the world you are trying to refer to and the s indicates an additional meaning that there is more than one) •Studied is 2 morphemes (the spelling can change - that's a separate issue we aren't focused on that - but study is the initial morpheme and then you are adding tense - whenever you change the tense that is an example of an affix and its usually done with a suffix) •Restudied here is changing the meaning - so this is 3 morphemes - you are adding re- to do again - so studied means the act of studying happened in the past and restudied is saying it happened again •Related is only 2 morphemes - the re- here (just because a word starts with re- doesn't mean it's a morpheme) the word relate doesn't mean to late again - relate is the root word and -ed pluralizes it

Phonemes

-Phonemes: the smallest unit of speech that can be used to distinguish an utterance from another (in a given language) -In English phonemes are made of a consonant or vowel n 44 in total -Different languages employ different sets of phonemes (and different types, such as 'clicks') -Children appear to be sensitive to any set of phonemes at birth •it's a minimal unit that distinguishes one thing I can speak from another •in English phonemes are usually made of a consonant or a vowel •so we have 44 in total •when you put a consonant and vowel together that's a syllable so if you want to speak a phoneme its hard to do it for the consonants without adding a vowel •ex. phoneme "p" is hard to say - I could say paw or pu so im always adding a vowel but the phoneme is "p" - so the consonant phonemes are hard to say without the vowel but it is something I can distinguish •but the difference between pa and p the difference isn't the vowel it's the consonant part of the phoneme •so there are consonant phonemes, vowel phonemes •44 in total in English - other languages have fewer and some languages have many more •different languages have different sets of phonemes - so English doesn't involve clicking sounds but that example of the Calihari bush language does involve clicking sounds and they convey meaning - and a particular kind of click is one of the phonemes that is important in that language •and children appear to be able to process and distinguish any potential set of phonemes at birth so you are not born restricted to only understand the phonemes of the english language or Japanese - or mandarin Chinese or Calahari bush people - you are able as an infant and young child to understand any set of phonemes and then as you get exposed to language your system begins to lose that receptiveness to any set of phonemes and you really only begin to distinguish phonemes in your language •so if you aren't exposed to certain languages or speech at a young age it becomes very hard later in life to be sensitive to those phonemes •for ex. English has a distinction between r and l which is very common and important for distinguishing many English words •but that distinction is not present for ex. in Japanese and it would become somewhat difficult if you weren't exposed to English at a young age to distinguish those later in life if your brain wasn't encountering them at this sensitive critical period -Phonemes are produced by modulating the flow of air from the lungs to the mouth and nose. -Phonemes can be classified according to specific features. •to produce a phoneme we modulate the way air is moving through our vocal apparatus - we are pushing air from our lungs up through our throat •and we have our vocal cords - our larynx (vocal folds) which is one way to restrict the air •and then the air is passing up our oral cavity and we can move our tongue which affects things •and we can move our lip •and we have the nasal cavity •and all of these allow us to modulate the way the air is being pushed through and change the way the sound comes out

Prosody aids sentence parsing

-Prosody refers to the patterns of pauses and pitch changes that characterize speech production. It is used to: -Emphasize elements of a sentence -Highlight the sentence's intended structure -Signal the difference between a question and an assertion •As we speak we are modulating which words we are emphasizing - some words we are saying more forcefully some times we sweep our voice up or down •This could signal the difference between a question and an assertion - so for ex. no, dogs are here vs. no dogs are here - by taking this pause which is a form of prosody you are saying no that's not true, dogs are here vs. no dogs are here like you aren't going to find dogs here •Heres another example im sorry (that's a question) vs. im sorry (which is a statement or apology) and you change the prosody the way these vocal sweeps •So this is the pitch and you can see with a question you sweep up and with a statement here you sweep down and as the last slide ill show you this valley girl speak •With valley girl speak what happens is that almost every statement ends with what sounds like a question •You see this up sweeps - which is called up talking and its becoming more and more dominant •Its not just so-cal valley girls - you see these up sweeps or up talking becoming a major dialect and maybe changing the way people communicate

Garden path sentences

-When we perceive a sentence, we must parse the sentence's syntactic structure -A garden-path sentence initially suggests one interpretation, which turns out to be wrong. -The horse raced past the barn fell. -Fat people eat accumulates. -The man whistling tunes pianos. •another way that we encounter syntactic ambiguity when we are reading sentences - some sentences can garden path us •if you imagine yourself walking along a garden path you are committing to one route and then as the sentence unfolds our interpretation of the phrase structure might be wrong •we don't know what is going to be spoken - we have to interpret the phrase structure as we go •the words are coming in and your brain is automatically mapping them in this unconscious way •you are doing that all automatically - the computations are behind the scenes and sometimes they are wrong - so when we perceive a sentence we must parse the sentences syntactic structure •a garden path sentence initially suggests an interpretation that turns out to be wrong •the horse raced past the barn fell - you are thinking oh the horse raced - thinking of the horse racing - and then past the barn - the horse is racing past the barn - why is fell there? •But here it's the horse fell - the horse that was raced past the barn by its rider fell •So there is an interpretation that is an alternative that is not apparent when you first parce it •So fat people eat accumulates - they are eating accumulates? That's not what is interpreted •You got garden pathed because you were thinking fat people eat - fat people are doing the eating - but its really the fat that people eat is accumulating in their bodies •The man whistling tunes pianos - that's another garden path sentences - the man whistling tunes? Whats he whistling he is whistling tunes - but really he is tuning pianos

What are concepts?

-a category is a set of things that are grouped together -a concept is the mental representation of a category •a category is a set of things that are grouped together by virtue of some shared attributes •there are different theories and ways of grouping things together - different types of categories but the category is the collective set of things that are grouped together •and the concept is the term we are using for the mental representation of the category •the category is a thing that could exist independent of the mental representation but the concept is really your mental representation of the category •concept is really the mental representation here •I say chicken and egg because the categories if they exist in the world - like living things for ex - is that something that exists first and then the mental representation allows us to model that or is it that we have a mental representation that certain kinds of things in the world have properties that we are going to call living and therefore we define the category and so this becomes almost a philosophical issue and we aren't going to delve into that

