Psych 120A Midterm

Analog Theory in a nutshell

-Kosslyn -image is a picture -if you imagine a triangle where the lines are bisected you would imagine something like this

Types of illusions

-ambiguities -distortions -paradoxes -fictions

Capgras Syndrome

-different parts of our brain work together and they are all integrated together so -what happens if we lose that integration? -what happens is capgras syndrome -can recognize a loved one's face, but no feeling of familiarity -cognitive appraisal intact (the fact that they can still recognize their loved ones shows that cognitive appraisal is intact - they cognitively have the knowledge that they know who the person is), emotional appraisal disrupted (discordance between both) -neural evidence: damage to temporal lobe, which disrupts circuits in amygdala (emotional evaluator) -also very correlated with schizophrenia and in schizophrenia you have a lot of brain abnormalities in prefrontal cortex (located in frontal lobe) -active during planning or careful analysis -requires multiple brain areas (this happens because there is a dissociation between these brain areas - these functions are happening at once so when there is damage to one you can see how our interactions would be abnormal)

The lobes of the brain

-frontal lobe (form the front of the brain- right behind the forehead) -parietal lobe (the brains topmost part) -temporal lobe (below the bottom edge of the frontal lobes) -occipital lobe (very back of the brain - connected to the parietal and temporal lobes)

other techniques for planting false memories

-imagination -dream interpretation -hypnosis -exposure to other peoples memories -false information -doctored photographs

The Face Inversion Effect (a.k.a. The Thatcher Effect - distortions)

-the face inversion effect illustrate how much of a difference it makes if you process faces upside down vs. right side up -this one looks grotesque because eyes are flipped and mouth is flipped but you just don't notice that for an inverted face -the face inversion effect illustrates that our visual processing mechanisms for integrating facial features only work properly for upright faces -this could be something genetically engrained or something having to do with experience probably largely due to our experience we have a genetically innate bias for faces

Double dissociation in support of the two-stream hypothesis

Double dissociations in neuropsychology: -lesions in brain structure A impairs function X but not Y, and a lesion to brain structure B impairs function Y but spares function X -strong evidence for functional independence -allows us to dissociate ventral and dorsal streams -lesion to one part of the brain imparts function X, other part impairs function Y -therefore, relate function and structure Optic ataxia: -impairment of visually guided reaching despite normal object recognition -lesions of parietal lobe (issue in dorsal stream) -What about the flip side? -impaired object recognition/spared reaching ability Visual agnosia -patient D.F. had diffuse damage throughout the ventral stream as a result of carbon monoxide poisoning -could not judge even basic aspects of the form or shape of objects -could not even describe a line as vertical, horizontal, or tilted (Normal visual acuity, normal ability to name objects - when placed in her hand) -patients with lesions in the what system show visual agnosia - an inability to recognize visually presented objects, including such common things as a cup or a pencil -However, these patients show little disorder in recognizing visual orientation or in reaching -There is evidence from patients with brain damage about this -Lots of different ways we know about this dorsal and ventral streams we can record from neurons what they are sensitive to in terms of object information and spatial information but we actually learned a lot from patients with damage to parts of their visual processing pathway -This is from branch of field of neuropsychology - studying cognitive processes on patients who have brain damage in certain regions -Double dissociation is a term where you are able to infer that a lesion to one part of a brain - brain structure A is going to impair function X but not Y and a lesion to brain structure B is going to impair function Y but spare function X -Its hard to know these are independent systems -But when we find examples of patients who have damage to one region that impairs one function but does not impair another function and we find other patients who have the inverse damage to a different region that spares the first function and impairs the second function this gives us strong evidence for functional independence -Optic ataxia just one example of impairment that can occur to the dorsal stream -Impairments of visually guided reaching despite normal object recognition -This generally occurs when you have lesions to the parietal lobe -Optic ataxia makes it difficult for patients to reach for object - they are not getting the right information about where it is in space and trouble coordinating movement to the object - because their spatial processing in the dorsal stream has been disrupted through this lesion -What makes this a double association is that there is a different class of patients that have been found -Perfectly in tact dorsal streams that have damage that affected their ventral stream and they have the opposite pattern -Their object recognition is severely impaired but reaching ability is fine -Example of this is visual agnosia - visual recognition deficit -Patient D.F. had diffuse damage throughout the ventral stream as a result of carbon monoxide poisoning -This was the part of the brain for her that got damaged -For her she couldn't even judge basic aspects of the form or shape of objects -When you asked her can you describe a line as vertical, horizontal or tileted they couldn't do it -The patient had normal visual acuity - you can measure this in different ways - like are there one dot or two dots and whether they are close together -The patient could name objects - didn't lose all knowledge of objects but had to use other senses - she knew what object was called when she touched it but couldn't recognize it in the visual world

Introspection vs. Behaviorism

Introspection: concerned with beliefs, wishes, goals and expectations -can NOT be directly observed or objectively recorded -viewed NOT as a scientific tool -that lead to the beginning of behaviorism Behaviorism: Concerned with objective behavior and stimuli (to them what was objective and a good index of whats happening in the brain was behavior) -CAN be directly observed or objectively recorded -Viewed as a scientific tool

Ambiguities

-Ambiguous illusions are pictures or objects that generally present the viewer with a mental choice of two interpretations, each of which is valid. Often, the viewer sees only one of them, and only realizes the second, valid, interpretation after some time or prompting. A retinal image - like any picture - is infinitely ambiguous. It could correspond to, or represent, an infinity of shapes and sizes and distances. -Thus, an ellipse in a drawing could represent a tilted circle, which might be small and near, or distant and large. The possibilities are endless; yet amazingly, we generally see just one of the infinite possibilities. -How the visual brain usually selects a single possibility is not fully understood. -The term 'ambiguity' is itself ambiguous. It can mean several possibilities though but one or a few are seen or chosen. -Alternatively, it can mean spontaneous changes of perception, as occurs with the images in Figures 1 and 2. The first is passive confusion of different stimuli or patterns; the second is active generation of alternative perceptions from the same stimulus or pattern. -Ambiguities are spontaneous changes between visual hypotheses, though there is no change in the object. -The primary ambiguity is between something and nothing: objects, and spaces between objects (Figure 1). Objects can also 'flip' in orientation; a well-known example is the Necker cube (Figure 2). -its ambiguous because it can take on two different interpretations - some of you may see it like this where this is the front face of the cube and this is the back corner - so this part is further away then this surface here -and then you can also see it where this is the front surface and this is the far corner and these corners are all farther back -what makes this bi-stable is you can kind of toggle in your mind between the two interpretations - sometimes you just have a strong interpretation so your high level processing is sort of saying it's a cube and maybe this is the front surface here and then your interpretation of all the other features is constrained by that and its hard to break free -this one can be interpreted in multiple ways -these are ambiguities and they can often flip back and forth in your mind and you kind of have to commit to one interpretation or another you cant see both at the same time A well-known example of change of object is the old woman/young woman image in Figure 3. -Dynamic ambiguities are useful for research into how perception works, as perceptions change though the objects remains unchanged. -This shows that perceptions are not directly linked to objects. More technically, they allow us to separate 'bottom up' signals from the eyes, from 'top down' knowledge of objects, stored in the brain. -Much of what is seen is psychologically projected into the world, especially colors, which are created in the brain although they appear to lie on the surface of external objects. -Anatomically there are more descending nerve fibers from the cortex than ascending nerve fibers from the eyes. -Vision therefore seems to be more 'top down' than 'bottom up': it depends more on knowledge from the past than on signals from the present. So, although not all of perception is in the mind, most of it is! -The most dramatic demonstration that knowledge is important for perception is the 'hollow face' (Figure 5). Although hollow, it appears to be a normal face with the nose sticking out - simply because a hollow face is too unlikely to be seen. -This is strong evidence of the power of 'top down' knowledge to challenge and beat 'bottom up' signals from the eyes

Event retrieval: reconstructing past neural patterns (hippocampus)

-And then later on lets say I go back to reno - I see the sign and think ohh when was I hear last - so maybe the sign triggers some visual memories - and this is just one slice (say the word reno) is feeding into the hippocampus and what the hippocampus does is when you activate a node that was part of the original experience it has this ability to reactive the other nodes in the hippocampus ( other neuron ensembles coding for different associate elements of your experience) might come back and so these are reactivated and when that happens these nodes being active can send this signals back to the cortex and this is the reason that when you revisit a past experience (what did I do last Halloween?) the details these perceptual qualities of where you were and how you were dressed etc. all are going to be reconstructed and it can be a vivid memory for some people but what you are doing is reactivating the past memory -Its through a process called pattern completion -So one node is activated - something reminds you of a past experience and then the hippocampus completes the pattern - so pattern completion

Previous ideas on how good/precise memory is

-DVD analogy: encoding: record on an electronic disk maintenance: store the DVD in a drawer retrieval: play the disk back on a DVD player -most people believe that memory records the events of our life much like a video camera

The Design of a Feature Net

-How might out CLOCK detector work? -one option is to wire this detector to a C-detector, an L-detector, an O-detector, and so on -then, whenever these letter detectors are activated, this would activate the word detector -but what activates the letter detectors? Perhaps, the L-detector is "wired" to a horizontal-line detector, and also a vertical-line detector, and maybe also a corner detector -when all of these feature detectors are activated as a group, this activates the letter detector -The idea, then is that there could be a network of detectors, organized in layers -the bottom layer is concerned with features, and that is why networks of this sort are often referred to as feature nets -as we move upward in the network, each subsequent layer is concerned with larger-scale objects; and using the term we introduced earlier, the flow of information would be bottom-up - from the lower levels toward the upper levels -but what does it mean to activate a detector? at any point in time, each detector in the network has a particular activation level, which reflects the status of the detector at just that moment- roughly, how energized the detector is -when a detector receives some input, its activation level increases -a strong input will increase the activation level by a lot, and so will a series of weaker inputs -in either case, the activation level will eventually reach the detector's response threshold, and at that point the detector will fire - that is, send its signal to the other detectors to which it is connected -within the net, some detectors will be easier to activate than others - that is, some detectors will require a strong input to make them fire, while others will fire even with a weak input -this difference is created in part by how activated each detector is to begin with -if the detector is moderately activated at the start, then only a little input is needed to raise the activation level to threshold, and so it will be easy to make this detector fire -if a detector is not at all activated at the start, then a strong input is needed to bring the detector to threshold, and so it will be more difficult to make this detector fire -what determines a detector's starting activation level? as one factor, detectors that have fired recently will have a higher activation level -in addition, detectors that have fired frequently in the past will also have a higher activation level -repetition priming is explained in similar terms. presenting a word once will cause the relevant detectors to fire -once they have fired, activation levels will be temporarily lifted (because of recency of use) -therefore, only a weak signal will be needed to make the detectors fire again -as a result, the word will be more easily recognized the second time around

Practical applications of cognitive psychology

-Human factors: cars, cell phones, web pages, airplanes ex. does talking on cell phone impair driving ability? -when driving in car using cellphone, airplanes and all the controls pilots have available- there are many factors that come up -One example does talking on the cell phone impair your driving ability? -We are trying to figure out what factors we would need to manipulate -What if you eliminate the need to fiddle around with your thumb and not dialing just using voice to activate it - does that help or are you still distracted? -Education: it helps to understand how people learn (testing effect) -Treatment of brain damage: Alzheimer's, stroke, etc. -Legal system: eyewitness testimony (people when they are asked about their memory for their particular event often have inaccurate information and are confident in that - how much trouble people have reconstructing information from the world and what if there was a way to eliminate that uncertainty and go straight to their brain?)

Context and the blue and black dress (illusion example)

-If you just strip away that photograph - this is the original photograph that you've been looking at all along and you just look at this segment of fabric and this segment of fabric -To most people they would say its blue and black - like this does not look white and gold even if you are someone that thinks the dress looks white and gold -So the context of the rest of the photograph must be doing something to be making some people see this white and gold -This is just taking these strips away here -There are other illusions that have this property - this illusion here makes you perceive this bar which is a single color going straight across here -The background here is a gradient - it is going from light blue to dark blue -This strip here is all one color and that's hard to believe that your visual system is doing something that is making you think that this is lighter and this is darker and it has to do with this local context -If I strip the background away a little bit you can still see the gradient but if I take it away fully you can see it's the same color -Your mentally brightening this section and mentally darkening this section based on the surrounding and the more context there is the more dramatic it is

Paradox illusion

-Paradox illusions are generated by objects that are paradoxical or impossible in "real life" or three dimensions, but look oddly convincing and perplexing in two dimensional drawings. -Such illusions are often dependent on a cognitive misunderstanding that adjacent edges must join. -these cleverly designed 2D illustrations that imply something in 3D that couldn't exist this is because the features make sense locally -all features make sense locally, but figure doesn't work when processed globally -So when your attention is here - you can only sort of process so much at a time so if your attention is here well this makes sense - if your attention is here well this makes sense no matter where you look everything is coherent and that's what is so paradoxical about it -When you try to figure out what is wrong how come this cant be created - why is this an impossible shape -Everytime you look at one feature of it it seems to be right its just the whole thing doesn't work when you put it all together -The only take home point is that for psychology it showcases the disconnect between your local perception of coherent realistic visual features and the global sense of whether this could work -It also in the real world showcases the property of inverse optics -ex. statue of triangle: Showcases inverse optics in which if you were standing here you cant know if you were witnessing something like this or something like this - that those are both possibilities for the world

The phonological loop and visuospatial buffer

-The phonological loop (a.k.a. articulatory rehearsal loop) maintains linguistic information in a phonological form -e.g. like rehearsing a phone number -The visuospatial buffer (a.k.a. visuospatial sketchpad) aids in the temporary maintenance of visual and/or spatial information -e.g. glancing at a map and then holding the routes in mind while looking at a fork in the road -The central executive is kind of like the conductor - its an executive control system that is interacting with these other two slave systems

Geons and structural descriptions

-The relations are going to be spatial and approximate -Not every ice cream cone will be exact but in general will be these certain goons Importance of spatiality: -Example of two objects that are similar geons both have same geons but the important part is the relation -If geon 4 is on top of geon 2 its likely to be a bucket if geon 4 is attached to the side of geon 2 its likely a mug

Lateral inhibition

-a pattern in which cells, when stimulated inhibit the activity of neighboring cells -lets say you project intense stimulation to those first three neurons: what will happen to the middle rod if they are all getting intense stimulation: its neighbors are going to inhibit it because you are getting such big stimulation and it'll be weaker -but here what happens to cell C it is not getting very strong inhibition from B and its not getting strong inhibition from D because D, E, and F are only moderately stimulated so because those two are not as stimulated its not getting as intense inhibition so it is more stimulated

The central executive system

-a set of control processes involved in the processing of information from the phonological and visuospatial stores goal management: -keeping track of goals at various levels Selection: choosing which aspects of a particular piece of information to work with Scheduling: deciding the order in which to perform a set of operations -these executive control functions are thought to depend on the prefrontal cortex -So the final thing here is the central executive: it's a set of control processes (that act on information that is available in the two slave systems in the visuo-spatial sketch pad and phonological loop) involved in the processing of information from the phonological and visuospatial stores -The executive can do many things: -One is goal management - so keeping track of goals at various levels (what am I trying to accomplish? What do I need to hold in memory? -Selection: Often there will be many different things you can attend to so you need to select which stimuli will be attended - and what you need to do with that information -Also Scheduling: so setting up rehearsal loop or planning for retrieval task, deciding what order of operations to perform if you are doing mental arithmetic - even if you are doing some task like making a sandwhich (making a plan of what you are going to do) -Largely (almost entirely) supported by the prefrontal cortex

Neglect - Attention Bias

-attention is biased towards the ipsilesional side -e.g. line bisection task -You are going to process things more on the side that is the same as your lesion -If you ask them to just bisect all the lines in this display (put a line through them) they are getting all the ones on the same side of their lesion but the left side of space that they are not attending to and forgetting to do it and they don't realize anything is missing -You can also see with individual lines they think this is a good bisection they are not fully attending to the size of this and bisecting this with a bias to the ipselisional side -And it's the same even when they are reading -If you ask them to read this text passage they fail to read any of these words on the left they are just reading these ones on the right -e.g. writing they are only writing on the right side of the paper -this also affects writing - if you ask people to write they aren't going to write half sentences (they are complete thoughts - patients are able to think clearly it doesn't affect their writing ability) but you could see the writing is just gradually skewing more and more to the right side of the page -their view of this paper is skewed and they don't process that there is more space here to work with and it doesn't look strange to them -Its not that their visual cortex cant see this - all this information is getting into their visual system - its just that they aren't paying attention to it - it doesn't occur to them that it is missing - they are blind to it at the conscious level -attention is biased towards the ipsilesional side and attention is deficient on the contralesional side

Distortions

-characterized by distortions of size, length, or curvature. -It is well known that size and shape may be distorted in simple illusion figures.