Classical theory

-a category is defined in terms of necessary and sufficient features -necessary: has to be there -sufficient: all that you need -these features define the category -this representation is abstract -it does not store any information about specific exemplars •Classical: in classical theory a category is defined in terms of its necessary and sufficient features •If you haven't encountered necessary and sufficient in philosophy ill just unpack it here •So necessary are features that have to be there - if they aren't there then you are not a member of that category •These are ingredients that are needed - you cant do without •And sufficient is all you need so if a sufficient feature is present that should be enough to ensure membership •So these necessary and sufficient features define the category - and according to classical theory the representation is abstract in that you don't have to store information about specific exemplars you just have to store a set of rules •Abstract meaning that membership of a category has to do with a set of rules and if those rules are satisfied then you are a member •Ex. is sunlight a necessary or sufficient condition for the roses to bloom? Necessary (not going to be sufficient - sunlight is not all that you need you would need all the right soil or water or temp or whatever other thing - having sunlight wont ensure that it will bloom other factors need to be met to - it needs to be healthy and so forth •classical theory is called classical theory because it goes back to the ancient greeks •these are the Aristotelian categories (i.e. classical theory) -defined by their necessary or sufficient conditions -so all members of a category are equally good - all of these triangles are triangles its not like this is a better triangle than this triangle •crisp boundary conditions - if you meet these two conditions - you have three lines (edges) and are a closed polygon than you are a triangle •if you aren't both of them - having three lines is necessary for being a triangle and being a closed polygon is necessary but neither one is sufficient you need both to be the case for it to be a triangle •so a square is defined by these features •it is a closed figure (necessary), four sides (necessary), sides equal in length (necessary), equal angles (necessary) •and none of those alone are sufficient here but all of those are necessary - concepts and categories are rule based and have definitions (necessary & sufficient conditions) - Issues: can't account for typicality effects -concepts and categories have definitions (necessary & sufficient conditions) -Categories can be defined by necessary and sufficient features, so members are either "in or out". •If an enclosed shape has 4 equal-length sides and four right angles it is a square. •If not, it is not a square.

Symbolic

-a referent is the actual object, action, or event in the world that a word refers to -Symbolic: just means for a given entity in the world like for this creature here (we refer in English to it as dog) •In other languages they have chosen a different way to refer to it •The written forms are here and each of these are an arbitrary symbol that is pointing to the same thing •When you communicate that or say it or write it you are trying to get the listener to think about this •So this is the thing in the world and this is the symbol that each language has chosen to represent that •So when we are talking about referent the referent is the actual object - so the dog is the referent and these are words that refer to the referent here -we even use different words to refer to the same thing within the English language (dialects) •Also its not just different languages spoken by different people in parts of the world but even within a given country many countries have dialects that have formed based on how people were raised within the country •With a lot of migration the dialects get weakened but still •This map is one example of how people refer to a soda •In the Midwest when these data were obtained most people if you put this picture up would call it a pop •Its not all Midwest but in general •And then in the south if you put this up many people will refer to it as coke - even when they know its not necessarily a coke - it could be a dr. pepper or something else •Dialects: the regional differences in how people refer to certain things - the words they use

Problems with both prototype and exemplar views

-both rely heavily on the idea of similarity -whats wrong with similarity? -take a blackbird, remove all of its feathers and clothe it in a suit of bat skin -is it more similar to a bat or a blackbird? -would you categorize it as a bat or a blackbird? -similarity is always relative -there are an infinite number of ways in which two things can be similar -we require some way of knowing what features are being compared •then there are problems with both views •both rely heavily on the idea of similarity and similarity is usually good for carving up the world into different categories - certain things are more similar than others and we can group them based on that but the book gives an example like this - take a blackbird... •so is it more similar to a bat or a blackbird? This is hard - it depends on what you mean by similar here •its more similar to a bat in how it looks because it is wearing a clothe suit of bat skin but more similar to blackbird due to its genetic makeup •so similarity is always relative •there are many ways that similarity can fail us when we are making category judgements - or if its not failing us there are different dimensions in which two things can be considered similar or dissimilar •and we need some way of knowing which features are being compared - so this brings us to the final framework theory based categorization • Cannot explain ad hoc categories - E.g. what to take with you if your house is on fire. • Both theories rely heavily on similarity - Always relative: depends on which dimension is being compared - Similarity requires some assumptions that aren't always met • Causality: - Completely alter the appearance of something, is it still itself? - How to account for this? • Both theories rely heavily on similarity - Always relative: depends on which dimension is being compared - Similarity requires some assumptions that aren't always met SIMILARITY IS NOT EVERYTHING.... • If you inject a lemon with sugar water, paint it red and white, and run it over with a truck, is it still a lemon? • If you print something that looks identical to a $100 bill in your basement, is it legitimate money?

How are categories organized?

-category: a set of items or concepts that are grouped together on the basis of something* -This is again a set of items (or even concepts - a category could contain multiple concepts - so that makes an extra layer of complexity here) that are grouped together on the basis of something •And something is an asterisk because it could be sort of cats vs. dogs (here we are grouping them together on certain attributes that we associate with cats and other attributes we associate with dogs) •This same illustration was confounded because it could be grouped on pictures that have a square or rectangular border vs. pictures on a white background - you wouldn't know with this grouping without more examples what this is based on •You could also group things based on surface features here - like color so white fur, greyish fur, orange fur and so forth •There are many different ways to carve up the world and divide it into categories you could do it based on whether the ears are pointed up and then it would include this one in ears up category - long tail vs short tail etc -Category: a set of things that are grouped together by virtue of some shared attributes - can exist OUTSIDE of mental representation - grouped together on the basis of "something"

The exemplar theory

-concepts are represented by all of the exemplars that have been experienced -the category bird is represented by memories of all previous experiences of birds -when we categorize something, we compare it one or more exemplars retrieved from memory, and decide the category based on the most similar exemplars -the category representation is concrete -there is not necessarily a summary of the category •exemplar theory is the other major similar based theory here that concepts are represented by all of the exemplars that have been experienced in your life •the category of bird for ex. is represented as all the different birds you have encountered and when you look at an animal and want to determine whether it's a bird there is this comparison process so we compare the individual instance to other encounters we have had •individual exemplars of that category retrieve them from memory and if it's a strong match we might categorize this as a bird •its more like other things we have encountered that are called birds than other things we have encountered that are called mammals, amphibeans, or insects •and the category representation is concrete in this case - there are actual things we have represented that we are comparing it to •theres not necessarily a summary of the category although there must be some way we make a category on the fly cause anyone can try to describe what a typical bird is or draw a typical bird and that's something that can be done as needed to take individual exemplar and blend them to create a prototype - no summary representation- a concept is a collection of individual instances - Issues: relies on similarity, can't account ad-hoc categories people seem to be able to have generalizations about what defines a category PROBABILISTIC THEORY - no summary representation- a concept is a collection of individual instances • Concrete • People don't produce summary/average prototype naturally. • Just store all the exemplars • Category judgments are made by comparing a new exemplar to ALL the old exemplars/most common exemplar • Statistical variance • You have remembered lots of specific instances of chairs you have encountered. • Categorization draws on knowledge of specific category members (exemplars). • To determine membership, ask, "Does this resemble an exemplar I have stored in my head?" EVIDENCE: TYPICALITY EFFECTS • Which of the following is a "better" bird? • Which of the following is more like a bird? • TRUE/FALSE: A ______ is a bird. • Name a bird.

Discrete infinity (generatively)

-john hates cheese -my roommate heard a rumor that john hates cheese -i told her that my roommate heard a rumor that john hates cheese -it disturbed mary when i told her that my roommate heard a rumor that john hates cheese -i was amazed that it disturbed mary when I told her that my roommate heard a rumor that john hates cheese -professor rissman spent way too much of his lecture talking about how i was amazed that it disturbed Mary when I told her that my roommate heard a rumor that John hates cheese -It really bothered me that Professor Rissman spent way too much of his lecture talking about how I was amazed that it disturbed Mary when I told her that my roommate heard a rumor that John hates cheese -This is discrete infinity - just an example of that -We have a finite set of words and we can just keep adding to it to change the meaning -The capacity here is limited by working memory - I could keep adding to this but eventually you'll lose track of what the subject of the sentence is, what the object is, where are the embedded clauses -Each one of these has multiple clauses embed -"it really bothered me" well now it is the subject and then bothered is the main verb and this is all the main stuff that bothered me -you can see you can just keep adding to it and create more complex ideas without needing to make up new words

Structure dependence

-language is governed by rules that impart meaning and define which combinations of elements are acceptable and which are not -John kissed Mary -Mary kissed John -Kissed John Mary -John Mary kissed •structure dependence language is also governed by rules that impart meaning and define which combinations of elements are acceptable and which are not •so john kissed mary (acceptable combination of these words) •kissed john mary (that in English is not acceptable) •the verb kissed requires the subject and object to surround it like this it doesn't work if you move it to the front or back of the sentence •this is an example from the book: what does it mean to know a word •it doesn't just mean to know what it means but to know the grammatical or syntactic implications or limitations of that word (knowing how to use a word) •the word place always needs an object - you cant say something like she can place on the table •it makes no sense to use the word place without an object explaining whats being placed •at minimum you would need to put a pronoun there like she could place it on the table •you need to be there to be this object for this verb •but in contrast the verb sleep doesn't take an object - it doesn't work if you try to give an object •"she can sleep the books on the table" you just don't usually use it in that way •sleep just refers to the subject who is sleeping •when you learn a new word your parents, school etc isn't teaching you that the word place is always going to need an object you just aquire this and as you develop fluency in a language you just sort of know intuitively or non-consciously what the limitations or the rules are for each of these words •it becomes part of your use of the word to go beyond the actual word but to know its structure dependence

Neurobiology of Cognitive Control: The Prefrontal Cortex (PFC)

-now most of the discussion of how cognitive controls are implemented in the brain focuses on the prefrontal cortex •its not the only part of the brain important for cognitive control: prefrontal cortex interacts with many other systems in the brain but we are going to focus on the frontal lobes •you can see this is an area that is greatly expanded in mammalian evolution - in other species cats and dogs they have a very small prefrontal cortex •in monkeys a little larger and in apes (chimps) they have a prefrontal cortex that is comparable to ours in regards to the proportion of their brain - ours is bigger in terms of absolute size and slightly bigger in terms of its size compared to our overall brain •but you can see in the human brain its not only going to be physically larger but more convoluted (convolutions refer to these folds - the cerebral cortex is like this sheet and its all folded up and you can see in other species like in monkeys there isn't a lot of folding - and here the intricate folding allows this sheet to have more neurons packed into a smaller space •everything in front of the motor cortex would be considered prefrontal

TheTowerofHanoitask

-other test to assess PFC function -patients with frontal lobe lesions are impaired at solving the problem -the frontal lobe is active when normal subjects solve the Tower of Hanoi •in this task the subjects have a number of blocks on pegs •you have an initial position and your goal is to achieve either this, this, or this •they see the initial position and they are told they need to achieve this (this one will take 5 moves) •these are just different goal states and they take a specific number of moves and the frontal lobe is very active while you scan people doing this task •the planning involved in this and mental simulation - involves strategizing, working memory so its very frontal dependent and patients with frontal lobe damage just have trouble representing the long term goal of what they are trying to achieve and then coming up with different subgoals to get there you need these intermediate steps

Dorsolateral PFC (DLPFC)

-regulation of non-social cognitive functions -executive control of working memory -another part of the brain - the rod passed through very crudely so in gage's case it also damaged parts of his dorsolateral cortex so he kind of got a double whammy in that sense which is involved in regulating non-social cognitive functions (particularly controlling working memory and in that sense keeping track of your moment to moment priorities - what is it that I need to do now - what information do I need to keep in mind - how am I going to use this information to accomplish my goals so the central executive part of working memory •as well as other functions - the dorsolateral PFC is involved in long term planning and evaluating pros and cons of different actions and outcomes and integrating information that is available to you from your senses to make efficient decisions •it's the cognitive decision making part of the prefrontal cortex

But maybe Whorf was onto something...