Context and Memory

-context helps retrieval -context may be many different things: -other words on list -environmental cues (e.g. visual, auditory, or olfactory) -internal mental states at the time of encoding -the more similar the retrieval situation is to the encoding situation, the better retrieval -context can aid retrieval - by context we mean lots of different things -so right now you are learning about cog psych in this particular classroom so the room is the context -you have environmental cues - what it looks like - ex. you have these three different screens -and you remember where you are sitting and who you are sitting next to and my voice - that is all part of the learning experience it has nothing to do with cognitive psych -its just part of the context -if you are learning a list of words other words on the list are part of the context they can help later facilitate memory if you get exposed to some of the other words that might later help you retrieve a particular word you are looking for -and then your internal mental state (how you are feeling? Are you tired? In a good mood? Bad mood? Are you motivated? Are you sober or drunk?) -the general principle here is the more similar a situation is to the context in which you learned it the better

Sir Francis Galton

-had subjects use introspection to study mental imagery (asking research participants to introspect and to reflect on your mental experience and then see if you could use those subjective qualities to understand how the cognitive system worked) -he asked various people simply to describe their images and rate them for vividness -self-reports suggested they could inspect mental images as pictures -ex. instructions given to his subjects: "before addressing yourself to any of the questions on the opposite page, think of some definite object - suppose it is your breakfast-table as you sat down to it this morning - and consider carefully the picture that rises before your mind's eye" -participants made ratings on a series of attributes 1. Illumination - is the image dim or fairly clear? Is its brightness comparable to that of the actual scene? 2. Definition - are all the objects pretty well defined at the same time, or is the place of sharpest definition at any one moment more contracted than it is in a real scene? 3. Coloring - are the colors of the china, of the toast, bread-crust, mustard, meat, parsley, or whatever may have been on the table, quite distinct and natural? -it gave some information about the individual differences that some experienced detailed mental images and some didn't at all

Another split brain experiment example

-if you flash a stimulus (picture of horse) to the left visual field and their eyes are fixated here its going to come into the right hemisphere the right hemisphere cant say what they saw - can't speak -if researcher asks what was it? - they would respond I don't know -if the researcher asks what goes on it? - they would respond I don't know -the talking part of their brain (left hemisphere) doesn't know what they experienced but if you ask them to draw it - and again this only works with their left hand -they draw a saddle to represent what they saw - so it shows that the right hemisphere does know what they saw but cant say it but can express it here using their left motor cortex

Other examples of a changed environment hurting recall

-if you study with noise and are tested in quiet you do worse even though quiet is good for test but not good in and of itself quiet is good for test if you studied in quiet -If you studied in noise its better if you are tested in noise

Attention deficits are not sensory

-impairments manifest in mental imagery -impairments also apparent at multiple frames of reference -space based (environmental) coordinates (everything we have looked at so far - so the right side of the visual field vs. the left) -object-based coordinates (relative to a particular object the left side of that object tends to be ignored) -so attention sort of when its diffusely allocated will show this impairment that you are less likely to process things on the left side then the other but when you focus your attention in to a particular object each object they are missing half -here they are focusing their attention on drawing the tree- but they miss half the tree -then they move their attention here at the house and miss half the house -give them line bisection task you can see patients when it is all one big array of lines they have this space based deficit they are neglecting to cancel out the lines on the left but when you divide this into two sets - here they go to the right set and then here they are neglecting to cancel these ones -and then you go to the left set and they are showing even more of an impairment but they still get the ones on the right -once your attention zooms in on an object then with respect to that object they are showing neglect

Decoding more complex mental images: imagining specific paintings

-in 2015 researchers were able to tell pretty accurately whether you were imaging a painting by el Greco, richter, gursky etc

The process of object recognition

-in order to recognize an object, we must: -represent the physical stimulus in short-term memory (STM) -find a representation in long-term memory (LTM) that matches the current one -the process of object recognition can be thought of as just as challenge as taking what you see and representing it in this transitory state - its short term memory - taking something from the world - now its in your cortex and actively processed and then trying to match it to long term memory -you have extracted some information from your senses and now you are just going to do this search of all the things I know about the world - maybe it's a face and what to compare it to long term memory

Broadbent's filter Model

-information is selected on the basis of physical characteristics (e.g. which ear's input to attend to) -the selected information is allowed to pass to later stages where it undergoes further processing -unselected information is blocked completely -an example of an early selection model -Filtering is what we do to reduce processing of irrelevant information -So in this model we filter based on physical characteristics so something intrinsic to the stimulus (So in this case which side of space auditory space the information is coming from or which ear the information is arriving at) -In this case you can think of the unattended information as the right ear the attended information the left ear but this kind of model applies to the visual system or other systems -You have something attended and something unattended - could be paying attention to thing son the right side of screen and ignore left side so that's a physical characteristic -The sensory register is just the first stage of sensory processing (cochlea on to auditory cortex in this case) basically processing the basic features of the input and then the filter is going to be applied prior in this case to perception -The reason people can tell you so little about what is happening in their unattended ear is this filtering blocks out the unattended information quite early - it's a robust filter that puts a halt on information from that right ear -Whereas the attended information just passes through -You have these two different channels we call the left ear or right ear - one of them gets thwarted by the filter but the other information the gate is open and it passes through it gets processed perceptually and then enters short term memory which just means what is currently occupying your awareness -So unselected information is completely blocked in this model -This is referred to as an early selection model -Selection happens early in the system

Backgrounds and ambiguity

-many of these illustrate is this a black figure against a white background or a white figure against a black background so you have to interpret that in this case is it a white word against a black background or white facial features against black background -this illusion illustrates the problem of segmenting if you see the word figure here its against a black background but then the white backdrop is against the black backdrop -and this is the basis of some kinds of illusions like this one: fed ex -within logo they have a hidden figure (an arrow)

Memory errors

-memory is NOT like a video camera -we often remember only the gist of what we experience rather than the veridical details

Misinformation effects

-misleading information can affect memory for the actual event -important for courtroom/eye witnesses testimony • you can have people experimentally form these false memories that they are convinced are real and showcase how malleable human memory is • Influential for eyewitness testimony - shows thatthings people think they remember can be false -experiment: subjects watch a film on traffic safety that contained an accident -subjects later answer questions about the accident -on 1 question subjects asked: "about how fast were the cars going when they hit each other?" -other subjects given same question, but with "hit" replaced by: contacted or smashed • Later they asked subjects to estimate speed of vehicle - people who got the word contacted or hit estimated speed in low 30's whereas people who got bumped, collided and smashed guessed it to be in the high 30's • That question has changed there memory so now when they think back to the event it updated there memory this idea of the cars smashing into each other • And now you think back and think the cars were drivin faster -did the question affect memory for other aspects of the accident? • One week later subjects came back and asked did you see broken glass in the video - there was none • Only 14% have false memory in this condition for broken class • And here 32% it doubled beliefed there was broken glass after hearing the word smashed • Just that one word affected them

Word Recognition

-one possibility for how the visual system recognizes words is though a system called a feature net -the initial layer, at the bottom, comprises detectors for features -subsequent layers detect more complex patterns like letters, and then words -visual words can be recognized with extremely brief presentations (e.g. less than 40 ms) under the right conditions: -words that are more frequent in the language are better recognized -their "detectors" may have stronger pathways and/or higher starting activation levels -words that have been recently seen are better recognized, a phenomenon known as repetition priming -priming = facilitated processing (e.g. faster reaction time) following a prior encounter -words in general are better recognized compared to nonsensical strings of letters -Word recognition - one possible way that the visual system deals with word recognition is using this sort of network of hierarchically organized detectors and this was proposed in the 50's - there must be some detectors of certain constiguent features that make up the letters in the alphabet -As you learn reading you build these up and each one is not supposed to be thought of as a single neuron that's just waiting for this curved shape and fires when it detects it these could be ensembles of hundreds or thousands of neurons that are kind of distributed over areas of the visual cortex -Each one of these as a processing unit is designed to look out for the stimulus and when it is detected it sends information forward - and so these feature detectors are sending signals to a letter detector and this letter detector is waiting for the letter C and so when the letter C has sufficient evidence from the visual world - when each of these component features is stimulated simultaneously this letter detector will get higher activity and will send information forward to word detectors and only one word detector is shown here but you can imagine every word that you know in the English language has a word detector that there is some representation of it some pattern of neurons that fire when that word is detected and then that is then passed on to the semantic system so you know what that word means ex. that's a clock -We are focused now on not knowing what the word means but just was this word present given the visual stimulus -This is kind of the general idea of the model -Visual words can be recognized by anyone with basic training in reading with very brief presentation - you can flash these on screen for 40 milliseconds or less and the word appears and disappears and then asks what people saw and they use these brief presentations because it challenges the subject and if you weren't paying attention you could miss some details and if you showed the word for 1 second everyone would get it correct and you wouldn't have interesting data -But with these quick flashes you can start to see some properties of word recognition and these properties are intuitive for the most part -Words that are more frequent in the English language will be better recognized then words that are infrequent because you just have higher base line expectation- you have seen that word more times you have a stronger memory representation itll be more easily activated - its more likely -So their detectors might have stronger pathways in this model -Or you can think of it as a higher starting activation level - this node or unit has some activation level (if thinking of it as neurons some firing rate that is firing at a baseline level) even the word clock has some activation level and the word table has some and the word auditorium may have some but it's a little lower because it's a less frequent word and each of these words can then be boosted by visual input that is consistent with that -If the word clock appears its going to boost that detector up but detectors of more frequent words will have a head start (that's the idea) -And words that you have recently seen (clock that you just saw on the slide) that's now been something you've sort of seen on screen and ive just mentioned it and all of these things sort of prime it -You've now heard and thought of that word and now if I flash the stimulus and it has a c and a k in it you might be likely to think it was clock -This is repetition priming - facilitated or more fluent processing of something you recently experienced -The fact that you've seen a word recently gives it a higher activation level or a head start on the detectors model -Words in general are going to be more readily detected and recognized compared to nonsensical strings of letters -If the model was just letter detectors and each detector was just looking out for letters it wouldn't matter if you saw a word that was in the English language and had some real meaning or just a string of 5 letters that were randomly arranged because the letter detectors would just be determining what letters you saw - but that's not the case -If you show letters that actually make up those words youll be more quick at detecting those letters

Practical message about attention

-our attentional window at any moment in time is quite limited -many hazards to multitasking -effects of cell phone use on driving performance: -many studies show that driving performance is impaired when the driver is on the phone -(Panel A) on the phone, drivers are more likely to miss a red light, and (Panel B) are certainly slower in responding to a red light (slower reaction time) -(Panel C) disruption is not observed, however, if the driver is conversing with a passenger rather than on the phone -thats because the passenger is likely to adjust her conversation to accommodate changes in driving - such as not speaking while the driver is navigating an obstruction

Lightness constancy (blue and black dress)

-our brain does not directly perceive the true brightness of objects in the world, but instead compares the brightness of a given item with others

Color constancy

-our brain does not directly perceive the true color of objects in the world, but instead compares the color of a given item with others in its vicinity

Object-centered neglect

-patients can neglect the left side of the object, rather than the left side of space -black lines show expected left-sided person-centered versus red lines showing actual point where the patient neglected -their attention is shifted and zoomed into the object -and with respect to that object they are showing neglect -this illustrates an important property of attention that attention can be thought of as relating to where in space you are attending but also where in the context of a local object your attention is focused as well

So is attentional selection early or late?

-psychological and neural evidence for both -some information may be filtered out at any early stage of processing, while other information may be processed more deeply before it is selected against • its not that Triesman was right or Duetsch Norman was right I think that there are situations where things get filtered out very early and that happens in some circumstances where you can really block certain information coming in from a specific sensory channel but the blocking isn't always complete so there might be some attenuation and then there are also late selection mechanisms ways that after information has gotten in you can engage in additional selection -Most people think there are elements of all three models that are valuable in thinking about this

What is an illusion?

1. -an erroneous perception of reality -an erroneous concept or belief 2. The condition of being deceived by a false perception or belief -departures from truths about reality -latin root of illusion is illudere which means "to mock" -optical illusions mock our trust in our senses -suggest that the eye is not a passive camera -they suggest that the eye cant possible be this passive camera that just takes in an input and tells us what we are experiencing because there are all these circumstances of how this isn't the case -we initially make one interpretation and then we discover another interpretation -perception is an active process that takes place in the brain and is not directly predictable from simple knowledge of physical relationships -what I hope this will illustrate is that perception is an active process that we are making judgements and assumptions that are cognitive - that aren't purely driven by bottom up information processing but are taking into account our assumptions of what we are experiencing -the contextual details - our lifetime of experience with certain kinds of things -we make these guesses which lead us to interpretations that are usually going to be good but in the case of illusions often illustrate that we sometimes jump to the wrong interpretation of what we are seeing

Slow presentation vs fast prevention and memory

-slower presentation yields better memory for early and middle list items, due to enhanced encoding/rehearsal opportunities -more rehearsal --> better storage in long-term memory -presentation rate has no influence on recency effect -if you slow down presentation time - I was giving you 2 seconds per word but if I gave you 4 seconds per word and compared - the more time you had per word the better your long term memory would be and that's because you can process each word more attentively you might even do elaborative encoding (ex. if the word was table you might imagine a table) you might connect the words together and build a story - I mean all these things you would have more of an opportunity to do if you slowed down the presentation rate -You are just more robustly encoding -So this is just a difference of a few seconds but you can see all along the whole curve up until this point you get an advantage -So slowing the presentation rate improves the primacy effect and the middle list items -So just proves all of this is long term memory retrieval - the only part that is coming from working memory is the last items (in this case the last three or four) and there you see no advantage -So slowing down the presentation rate is not helping with working memory you still have the last three to four maybe 5 items available in working memory

The Human Hippocampus

-talking about the consolidation process and what happens in our brain that allows on a moment to moment basis what determines what gets into long term memory -and we are going to really focus on the human hippocampus - in the medial part (toward the mid- line) of the temporal lobe -seahorse shaped structure is hippocampus -whats special about the hippocampus? -this is a wiring diagram of the brain of a monkey -the region at the top of the diagram is the hippocampus - it really is there because its kind of in this unique position in the brain - its wired in such a way that even though it doesn't directly connect with a lot of regions - it receives information from all these other parts of the brain (frontal lobe, parietal lobe, visual areas - highly processed information - information that has made it through all these different stages and is now being processed at a pretty abstract level and is feeding into the hippocampus -its listening in to everything that's going on - the sights, sounds, and spatial information and emotions and thoughts and taste -the hippocampus receives that information and can do stuff with it

Sensory memory trace

-the initial sensory trace contains more information than we can ultimately remember -iconic memory traces decay by 300 to 500 milliseconds -echoic memory traces might last 2-4 seconds -So if you aren't really paying attention you cant go back 20 seconds and reconstruct what someone said

Two Tabletops Illusion (distortions)

-the table surface's are the exact same shape -the only reason it doesn't look that way is because the legs are giving your brain sort of the wrong cues about depth -it's a trick because in the real world 3D shapes have a certain appearance - things that go in the distance change their size and get distorted and here you are making those assumptions because of the legs -your initial interpretation of them being two different shapes is wrong its being influenced by the assumptions you are making about how these objects would proceed in the distance in a real 3D world -Why do people misperceive these shapes? -the answer involves the normal mechanisms of shape constancy -cues to depth in this figure cause you to perceive the figure as a drawing of three-dimensional objects, each viewed from a particular angle -this leads you - quite automatically - to adjust for the (apparent) viewing angles in order to perceive the two tabletops, and its this adjustment that causes the illusion -notice then, that this illusion about shape is caused by a misperception of depth: you misperceive the depth relationships in the drawing and then take this faulty information into account in interpreting the shapes

Blue and Black dress illusion

-this is a photograph that due to different factors of the exposure created this perfect storm -This doesn't happen very often where the lighting and the cropping together give it this ambiguity that roughly half the population sees it one way and half the population sees it the other way -It's a little skewed most people see it blue and black

fMRI study and the blue and black dress

-this one showed people in the scanner they looked at these little 12 and a half second blocks of trials where they saw these colors or they saw the dress and the images were each shown multiple times to the participant -What they found is that when people were looking at the dress who saw the dress as white and gold they consistently had more activation in all of these regions shown here -The individual dots here are the participants and this is the amount of bulb activation -And you can see in each of these regions there is more activity in the gold group then the blue black and they argued that the fact people saw the dress in white and gold show more active brains is maybe something to do with a top down mechanism using more cognitive processing to do this mental brightening of the dress -For someone that sees it as white and gold makes that assumption that it's a dress in the shadow that mental brightening is taking more effort in this specific brain regions in order to do that -White and gold group have more going on in the frontal and periatal lobes could be a top down intentional effect

Levels of Processing

Craig and Lockhart -Gus Graik and Bob Lockhart proposed a processing framework for memory -highly influential view of "levels of processing" -encoding of information into memory is a dynamic process -what we remember is a function of how we process information -Craik and Lockhart were developing these influential theories of what determines information and the likelihood it ends up in long term memory -And they came up with this frame work called levels of processing -This framework basically argues that encoding information is a dynamic process -So what we remember is really a function of how we process the information not just the case we encounter things in the world and then our brain stores them into memories -Its much more active then that - the way we think about information and the connections we make in our minds really make a difference in how we remember that -And they argue there is this continuum of levels of processing ranging from the most superficial or shallow types of processing (like if you were studying the words in the demo we did in class and just saw bowl you said oh bowl I need to remember bowl and the next word comes up and you say oh dance I need to remember dance - if you weren't thinking about it and imagining it connecting it in some way that is shallow processing -Just the word comes in and you think about it -Deep processing is going to be much more elaborative

Lobes of the Human Brain

Frontal: reasoning, planning, speech Parietal: spatial, body awareness Temporal: auditory processing, object recognition Occipital: vision Cerebellum: movement coordination Central Fissure: divides F and P Lateral Fissure: divides F and T -C comes before L so C is on the top and L is on the bottom

What is the capacity of STM?