-speakers of different languages do talk about time differently -In English, we most commonly use words referring to front vs. back to talk about the future vs. the past: -We talk about the good times ahead of us and the bad times behind us ¤ We look forward to meeting a friend for lunch tomorrow -We fall behind schedule; We think about going back in time. -Speakers of Mandarin Chinese use up vs. down metaphors to talk about the temporal order of events: -Earlier events are said to be shling or 'up', and later events are xi`a or 'down' -Boroditsky (2001) examined whether speakers of English and speakers of Mandarin Chinese think about time in different ways. -English speakers were faster to correctly answer questions such as, "Does March come before April?" after seeing a row of items organized horizontally. -Mandarin Chinese speakers were faster at correctly answering the same questions after seeing a column of items organized vertically •But maybe whorf was still onto something with this idea •And it does seem that speakers of different languages talk about time differently •So in English for ex. we most commonly think of time as a front to back timeline - so the future is ahead of us and the past is behind us •So good times ahead of us - past is behind us •So this is kind of a metaphor for the passage of time that is determined by the way our language has taught us to think about this •We look forward to meeting •So the meeting that is coming up we would look forward to it •We fall behind schedule - so we think about going back in time •But speakers of mandarin Chinese use up and down metaphors to talk about the temporal order of events •So earlier events are said to be shling or 'up', and later events are xi`a or 'down' •Boroditsky examined this experimentally - she took speakers of English and speakers of mandarin Chinese and she found that English speakers were faster to correctly answer questions such as, "Does March come before April?" after seeing a row of items organized horizontally. •They have just a row of items arranged horizontally and that's supposed to be something that primes you - this is a priming study - so you look at the horizontal now you are thinking about horizontal organization - and then you get a simple yes/no question - does march come before april? And then you press a yes button or no button and English speakers were facilitated after seeing a row of items horizontally vs. a vertical column •Mandarin speakers showed an opposite affect so you prime them with vertical organization of objects it facilitated their judgements about time - like the order of months or dates for ex. •So this is just one example that language might influence the way you think about time even if it doesn't fundamentally constrain it •So Boroditsky and some of her collegeus would argue for this linguistic relativity perspective or this more mild interpretation

Cognitive Control: Overriding Competition from Automatically Retrieved Information

-the Stroop Effect -Stroop task a classic task to assess how well people can override a dominant or prepotent response •You've probably experienced this task before and what you noticed subjectively and what you would notice objectively if you were measuring reaction times (how quickly you can make a button press for which color it was or how long it took you to vocalize it) it's a very robust result •For anyone that knows how to read English these words - the printed words are going to be read almost automatically •You are going to read these words automatically and you have to override the fact that this says green to name the ink color - we name ink colors much less in our lives than we read words •So the reading is happening faster here - that is the prepotent response to want to say what you read and the PFC in conjunction with its interactions with other areas is going to be critical for overriding that and helping you to produce the correct response given the goals of this task which are to name the color not the word •So heres another schematic model - you can think of the stimuli here being a combination of colors and words and there are different parts of the brain that process the word information and different parts of the brain that process color information •These nodes represent activation of different processing steps related to the stimulus - so here you see the word blue and its in red color so the activation here in yellow - red ink gets activated and the word blue gets activated •So these are both activated - you are processing the color and you are reading the word •And your task is going to be to say red here so ultimately the right verbal response you want is red but this blue node here being activated is going to conflict with that so you have to override that •You can see this is the dominant response that's why it is highlighted in the bolder line (the word Blue) •And this is a nondominant response (red ink) and what you end up getting is more activation of blue than red so if you are not thinking and not engaging in cognitive control (doing this too impulsively or quickly) you might mistakenly say blue or at least be pretty slow to overcome that •With cognitive control you have again a context - in this case your context is the task goal - the context is that you are doing a task that some experimenter told you to do and Im telling you to say the ink color - so if you are representing that task goal and saying im not trying to read the word im trying to read the color - now this context can send signals down and here it is going to want to enhance the representation of this layer - this intermediate step from recognizing the colors to the action of speaking the colors you want to enhance what is happening here and you want to filter out to the extent you can - this response to sort of shrink that (words) •These circles here are going to shrink and these are going to grow and in the end you are trying to increase the probability that you are going to make the right choice here •And this sort of context control module is supposed to represent the PFC (what it is doing as an intermediary between the stimuli and the response) its allowing you to update your prepotent response in concordance with your current goals

Phineas Gage and Prefrontal Cortex

-the famous case of Phineas Gage (1848) -Rod goes through skull and lands 25 feet away -Gage is dazed but does not lose consciousness -Gage is taken 3/4 mile to Dr. John Harlow -Harlow shaves Gage's head, removes bone fragments -treats the wound with magnesium sulfate, calomel, and rhubarb (no antibiotics in 1848) •one of the most famous cases in neuropsychology as a person whos brain damage taught us a lot about cognition and the role of a certain area of the brain that we didn't know prior to this case •in the mid 1800's gage was a railroad worker and his task was to use this big rod and drive it into the ground to put explosives into holes - pack them down and they would blast away the rock to create the railways as they were building these railroads •due to an accident as he was tamping there was live explosives and it blew up the explosives before it was time to do so and this rod shot through his skull and landed 25 feet away •he survived it - didn't lose consciousness •he was taken to nearby doctor and they were able to stop bleeding •he survived - it didn't get infected and the area of his skull somewhat healed over •in his case he survived and we went on to learn a lot about the frontal lobe which was the area that was damaged •this is just a depiction of the rod next to his skull and you can see here how the skull was pierced by the rod which came out here right through the frontal part of the brain •this is with diffusion tensory imagery showing how the rod went through his skull •the DTI shows all the white matter pathways in the brain and those were going to be damaged and it would affect the ability not only of the frontal cortex to do its job but also for it to send signals to other parts of the brain •so this is the memorial plaque at the sight of the accident •but this wasn't just a lucky break for him it did affect his personality •this quote is from harlow who studied him "..." •he lost the essence of who he was in a way • he was no longer acting the way he used to - his personality had changed he was much more impulsive - he would go from one option to another •he wasn't able to keep long term goals in mind and make efficient plans for the future •he would act more instinctually and wasn't regulating his social and emotional behavior and that's largerly because the rod that passed through his head damaged this area of the prefrontal cortex that we now know is very important for regulating our social and emotional behavior in different contexts certain actions are appropriate and others are inappropriate

Syntactic ambiguity

-the girl looked at the boy with the telescope •now there are situations where the syntax is ambiguous - where the phrase structure is not readily implied in a way that everyone would agree on from the sentence •so if I say "the girl looked at the boy with the telescope" it could have two different meanigns if you think about it •one interpretation is like this: the girl here is the noun, looking is the verb, and the verb phrase here that includes the verb has this big prepositional phrase "at the boy with the telescope" so that's what the girl is looking at •so the boy has the telescope - the girl is looking at the boy with the telescope •other interpretion (others may parse the sentence like this): now the girl is looking at the boy and what she is using to look at the boy is the telescope •she is using the telescope and looking at the boy •and so how do you determine which one is being conveyed - well you need to use other information - you need to know does the girl have the telescope - does the boy have the telescope - why would she be looking at the boy with the telescope (was she really far away?) •all these things could help you resolve the ambiguity -more examples of syntactic ambiguity Complaints about NBA referees growing ugly -Dr. Ruth to talk about sex with newspaper editor -Teacher strikes idle kids -Juvenile court tries shooting defendant -Squad helps dog bite victim -NJ judge to rule on nude beach

Properties of language

1. Symbolic: make us of arbitrary relation between sounds and meaning 2. Discrete infinity: a finite set of elements can generate a (potentially) infinite set of 'meanings' 3. Structure dependence: meaning is conferred through a specific arrangement of symbols 4. Displacement: language allows referring to ideas/elements that are not "there" 5. Organized at multiple levels: Sounds, words, sentences, paragraphs, and text -Language fundamentally is symbolic - it is making use of arbitrarily chosen symbols and these symbols could be written (in the case of written language we have certain letters or characters that are going to represent different things we are trying to communicate - different words or ideas) then we have sounds (somewhat independently of the written symbols we have we have a decision of how we are going to speak them - what sounds different letters or characters will make and how to articulate them and understand them - and those sounds and the written symbols really have nothing to do with the meaning - in some cases we pick a word that has certain qualities phonetically that are like what we are trying to convey (the sound hsss is designed to match the sound of a snake) but in general the word that we use (the fact this is called a podium - there is nothing about it that demands that it would be called that - and in any other language it would have a different word so its arbitrary why that's the thing we have chosen to convey) •So its symbolic mapping between some written or spoken form and some meaning we are trying to convey •Another property is called discrete infinity: its both finite and infinite in the sense that there is a finite set of elements that we have (we have a lexicon like a mental dictionary of some number of words we know) with those words and some sort of grammatical rules you know about how to use those words and how to connect them you can create a seemingly infinite set of meanings and there is no limitations with what you can do with that fundamental building block of language •You have the words and grammar and you can put them together in what seems like an infinite amount of ways •It seems infinite potentially •Structure dependence: this is the idea that the meaning that we convey through language depends on the structure in which we organize the elements •Here im saying symbols but they can be written or spoken words so the order usually matters •Displacement: language allows you to convey or communicate ideas or talk about things that are not there or present in the current environment in the current moment so it allows you to teleport beyond your surroundings and talk about something else - talk about something from the past/future something that you want to invent - some idea you had that may have never been expressed before in the world and you can travel there - you can do that with language •And language has this multi-level structure - the inputs are usually going to be auditory sounds or visual words (could be tactile like brail) and then sentences are constructed of words and paragraphs of sentences and then text or discourse out of paragraphs

Example 4: Grammatical gender

English -Does not assign a grammatical gender to most words Spanish (as well as many other languages) -Marks gender with morphological info carried by pronouns, determiners, nouns, and adjectives -"una niña alta" vs. "un niño alto" -"luna" = feminine, "sol" = masculine •So in English we don't use grammatical gender - words are the words and we don't have words that are designated as masculine or feminine but other languages like Spanish and german have genders assigned to every noun •So you could take an example form Spanish una nina alta - so language can also confer the gender to the adjective that describes the noun so a tall girl or a tall boy - you can see these words take on a gender •But there are also nouns that are seemingly irrelevant to gender like moon - which Spanish is calling la luna and sol - like el sol = masculine so the language has decided at some point when these words were being created which words are going to be associated with masculine gender and feminine gender and then every object in the world has this gender assigned to it by the language Lera Boroditsky's 2003 Experiment: -Do people include gender in their conceptual representations of objects? -Are people's ideas about the genders of objects influenced by grammatical genders assigned in their native languages? Boroditsky et al. (2003) experiment: -Taught Spanish and German speakers object-name pairs (e.g., apple-Patrick) -Name was either consistent or inconsistent with the grammatical gender of the object in their native language -Measured memory for the pairs; all testing was in English -What were the results? -Both Spanish and German speakers remembered object- name pairs better when the gender of the proper name given to an object was consistent with the grammatical gender of the object name in their native language. •So Boroditsky asked how does this influence peoples perceptions of objects •Is there language changing the way they are thinking of suns and moons or anything else in the world •Are peoples ideas about the genders influenced by this very arbitrary assignment that their language has done for them of taking some things and calling them feminine •So they taught in this experiment spanish and german speakers object-name pairs in English •The experiment was done in English but the speakers were bilingual they were either Spanish/English speaking or german/English speaking •And the name was either consistent or inconsistent with the gender of the object •So for the apple-patrick example - this is consistent in German cause the word for apple is a masculine term and Patrick is a boys name •This would be a consistent pairing in Spanish this would be an inconsistent pairing •Then they looked at just peoples memories for these pairs so which name went with apple? Which name went with whatever other words they had •And what were the results •German speakers remember this apple-patrick name better and Spanish speakers would remember like apple-Susan better - a female name would be better for them •So theres this dissociation that when the pairs were consistent gender it facilitated their memory

The wide range of availability effects

Schwarz et al. (1991) -Asked subjects to recall examples from their lives in which they had acted in an assertive fashion ¤ Half of the subjects asked to recall 6 examples ¤ Half of the subjects asked to recall 12 examples -Then participants were asked some more general questions, including how assertive they thought they were -Which subject group rated themselves as more assertive? A) The group that had been asked to generate 6 examples B) The group that had been asked to generate 12 examples C) No difference •heres some other experiments that have been done to show how availability influences our judgement in different kinds of contexts then that •in this experiment by shwartz - they asked participants to recall examples in their own lives in which they had acted in an assertive fashion •the participants were told to recall either 6 examples or another group of participants were asked to recall 12 examples •so if you were in the study you would have a sheet of paper and tell the experimenter different examples of things you've done that were assertive •then participants were asked more general questions including how assertive they thought they were just as a personality trait •when asked to generate 12 its hard to come up with 12 •the challenge of coming up with 12 leads you to think well I must not be that assertive because if I was surely I could come up with 12 examples of it •so it's a small effect - the average assertiveness rating on this 10 point rating scale was influenced by about 1 point by this manipulation so many psychological effects like this - this kind of social and behavioral psychology phenomenon do tend to be small - its not like it fundamentally changed your appraisal of yourself - but it did influence it significantly •so people who recalled 6 examples judged themselves as more assertive like a 6.3 then people who recalled 12 •and that's due to the difficulty of coming up with more examples leads them to underestimate its quality in this case as a personality trait