George Miller (1956) -the maximum number of items recalled without error is 7 plus or minus 2 -but what is an item? -memory span is not limited to a certain number of items per se, but rather by the number of chunks -Chunking: grouping a series of apparently random items into a smaller number of meaningful segments to enhance recall Nelson Cowan (2000) -when rehearsal and long-term memory are factored out, our true STM capacity is actually only about 4 chunks -He wrote this paper that argued that the capacity of memory is really 7 plus or minus 2 -So for some people its 5 for some people its 9 for some tasks it may be 5 or other tasks it may be 9 but generally whenever he looked at different kinds of experiments with different stimuli this number 7 just kept coming back like that was the capacity of working memory -But the question that then became apparent as researchers grappled with why is this 7 what does it mean? 7 what? -7 numbers? Well you all just remembered 16 so what item is it? the year that is the item in that case you just remembered 4 maybe you could have done 7 years or 28 individual digits if you were using that strategy -the important point that became apparent after miller's work - is that what you can remember is 7 plus or minus 2 chunks -so miller and then others realized that by chunking information - by grouping individual stimuli or features into some bigger whole that then becomes a single item and is only using one of your 7 slots -so if you group these seemingly random items into more meaningful segments you are able to enhance your recall -Nelson Cowan commented that when you take out other factors like rehearsal and the use of long term memory (so being able to process these three letters as UFO is using longterm memory to scaffold you know that UFO is a meaningful three letter acronym for you so being able to use that to chunk that together is lightening your load which is what miller was saying) -But now Cowan is saying when you take this out of the equation particularly rehearsal and the use of other strategies it seems like miller over estimated that 7 plus or minus two is an overestimation -And that most people can only maintain only 4 chunks -The capacity is probably a little less in reality

Attention's Properties

Selective: in that case you were supposed to be selectively focused on a sub set of the people in the video and what they were doing Divisible: you need to be able to attend to multiple things: the players in the white shirts were moving around and so you were kind of dividing your attention between multiple players and tracking them Shiftable: so everytime the ball passed your sort of moving your attention to a new special location - need to be able to shift it rapidly Sustainable: a lot of times you need to attend to one thing (security guard needs to be vigilant and sustain your attention so your not having a lapse ) or if you are looking for something you are sustaining your attention in this fixed way

Kosslyn's image scanning experiment

Task: -memorize map -map taken away -focus attention on named location -hear other location -press button when attention at 2nd location -And if images are represented like pictures and have this spatial relationship that preserves the metric details (the distances between objects) he argued you should see a linear time course of mental image scanning -The longer it takes to get from one location to the other in the picture the longer to move your attention through the mental image -This was true -"scanning" farther takes longer -This was support for the analog view - this functional equivalence -That scanning your attention from location to location in the image was just like moving your attention in space -Lots of evidence that support analog visual images -mental images are internal representations that operate in a way that is analogous to the functioning of the perception of physical objects

Two viewpoints on the nature of mental images

The Analog Viewpoint: visual mental images are analogous to pictures in the head ("functional equivalence") -the analogy of a mental image to the actual picture -imagining yourself sitting in your living room is just like sort of looking at a picture of it -might not be quite as detailed but it should share all the same properties as viewing an image - should be functionally equivalent -Championed by Stephen kosslyn The Propositional Viewpoint: although we may believe we experience images as "pictures", the underlying mental representations are actually non-pictorial abstract concepts -this subjective experience of pictures is not actually the mechanism that allows us to use imagery to make decisions -but instead mental representations are these abstract non pictorial representations or structural descriptions -Championed by Zenon Pylyshyn

Basic properties of geons

1. they are sufficiently different from each other to be easily discriminated -You need to be able to discriminate geons from each other 2. they are view-invariant (Look identical from most viewpoints) - if you are looking at it from one angle or another the geon will look the same -You can extract what you are seeing regardless of how you are seeing it 3. they are robust to noise (can be identified even with parts of image missing) -perceptually degraded pictures are better recognized if geons are preserved

Example of hierarchical organization in the hippocampus

-And if you go one stage further to the hippocampus which is receiving inputs from areas like IT - here youll find neurons who respond selectively to individual stimuli like this neuron likes Jennifer anistons face and whenever it is present it fires a lot -This neuron likes halle berry but not to other stimuli -You see this hierarchical organization - information is getting passed form the retina to the thalamus then the visual cortex - at each level it is getting integrated -Neurons with small receptive fields are sending signals to neurons with large receptive fields till you are able to extract what it is in the world you are seeing -there is a hierarchy of detectors, with each successive layer within the network concerned with more complex aspects of the whole -thus, low-level detectors respond to lines at certain orientations; higher-level detectors respond to corners and notches -at the top of the hierarchy are detectors that respond to the sight of whole objects -these representations are probably supported by tissue in the inferotemporal cortex (IT) -recordings from cells in this area has shown that many neurons here seem object-specific - that is, they fire preferentially when a certain type of object is one the scene -ex. the researchers located cells that fired strongly whenever a picture of jennifer aniston was in view - whether the picture showed her close up or far away, with long hair or shorter

Cognitive neuroscience

-Cognitive neuroscience is the term that has emerged for the neuroscientific study of cognition -Anyone using tools - brain imaging, brain stimulation, electrical recordings or various neuroscientific techniques to try to understand perception, attention, memory language - that's all cognitive neuroscience

Ventral stream lesion in patient D.F.

-This is just an example for patient DF for where the damage was -You can see it is these areas in the ventral stream - left and right occipital temporal boundary here -Mostly occipital cortex -If you look at subjects who have healthy brains and you scan them while they are doing an object recognition task and you see which areas are most active you get strong activity in these regions -Area LCC Its an object selective processing area - in the healthy brain these are the areas of high activity for objects

What do illusions tell us about?

-it shows us where perception fails

fMRI setup

-lying on back looking at video screen -Gradients are working hard changing polarity of magnetic field to know where in brain signal is coming from -Like PET scanning you can have subject do one task here like A stage is just resting and B stage is they are tapping fingers in certain sequence and you contrast B stage images of brain and A stage images of brain and see which areas are more active

Diffusion Tensor Imaging (DTI)

-these are color coded images of white matter in the brain obtained by an MRI scanner -they are showing the direction of information transmission - the highways of the brain you can see all these pathways in the brain in exquisite detail

Example of representations

-version of something in a different format -e.g. maps as representation -This is a photograph of San Francisco which is one form of representation -The actual san Francisco is the city that is there - the photograph is one way of representing it but here we are talking about the actual city -Heres a map of San Francisco - this is one way of representing it - and this is a physical map that one could draw that sort of illustrates where the streets are and where different buildings are located and then there is also the mental representation one could have that if you were standing there and trying to find your way around how do you make your navigational decisions - that's going to be a little different maybe for you its map like - or based on different things like based on land marks

Support for RBC

-we can recognize partially occluded objects easily if the occlusions do not obscure the set of geons that constitute the object

Potential difficulties with RBC model

1. Structural description not enough, also need metric info -these two animals have similar structural descriptions but there are differences in snout size, neck size, leg size etc -don't need to just know what geons are where but how elongated they are 2. Difficult to extract geons from real images -this food cart its hard to know what geons it is -and if you turn it into geons it doesn't clearly represent what you are seeing 3. Ambiguity in the structural description: most often we have several candidates -this letter A can be represented by a collection of geons but maybe it is two vertical ones and one horizontal ones there are just many different candidate systems and how does our visual system know which one is best 4. For some objects, deriving a structural representation can be difficult -this shoe its hard to know how you would build it out of goons 5. RBC model also does not include a mechanism for contextual influences on object recognition -these letters here are visually identical and if you were going to recognize them with templates or geons you should get the same answer but when you read it you quickly read it as the cat -even though it is the exact same stimulus you are interpreting them as different letters -so the context, the letters that surround it and your knowledge of the English language is affecting your perception and any good theory of object recognition needs to take the context into account -RBC model does not include a mechanism for contextual influences on object recognition -This is a problem for both the RBC model and template model and really any theory of visual perception should try to grapple with this fundamental issue which is we are not experiencing individual stimuli or features in isolation - we are experiencing them in the context of other surrounding stimuli so in the case of word reading shown here - this shape here is perceived in a completely different way depending on if the local context is the T and the E vs. the C and the T at least in the English language for people who are unfamiliar with the word the and cat you wouldn't necessarily have this problem -If you just recognized the alphabet you may think oh these are just both A's. You don't know that TAE is not a word but if you do know that you immediately jump to the conclusion that this must be THE

Whydoyoudobetterwhnenthecontextofstudymatchesthecontextoftest?

-All of these things you are thinking about underwater - the noise of breathing, coldness - these are all environmental cues and internal cues that are part of your context -And later when you are doing the retrieval task even though the context is not what we are testing we only need to remember which words were on the list - thinking back or re-experiencing that context can activate some of the memories -So the context can sort of help facilitate retrieval by giving us a root to it -You might have places on campus where you have experiences and then those memories from the past come back to mind when you go back -Being in a certain context can bring back certain memories

Brightness Illusion and the blue and black dress

-And the illusion is that this square A and this square B are the exact same darkness -One way to prove it is to put these bars through - they blend effortlessly here and here -Another way is to remove the surrounding squares and you can see it -Why does this work? there is a few things happening in this image that make it so dramatic - one is that the contrast of the white and black squares here -The A square will be mentally darkened because it is surrounded by lighter squares and the B square will be mentally lightened because it is surrounded by darker squares -Another is the presence of this shadow here which is causing you to mentally brighten square B -Lets talk about the shadow which is the main thing - without this cylinder here you wouldn't see this effect -what has happened here? The answer again involves the normal processes of perception -First, the mechanisms of lateral inhibition play a role here in producing a contrast effect: the central square in this figure is surrounded by dark squares, and the contrast makes the central square look brighter -the square marked at the edge of the checkerboard, however, is surrounded by white squares; here, contrast makes the marked square look darker -but, in addition, the visual system also detects that the central square is in the shadow cast by the cylinder. Your vision compensates for this fact - again, an example of unconscious inference that takes the shadow into account in judging brightness - and therefore powerfully magnifies the illusion

Hierarchical organization of visual processing

-And when we record from different parts of the visual system going from area V1 which is the earliest stage of visual processing up this hierarchy of V2 which is the next stage and then V4 and then IT (inferior temporal cortex) -and what you see in this diagram (cartoon of how the visual system is wired) is this pattern of convergence (each V2 cell receives input from many V1 cells, which expands RF of V2 cells..etc -neurons in V1 (which is the first stage of cortical processing) have small receptive fields and each neuron is going to project to a cell in V2 but each V2 is receiving inputs from many V1 neurons and that has the useful property of expanding the receptive field size of V2 cells -Area V2 is going to listen in on information from a slightly larger part of the visual world -Because this neuron has inputs from all these different V1 cells so it has expanded the receptive field and allowed it to pick up on features that are more abstract -Whereas V1 might only care about is there a diagonal bar in my receptive field and this neurons whole purpose in life is to fire when there is one -It likes this orientation in a certain region of space and when that's present it tells the next level up I see something here you should be aware of it but this cell in V2 is getting input from lots of neurons and this neuron might like bars that are somewhat like this (acute angle) this one might like bars that are in a different orientation in a slightly different region of space -Area V2 can process shape features is it an acute angle etc.. (is it straight or curved) -Area V4 theres also convergence -All these V2 cells are projecting to V4 neurons so we are going up the hierarchy and now these cells if you measure from the V4 cell might respond very selectively to a particular shape - you might find a neuron that likes triangles, squares, circles or other kinds of complex shapes in the world -And if you record from an IT cell these neurons have the largest receptive field - could be the entire visual field - they dont care if you show a tea kettle here, here, here or here because neurons are receiving so much information -You can see these neurons are processing most of the visual space and are sensitive to this high level integrative feature - so this neuron might respond selectively to tea kettles (look at diagram)

Evidence for the Analog Representations

-Anecdotal evidence: the experience of imaging feels very much like seeing a picture in one's mind -Experimental evidence: one of the classic tasks used for this was this mental rotation task - for this task you see two shapes and they are either identical but one is rotated or they are different shapes -the question is not about whether you can make these judgements accurately but how long it takes you to do it -how this experiment works is sometimes the objects are exactly the same and with those trials we can measure your reaction time but the interesting question is really what happens when its different and it depends on how different it is -so 0 degree rotation means that it is the same - youll quickly be able to say it's the same -but looking to see how long it takes when the object is 20 degrees, 40 degrees, 60 degrees etc... -reaction time gets slower with increasing rotation angle -and in fact they do so in a beautifully linear way -this works whether you rotate it in the picture plane (which we just saw) or rotating in depth (rotating towards or away from the viewer) -you can see that when the rotation is 0 degrees the reaction time here is just over a second so that's their baseline response for how quickly you can identify the two images side by side as exactly the same -but as you add rotation their reaction time is increasing a certain number of seconds for each degree so in these 20 degree steps you can see the number of degrees per second it is taking them to mentally rotate them -and this is just like if you had an object and you had to physically in front of you turn it a little bit until it matches -and hopefully for many of you while you were doing this your subjective experience was like that that as you were making the decision you felt like you were mentally manipulating the object -what these experimental results are showing is that the more rotation that's needed the longer it is going to take you and for the analog viewpoint this was just the evidence they needed that the degree of mental rotation was completely proportional in this linear way to the reaction time -its as if subjects are mentally manipulating mentally rotating these in a picture like way -and they got this even without instructing subjects to use imagery -reaction time increases linearly with distance -as if have image in head that can gradually update -analogous to real physical rotation -these results are observed even without instruction to use imagery, suggesting that participants spontaneously form mental images and scan them

Physiological basis of the BOLD signal

-Normally the blood has a mix of oxygenated and deoxygenated hemoglobin and then once this neural activity occurs - the capelleries will deliver more oxygen rich blood and it just changes this ratio and that's what we are measuring -the ratio of oxygenated to deoxygenated hemoglobin increased, BOLD fMRI signal increases -All you need to know is the signal is not directly measuring neural activity its measuring this blood flow related correlate -Measuring activity at a slight delay but we can still tell when the activity occurred -We can assume that 4 to 6 seconds earlier there was neural activity that caused this BOLD signal

Our Brain in the cloud

-Another way that cog psych is becoming relevant in our lives -This is the increased reliance on the cloud -With google and the full network of knowledge available to us - the internet in general there is so much info at our fingertips -You have this phenomenon- thanks to the cloud all your devices are in sync with one another - all the information is there and its only your brain that doesn't know whats going on -This infographic just illustrates some of the ways in which google for ex is changing the way we are using information - so we don't need to keep a physical calender anymore we can just use our phones -In the past the world as it existed pre-internet and google - if we wanted to learn information we had limited options available (had to go to library and read books) and we found ways to take the knowledge and try to commit it to memory and have ways that were useful to us to access them later- we developed strategies to store information and retrieve it and people were really good at memorizing things and that was a big part of education for most of the century until recently -The next time the information was available we would take time to sort of strengthen the memory -Now things are different - with the internet things are just a click away and we can access the information we need -With search engines we don't have to encode it in our brains we just have to remember where to get it - We are offloading some of the demands of storing all this knowledge which might not be important to us to form in memory as long as we remember where to get it -This is changing education because the emphasis on memory is outdated - focus on conceptual knowledge and how to use it -One ex. of study that was published a few years ago - in this study these results suggest that the processes of human memory are adapting to the advent of new computing and communication technology - we are becoming symbiotic with our computer tools, growing into interconnected systems that remember less by knowing information than by knowing where the information can be found -The researchers showed that people when they know they can look something up later they intuitively make less of an effort to store it -They are offloading it from internal memory to external memory

The Modal Model of Memory

-Atkinson and Shiffrin proposed this formal psychological model - worked like this -Information from the world comes into your mind through sensory registers (so visual information comes into visual register, auditory into auditory register etc) and then things that you are paying attention to and things that are important to focus on will enter into short term memory -And we sort of already talked about this arrow in the attention lecture - there are a bunch of different theories about what determines how you get from sensory registers to short term memory -But assuming your paying attention to something and it makes it into short term memory now you have these other things you can do with it (you can rehearse it, you might recode it - so if I present you with a 7 digit number on the screen and I take it off the screen and ask you to rehearse it you probably wont want to just close your eyes and visualize how the numbers looked on the screen that's not going to be effective it would be more effective to recode it into a verbal code - so instead of thinking about it as a visual stimulus you could just change it into a verbal code and remember it as a 4,6,7,9,8 so you've recoded it into a phonological code, you can also use different retrieval strategies remembering things from your past that might bolster your maintenance -And then some of this might make it into longterm memory - most of the things that we experience on a moment to moment basis don't really get stored in a robust way - you are just navigating around the world and its not like every moment of your life is getting stored as a photograph or video in LTM but some of it will and so that's happening dynamically here

Neglect and mental imagery

-Bisiach and luzzatti asked Italian neglect patients to imagine themselves in the particular location at Piazza de Duomo and to report what they could "see" -another illustration that this isn't purely a perception deficit its really something that affects their attention in every way -these patients were familiar with milan and the layout of the piazaa del Duomo plaza -you could ask the patient to imagine themselves in front of a particular landmark - so in front of this church or opposite the church -and then when you ask them to just report what they see they are neglecting even retrieving certain sides of their visual field from memory and imagining it to report things from space -this shows that its not just about how well you can attend to things in your perceptual environment but even these mental images you are pulling out of memory are showing a bias of your attention towards one half of your visual field

Cerebral Hemispheres

-Corpus Callosum: major white matter pathway that connects the two hemispheres of the brain -left hemisphere: language -right hemisphere: spatial tasks -the left hemisphere is not purely for language and the right hemisphere is not purely for spatial tasks its more that its lateralized its predominantly for language in general

The different levels of the brain (general)

-The brain exists at multiple levels of analysis also - people who study the brain may study the molecular level (how individual neurons communicate - how certain genes code for the expression of receptors) -And then there is cellular circuitry ensembles of neurons that fire with certain properties and those are part of a broader networks and the networks allow communication across different regions and the regions might be areas like the left prefrontal cortex and we can look at them on a brain -Much of what we talk about is going to be on this level - of sort of which regions are involved - the hippocampus etc.

The carpentered world hypothesis

-Does the Muller-Lyer illusion depend on experience? -our perception of this illusion (Muller-Lyer) is due to our extensive life experience using corners and angles as distance cues -which is that our perception of this illusion the fact that we see so many of you experience this size distortion is due to our extensive life time of experiencing these kinds of angles and shapes in the real 3D world -all the time in this room look how many right angles there are and you can always tell you are looking at something concave or convex -country-dwelling Zulu people who live in round huts with rounded doors and windows DO NOT experience this illusion -city-dwelling Zulus are fooled by the illusion like everyone else -this is just a feature of the modern world but in some cultures they found these country dwelling zulu people who live in round huts and had only experienced these round huts and hadn't experienced that many angles and they didn't experience these illusions -when they look at those muller lyer type figures the lines look the same for them whereas city dwelling zulu's were fooled by the illusion like everyone else -so this shows that your experience with the ques provided in a 3D world of structures with right angles that either indicate something is an ingoing corner or ex going corner are very reliable in the real world and we use these to constrain our interpretation that this must be shorter than this

Donald Hebb and memory

-Donald Hebb: proposed distinction between short-term memory and long-term memory mechanisms -STM: relies upon temporary activation -LTM: relies upon structural changes in neurons/connections -Donald hebb came along with more of a neuroscientific perspective and proposed this distinction between short term memory and long term memory -Introducing these terms and replacing primary and secondary memory -And for hebb he went at it from a little bit more of a neuroscience or mechanistic perspective that short term memory is this temporary activation (that neurons holding information in mind should be actively firing in this transient way to keep that information active - so if I show you a picture of a face and then take it off the screen to keep that face image in mind and judge a few seconds later whether another face is the same or different you would keep neurons in visual areas and face related processing areas consistently firing) -Whereas long term memory is this kind of plasticity of this neural connection (neurons are all interconnected - the synapses having some degree of strength - so right now your brain has a particular state and 5 minutes later your brain is going to be a little different because the things that im telling you - some of it will make it to long term memory - some of your experiences will change your brain and certain synapses will become a little stronger others a little weaker and all of these synaptic strengths in the brain encode the entirety of the information you know so these are structural changes in neurons and connectins -This connectivity matrix which is ever changing which is long term memory -You cant scan someones brain and find all their long term memories if they aren't thinking about the memory actively - if they aren't retrieving a particular event from their past -The only way that its encoded in the brain is by which synapses happen to be stronger or weaker and I don't think scientists will easily be able to go into someones brain and weed out memories that they aren't actively thinking about

Socrates concern about the advent of written language

-Even Socrates had concern about what the new technology was -Even at that time something that was becoming increasingly prominent people writing was starting to change the way people are using their memory -"if men learn this writing, it will implant forgetfulness in their souls; they will cease to exercise memory because they rely on that which is written, calling things to remembrance no longer from within themselves, but by means of external marks."