Another Boroditsky experiment

Spanish and German speakers asked to write down the first 3 adjectives that came to mind to describe various objects -More masculine properties produced for masculine objects (defined by native language), and more feminine properties produced for feminine objects Example: key -Masculine in German ("Schlüssel") feminine in Spanish ("la llave") -German: hard, heavy, jagged, metal, serrated --Spanish: golden, intricate, little, lovely, shiny -Grammatical gender focuses speakers of different languages on different aspects of objects •So there was another experiment: they used again Spanish and german bilinguals •Everything was done in English •So one thing that's cool about this experiment is that it works even though they aren't asked to think about the translations - they aren't asked to think about oh what word is apple in your other language •The idea is that Spanish speakers just think of apples as more feminine and germans speakers think of apples as more masculine and that makes it easier for german speakers to associate it with Patrick and harder for Spanish speakers to associate it with Patrick •In this experiment they just asked what are the first 3 adjectives that come to mind •So they gave subjects a pic of key then they just had to come up with adjectives for key everything was done in English •But some of these objects like key are feminine in Spanish and in german it is masculine •And so in german the words they came up with were hard... •And in Spanish they referred to it as golden... •They did this with a lot of words and then they had people rate the masculinity and feminity of these terms and they found that grammatical genders... •So even though this was done in English you were just given an object and asked to talk about its attributes - Borotsky argued that the language that you learned and its very arbitrary assignment of grammatical gender to different words influences the way you perceive them •Like is a bridge masculine or feminine? And will that affect the adjectives you use for bridge

Judgment

Types of Judgments: -what is the relationship between X and Y? -which is further west: Florida or Chile? -how often does an event occur? -how often does UCLA beat USC in football -accuracy is important...but uncertain -how often will you go to the hospital next year? -what insurance plan should you choose? •There are various types of judgements •2 example types: one judgement we make all the time is just understanding how two things in the world relate to each other as an example and its not an example of an everyday judgement but one to think about for now is which is further west: florida or chile? •Many think chile is further west and that is a typical answer and the logic that those who picked that would have with this intuition is that well chile is on the west coast of south America and florida is all the way on the east coast of north America and north America and south America are roughly above each other so that makes sense - it's a useful mental strategy for thinking about this but in reality its actually that florida is further west •You just don't necessarily think about this and it just illustrates that when we make judgements we need some way of reasoning - the heuristic you used here was thinking about where each of these two places was in relation to the broader country or continent •How often does an event occur? How often does UCLA beat USC in football •If you have to guess and you have little information you would use some sort of strategy you need to come up with some estimation - so you would think what happened this year? Didn't win this year - what happened last year? And would have to come up with that and so forth and would begin to come up with your estimation •In reality UCLA won 35% and tied 9% •So these are just examples of kinds of judgements •In judgements - maybe not these but in many judgements accuracy is important •So which health insurance plan to chose? Maybe you don't have a lot of options to chose now but as you go out into the work world and move away from the student insurance plan there are so many different options and the options generally vary in terms of the amount you are paying each month vs. the deductible and other cover things •It's the same for car and life insurance you are always making gambles about how much you want to spend for things that are unlikely to happen - like having a major injury or having a $100,000 in medical treatment for some ailment that you come down with and these things for most healthy people seem like remote possibilities but if they do they are going to be costly and you are always trying to weigh whats the chance im going to crash my car and so forth - how often do I get sick •And its impossible to have all the information - all that we can know is estimate some set of probabilities and try to weigh the various options and because of the uncertainty in many of the decisions we make we have to use certain rules of thumb - you all decided where to go to college - you decided to come here - that might have been an easy decision for some for others you might have agonized about it and other options had certain appeal and others had other appeal •These are judgements that we have to make and we really don't know the future so we can only make a best guess for many things

Where do categories come from?

the rationalist answer: -we are born with them (plato, kant) the empiricist answer: -we learn them (locke, hume) a middle road -we learn categories, but the kinds of categories that we learn are constrained •This is a little bit of the philosophical background although we aren't going to go sort of as deep into the ancient greeks as we sometimes do •Plato and much later emanial kant thought that categories were something we were born with and that they exist in this pure form and they are sort of there in a way that we don't have to learn •Whereas an empiricist would say we do have to learn them - we have to aquire them •Around the time of Kant - at a much later time than plato we are talking about Locke and hume saying that categories are something we learn through experience •And a middle road here is that its kind of a combination - that there are certain kinds of categorires that we are innately predisposed to learn certain things in the world that all people regardless of which culture they grow up would categorize one in a similar way - and that reason for that is maybe our brains evolutionarily is designed to categorize certain things - things we can eat - things that are going to be safe - things that are dangerous - things that move - things that are still and so forth •So we do learn things through experience but maybe the wiring of the brain over millions of years is designed to learn these things and say orient to a face because that is an important category to learn •But many of the things we categorize in the world are things that we learn •Computers, cell phones things like that obviously we aren't born knowing what those are -Item set that is grouped by some dimension/feature • How is this ability acquired? - Plato/Kant: Born (rationalist) / Locke/Hume: learnt (empiricist) / Middle Road: both

So...which is correct?

• Both! • Prototypes are very economical but more general (can lose some information) • Exemplars are less economical but might allow more information. • Exemplars don't account as well for ability to extract general properties of categories. • Prototypes don't convey info about category variability.

Broca's Aphasia

•Broca's area in the front of the brain - the left inferior frontal lobe as shown here in the brain post-mortum in a famous case that Paul Broca had come upon •Is an impairment in speech production - the speech becomes non-fluent the output is affected but the comprehension seems to be intact •So in this patient he had a really extreme case following a stroke that damaged this area of the left inferior frontal cortex and could only say the one word "tan" •Everytime he tried to speak that was the only thing he could muster which shows that this area of the brain is critical for producing articulate speech

Anterior Cingulate Cortex and PFC

•Im going to add the ACC to this model - we already showed this model where the dominant response here is to read the word and say blue which is not what you want to do - how do you overcome that •Well according to this model the ACC is going to detect the conflict - saying the response I want to make is inconsistent with the one im being told to make - there is multiple outcomes that are activated here simultaneously - ACC is saying conflict we need to do something and then the PFC kicks in even more •So you can proactively regulate conflict - you can say im doing this task im going to be really careful im going to be slow im not going to say my answer till I know that it is right and in that case you might not need to do so much of this conflict monitoring - but in most cases you are just doing the task and something is telling you oh there is conflict here and on the trials where there is conflict the ACC is telling the PFC you need to engage in control - this is more of a reactive form of control - where you are reacting to the conflict and saying ok for the next trial im going to slow it down im going to increase this is kind of top down modulation - so PFC is sending signals down here to areas involved in the perceptual processing and decision making in order to create the right response

Phonemes: segmentation errors

•Now this is a problem that we usually solve pretty well - its not usually a problem to understand what someone is saying to you but there are situations where it could fail and those situations are interesting in the same way visual illusions are interesting for showing limitations of visual processing •So this sentence: "this guy is falling, this guy is falling" •It is completely ambiguous with the alternative interpretation of "the sky is falling" •This is a good example of something where the only way to tell whether you are saying the sky is falling or this guy is falling is either based on your prosity (how you modulate the delivery of the words - so if I say this guy is falling vs. the sky is falling and enunciate the specific words I could maybe make it a little easier - but in general you could say these two sentences and have exactly the same spectral wave form that would be indistinguishable and so the only way you would know what you are conveying is with the context (whats called the extra-linguistic context) •There are other things going on that might be relevant - for ex. chicken little is looking up at the sky and saying the sky is falling but if there was a guy in the scene who was falling you might take the other interpretation •This is an example of more real life-ish example of a phoneme segmentation error that was very commonly experienced a few years ago •Taylor swift: blank space •What does it sound like she is saying? Starbucks lovers •The actual lyric- gotta long list of ex lovers •So its kind of like the sky is falling vs. the guy is falling - all the lonely starbucks lovers •So this was just a common segmentation error •This actually happens in a lot of songs •Just a common segmentation error •Where you had a lyric wrong in your head for a long time and realized it was something else and its just about segmenting the phonemes - its hard when its in the context of a song

fMRI evidence for a functional dissociation between ACC and PFC

•So this is an experiment to show evidence with fMRI for this model •What they did was stroop type task •Where subjects in scanner looked at words that were either consistent or inconsistent with ink color - so you have congruent trials and incongruent trials and they are showing activation during two different phases of the task •The first phase is the instruction phase which is telling you what your task is - sometimes the task was to read the word which is easy because it's the dominant response and sometimes the task was to say the ink color which is the classic stroop experiment •The first set of graphs youll see are during the instruction phase of the task and you see the BOLD signal relative to baseline over time • These 1,2,3,4,5 are representing about 15 seconds of activity and you can see the ACC doesn't care whether the task is the color task or the word task it shows the same low level of activity for both •but the lateral PFC does care when you are told to do the color task which is the hard task (naming the ink color) it shows this ramp up of activity in the instruction phase but when you are told to do the word task it doesn't •its preparing you to engage in cognitive control •now when you get the stimulus and the word actually appears on the screen you see this dissociation now the ACC shows a response that's different depending on whether the word is congruent or incongruent •so if the word is incongruent to ink color this region of the brain shows greater response - its detecting conflict whereas the lateral PFC shows an equivalent response so its involved in preparing yourself for what your current context is and what your goals are in that context but once the stimulus actually comes you've already done the regulation you need in PFC so it doesn't show much of a difference whereas ACC is involved in detecting the conflict and in fact this research team and others in the years that follow showed that the degree to which the ACC engages or activates on trial 1 will affect the degree of pre frontal control that's allocated on trial number 2 •so this conflict detection will affect the next trial and youll get more of this pro-active ramping up of the prefrontal cortex

Categorical Hierarchies

•So this is just showing that categorization generally has hierarchical properties that there is these levels of categories where one higher level of category includes many sub categorys -the category of animals includes many different types of animals and each of these types (dogs) will have different breeds or species •And you can divide these up even more (ex. with animals you could have mammals and mammals could include dogs and so forth) •So this is not trying to show you the scientific taxonomy of living things in a complete way but just showing you that certain categories include sub-categories •Living things could also include other categories - so plants is another one •You could make that diagram from there and you could draw it in this hierarchical representation or you could take the same thing and represent it in a venn diagram •So these are just different ways of showing which categories are subsets of others •So the category of animals includes multiple subsets and within animals is dogs and within dogs are different kinds of dogs and within different kinds of dogs (beagles) could be different (a pet beagle, or snoopy)

Making Judgments: how heuristics guide (and misguide) our evaluations of the world

•We are talking about judgements - particularly we are going to focus on how heuristics guide our judgements and help us in many ways to make efficient and fast judgements about all kinds of things in the world •But cognitive psychologists have explored and found interesting quirks in our judgement and decision making that we will talk about today and Thursday that showcase limitations of this •That the heuristics or mental shortcuts that we use sometimes lead us astray

Wernicke's aphasia

•Wernicke's aphasia is caused by damage generally speaking to the posterior temporal lobe sometimes extending into the inferior parietal lobe here •Temporal - parietal junction where they meet - so wernickes area is sometimes considered this broader region here •And Wernicke here discovered this patient with damage that turned out to be in this area of the brain and in each of these cases neurologists didn't know where the damage was till the patient died and they looked at the brain post-mortum •But this case was an impairment of speech comprehension and spared speech production •So together with Broca's aphasia and Wernicke's apashia it's a double dissociation where Broca's aphasia is associated to damage to the frontal lobe and impairs speech production but not comprehension and Wernicke's apashia is the opposite •Its damage to the posterior temporal lobe and impairs comprehension but spares production - generally speaking •These aren't pure dissociation - Broca's patients have some trouble with certain kinds of comprehension tasks and Wernicke's apashia patients have some trouble with aspects of production •But more or less it shows that these different regions of the left hemisphere play different roles in language

Assessing color perception with Mansell color chips

•You could do this in a more intricate or experimentally rigorous way by having a full range of color chips so you could have squares like this that span the full spectrum of color and just ask people to name colors and when you do that with English speakers you generally find this •Now everyone would have slightly different boundaries: but in general across many English speakers this is what you find •So if I ask you to label this most people would say pink if I ask you to label this most people will say red - some of you might call it purple it might extend a little bit in here but for the most part this range is purple, this range is blue, a large range of green here, yellows oranges and so forth •So this is the English naming scheme for these colors •Now other people who speak different languages for ex. the Berinmo tribe in New Guinea they divide the colors differently •So their language has fewer color terms - they have a term for dark and a term for light and they have a term for red-ish things •One term of nol which encompasses blues and greens and then this wor term which is more yellows •And you can see its going to be a very different mapping that they don't have this term purple and so they tend to call these things dark up to a certain point and then they end up calling them wap which translates to light •So they don't use color terms for things that are brighter •An interesting experiment that was done comparing English speakers to Berinmo speakers is showing them colors that are in this range - and whats interesting is the Berinmo have one word for everything in here they call it nol •But in English we have this boundary - some things in here we call green and at some point we start calling them blue •So we have two different terms for this range whereas they have one term •So this was an experiment that was done -do berinmo speakers perceive color differently? •The berinmos perceive colors differently and the hypothesis is that if categorical effects are restricted to linguistic boundaries, these groups should show different responses across the two category boundaries (green-blue and nol-wor) •English speakers should show sensitivity when the color chips straddle our boundary between green and blue berinmo speakers should show affects when the color chip straddles the boundary between nol and wor and those boundaries are different •On the other hand if the categorical effects are determined by the universal properties of the visual system, then both populations should show the same response patterns