Experiment representing double dissociation

-Experiment: in the perception condition patient looks at slot and has index card and is asked to orient card so it lines up with slot -they can't do it -In the action condition when you ask them to mail the letter and put it in slot they could do it -or if asked from memory to do it, can do it -in terms of object recognition you need ventral stream to be able to tell which orientation the line is -In terms of action and the dorsal stream which is in tact you have the information you need to have the reaching motion that lines up the index card to the right orientation -The knowledge of what you would do is in tact IMPAIRED VENTRAL STREAM

Fiction Illusion

-Fictions are when a figure is perceived even though it is not in the stimulus like the Kanizsa Triangle -What is the ghostly triangle - fiction -This is considered a fiction because you are seeing something that is not actually there -There are no lines here and yet you see this white triangle that is including the black circles that is including this underlying triangle -And it feels like there is this white object that is just kind of sitting there and you can just fill in in your mind these edges - they aren't super strong but that is a figure we would call a fiction -The Kanizsa triangle is one of the most famous fictions -Due to some of these gestalt principles where you want closure -you want in the real world things that happen to have these properties don't happen by coincidence there must be some shape that is causing this to happen -The fact that most of you see something line like here is whats known as an illusory contour

Recognition by Components Theory (RBC)

-Geons: geometric elements of which all objects are composed -36 different different geon shapes -another theory that has been proposed that has some similar elements to templates - its thinking about this kind of process of extracting information and matching it to something you store but instead of a template we have these constiguent elements we call geons -recognition by components theory by irving biederman proposes that we have 36 different geon shapes stored and that all objects can be recognized in decomposing them into geons -each one of the 36 comes from 6 different kinds of geons -each one of these geons has 6 different kinds of variants -this model includes several important innovations, one of which is the inclusion of an intermediate level of detectors, sensitive to geons -the idea is that geons might serve as the basic building blocks of all the objects we recognize; geons are, in essence, the alphabet from which all objects are constructed -geons are simple shapes, such as cylinders, cones, and blocks and only a small set of these shapes is needed: according to biederman we need (at most) three dozen different goons to describe every object in the world, just as 26 letters are all we need to spell all the words of English -the RBC model, like the other networks we've been discussing uses a hierarchy of detectors -the lowest-level detectors are feature detectors, which respond to edges, curves, vertices, and so on -these detectors in turn activate the geon detectors. -higher levels of detectors are then sensitive to combinations of geons -more precisely, geons are assembled into complex arrangements called "geon assemblies" which explicitly represent the relations between geons -these assemblies finally, activate the object model, a representation of the complete, recognized object

Cueing Sleep-Dependent Consolidation

-Had subjects learn in these concentration matching game - they learn where matching pairs are -Then subjects sleep in scanner - and monitoring sleep stages but during learning participants got either an odor or nothing -And during sleep they either got presented with same odor - or they get presented with vehicle which is puff of air with no smell -Then they wake up and do a retrieval task - question is how much of a difference does it make when you represent the odor that was present during learning and now that's coming up during sleep does that impact performance -Why would it impact performance - because odor was context of learning -During presentation of odor - when it comes on you get activation of hippocampus and that's evidence that odor is triggering this hippocampus reactivation process by reminding them of context and encouraging them to focus in their sleep of this particular aspect of their day (the experiment) -The odor encourages the hippocampus to be focused on that -The amazing thing in this study is you get this big increase - participants with vehicle remembered 85% of card locatin - so did well -But if you present the odor during sleep they are remembering nearly 100% -And if you just present odor during sleep there was not significant difference here - so only when odor was present during learning -Sleep seems to be this active process where the hippocampus in coordination with cortex is helping to consolidate experiences of your day -Say you are studying for exam for Thursday and there is a smell in the room can your alarm clock be triggered to release that smell during sleep -Targeted memory reactivation - smells or sounds during sleep that then influence what it is your sleep dependent consolidation is prioritizing -If you don't do this then what determines what get rehearsed in sleep - its spontaneous -Other things like salience - something happened that day that was big that you were thinking a lot about -Or intention, novelty, reward etc

Henry Mollasson (HM) and the Amnesic Syndrome

-How do we know all this? -Patient HM •When he was young he had seizures - he pursued surgery for it •They were doing new kind of experimental procedure where you take area where seizure is originating from and in memory people with epilepsy it's the bilateral medial - temporal lobe that is the site of your activity so you get this abnormal neural firing there that is causing the start of the seizure •So surgeons thought if we cut that area out you could eliminate seizures •Remove the hippocampus •What became apparent quickly after removed he could talk walk interact but he couldn't remember experiences and it never got better •He had conversations - met people and then a few minutes later or the next day he would have no recollection of what happened - they are living in the moment and anything occupying short term memory that is accessible to them but once that info fades from working memory the chance of it being held on an accessible days or minutes later are very low •His experiences were not getting laid down in long term store and from that we learned a lot about how memory works •Medial part of temporal lobe bilaterally both left and right was removed •This is a quote from him showing what its like •Sue corkin who studied him wrote this book called permanent present tense about his life •Put it up for the title he is living permanently in the present tense the things he are experiencing aren't staying with him •Things happened before surgery he could remember but things after he couldn't

The inverse optics problem

-If a round object appears on the retina how do you know if its something very large and far away like the moon or something smaller but closer (is it a disc, or a cone) -There are many possibilities that cast the same image -This looks like an oval but we can kind of infer that it is probably a circle that is tilted towards us and so forth -And there is really no way for the eye to know -Here is just another example with shapes - could have squares, rectangles, tilted in different ways and these will all cast the same image on the retina and how do you know if its something that is sort of square and close or rectangular but further away and tilted towards you and so forth -How could one possibly build a physical version of this object? (look at diagram) -And you cant really build this in the real world but you can trick the visual system into thinking that you can -This is the view of this impossible object if you were just looking at it from this vantage point that this photograph was taken from but they put a mirror behind it so you can see this is the actual object that they built - this is a possible object (with the leg sticking up) -this is an impossible object (the triangle) and when you look at it from this specific vantage point that the photographer took you can see that this part and this part fuse together perceptually so your perception of this is really coming out of a 3D world that in truth is like this - so this is the illusion

Apperceptive Agnosia

-The particular kind of agnosia DF had was apperceptive agnosia -Given line drawings and asked to copy them - did very poorly -And yet if asked to draw from memory the drawings are much better because they are using stored knowledge they are using from before the accident in addition to dorsal stream movements to control the arm movements -The patient might not be able to recognize this afterwards but the patient can draw it Dissociation between object perception and spatial processing -patients who suffer from apperceptive agnosia can see, but cannot organize the elements they see to perceive an intact object -this deficit was evident when patient D.F. was asked to copy the drawings shown in Panel A -his attempts are shown in Panel B -the problem is in D.F.'s perception, not in his drawing ability, because his performance was much better, when he was asked to draw the same forms from memory, rather than from a model

Dissociation between phonological loop and visuo0spatial buffer

-Ill give you two patients that show this -Each one of the patients the case in and of itself is a single dissocation where they have damage that is going to impair one aspect of memory but spare the other and together with the two patients what youll see is we have a double dislocations -Stroke patient P.V. -damage to left hemisphere temporal lobe and frontal lobe regions -normal intelligence -but unable to repeat back even short sequences of spoken digits -and unable to perform mental arithmetic -So patient P.V. had an impairment that affected only her phonological loop and spared her visuo-spatial buffer -She had damage to some left hemisphere temporal lobe and frontal lobe regions (damage was broad) -Normal intelligence but unable to repeat back even short sequences of spoken digits so her phonological working memory was almost non existent -And unable to perform mental arithmetic which requires thinking about whats like 46 plus98 you sort of have to think about the numbers and keep track of what you are adding and whats going to go to each place - that requires phonological working memory -So this is just one example of data from this patient on a single consonant memory task- what you see here is percent correct for visual information - this is just maintaining a consonance (memory of consonance) -So letter is presented on screen if it is presented visually she is 100% correct doesn't matter if you are tested immediately 3 seconds later 9 seconds later or 15 seconds later - perfect performance -With auditory if she is tested immediately - so you say C and she says C you say D she says D so she can hear you and repeat back if its in echoic memory but once you wait a few seconds by 3 seconds its basically chance

The first "brain imaging" experiments

-Italian physiologist Angelo Mosso had a patient come in who had a wide skull fracture -he came in and they removed the fractured skull bones and there was this exposed patch of cortex (2 cm cerebral mass was exposed in the right frontal region) -the skull never fully formed over it - soft spot -mosso had this idea to put a little pressure sensor there and measure the wave forms -how much pulsation there was -riding ontop of the heart beats is the slower fluctuations - sometimes signal is higher and lower -he just asked him to do cognitive things like multiply - when he did that the signal increased temporarily (brain pulsations increased) -William james was aware of this: "In Mosso's experiments the subject to be observed lay on a delicately balanced table which could tip downward either at the head or at the foot if the weight of either end were increased. The moment emotional or intellectual activity began in the subject, down went the balance at the head-end, in consequence of the redistribution of blood in his system" -Mosso took subject on table and gave him difficult mental tasks to do and whenever he was engaged in deep thinking the table dipped backwards and mosso concluded that more blood flow was going to brain -This was an important observation that the amount of blood flow going to brain could be a correlate for cognition -Key insight: increased brain activity leads to increased blood flow -the brain necessarily has a rich intricate blood supply

Scaling of Mental Images

-Kosslyn -examined how participants scan and use mental images -some participants imagine a rabbit next to an elephant -others imagine a rabbit next to a bee -then answer questions about the rabbit: -does the rabbit have whiskers? -does the rabbit have ears? -does the rabbit have a beak? -reaction time to answer is measured -According to kosslyns framework you have to preserve these metric spatial relationships -So when you imagine a rabbit next to an elephant you don't have a giant rabbit that's the same size as the elephant -In order to imagine these two things one is going to be big and one is going to be small and in order to answer these questions - like does the rabbit have whiskers? -When people answer these questions they are kind of primed by this size relationship and they might imagine does it have eyebrows? - they would have to zoom in -So kosslyn thought this act of zooming is going to take time and slow people down -Whereas if you imagine the rabbit next to the bee - the rabbit is already big so you need less zooming -This is called mental image scanning - moving attention around image either from location to location or zooming in and out -It took longer to respond to rabbits paired with elephants than to rabbits paired with bees -That was some evidence for his perspective

The human visual system

-Low level vision starts at the retina - light is going to come into eye - at the back of the eyeball are retinal cells that are sensitive to the incoming light and give information about the intensity and luminious of light in different positions of space -This information will be transmitted to the thalamus here that receives inputs from retinal cells and then sends that to the primary original cortex

Retrograde amnesia

-MTL damage impairs not only new learning, but also disrupts memories acquired before the injury • But patient like HM would have retrograde amnesia • You usually get these hand and hand • And retrograde amnesia is just this little orange gradient here - its an impairment that is most noticeable for things that happen recently from the surgery • If I told hippocampus out today you wouldn't remember things in the last 30 min, past week depending on how severe damage was past month or even past year • That's why this is shown as a color gradient • Things that happened 2, 3 10 years ago will be intact and consolidated • But consolidation takes time so if you remove hippocampus right now anything that hasn't been consolidated still depends on the hippocampus and without hippocampus you are going to have a hard time retrieving that information • Forget things shortly before injury • Ribot proposed this principle of memory that said the vulnerability... • It wasn't hippocampal damage specifically but patient came in with head injury they realized the older memories were the less vulnerable they were • That's what you see in amnesia • When you have damage to hippocampus the remote memories are more likely to be available to you and more recent memories are likely to be impaired • Normal memory we have this gradient • Normal memory things that happen today will be accessible to you more tomorrow than things that happened to you at birth or childhood • So in general the further you go back in life the harder it is to recall things because of this recency curve • Things that happen more recent is more accessible - this is normal memory (not always a perfectly straight line) • If I damage HC today things that happen going forward you wont remember at all and the rest of this for anything else that happened before injury will show this normal curve depending on the time you got the damage • That's not all that happens though Patients will often show both anterograde and retrograde amnesia and the retrograde amnesia is graded its not all or none so from the time of trauma you have complete anterograde amnesia and retrograde amnesia is time dependent graded meaning its gradual over time eventually you get back here and if you tested patient like HM • Who had HC taken out early adulthood wouldn't remember things post surgery but would remember things at childhood and then at some point they show graded decline as you get closer to time of damage • So temporally graded retrograde amnesia

McClelland and Rumelhart Model

-McClelland and Rumelhart's model of word recognition included two additions: -excitatory and inhibitory connections between detectors -top-down connections from words to letters and letters to features -in the 80's there was this new wave of computational modeling - mclelland and rumelhart added in this extra importance processing layer of inhibitory connections so now the fact that you think you see the word trip will send inhibitory signals to other word detectors and make you less likely to think you saw the word trap for ex. -adding excitatory and inhibitory processing makes it more brain like and also helps resolve competition so once you settle on one interpretation you kind of diminish the likelihood of alternative interpretations -and each of these can feedback - so once you think you see the word trip now the TR bigram would become more activated - kind of top-down -excitatory connections - connections that allow one detector to activate its neighbors - are shown as arrows -for ex. detection of a T serves to "excite" the "TRIP" detector -other connections are inhibitory, and so for ex. detection of a G deactivates, or inhibits, the "TRIP" detector -in addition, this model also allows for more complicated signaling than we've used so far -in our discussion, we have assumed that lower-level detectors trigger upper-level detectors, but not the reverse -the flow of information, it seemed, was a one-way street -in this model, higher-level detectors (word detectors) can influence lower-level detectors, and detectors at any level can also influence other detectors at the same level (e.g. letter detectors inhibit other letter detectors; word detectors inhibit other word detectors) -to see how this would work, lets say that the word TRIP is briefly shown, allowing a viewer to see enough features to identify, say only the R- I-, and P- -detectors for these letters will therefore fire, in turn activating the detector for TRIP -activation of this word detector will inhibit the firing of other word detectors (e.g. detectors for TRAP or TAKE), so that there other words are less likely to arise as distractions or competitors with the target word -at the same time, activation of the TRIP detector will excite the detectors for its component letters - that is, detectors for T,R,I and P. -The R-, I-, and P- detectors, we've assumed, were already firing so this extra activation from above has little impact -but the T-detector we've supposed was not firing before -the relevant features were on the scene but in a degraded form (thanks to the brief presentation); this weak input was insufficient to trigger an unprimed detector -however, once the excitation from the TRIP detector primes the T-detector, its more likely to fire, even with a weak input -In effect, then activation of the word detector for "TRIP" implies that this is a context in which a T is quite likely -the network therefore responds to this suggestion by "preparing itself" for a T -once the network is suitably prepared (by the appropriate priming), detection of this letter is facilitated -in this way, the detection of a letter sequence (the word TRIP) makes the network more sensitive to elements that are likely to occur within that sequence -that is exactly what we need in order for the network to be responsive to the regularities of spelling patterns

What does the hippocampus do?

-Memory formation (event encoding) -When you form a memory of arriving at a place you are going to have different event details so the sights and sounds - in this case what kind of day, what the hotel looked like etc - these are details present when encoding the experience -Just showing visual elements here - coming into the eyes, the thalamus, V1 and then you have dorsal and ventral stream processing -These circles are nodes or pieces of your brai that might respond to some aspect of the visual experience and then some of these areas might send signals to the frontal lobe (bottom up signals - there might be top down signals) all of these areas are active in part of your encoding -When you are having these experiences you have these cortical regions that are each dealing with one facet of the information you are experiencing -And so what happens in the hippocampus - just a theory but lots of evidence - all of these areas of the brain that are part of the experience - right now for you its where you are sitting in the room, what the classroom looks like, you might remember later that I was dressed as waldo so it was Halloween so the temporal context - these are all details processed at the cortex and feeding into the hippocamps -And the hippocampus has many neurons and you can think of the information arriving simultaneously and what this region seems to do is to bind it or link it together as if its indexing that all these different disperate elements of your experience are together -Hippocampus is this glue that is indexing how all these different attirbutes are co-occuring

State-dependent recall

-Now we talked about internal states as context -So state-dependent recall if you are experiencing information and you are a little drunk and later you have to testify in court about what happened do you want to be sober or drunk? What is going to facilitate memory? -It turns out if you study drunk and are tested drunk you do better -You might think being sober for test your memory system is working you know how to find the details you are looking for but studies show different -Sober during study and sober during test was best numerically but drunk at learning and drunk at recall wasn't too bad -Drunk at learning and sober at recall was surprisingly bad -also did a version of this experiment where subjects either smoked normal or marijuana cigarettes during study and/or test -Smoking a regular cigarette during study and then testing - you do get a benefit if you learn with a cigarette and test with a cigarette you do better -if you learn with marijuana cigarette and are tested with marijuana cigarette you do better State dependent recall: effects of mood -easier to remember happy memories in a happy state and sad memories in a sad state --> mood primes certain memory contents -happens with mood too - your state can be a mood of happiness and sadness you can induce happiness and sadness by showing people movie clips (joyful ones, and sad ones) and then you test them -if you are in a good mood when you study and are in a good mood when you test you do better -if there is a mismatch you suffer

Hippocampus and memory consolidation: transferring memories to neocortex

-Over time we know that the hippocampus becomes less and less important - initially it's the glue that's holding together the features of the memory and binding them but somehow over time its importance diminishes and eventually maybe after a few months or a year its not needed at all -Whats happening is that the hippocampus is thought to be synchronizing its activity with these cortical regions and getting them to occilate in synchrony and form connections between each other - so these cortical connections with other cortical regions now link up the elements of the memory and the hippocampus becomes unnecessary -Transferring memory of the hippocampus and then it eventually fades away - it can be used to encode new experiences and retrieve other recent experiences but it is not going to be necessary to retrieve multi- featural memories after they have been consolidated

PET:Pros and Cons

-Pros: -It has good spatial resolution (how fine grained in space we can map where activity is coming from) - within half a centimeter to a centimeter we can resolve where activity is coming in the brain -It can also measure neurotransmitter metabolism - other kinds of PET injections not just to bloodflow but binding to particular neurotransmitters can show us where those neurotransmitters are concentrated (where dopamine is in high concentration) -Can also aid in diagnosis of alzheimer's disease - unstable carbon 11 that can be bound through a specific metabolic trick - allows you to see where amaloud plaques are -PET is also very quiet - not noisy like MRI - basically a camera detecting these photons -Cons: -its invasive - you need radioactive injection (not a lot of radioactivity though) -Its very expensive -Poor temporal resolution - bad ability to resolve events in time - cant isolate when the signal is coming from (Is it happening when they first saw the word or spoke the meaning - we can only integrate across a minute of time and look at signal related to that minute) -Cannot detect the neural response to discrete cognitive events - if you want subject to look at face and hold it in mind for a few seconds and then decide if it matches face a few seconds later you cant distinguish the initial processing of face from holding it in memory from making decision it all just blends together

fMRI Pros and Cons

-Pros: -good spatial resolution better than PET (1-3 mm) allows precise localization of brain activation -Lots of information represented within spatial patterns of brain activity -Non-invasive (No known harms to subjects) -MRI scanners are widely available at medical centers and research universities -Cons: temporal resolution (1-4 seconds) better than PET but slower than measuring actual neural activity -Very expensive -Scanner noise is very loud

The problem of inverse optics

-The problem with visual perception is that the world is 3 dimensional and yet the retina (camera photo detector in the back of our eyes that is detecting photons simulating at different positions) is 2 dimensional -Just like a camera - you click the shutter and the light is let in and the photo detector tells you what is there - our eyes camera the retina is a 2-D sheet - its going to have to infer something about the 3-D world from this 2-D image we get in our retina -And how does this work how can we know what is really out there in the 3-D world with 2-D data (How can the brain reconstruct the 3D world based on a 2D retinal image?) and the answer is you cant -You cant perfectly infer whats out there in 3-dimensions given fact that our retina is 2 - dimensional -And it turns out there is an infinite number of 3D worlds (possible stimuli) that could produce the same 2D image on the retina -So all of our perceptions are going to be the best guess of what we are experiencing -the problem is fundamentally ill-posed

Dichotic listening/shadowing tasks

-Psychologists figured out if you had headphones and could control what input was coming in the left ear or right ear and you could play different input in the two ears you could create this task where subjects are told to either attend to the left side of space or the right side of space and by space we mean auditory space -In this case the subjects task is to attend to the left side: and repeat back -And then in the other ear you could play nothing, something different, something related so lets you see how selective are they being -early studies of attention often used a setup called dichotic listening: participants worse headphones and heard one input in the left ear and a different input in the right ear -the participants were instructed to pay attention to one of these inputs - the attended channel - and told simply to ignore the message in the other ear - the unattended channel -to make sure participants were paying attention, they were usually given a task called shadowing -the attended channel contained a recording of someone speaking, and as participants listed to this speech, they were required to repeat it back, word for word, so that they were, in essence, simply echoing what they heard -What gets through from the unattended message? -It turns out not a whole lot -Surprisingly people almost entirely block out the information coming in the ear they were not paying attention to -They remember when you ask them afterwards lowlevel details - like was it a human voice, changes in gender of the voice -But not a change in language -There is also experiments where the same word gets repeated over and over again and if they are doing the task well and complying with their goal to just listen to one ear they don't know its repeated -So our ability to filter out information from one side of space or another is quite good

Propositional Theory in a nutshell

-Pylyshyn -Image is a description -if you imagine a triangle where the lines are bisected you would process this as a description: 3 sided shape, 3 angles of 60, small lines bisecting -for all of you who experience mental imagery and think it's the analog viewpoint here is what pylyshin would say: even though you can imagine a mouse biting a cats tail, or a cat under the table and there might be some subjective experience that you think is important for making judgements about how big the mouse is or whether the mouse is to the left or the right of the cat - that you are doing this from a pictorial perspective - pylyshin would say that is not how the brain is organized that actually you have these propositional representations of concepts and relationships -so bite is an action - the agent of the action is mouse, and cat is the object of the action -so a proposition is sort of a statement or fact of the entities here that you are picturing and their relationship to each other -mice are furry so furry is a structural characteristic of a surface - its an attribute of the mouse which is an object -a cat under the table is like a relationship of a table to a cat where the former has a vertically higher position than the latter -if a cat had mental imagery and was picturing this corner of the room the analog representation would look just like the actual thing the propositional one is a set of relations of Wall 1 and Wall 2

Craig and Tulving Levels of processing experiment

-So the different ways that they would study this in the lab to provide evidence for this theory is giving people different tasks -So if you have a participant come in and you have them do certain things with the list of words and you don't tell them they are going to have a memory task you can see which factors will impact the likelihood that a word will be maintained -So for ex. in this task subject had to decide is this word written in capital letters - some of the words were some weren't -So you go through a list of words and you are doing this task - yeah that's lowercase yeah that's capital etc. -So they are thinking about the most superficial characteristic - they might not even bother to read table -Maybe it gets in because its hard not to read word that is in front of you but they are just looking at the letter structure here - the physical characteristics -So that is very superficial - very shallow -If you give them a task where they say ok here is a list of words I want you to decide for each word if it rhymes with mat and so - cat you say yes - fish you say no etc. -It doesn't really matter so much if the word does rhyme with cat or doesn't rhyme with cat thinkinga bout the sound of the word is going to be better than just thinking whether it was written in capitals or lowercase- itll make it a bit more memorable -So its going to be intermediate processing and then the deeper processing happens in this task where you get a list of words and for each word you decide whether it would fit in this sentence -You get a sentence "I know it is springtime... -So a bunch of words come in and for each word you decide would it fit in the sentence so now you need to think about not just what the word is and how it sounds but what it means -Does a daffodil go in a garden? So that little extra processing -Thinking about a daffodil being a flower think about it blooming is going to make the memory more sticky (more likely to be encoded in a way youll letter retrieve on a test) -These are the results - this is percent of words recalled and this is the activity subject performed when they first saw the word -So if they judged whether it was upper or lower case they are only recalling 10% of the words -If they judged whether it rhymes with something they are recalling around 15% and whether it fits in this sentence around 20% -you certainly will retrieve more if you process them deeply

Positron Emission Tomography (PET)

-Subjects injected with radioactive isotope, which emits positrons (tiny particles which are made as the radio-tracer is broken down inside your body - create gamma rays which create a three-dimensional image visible to scanner) -positrons collide with electrons, emitting two photons (gamma rays) in opposite directions -detectors surrounding brain reciter simultaneous photons and compute likely source -neural activity -> increased metabolic demand -> local increase in blood flow the active region -compares regional cerebral blood flow (rCBF) between cognitive states (experimental conditions) -integrates signal over 45-60 seconds -need to wait a number of half-lives before next injection -The first brain imaging technique to study cognition in an effective way is called PET scanning -100 years later working off mosso's insights researchers had the idea to put someone in this PET camera -first a participant will be injected with a substance containing a radioactive isotope -this goes into the blood anywhere blood is going you have this isotope -and because it is radioactive and unstable it is going to decay -as it decays it omits positrons -these positrons just shoot out of these oxygen 15 molecules -when a positron collides with a free electron - these collisions are inevitable -it gives off two photons (Light energy in the form of gamma rays) -this is just a camera - radiation detector that is going to detect when the two photons hit -PET works by detecting the energy released (decay) by positrons -the detectors register these simultaneous photons and compute the likely source -the key idea here is that neural activity leads to increased metabolic demand which causes local increase in blood flow the active region -when an area of the brain is active and neurons are firing away the brain knows this area has been working hard and we need to give it more glucose and oxygen and the brain delivers -it gets this boost of blood and the more blood the more of these positrons will be hanging out and more likely to detect photons coming from that area -you get these images of the density of the radioactive enilation events happening -you can do this while the subject is thinking of something -have them do some numbers of trials of one thing -you can subtract the differences between all these images -over the course of many trials you can subtract these images and do statistical analysis of which areas of the brain were more active -PET scans start by introducing a tracer substance such as glucose into the body; the molecules of this tracer have been tagged with a low dose of radioactivity, and the scan keeps track of this radioactivity, allowing us to tell which tissues are using more of the glucose (the body's main fuel) and which are using less -the primary data is collected by a bank of detectors surrounding the head; a computer then compares the signals received by each of the detectors and uses this information to pinpoint the source of each signal -the computer reconstructs a three-dimensional map of the brain -for PET scans, the map of the brain tells us what regions are particularly active at any point in time

Treisman's Feature Integration Theory

-There are these two stages - the first stage when you see an object is the preattentive stage, where certain features that are unique in the display will pop out effortlessly to you (you separate out those features that are unique from surrounding features) -during the pre attentive stage, features pop out effortlessly -search occurs in parallel -attention is not required -During the focused attention stage features are combined together to create object representation -search is serial -attention is required and serves as the "glue" that binds spatially associated features -the question of how attention fulfills this role is called the "binding problem" -Initially when you process the world all these different feature maps are going to be handled separately (color information, shape information, line orientation and so forth) and its only with focused attention that they can be bound together -During the second stage which is necessary for many kinds of searches search has to be serial -During the focused attention stage features are combined together to create object representations -So attention has to shift from location to location and when your attention ends up at a location - the magic quality attention has is when you allocate your attention to one position all the features in these different maps that co-occur there are sort of bound together - its like attention is this glue that causes maps to converge -When attention is at this location any colors that are present at this location, any line orientation information any motion information (are things moving left, right, up down) other features curvature depth, there are many kinds of features we seem to have spatial maps for and when you allocate your attention here any features that co-occur are sort of bond together -There is nowhere in your brain where you have a green square detector necessarily you have a map for where green is and a map for where certain shapes are and the only reason you perceive this as a green square is because these two maps that are represented in different parts of your visual system are bond together -So attention when it hits this part of space sort of fuses those and says oh the greenness is co-occuring with this shape (it's a green square) -That's what we call the binding problem we kind of take it for granted that when you kind of look at a display every feature is kind of uniquely represented but that's not how our visual system works we send all the information out to these different processing systems to deal with shape and color and depth and motion and different neurons in different parts of the visual system are processing all these different features and then spatial attention is sort of the glue

Synaptic neurotransmission

-These neurotransmitters could be many different types - glutamate, dopamine etc - different neurons are sensitive to different neurotransmitters -They will arrive at the dendrites and if they bind to the receptor channels they can cause the cell body to receive an influx of current and if it's a sufficient level the neuron will fire an action potential - so send this sort of electrical wave form that is transmitted across the length of the axon fairly quickly -And then that (action potential) arrives at these axon terminals and these terminals contain little vesicles filled with neurotransmitter that they can then release to the next neuron -This is sort of the basic flow of information through a neuron -when the action potential reaches the axon terminal, it causes the release of neurotransmitters into the synapse (space where the axon terminal contacts the dendrite) -each neuron has 1,000-10,000 synapses -neurotransmitters bind to receptors on the post-synaptic dendrite and influence its activation state -The space between two neurons is called the synapse - the synapse is not actually a part of the neuron it is the gap but it is sometimes the whole unit is called the synapse because it has the terminal of one axon which contains these vesicles filled with neurotransmitters and those will come to this post synaptic terminal and release and then some of these neurotransmitters will bind with receptor channels and allow sort of ions to flow through stimulating the next neuron -Each neuron has 1000 sometimes 10's of 10000 synapses generally and this creates a massive amount of information processing potential that the neuron communicates with many other neurons and that neuron communicates with other neurons -There are about 85 billion neurons in the brain and many more connections -when a neuron has been sufficiently stimulated, it releases a minute quantity of a neurotransmitter -the molecules of this substance drift across the tiny gap between neurons and latch on to the dendrites of the adjacent cell -if the dendrites receive enough of this substance, the next neuron will "fire" and so the signal will be sent along to other neurons -neurons don't usually touch each other directly -instead, at the end of the axon there is a gap separating each neuron from the next -this entire site- the end of the axon, plus the gap, plus the receiving membrane of the next neuron - is called a synapse -the space between the neurons is the synaptic gap; the bit of the neuron that releases the transmitter into this gap is the presynaptic membrane, and the bit of the neuron on the other side of the gap, affected by the transmitters, is the postsynaptic membrane -just know about the synapse and that it contains neurotransmitters that allows communication between neurons -neurotransmitters are "dumped" into the synapse (gap) -NT's bind to post-synaptic dendrite -what the actual signal is is a neurotransmitter - they are dumped into the synapse and those neurotransmitters bind to the postsynaptic dendrite which is the neuron receiving the input from the output

What are not mental images?

-They are not symbolic or linguistic representations -Ex. Structural description: if you had a structural description of a dog for ex it could be something like this - pointy ears on top of head, head on top of body, four legs beneath body etc.. -This is an abstract set of relationships that can define an object or anything in the world without explicitly putting these into pictorial space -This is just an example of a way of imagining a dog that wouldn't require a picture -If you sort of pull up from memory the set of features and relationships in this abstract sense - the ears on top of head head on top of body... -You don't have to have a picture in mind to answer certain questions about a dog that you have imagined

Center-surround cells

-They documented the existence of specialized neurons within the brain, each of which has a different type of receptive field, a different kind of visual trigger -for ex. some neurons seem to function as "dot detectors". -these cells fire at their maximum rate when light is presented in a small, roughly circular area, in a specific position within the field of view -presentations of light just outside of this area cause the cell to fire at less than its usual "resting" rate, so the input must be precisely positioned to make this cell fire -these cells are often called center-surround cells, to mark the fact that light presented to the central region of the receptive field has one influence, while light presented to the surrounding ring has the opposite influence -if both the center and the surround are strongly stimulated, the cell will fire neither more nor less than usual; for this cell, a strong uniform stimulus is equivalent to no stimulus at all -For some cells you see center surround properties - neuron has area of space that it is sensitive to that is a fairly large region -there's an area within its receptive field it responds strongly to and area around that central part that it responds not only weakly to but actually light energy in this part inhibits neuron - so it reduces its firing below its baseline rate - shuts the neuron down -So if there is light that is hitting the neurons receptive field in its outer ring its going to fire at its lowest level - inner ring its highest level and if its at the boundary it will just fire at its resting baseline level -This is a mechanism that is useful to enhance contrast of visual world - neurons don't just have one area of space they care about and respond to but they also have this surrounding ring where if light comes there the neuron will be inhibited so you can tell with higher contrast what is happening in the world because the central part of the receptive field is accentuated by this inhibitory ring of the fact that the surrounding region dramatically reduces the cell's firing

Neuroscience data and the Mental Image debate

-They had subjects imagine a visual display - a stimulus would be on for some period of time and they were measuring here activity in the visual cortex -When stimulus was on BOLD signal goes up and stimulus goes off BOLD signal goes off and then they had them imagine stimulus so nothing was on the screen and you can see when they are doing that they are activitating the visual cortex (Not quite as strongly as when they are looking at a picture) -But this showed for the first time that the act of imagining something activates the visual cortex in what must be a top down way - they are retrieving from memory this visual image and the feeling they get subjectively when seeing something is associated with increased activity in the visual cortex -This was the first evidence -proof for analog viewpoint

Single-unit recording (monkey example)

-This is what it would look like if you were doing it in a monkey - electrode system implanted and there would be an electrode going to part of brain you are trying to record from and the signals will come into sensor and you can measure everytime you get a line here its an action potential - the neuron firing an action potential (spike) those happen many times a second - neuron is experiencing world and when things happen neuron fires more rapidly than other things -Here this bar of vertical light is being moved around a visual field and researchers are trying to figure out what is this neuron care about - when light happens to be near neuron is firing rapidly and if you move it slightly over its firing rapidly but less so and you move it down its barely firing -Neurons have a resting baseline of action potential - they will always fire at a baseline rate - always going to get some spikes -This neurons receptive field is relatively small - now you can know something about that neuron and study it further

Decoding Individual Mental Images from fMRI activity

-This region of the brain loves faces (fusiform) -The PPA is location in brain that is going to respond strongly to landscape scenes, buildings, landmarks and it doesn't respond very much to faces -They localized these areas in participants - and what you see here is a time course of the BOLD signal in these two regions -The red is FFA blue is PPA -What the task was was to imagine a building on MIT campus or imagine face of celebrity -Each one of these arrows is a time in the experiment when they were told to imagine something - you can see at this time they were told to imagine something and their FFA is very active and PPA was less active (imagining celebrities) -this shows that when you imagine a face you activite FFA(even though no face was present) and when you imagine a place you activate PPA -so this is showing that something must be picture like because the same areas that perceive faces and scenes are active when you imagine them

Inattention blindness and unconscious perception

-This was a clever experiment and in this experiment subjects looked at this display and there were these two lines on it and these dots flickered around and all they had to decide was which line was bigger -The background dots were arranged randomly from trial to trial but because they were so focused on these lines they actually didn't notice that on some trials the random dots lined up and created this illusion (muller layer illusion) so they consistently when these lines were arranged like this on this critical trial they experienced the illusion even though they didn't consciously notice the fins - they believed this line was longer then this line even though they were the same -That shows that unconsciously this illusion is increasing their estimate of the size of this line -Its having an effect even though they were unaware of it -Goes back to this point that stimulus might not be consciously perceived but nonetheless have an influence on the perceiver in this case creating that lengthing effect of the muller layer illusion -on the one side, there are cases in which people seem unaware of distractors but are nevertheless influenced them - so that the (apparently unnoticed) distractors guide the interpretation of the attended stimuli