Typicality and attractiveness

•and this can also affect attractiveness - things that are more typical tend to be more attractive to us •and with faces this has been studied - if you look at the contestants of miss Germany you can see if you average the faces (on the left is real miss Germany and this is virtual miss Germany who was created by blending all the contestants of the other regions) so if you just average the faces together and in most cases things that average across exemplars that show the most typical features people will view as more attractive •these are averages of faces that were taken at each of the rating categories - a 1 to 10 scale •these were the average of faces that got rated between 9.5-10 etc •even at these lower levels of hotness ratings when you average together the faces they are all pretty attractive •so even if the individual exemplars were all getting low ratings the averages are very attractive and the average of all faces •this is cross cultural - so if you take faces in any region or ethnicity and average them you can see the average face carries what most would consider high attractivess •not the only thing for attractiveness - averageness isn't the whole story - other factors are important too (ex. distance from pupil to pupil, distance from midpoint of eyes to midpoint of mouth)

Prefrontal cortex development (homosapiens, modern humans)

•even in the course of homonin evolution as we got toward anatomically modern humans or homo sapiens there was the development of the prefrontal cortex - it is one of the main things that has changed about our brains when you look at the skulls of other species •you can see that the slope of the forehead changed and that is due to the expansion of this part of the brain •that is very correlated with our growth and intelligence - the ability to breakfree of stimulus response type behaviors where things happen in the environment and you react to them •there is prey available and you are hungry so you pursue it - there is something chasing you and you run away •many animals work in more of a stimulus response fashion and there is some degree of planning and decision making and evaluating pros and cons of other options •thinking about your future - you see this in other species •they have a certain degree of cognitive control but certainly in humans especially the homo sapiens that we are today this expansion of the prefrontal cortex was one thing that really allowed us to advance to the level of greatness we have achieved •you also can see this also in the course of brain development within modern humans if you look at a 5 year old brain and compare it to a 20 year old brain and just color code different areas of cortex depending on their relative maturity •in this figure - you see areas in the deeper blue and purple colors are the most mature they've achieved their adult level maturity and areas in the warmer colors (reds and oranges) are less mature •and you can see as the brain matures one of the last regions of the brain to mature is the frontal lobes •at this stage you are at the 20 year old brain •and it actually continues to mature into the early 20s so even though we think of adolescence as the period where the brain is achieveing that maturity and then once you get to 18 year olds you have a mature adult brain but it actually continues a bit beyond and the prefrontal cortex is still developing - one of the last areas not necessarily the very last but one of the last

Another oppenheimer example of availability heuristic

•heres just one more example from work Oppenheiimer did •he showed people an image like this and asked how much could you buy with this dollar bill and they had different choices like how many sheets of paper could you buy •how many paper clips, gumballs, thumbtacks things like that •and so everyone knows what a dollar is people don't necessarily know the prices of paper clips but they can estimate that and so one group saw this dollar bill another group did the same thing but saw this dollar bill •they look basically the same except the difference is George washingtons face is flipped •so this shouldn't influence you •but seeing this even if no one realizes this is an altered dollar - it looks a little bit off and so this could unconsciously influence people •and so he did three different experiments •1 was the example I showed you with real dollars and then altered dollars •1 was with 2 one dollar bills vs. a 2 dollar bill - everyone knows a 2 dollar bill is worth 2 bucks but they are not as common so would people estimate they could buy more with 2 one dollar bills than a 2 dollar bill? •And then also the susan B Anthony dollar coin which is rare vs. the 1 dollar bill •And then combining across all three experiments and in almost every case or every case you see an advantage •So the number of paper clips that you could buy per dollar people estimated over 80 when it was a familiar dollar or the dollar bill vs. the dollar coin and in the unfamiliar they estimated half - even though they didn't realize that this was influencing them •So that's just another example of how this plays out

Levels of language representation

•language is represented at multiple levels - these are the levels we will be referring to on this slide: •the basic level of language that words are composed of are phonemes these are the minimal sound elements of a language •phonemes are the constiguent elements of morphemes which are the constiguent elements of words •a morpheme is the minimal level of meaning in a language - so the word chef has one morpheme of chef •but the word burned has two morphemes (burn, and this modification - a suffix -ed that indicates past tense - so there are two forms of meaning in this word you have burn which is the lexical elements of the primary thing you are trying to convey and then you have a modifier) •and then from the morpheme level you have words and from the words can be organized into syntactically correct sentences that can be parsed in terms of what the subject is and the object and the nouns and verbs •and then above that is this discourse level - just the idea that you can take sentences and put them in a broader discourse of conveying ideas beyond this one sentence (it can be part of a paragraph or a speech or something)

Prototypes vs. exemplars

•prototype theory views categorization as taking all these instances we have encountered and making some mental prototype •just a representation of the average and we have a prototype in our mind for what a fruit is, cat, vehicle and so forth •and through all of our experiences we just generalize these prototypes •and we make category judgements by taking an instance of something we encountered - what is this? - and compare it to a prototype and it matches a prototype of water bottle pretty well so you might say water bottle -category judgements are made by comparing a new exemplar to the prototype •exemplar model is similar in a way that you encounter all these instances but instead of averaging them mentally and storing this abstract representation of a prototype what you are going to do is having memories of all these encounters - individual members that are stored in your brain and these are all labeled through your experiences (this is a dog) •for a child learning world - child might not know which is which - they may call all living things dogs until they realize oh that's a cat, that's a horse and so forth •but once you get these instances and have labels for them then in your mind you encounter a new thing and you quickly have a mechanism for comparing it to all the examplars •and the exemplar model can even posit that people generate a prototype on the fly •you see in this illustration here - derive a prototype if needed •so even though you are making judgements by storing individual exemplars if you were asked to draw a prototypical draw - people can do it - people can come up with a description of what an average or prototypical vehicle or computer or anything would be -category judgements are made by comparing a new exemplar to all the old exemplars of a category or to the exemplar that most readily comes to mind