Serial position Curve

-U-shaped curve describing the relation between position within the series - or serial position - and likelihood of recall -research participants heard a list of 20 common words presented at a rate of 1 word per second -immediately after hearing the list, participants were asked to write down as many of the words on the list as they could recall -the results show that position in the series strongly affected recall - participants had better recall for words at the beginning of the list (primacy effect) and for words at the end of the list (the recency effect) -another ex. of serial position curve: -this is a 12 item list so for shorter lists sometimes the primacy effect is more robust cuz there is less forgetting -This slide is illustrating three different curves so three different versions of this experiment - the one that we just did in the demonstration was like the red line: immediate recall -So as soon as the list ended I gave her the opportunity to list the items -It wasn't distraction free but it was fairly immediate -Now if you do this and I made her wait 30 seconds and then said go the result ends up being about the same -It doesn't matter that much if you wait about 30 seconds or so (30 seconds not enough to eliminate the recency effect) -another intervening task is needed to do so -If you waited longer like a minute or two minutes you would start to lose this recency effect for the most part -Unless the participant kept rehearsing and was really focused on keeping those last items that were in working memory active -And then you could keep the recency effect as long as you want like if those last few items on the list were fresh on your mind and you know you have to wait 5 minutes and you have nothing else to do you can just keep rehearsing them and you would get nearly perfect performance for them as long as you don't get distracted -But its when you have this 30 second filled delay - usually filled with solving math problems or counting backwards etc. - that wipes out the recency effect -An intervening task abolishes the recency effect but has no effect on the primacy effect (no influence elsewhere in the list) -And that shows you that this sort of active processing that is happening during this filled delay is using up your working memory resources and forces you to lose whatever words you were holding in working memory -So at that point since you've wiped out working memory for the verbal materials on the list all you have left when asked to retrieve 30 seconds later is your long term memory and your long term memory shows a kind of standard serial position curve up until this point -So the recency effect is really just this bonus you get if the items are still accessible in working memory

Example of BOLD signal change during presentation of static and moving visual stimuli

-Visual cortex - data shows when subjects looking at static dots on screen or moving dots on screen activity increases and when nothing on screen its low -This area of brain doesn't care if its moving or static just likes dots but MT (middle temporal visual area) cares that its moving -Can dissociate that this area (MT) is responsible for motion and other visual processing -FMRI is powerful technique - one of the cool things we can do is localize individual regions and find these different information processes regions that are highly specialized to specific information -We can also look at how areas of brain communicate with one another

The costs and benefits of selection

-When you look at a display here and you focus your attention broadly you can tell certain characteristics - like there are horizontal lines and there are vertical lines and some are black and some are red -So if you flash this display quickly and the persons attention just has a broad focus and then ask them some questions about this - where there multiple colors? What colors? Where the lines all vertical? -They can kind of get the idea that there are both colors and both lines but if you ask them whether all the vertical ones were red or were all the horizontal ones black or what was the one in the bottom most part of the screen was that a red vertical bar? People will answer those questions poorly because even though when you flash the display people feel like they perceive it accurately without spatial attention being focused on each of the individual stimuli you are actually not binding them together - so the world doesn't feel as fragmented to us subjectively as it actually is -If you do these experiments people will get which features were present in display but won't get which features co-occur at locations -When you focus your attention narrowly - you are focusing here and you can shift attention quickly (so looking here, here, here) and if display is up long enough you might be able to answer questions about any of these positions but if you flash this very quickly unless your attention is narrowly focused on that specific area you aren't going to be able to answer whether this is a horizontal red bar -BOOKS explanation: focusing your attention involves a trade-off: if you focus your attention broadly, you can take in many inputs at once, and this is faster (because you don't have to search through the items one by one) -but since you're taking in multiple inputs simultaneously, you may not know which feature belongs with which -in contrast, if you focus your attention narrowly, you'll be slower in your search (because now you do have to search item by item), but with information coming from just one input, you'll know how the features are combined -You can do these experiments and show illusory conjunction - if you have letters that are different colors and then you ask what color was the B and it was actually green they may say it was red or blue -They are taking some color that was present in a different part of the display and incorrectly fusing it with the letter because their spatial attention didn't have enough time to linger at that location and they are taking these two maps and just taking a guess at what might have co-occurred -When you narrow attention you fuse the features that are co-occuring at that attention when your attention is broad the features themselves are present but you don't know which features go together

Solving the binding problem through temporal synchrony of neural ensembles

-When you see it one way certain features will go together - like these two eyes are part of the same object -When you see it another way as two faces these two features are parts of separate objects -So if you had 4 hypothetical neurons that had receptive fields in these 4 locations - and these are just spike rates so lets say you found neurons in the early visual system that represented these locations - so this neuron responds when something is present here this one here, here - and these are their spike trains or how much they are firing - you can see that if your perception of this object happens to be that this is a single face and therefore these two eyebrows are part of the same object the two neuron's at that location are firing here synchronisly - the neurons are dancing to the same beat - they are firing at the same time -That's going to be an important characteristic that sort of binds these together -These neurons don't know to do this on there own this neuron is just sort of listening in at their location and this neuron is listening in to their location but attention in this case by processing this as a single object is causing those neurons to sort of exhibit synchronis firing which is binding these features together to be a common object -This is a little bit different from the spatial spotlight of attention - this is object based attention where attention binds together features that co-occur in a common object -In 3 and 4 you see these neurons are firing in synchrony meaning these two things are represented together -When you perceive it differently - you decide its two faces separated by a gap you break the synchrony between neurons 1 and 2 (so these neurons are now firing asynchronisly) and neurons 3 and 4 are firing asynchronisly -But neuron 1 and 3 representing 2 parts of this face are now in synchrony and neurons 2 and 4 are now in synchrony - so attention is allowing you to cause different aspects of your perceptual experience that are related in some way either because they co-occur in space (as you saw in the previous slide) or because they are co-occuring in a common object -To dance to the same beat or show this synchronized firing patterns and that's sort of our best guess as neuroscientists as to how attention binds together certain elements of the world and fuses them into a coherent percept

Reed and Johnsen evidence for proportional representation

-Which of the pictures on the next slide are part of this picture? -reed and johnson (1975) found that subjects only succeeded 55% of the time -suggested that visual images often lack sufficient details for direct comparisons of part-whole relationships -perhaps subjects don't really store images as "pictures" -They would argue that the fact that most people don't recognize this show that mental images are not as detailed as we think at minimum -Reed and Johnson found that individuals succeeded roughly half the time and they thought that visual images even if we have them and use them aren't detailed enough to be picture like -They lack these part-whole relationships -They argue that perhaps subjects don't really store images as pictures if they did, this shouldn't be as hard

Hierarchical Organization of visual perception

-With the visual system there are different levels you can study -The general terminology we adopt in the field to distinguish the relationship between visual input and the knowledge we extract from the visual world is divided into low level and high level vision and inbetween there's mid-level vision -For low level and high level - when we say low level we mean the basic features of the visual world (oriented bars/edges - if you think of an object as having different orientations ex. a vertical bar or horizontal bar - any object can be understood as a collection of lines that are going to be oriented in some way - they might be curved, they might be straight but these basic features are what we mean by low level - and then theres these other features like motion (is it static is it moving), texture and depth -High level vision which is extracting what it is we are seeing - getting beyond the basic features of these constiguent elements that make up an object and recognizing sort of is it a clock, a face, a letter of the alphabet etc. -high level vision: object/face/scene/word recognition -Extracting that higher level information involves integrating all this lower level information at a more extract way -visual input --> low level vision <--> high level vision <--> knowledge

Explanation for the blue and black dress

-You have to resolve when you are looking at this dress resolve the ambiguity by making one of 2 assumptions- you can assume its in the shadow -If something is photographed in a shadow the image is going to look a little bit darker - you know that things that are in fallen shadows tend to be darker and what we do is we mentally brighten them we assume oh its in a shadow so the thing in the shadow must have had a brighter color then it appears to me and you don't have to think about it - your brain is doing this correction on the fly -We adjust for shadows so people who may preserve it as white and gold may be under the mental assumption that its in a shadow and the shadow is darkening it so its appearing more like this but really the color of the dress is white and gold so that's what we preserve it as white and gold -You kind of mentally subtract out the shadow whereas if you preserve it in bright light and its just a blue and black dress that happens to be washed out by really bright sunlight which actually was the case in this photo -If your brain is making that assumption your going to see it in this darker format here that makes the blue look even bluer and the black even blacker

Questions of Cognitive Psychology

-how is our knowledge of the world represented? (how is information in the world represented mentally - how does our minds take things that we experience and form some knowledge that is kind of a symbolic link to things that are out there in the real world) -what mental processes are involved in acquiring, accessing, and using this knowledge? (what actions can our minds take to influence this to make sure we are acquiring the information we learn and access the knowledge we are trying to retrieve - that we use knowledge in a flexible way to guide our behavior) -how can we measure behavior to make inferences about mental representations and processes? (how can we create tasks that measure people's reaction times or their eye movements or their decision-making choices or their responses on questionnaires or navigation in the real world - to make inferences about what mental representations they have in their brain accessible to them how they are using those and what kind of processes they are engaging in - and whether these processes are dependent on one another or independent) -cognitive psychology lets us evaluate this as long as you have carefully designed experiments

Hemispatial neglect

-a disorder of attention in which stimuli or parts of stimuli presented to contralesional side (the side opposite the lesion) are undetected and not responded to -hemispatial neglect is most often associated with damage to the right parietal lobe -right parietal damage leads to neglect of the left side of space -neglect patients will therefore read only the right half of words shown to them: they'll read "threat" as eat, "parties as ties, -if asked to draw a clock, they'll likely remember that the numbers from 1 to 12 need to be included but jam all the numbers into the clock's right side -There is a phenomenon with brain damage particularly to the right parietal area of the brain that causes a failure to attend to the other side of space -What we call the contralesional side so the other side of the lesion -This is a disorder of attention where you are not paying attention to anything that is on the opposite side of your brain -This just shows you patients who have this problem (mostly with right sided damage) are going to fail to attend to the left side of space -If you ask these patients to copy shapes like this you see things like this - they will copy approximately half of it and then will leave it out and look at it and think its complete - it doesn't look wrong to them -Its not like they cant see whats on the other side of space its just their attention doesn't grab them it doesn't occur to them that it would be useful to look over there - the same with this house here - they are leaving out a window and the door and the other side of the wall -That's with copying -Spontaneous drawing if you just asked them to draw a face you see here they are leaving out an eye and the hair on this side -If you ask them to spontaneously draw a clock they start with 12 here at the right space and numbered it down here

What are visual images?

-a kind of mental representation, sharing certain properties with pictures: 1) preserves metric spatial information (the spatial layout - the relationship of different features in the image to each other) 2)changes with viewpoint (when you are looking from one perspective you experience one view and then you can move yourself to a different perspective and it looks different) 3) empty space explicitly represented (if you are imagining something and there is a whole section of the image where there is not a lot going on we don't compress it and conjure up the details of the parts that have objects and details in it - we represent the empty space) 4) experienced using spatial attention (the same mechanisms that allow us to shift our attention to different locations in space - also are important for shifting our attention around when we are visualizing an image of something that is not present in our environment) -mental imagery may involve any of the sensory modalities (imagine a sight, sound, taste, touch, or smell)

Representation

-a set of objects that stand for another set of objects by virtue of having the same causal relational structure (how these representations relate to one another in the environment) -we sort of have the represented world which is happening in objective reality and the representing world which is happening in our minds and there is going to be some mapping so certain things that are happening in the world that we are perceiving through our senses through our sight our hearing or that we infer to be sort of out there even if we can't observe them -we can then create mental representations and these could be one to one mappings where sort of this item A in the world (that clock in the back of the room) and then I represent it somehow in my mind - mental representation of a clock or they could be many to one - multiple things that are grouped together -or one thing represented in multiple ways - the clock could be sort of a visual object and it could be my knowledge about what time it is and my sense of how much time I have left to finish the slides for today and those might be represented differently -we have representations in our daily environment all the time but we also have those same representations in our brain -ex. stick figure could represent a person

"THE CAT" and bigrams

-a similar mechanism explains why an ambiguous stimulus can be perceived as an "A" in some contexts and an "H" in others -its unlikely this is an A here because AE is an uncommon bigram in the English language where is TH is very common and HE is very common -likewise this is probably not an H because even though CH is a common bigram HT is much less common

Parallel processing

-a system in which many different steps are going on simultaneously -ex. one advantage of this simultaneous processing is speed: brain areas trying to discern the shape of the incoming stimulus don't need to wait until the motion analysis or the color analysis is complete -instead, all of the analyses go forward immediately when the input appears before your eyes, with no waiting time

Primary sensory cortices and association cortices

-another important distinction when talking about brain areas is thinking about the type of sensory information they receive -certain regions of the cerebral cortex serve as the "departure points" for signals leaving the cortex and controlling muscle movements (motor cortex) and other areas are the "arrival points" for information coming from the eyes, ears, and other sense organs -each of the senses has a primary sight where the information from the sensory organ (eyes, ears, nose) arrives and then there is secondary or association cortices that are near by -so for the visual processing the primary region is in the occipital lobe and then near V1 is all this additional cortex here that is going to process visual information but at a higher level so that the information is initially arriving transmitting from your retina to V1 - and then these other areas will extract more abstract understanding of what you are experiencing -so V1 is detecting some of the basic understandings of line and contrast and orientation of bars of light as you get further here you are processing shapes, angles, depth and eventually things like faces, letters, words, objects - that's all visual association cortex -same goes for auditory processing - you have primary area that is receiving input from cochlea and then there is all this auditory association areas that might be able to detect not just what the frequency of sound is but things about pitch and whose voice it is more complex higher level processing of that auditory single is happening in those near by cortices

Anatomical terminology

-anterior/rostral: term for front of brain -rostral = towards the nose -posterior/caudal: term for the back of the brain -caudal - towards the tail -left and right: left and right of the brain (hemispheres) -superior/dorsal: upper part of the brain (dorsal fin of dolphin - on top) -inferior/ventral: lower part of the brain (ventral = towards the belly) -lateral: sides of the brain (from top down) -medial: middle of the brain (from top down)

The Primary projection areas

-areas of the body that we can move with great precision (fingers, lips) have a lot of cortical area devoted to them; areas of the body over which we have less control (e.g. the shoulder and the back) receive less cortical coverage -the primary motor projection area is located at the rearmost edge of the frontal lobe, and each region within this projection area controls the motion of a specific body part -the primary somatosensory projection area, receiving information from the skin, is at the forward edge of the parietal lobe; each region within this area receives input from a specific body part -each one of these areas - these somatosensory cortex and motor have this topographic representation (maplike) of different parts of the body -its not organized in the exact same spatial arrangement of your body -its not like from head to toe per say -the tongue is represented down here and other parts of mouth and face are represented over here -its actually a big part of this here for experiencing sense of touch as well as for movement is devoted to facial movements and facial sensations and those are very important for us for the motor cortex for speaking, articulation, eating, for taste and all the sensations we have in our face -but then these parts up here are devoted to your arms, hands and at the very top and kind of tucked within the medial surface is your legs, parts of your body like this -and some of these are heavily represented - like your hands have a large amount of cortex devoted to them and this could increase even more like if you are a concert pianist or violinist you will have more neurons devoted to that part of your body and you can see that in measuring cortical thickness in an MRI scan -even just learning juggling for a few months can change the representation of certain parts of the body -but you can see its asymmetric some parts of the body are over represented others are under represented - your leg and arm don't have as much as your hands and toes -so if you were to take this and try to map it on to a sculpture you can see this asymmetry -proportionally speaking your hands mouth and tongue have much more representation in cortex - more neurons in your motor cortex and somatosensory cortex devoted to them than other parts of your body -they are weighted by importance or the degree to which we need fine control of those parts of your body in the situation of motor function or fine grained information about what we are experiencing in the domain of sensory input -cortical representations of information is weighted by "importance"

The role of similarity and search times

-as the similarity between target and distractor increases, so does the difficulty in detecting the target stimuli -factors influencing search: -similarity between the target and the distractors -similarity among distracters themselves (The similarity of distractors when they are all the same makes it easy) -its NOT about the number of features per se but just about the confusability of items -Its not just the number of distractors or features in general but how confusable the items are -But...the story is actually a bit more nuanced - there is another factor its not just the number of items present but the similarity between the target you are looking for (what you are allocating your search for - if im telling you to look for something red but if I tell you to look for something that has a lot of similarity to other features its going to affect you) - also the difficulty in detecting the target since the target is not a feature that can be summarized so simply (detect right angle, acute angle, strange symbol) that's going to also affect performance

Ancient Greeks and optical illusions

-as with many aspects of science, our knowledge of optical illusions can be traced back to the ancient greeks -the greeks appreciated the importance of visual angle -They knew if they wanted to create a structure of beauty like the Parthenon they couldn't just build it the way they wanted it to look with right angles and symmetry because it wasn't going to come out right -so what they realized is to an observer looking at structure if they want the structure to have even segments on each of these columns they need to take into account the visual angle -for an observer standing down here in order to make this segment look the same size as this - they actually have to make it a little bigger because its going to be a slightly wider angle then this one here - it is a bit more further away -If you wanted the temple to appear like this and built it intuitively the way you think you should it would actually appear more like this (it would bend towards you) -If you want to make it look like this and appear with the right angles and symmetry you actually have to build it like this (bent away from you) - there is hardly a single straight line or right angle in the Parthenon -The architects were aware that you had to add curvature had to change these sizes in order to create a structure that appeared to an observer in the 3D world to look proper -They are aware that if you want to create something that looks more beautiful you have to adjust for the limitations and quirks of the perceptual system

Inattention blindness

-attentional weakness -So one example in your book is from a task in which the subject is asked to hold their attention here - attention is not to be thought of where you are looking necessarily - the subject are holding eyes here but their attention is in this corner and they are trying to look to see if something is changing here - a faint change to this display -That's their goal according to instructions -Unbeknownst to subject this circle they are to stare at changes shape - changes from dot to triangle and then square and then afterwards they want to know did you notice this change - they were looking directly at it but because if they are doing this well their attention is in the corner and not in the center - so they might miss it -If there are no warning that something might change 90% of subjects failed to detect the change -If they were warned that something could change most people did detect the change - they divided their attention because they knew they might have to report that -Usually we aren't aware of these things and here the person was looking right at it and missed it

The Deutsch-Norman Selection Model

-attributed to Deutsch and Deutsch (1963) and Norman (1968) -proposed a second stage of selection that occurs later in the processing sequence -late selection model -selection at this later stage is based on the semantic content of the message -"it is as if attention provides the key that unlocks the gate dividing unconscious perception...from conscious processing. Without this attentional key, there is simply no awareness of the stimulus" -They proposed that theres this second stage of selection that occurs later in the processing sequence and this is a so called late selection model -Triesman's is still early but instead of filtering you attenuate but here they said well information gets through and they are trying to account for the same sort of data why people would complete that peanut butter sandwhich for ex. - why is all this information getting through the unattended channel and they said well really we sort of implement the selection later based on the sementic content of the message -So if the sementic content is relevant to you or congruous to whats happening in the unattended ear then you will select it -Their model can also account for that phenomenon because now you are sort of letting everything in and filtering things at this later stage -All the information gets processed perceptually and then there is sort of this late stage filtering which can be based on which ear its coming from but also based on semantic information

Testing as Learning

-being tested for information serves to boost memory -retrieval is a form of learning -learning effective mental routes to recover the information -strengthening the memory trace -weaving new contextual details into the memory -The Testing Effect: the deep processing needed to take a test helps create longer lasting memories of the material -testing is really a form of generating information from memory -so being tested for information serves to boost memory • and we made this point on the first lecture that retrieval is a valuable form of learning that's why we have the quizzes • everytime you have to think about something we learned in this course and retrieve it from memory even if you don't succeed the act of searching for it and then I give you the correct answer and provide you feedback all of this will strengthen that memory and make it more likely that next time it comes it will be there in your memory right where you left it -this process of searching your memory and trying to find these connections or make these associations strengthens those connections as you do -it strengthens the memory trace and you can weave new contextual details into the memory -so if you got it wrong and I elaborate to why the answer was correct now you can take that old memory and make it a little stronger and add some new things to it - update it -so this is the testing effect and it's a very deep form of processing one of the most effective ways to strengthen a memory is to actually test yourself on it

Primacy effect

-better memory for the first few items -based on retrieval from long-term memory -The primacy effect is a little bit weaker so its this benefit you get for the early list items and the middle list items are generally the poorest recalled -The primacy effect really only includes the first few items and then drops off pretty rapidly -And this is based on retrieval from long term memory at the beginning of the list there are a few reasons why the primacy effect exists -One is you are starting fresh you don't have other words you are juggling in your mind so when the first word comes in you are like im going to remember this! Your motivation is strong and you get some chances to rehearse it and then the next word comes and now you are only rehearsing two words so you kind of go again - you are rehearsing those two -You are getting more rehearsal opportunities and there is less interference but as you get into the middle list you think wow this is hard, you are frazzled you cant cycle through them anymore in your mind so you don't get that benefit

Recency effect

-better memory for the last few items on the list -based on working memory -The recency effect sometimes includes the last 4 or 5 words -The recency effect is better memory for the last few items on the list -And the reason for this is that those items are still accessible in working memory -So we finish the list - I said recall and she can sort of go back to the words she was recently rehearsing or keeping active and they are still for the most part at least the last few there and they are rapidly accessible -So most people will dump those out of working memory first they are the freshest in their mind and when you say go they start with the back of the list and get those out of the way

Change blindness vs. inattention blindness

-both are failures of visual awareness -change blindness is the failure to notice that some aspect of a scene has changed -focused attention to a given feature is needed to notice that the feature is changing over time -It requires that there are two or more different options that are changing and when you are experiencing change blindness you need to search around till your attention gets to a particular place or feature to notice that it is changing - if you aren't paying attention to that feature carefully you aren't going to process the change -inattentional blindness is the failure to notice the existence of an unexpected yet fully-visible item -Unexpected could be you didn't expect there to be a shape change, didn't expect there to be this muller lyer like features that are connected to those lines didn't expect there to be a gorilla Not being aware of something that is right in front of you that you are perceptually processing because your attention isn't there you are focusing on something else -unless an item is distinct enough from its background to "pop out" focused attention is necessary to perceive it -inattentional blindness is similar to change blindness that people fail to 'see' an object -unlike change blindness, inattention blindness occurs while attention is engaged in some demanding task -where change blindness reflects an inability to identify how the visual world changes over time, inattentional blindness refers to the failure to notice that a fully visible item exists at all

Motor association cortex

-certain regions of the cerebral cortex serve as the "departure points" for signals leaving the cortex and controlling muscle movement -in everyone of these regions you see the information goes from primary to association except for motor cortex - motor cortext is not input sense its output -what actions are you going to take - so here primary motor cortext is area that is going to send signals through spinal cord if you are going to move arms or legs and sort of telling muscle groups what to do and how to do it -and the planning of this if you are thinking about how to swing your golfclub but don't actually do it that planning is all happening in your motor order association cortex -and all the other areas in grey are sort of higher order association cortices they are not associated with any particular domain but rather they process information at a higher level of abstraction -most of the prefrontal cortex is in this category it is processing abstract concepts and things about the world not tied to certain modality (look at picture on slides)

Reconstructing visual experiences from fMRI brain activity patterns measured during movies viewing

-clip of movie patient was watching in scanner and right a reconstruction of what they might have been viewing based only on brain activity pattern -did raise idea that maybe if we could do this on imagery when people are imagining what if we could do this: DVR our dreams? -In 2012 scientists read dreams: decoding 15 attributes about dream (whether person in dream or not etc...) not able to reconstruct it as a movie but could still see specific things -The neuroscientific evidence has showed that mental imagery at least in its representation is fairly picture like

Bigram detectors

-detectors of letter pairs -these detectors, like all the rest, will be triggered by lower-level detectors and send their output to higher-level detectors -and just like any other detector, each bigram detector will start out with a certain activation level, influenced by the frequency with which the detector has fired in the past and by the recency with which it has fired -to explain the word-superiority effect, we must add another layer to the network that detects bigrams, or letter pairs -here, only some letter "O" features were detected, but this is compensated for by the higher baseline activity of the "CO" detector -the bigram layer helps the system recover from confusion about individual letters - this basic network that started with feature detectors, then went to letter detectors, and then to word detectors had some limitations -the model was augmented by adding this extra layer which has this nice property - this is called bigram detectors, a bigram is just a pair of letters - two letters next to each other is a bigram and according to this model you would have bigram detectors for every possible pair of letters - ex. CA, CB, CC etc -if you see this word corn now we have SP4, all of the individual letters stimulating feature detectors, and those feature detectors are saying is this feature present or not, and each of the feature detectors are sending signals to other detectors but now we have these bigram detectors so theres the CO bigram detector, that's present in the word, these other ones aren't -if you flash this word quickly, not all the features are going to get processed perfectly, so what the quick presentation does is means you have a somewhat impoverished coding of the features and so you get a fill it in - fill in the blanks of what you missed by virtue of having these detectors here -so the CO detector has a higher baseline activity because CO is a common letter string in the English language and CU is less common and CQ is even less common -and so each of these bigrams just start with a baseline level based on how likely it is to occur in your language and then here just a little stimulation of the C (I think there was a C and there might have been an O) that is enough given this sort of higher starting activity level to make you guess that its probably a CO -this helps you recover from confusion - these bigrams help make best guesses based on their assumptions or their baseline state which is really an assumption that comes out of your experience with the English language to help you even when the stimulation is degraded and the stimulation could be quick presentation but it could also be blobs of ink over certain letters or could be viewing word over noising screen to show these effects

Problem #2 with early selection theory

-early prediction theory predicts no semantic processing of information from the unattended ear -but experiments have shown that some unattended words do get processed at the semantic level -Anne Treisman's experiment: • subject asked to pay attention to the right side: "in the picnic basket, she had peanut butter book, leaf, roof, sample always -But in this ear (unattended ear) it starts out with random words: cat, large, day, apple, friend, every, select sandwiches and chocolate brownies -She is asked to shadow this - if she gets it correct she should be saying just what has come out of the right ear but sometimes the subject will actually switch and start finishing the sentence with whats in the unattended ear -Participant wont even realize they are doing this - its just that this information coming from the unattended ear connects well semantically in terms of its meaning with whats happening in the attended ear -And so unattended words are not being filtered out as much as we thought -In the Broadbent model there was no way for accounting for this - this should have all been blocked out -Not making it to perceptual analysis she wouldn't make the connection that sandwhiches would be a good next word in this sentence because it never even got to that stage -And it does which implies that the information was in the system at a higher level

The propositional Theory's response to mental rotation

-elaborate structural descriptions can explain rotation results -angles of features are specified -e.g. partial description of mental rotation task -pylyshyn said well this result is great - it clearly is true that the degree of mental rotation is proportional to reaction time but that you don't need the analog perspective to get this result -he thought elaborate structural descriptions could explain the result -so even if you aren't thinking through this in your mind in words (like a 45 degree segment, on top of vertical segment) representing each shape as a set of angles and relationships -that this is the way your brain is dealing with it - when your brain perceives the shape it is parcing it into these connected segments with different lengths and angles and then the rotation is just this computation that you know when you mentally rotate something you increment the angles in a proportionate way and you just keep doing this iteratively -ex. it originally started as 45 degree segment on top of vertical segment 37 degree segment beneath vertical segment (then rotation) 46 degree segment on top of 1 degree segment 38 degree segment beneath 1 degree segment

Figure-ground

-elements are perceived as either figures (distinct elements of focus) or ground (the background on which the figures rest) -Is the background white? - good assumption here because everything around it is white and there's two black turtles -Another ex. the cup on a white background is more likely given the rest of the stimulus but you could also perceive the two faces -We usually quickly extract in any visual stimulus what is at the front and what is behind -your perception contains information about which part of the form is figure and which is ground - that is not contained within the stimulus itself -apparently then this is information contributed by you, the perceiver

The Self-Reference Effect

-encoding information with respect to oneself increases memory -another form of deep processing -so one way to make sure information is going to be memorable is to relate it to yourself -So again these are four different conditions -You get a series of words and one group of subjects you just have to decide are there capital letters or lower case, another task is do they rhyme with something (so these are both pretty superficial), this one is a little bit deeper "does it mean the same as" so now you are thinking about the meaning or the semantics of the word, even above and beyond this if you think does this word describe me - how does it connect with you now you get this huge advantage) -Don't need to worry about the yes or no here that is whether you say yes it describes me or no it doesn't - it doesn't matter so much if the answer is yes or no -In the yes case you get a little bonus but the bigger point here is regardless of whether or not the word describes you just thinking about if it describes you is enough to give you this huge advantage -So if you are trying to learn something and relate it to yourself youll benefit from the self reference effect

BOLD signal graph and viewpoint invariance

-ex. that shows in the brain something like this -when you see objects the first time - this is BOLD signal measured in two different regions -second time you see image the fusiform shows less activity which is neural priming -facilitated processing - its taking less BOLD signal so less activity to recognize the object the second time you see it -priming effect - reduced activity because you are processing it more efficiently -doesn't matter if it's the same view or different view and that's what you are seeing in this fusiform cortex that is processing identity of objects and faces -in occipital area that is processing basic features it activates this much the first time you see it - and then if you see it in the same view it shows priming so it is sensitive to the identity -if you see it in the different view its activating just as much as a new object -occipital cortex is not where you are getting this viewpoint invariance -when you see object in rotated orientation it responds strongly -the viewpoint invariance is clear by the time you get to fusiform cortex it doesn't care about orientation it is just interested in object identity where as occipital cortex is sensitive to lower level features - if it is the exact match you get priming and if not then you don't

The Attenuation Model (Treisman)

-formulated by Treisman (1964) -Unattended message is not blocked completely but attenuated -the likelihood of information getting through is determined by its threshold -threshold = minimum amount of activation required to produce conscious awareness -important words (e.g. your name, fire "help") have lower threshold -So treisman's revised model of Broadbent's model and instead of blocking completely the irrelevant input now she thought of it as a filter its kind of a faucet where you can adjust the level of flow -So the attended channel you want to keep it wide open so its full flow - and this information makes it through the register -This is attenuation where you can weaken the flow if you wish and then perceptual processing where you are extracting information about what you are hearing and makes it to short-term memory -But the modifications instead of this being an X that just blocks completely irrelevant input now it attenuates it - you just sort of dial it down reduce the flow so the unattended information doesn't get in very strongly - you sort of weaken it -Its not completely blocked and then in certain circumstances like the cocktail party effect where its something interesting to you or in that sandwhich example where the two halves of the sentence are semantically connected this theory allows that to happen -So even if the unattended information is processed weakly when it connects with something that is relevant to you you are more likely to notice it -And according to Treisman now you need to think about why it gets through and why sometimes it doesn't -She said well that must have to do with a threshold some level of processing of the irrelevant input above which you've become aware -So if information is sort of activated above threshold itll reach awareness itll make it on short term memory -If its below threshold it wont -Threshold: the minimum amount of activation required to produce conscious awareness -One idea here is that certain kinds of words ex. your name, names of loved ones, as well as other salient words, help, fire, words that grab your attention those just have lower threshold -So if suddenly in the unattended ear the word fire came up that would be more likely to make it to conscious awareness then if the word ceiling came up -And this is an illustration of that - the Y axis on this graph shows the signal strength (threshold) needed to activate -Something like your own name has a low threshold - low threshold means it doesn't take much to make it to awareness -Your name has a heightened likelihood of getting passed on because it has a low threshld -Something like rutabaga which is infrequent and not often relevant to you would have a high threshold and boat would be intermediate

Functional magnetic resonance imaging (fMRI)

-functional MRI is scanning it to look at brain function -measures the oxygen content in the blood flowing through each region of the brain; this turns out to be an accurate index of the level of neural activity in that region -lower resolution image compared to MRI but can get one of the images pictures (on slide) every second -The subject is lying in scanner doing whatever task researcher designed and every 1 or 2 seconds we get snapshot of what's happening in their brain -We can use this amazing trick that lead to this being possible which is the fact that when an area of the brain is more active or there is more blood flow it slightly disrupts the magnetic field in a way in which we can measure called the BOLD signal (blood oxygenation level dependent signal) -indirect measure of neural activity -which gives us this correlate (not directly related to neuroactivity) - when neuroactivity goes up the blood oxygen goes up and the fMRI signal is going to go up and we can measure these changes even though you cant see it by eye image by image -statistically whenever an area of the brain is more active the signal is going to increase because the blood contains hemoglobin that has iron which disrupts the main properties and gives us the BOLD signal -the results of an fMRI are highly variable, because the results depend on what task the person is performing -this linkage between the scan data and mental activity once again confirms a fundamental point: different brain areas perform different functions and are involved in different tasks -more ambitiously, we can use the neuroimaging procedures as a means of exploring brain function - using fMRI scans to ask which brain sites are especially activated when someone is listening to Mozart of when someone is engaged in memory rehearsal -in this fashion, the neuroimaging data can provide crucial information about how these complex activities are made possible by specific brain functioning

Gradual recovery from neglect

-here is a nice illustration of someone that was an artist before he got his severe right parietal lobe stroke - as he recovered from the stroke -you see this patient recovering after stroke - 3 months after stroke his self portrait shows really strong signs of this right sided bias - he is neglecting a lot of details and this whole side of space is empty -and then 6 months later his portrait is already getting a little better and more complete -by 9 months it looks pretty reasonable but there is not much going on here -and by 12 months its symmetrical and a regular portrait in his artistic style -this is typical of neglect - these patients usually have very severe attentional impairments right after their stroke and for most people it recovers as the swelling goes down and the brain heal itself the attentional system can rebalance -and even if some areas are permanently damaged the attentional system gradually rebalances -very few people have neglect for their whole lives

Hippocampus and memory consolidation: the importance of sleep

-hippocampus "teaches" the cortex during sleep -We think the consolidation process critically depends on sleep -When you sleep your brain is not just processing random things - there are a lot of things that happen physiologically ad restorative properties that happen - but in your brain it seems that (this is just a schematic so there might be a bunch of cortical regions connected to lets say one memory - the hippocampus is initially linked to them then during sleep the hippocampus is entraining these cortical regions through these slow waves -The hippocampus is activating these nodes in a synchronis fashion and that activity that synchronizes them makes the connection between each other closer -And eventually when you wake up (or after a few nights of sleep) the memory is consolidated and make it independent of hippocampus and fress up hippocampus to do other things -Its keeping track of recent experiences and all the different pieces and keeps them together and as you sleep it consolidates

Example of using both Working Memory and LTM to accomplish a task

-how is memory involved in doing mental arithmetic? LTM: -rules of arithmetic -learned strategies for solving problems Short term memory/Working Memory -holds information about the particular problem -applies the rules and strategies retrieved from long-term memory to the present information -transiently stores intermediate outcomes and final solution -if you get a problem to solve and its not a simple problem (2+2 which you can just retrieve as a fact) you usually need to retrieve in long term memory the rules of arithmetic -Like if it's a 2 digit multiplication problem where you have to mentally multiply and remember if you are dealing with 1's, 10's, 100's or carrying numbers so you are using these strategies you've learned -And as you are doing it in your mind because you aren't writing on paper you have to hold the intermediate terms in short term memory -Even just simple 2 or 3 digit subtraction if you have to borrow from the 10's column to the 1's column you are holding information in working memory and retrieving all this semantic knowledge -The factual knowledge about how numbers combine or subtract from one another and these strategies or algorithms for long division - like how you are going to take this 2 digit number and see how many times it goes into the 70 digit number you've learned strategies for doing that which is LTM and you are using working memory to keep intermediate products in your mind as you go

Cognitive subtraction

-in all of these studies a key experimental technique that we use is cognitive subtraction -have to take one mental state and measure activity levels during that and then we have control a base line state that has all the same components except one -if the experimental condition is passively viewing words we need a baseline -baseline is passive viewing of fixation cross (+) -so the difference between these is word processing -through clever experimental designs we can have task and experimental conditions that are well matched other than one aspect so we can measure the differences in one specific area -based on logic that we can find two tasks (experimental and baseline) that differ in components, to find effect of experimental component -ex. if you have a basket with apples and pears and a basket with just apples and you subtract them then you just have pears and that is your item of interest and you know how much pears contribute to that basket

Examples of the powerful role of context

-if I ask you to read these numbers (12,13,14) easy task -But if I had started with this and asked you to read these letters you would say A, B, C and yet this stimulus was the same -Another example like cat simulation where a stimulus that is perceptionally identical and in the bottom up stimulation of your retina it is no different when you see it in the 12,13,14 vs the A,B,C it's the same information coming in but your interpretation is different and the way you process it is going to be constrained by the context -So if you were expecting numbers then that fits in with numbers -You parse it differently a 1 and a 3 -There is a lot of processing in terms of reading that takes into account the surrounding letters and constrains your interpretation -Not entirely accurate letter order does matter -There are certain swaps that preserve your ability to read the words and other things you need -First and last letter are important but there are other things that affect your ability -Context does help -The letters within the words don't need to be in order for you to guess which English word would be the likely candidate there because of the context because it fits in with the previous word and the following word and makes sense semantically - meaning can be extracted so you make these guesses -There aren't a lot of other words that have these letters like university for ex.

The Human Brain External Structure

-if you look at a brain outside of the skull this is how you are used to seeing it (lateral view - looking at side of head with skull removed) -if you were to split the brain in half and look at the inside surface you would have to cut the section here it's the corpus collusum (big white matter bundle that connects the two hemispheres) -if you cut that and look at the inside surface of the brain youll see the medial view -top-down view: looking at the brain from the top down (from above)

Neural processing of illusory contours

-illusory lines "detected" by V1 neurons -because line is not present in stimulus, must be due to top-down feedback signals from higher cortical areas -neural response to illusory line is weaker and delayed relative to regular visible line -Its an edge that is not there and if you were to record from neurons in the early visual cortex -So lets say area V1 you can actually see - so recording from the first stage of cortical processing of vision - you can actually see activity when you experience this illusory contour -What you see here is neural activity over time - this is the time the stimulus comes on this is 400 milliseconds later - this is the average firing rate so just number of spikes per second - number of action potentials per second the neuron is firing -You can see that when a line is actually present - so there are four conditions -If this neuron receptive field is actually looking at a line on the screen the neuron has this big response that's this black line if the neuron has a grey square the edge of which is in its receptive field so its not as striking as this single black line it's the grey square against the white background it also has a strong response - not quite as strong because the grey squares lower contrast -If there is a white square against in this case probably a grey background so lower contrast you see again this big response here and then it fades off -This neuron detects edges - it prefers the line over the squares but also responds to the squares so these three conditions we have looked at are not illusory those are real edges that this neuron is seeing in its receptive field -The interesting thing here is the red case -If the neurons receptive field is listening in on this part of space there actually is no visual stimulation coming to the neuron - the signlas from the retina that are arriving from the V1 neuron are saying there is nothing here -There is no stimulation - its an absence of information in that neurons receptive field and for this neuron the only thing it knows about the world is if there is a stimulus present in its receptive field and so why does it fire -Why does it show this sort of increase in activity? -Even though as far as the inputs its getting there is nothing -And it has to be due to top down feedback signals so higher level areas that are perceiving all of this other information outside of the neurons receptive field are sending neurons back to areas as early in visual processing as V1 saying we think theres a square of about this size and then all of the individual neurons that are representing different parts of this in V1 gets stimulated from V2 for example -And so this is a feedback signal and that's why its delayed -It takes about a 100 to 150 milliseconds to peak where as these ones happen very quickly so the stimulus appears within 50 milliseconds you have responses -If there is actually something present but it takes another 50 to 100 more milliseconds if this is happening through feedback -The pointhere is that the fact that you actually experience something line like here is not just because your brain is making a guess that there is a square but actually that guess is causing neurons in your visual cortex (part of your brain responsible for basic visual perception) to start firing and its weaker - if I drew a black line here and this square had a boundary it would be more contrast -So you see this kind of subjective experience because of this neural activity

Illustration of top-down modulation of visual processing

-illustration: information arriving in early visual cortex -And then it is going to travel through this V4 pathway through the ventral stream this is one example of bottom up processing -Bottom up processing can also occur through the dorsal stream -And then different regions of the brain (Here it is illustrating parietal lobe and prefrontal lobe) can send signals down and these signals can be sent to different levels you can modulate processing early -At any level you can send signals that will influence processing -Look at this and see all the blue arrows are bottom up pathways going from the basic cortical processing (vision from V1) moving up to different levels V2, V4 and so forth and then every connection you see these feedback pathways which are top down connections going from higher levels of processing particularly here in the prefrontal cortex coming back down and influencing this

Loftus et al

-immediately after viewing these slides, subject filled out a questionnaire of 20 questions -for half of the subjects, question #17 was did another car pass the red datsun while it was stopped at the stop sign? -for the other half, the same question was asked with the words stop sign replaced with yield sign -some time later, shown 15 pairs of slides and asked to judge which image in each pair was the one they originally had seen -the critical trial was the one where these two images appeared -subjects given misleading information after encoding had false memories for the details of the visual scene

Split brain patients experiment

-in a patient with a split brain - the subject is asked to look at dot here -that ensures ring will be coming into their left visual field and key will be coming into their right visual field -they are looking at this screen and two words are presented and are asked what word they see -the patient when asked says ring here - the reason is ring is coming into their right visual field which is going into their left visual cortex -this word is being perceived by their left hemisphere and this word is being perceived by this hemisphere and the hemispheres cant communicate what word they experienced -in almost every example speech processing is dominated by the left hemisphere - the right hemisphere is incapable of speaking so if you ask the person which word did you see - the speaking hemisphere will be the one that answers and will say ring -the left hemisphere wont know that the word key was presented the right hemisphere will know that but cant speak -its kind of like these two brain in one head -if you ask the person to speak what they see they say ring but if you ask them to reach for what they saw and force them to use their left hand (the left hand is controlled by the right hemisphere) the right hemisphere only saw the word key and so the left hand which is controlled by this hemisphere is receiving information to get the key -the right hemisphere is able to read the word key its just not able to speak it but when you ask them to reach they grab it - really interesting phenomenon (look at diagram)

What happens if the connection between the two hemispheres is severed?

-in some cases people are born with corpus callosum that never develops -or it is surgically cut to elevate seizures -that gave us insight to what the hemispheres do -brings us to this contralateral organization which exists for both vision and motion -contralateral means on the opposite side of -the information from your right visual field is being sent to left hemisphere and vice versa -this doesn't depend on which eye you are looking at both eyes see all parts of the world -having two eyes is useful for other reasons but it is not what determines what information is getting to which side of space -each eye sees both halves of the world and here information is color coded according to whether its in your left or right side of space -the information travels both eyes are seeing both sides of the worlds but once you get to the optic chiasm (cross over) here the information is divided up -by the time you get to the thalamus its all lateralized (look at diagram)

Template theory

-in terms of different theories of how object recognition works - one prominent idea that has been proposed is this template theory -we have templates stored in long term memory and those templates are what we are going to use to figure out what is the most likely thing we are seeing -if we have a template for letter A and then flashed with stimulus that matches the letter A we compare the short term memory to our long term memory and here we have a match -this correlation between A and A will be higher than L correlation -you correlate short term memory with every possible template in long term memory and the one that has the highest correlation is the best guess of what you are experiencing -representations are mental images -those in LTM are templates -matching process is correlational -overlay input and template -highest correlating template wins

Somatosensory cortex

-information arriving from the skin senses (your sense of touch or your sense of temperature) is projected to a region in the parietal lobe, just behind the motor projection area: this is labeled the somatosensory area -motor cortex and somatosensory cortex is divided here by this boundary between the frontal lobe and parietal lobe

Sensory Registers

-information from the different sensory modalities is initially stored in separate sensory registers Iconic memory is a visual sensory store: -short duration (less than one second) Echoic memory is an auditory sensory store -longer duration (several seconds) -The visual sensory store has this very limited capacity to hold information in mind for less than a second without attention -So if I flash a stimulus on the screen and it goes away for the next 3-500 milliseconds you have the ability to access it in a pretty good way that's known as iconic memory -Iconic memory is just this lingering buffer or transient representation of what you just saw and it really is very short lived -This is way less than that 20 sec short term memory capacity -We are not thinking of this as STM this is just the sensory register how long your visual system holds on to something you just saw regardless if you were paying close attention -Echoic memory is a bit longer - could last a few seconds -This is just if I say something and you were spacing out and they were like what the heck you weren't paying attention and I say yeah I did and repeat back what they just said even though I wasn't paying attention at all -Its in your echoic memory - your auditory system keeps a somewhat longer store independ of whether you were trying to encode it -This is sort of the first stage information getting into the sensory register then you are going to need this other ingredient - some degree of attention or importance for information to make it into STM

A Brief history of psychology

-introspection (late 19th century) -Pioneers: Wilhelm Wundt and Edward Titchener -concluded that the only way to study thoughts is for each of us to introspect, or "look within", to observe and record the content of our own mental lives and the sequence of our own experiences -methodically examined consciousness, thoughts, feelings, and perceptions Pros: important topics! "meticulously" trained (introspects had to be meticulously trained: they were given a vocabulary to describe what they observed; they were trained simply to report on their experiences, with a minimum of interpretation Cons: highly subjective (someone could describe having the exact same headache using very exaggerated terms and someone else could describe that same headache saying it is nothing - there is no systematic way to collect data), generally not falsifiable, observable, or measurable; some thoughts are unconscious Bottom line: does not follow the scientific method

The Modal Model of Memory: short-term memory

-longer lasting than sensory stores (18 seconds without rehearsal) -available to conscious access -processes to regulate flow of info to and from LTM -This is available to conscious access and will determine whether information in STM can make it into LTM and information into LTM can come back into STM

Resolution of the imagery debate

-lots of evidence for analog visual images -but propositional theory never completely ruled out -assumes that the mind can rapidly work with extremely elaborate structural descriptions

Corpus callosum

-major white matter pathway that connects the two hemispheres of the brain -the two cerebral hemispheres are connected by an important and dense fiber bundle called the corpus callosum this is a massive white matter pathway that connects from the anterior parts to the posterior parts - all parts of the brain allowing the left and right hemisphere to communicate with each other in a very fast and coordinated way

Second dissociation patient

-suffered stroke to right hemisphere frontal/temporal lobe region -reported difficulties in finding her way home, and memory problems for unfamiliar material -severe deficit in visuospatial memory -impaired at Corsi blocks (spatial span task) -normal auditory short-term memory -Now patient E.L.D. had damage that affected the visual-spatial buffer and spared the phonological loop -So example of a kind of task that this patient would be bad at is this Corsi block tapping task -In this task you get a series of blocks arranged on a table and the experimenter taps them in a certain order the participant doesn't get to see these numbers they are looking at it from the other side -The experimenter knows what the right answer is cuz they are looking that they are supposed to tap on this trial 2,8,4,7 and the subject has to repeat it back - so they have to keep these spatial locaations in mind in order -And this patient couldn't do it very well at all but had normal auditory short term memory

Other ways to improve memory: spaced vs. massed encoding

-massed encoding: table table table apple apple apple fair fair fair -spaced encoding: table...apple...fair...table...fair...etc -spaced practice is better than massed practice -why? -in massed practice, the context at encoding is similar for all repetitions -in spaced practice, the context will differ on each repetition -some of this context is likely to match what is encountered at retrieval time (i.e. more potential "retrieval paths") -other ways to improve memory is by spacing encoding: if you study a list of words and you study table table table apple apple apple fair fair fair so you mass it together vs. table apple fair table apple fair which is spacing it -this is going to be much better - now usually its even more dramatic then this -instead of studying all of my psych stuff then chem then history if I do a little psych a little history a little chem and just space it out which is sometimes called interleaving that also is going to be better -this interleaving is helpful -but even just waiting - instead of table table table saying table then wait a little bit and then study table again just spacing it apart will be an advantage -so space practice is better than mass practice -spacing is better than mass practice because -in mass practice the context is going to be similar: so really comes down to context with mass practice your context and internal state is similar: so you just say table table table your mind and your mood and environment is all the same -but when you space or interleave your context is changing - the surrounding items are part of the context your mood, your location -so as things differ you just have more and more contextual variability and more retrieval paths

Template + transformation approach

-matching process involves an initial transformation step: -mental transformations: changes in orientation to rotating a stimulus are difficult for template theory -rotation - rotating stimulus -translation - moving something to a different location in space -scaling - things that are large and small -then compute correlation -allows for invariance to retinal position, size, and orientation -problem: still not enough invariance -the challenge is one of invariance - the visual system needs sufficient invariance - needs to not be specific to a particular size, orientation, location needs to be able to recognize things in different settings (Lying down or sitting up) we need this invariance and the template theory struggles with it -we can recognize many different fonts even though some of them we may have never seen before - some of these letters are descent matches to the template but some of them are very different from what we are used to -so we go through a transformation process but it is not enough to account for all the possible templates

Single-unit recording (usually in animal studies)

-measure activity of a single neuron with electrode probe (map out the cell's receptive field) -part of what we know about the visual system - indeed, part of what we know about the entire brain - comes from a technique called single-cell recording -as the name implies, this is a procedure through which investigators can record, moment by moment, the pattern of electrical changes within a single neuron -when a neuron fires, each response is the same size; this is again the all-or none law -however, neurons can, we've said vary in how often they fire, and when investigators record the activity of a single neuron, what they're usually interested in is the cell's firing rate, measured in "spikes per seconds". -the investigator can then vary the circumstances (either in the external world or elsewhere in the nervous system) in order to learn what makes the cell fire more and what makes it fire less. -in this way, we can figure out what job the neuron does within the broad context of the entire nervous system -in a typical procedure, the animal being studied is first immobilized -then, electrodes are placed just outside a neuron in the animal's optic nerve or brain -next, a computer screen is placed in front of the animal's eyes, and various patterns are flashed on the screen: circles, lines at various angles, or squares of various sizes at various positions -researchers can then ask: which patterns cause that neuron to fire? to what visual inputs does that cell respond? -what kind of detector is a given neuron? is it responsive to any light in any position within the field of view? or is it perhaps responsive only to certain shapes at certain positions -with this logic, we can map out precisely what it is that the cell responds to - what kind of detector it is -more formally, this procedure allows us to define the cell's receptive field - that is, the size and shape of the area in the visual world to which that cell responds -If we continue on to the primary visual cortex and stick an electrode there and record from an individual neuron in the earliest level of cortical processing (area V1) and you show this organism a simple stimulus here (an oriented bar of light in visual field) we can measure the response -Single-unit recordings - when we do this we can move the stimulus around in space and understand what is this neurons receptive field (term for the area of space this neuron is going to have access to) -Not every neuron in the visual cortex receives information about every part of space if we are measuring here in the left hemisphere this neuron is only going to receive information about stimuli in the right side of space -You want to know where is this neuron sensitive to -This is what it would look like if you took a slice - this is the electrode tip coming into the cells - it is listening to a recording of electrical signals that are emitted from one neuron

Electroencephalography (EEG)

-measures the electrical activity in your brain -by applying electrodes to your scalp (its non-invasive) -its really good for timing - you can measure brain activity

The operations of mental processes

-mental processes are operations performed on mental representations -creation -storage -retrieval -use for action -transformation -evaluation -If you have some piece of information that you are trying to get from the world - first you need to create a representation so you experience the flower now the flower is sort of active - you have sort of activated the flower concept in your semantic network you've activated certain visual features and then you could store that into memory or you could retrieve details (try to remember what that type of flower is called) you can use the information for action (now I know where it is and what kind so now im going to pick it) -You can do transformations I could navigate around campus and think about how it looked as a map and then realize where I am and kind of shift it so now it is in my perspective -Evaluation - thinking about the information and deciding what to do about it -this all happens in our brain!

Behaviorism

-mid 20th century -Pioneers: Edward Thorndike and B.F. Skinner -studied stimuli and behavior almost exclusively Pros: reliable and objective measurement, many findings are still pertinent today (ex. if someone reaches for sandwich you would interpret that as they are hungry because why else would they grab it) Cons: dead wrong about psychology! -we have all kinds of cognitive processes: thoughts, feelings, and motivations -without appreciating this, many behaviors (e.g. language) cannot be effectively modeled -life is not always so objective -cant say behavior = cognition

The Muller-Lyer Illusion (distortions)

-they are the same length -in the real world you can experience this also - this line here which has these tails seems longer than this one here even though they are exactly the same -the illusion is so powerful it almost makes you doubt if the red lines are parallel - this must be tipping up and down a little bit -heres another one - this red line here is the same length as this one but we perceive it to look so much longer -The line between the outward-pointing arrows looks longer, even though both lines are the same length, as may be checked with a ruler - though the ruler may appear distorted! The illusions in Figure 7 are associated with perspective, signaling depth in flat pictures. The outgoing fins of the Mu ̈ller-Lyer image may represent an inside corner of a room, the ingoing ones an outside corner of a building

Neural impulses

-neurons process information by electric impulses that travel down the axon -if the incoming signal reaches the postsynaptic cell's threshold, then the cell fires, that is, it produces an action potential - a signal that moves down its axon, which in turn causes the release of neurotransmitters at the next synapse, potentially causing the next cell to fire -action potential propagates along axon due to the entry and exit of ions through channels in the cell membrane -These neurotransmitters could be many different types - glutamate, dopamine etc - different neurons are sensitive to different neurotransmitters -They (neurotransmitters) will arrive at the dendrites and if they bind to the receptor channels they can cause the cell body to receive an influx of current and if it's a sufficient level the neuron will fire an action potential - so send this sort of electrical wave form that is transmitted across the length of the axon fairly quickly -to summarize this you have the input section of the neuron - the decision making - so if input is significantly high and reaches some sort of threshold level that the neuron is waiting for then it will decide in an all or none fashion (if the signal is sent - it is always of the same magnitude) to fire an action potential or not and if it does it will send a signal down the axon that is transmissioned within the neuron and then through the axon terminals -neurons transfer information to each other via axons (output) and dendrites (input) -when input to dendrites crosses a threshold value, action potential (all or nothing) fires and propagates down the axon -neurons have input that comes into dendrite it flows out of the axon which is the output it goes to the next neurons input which is the dendrite and basically when that input crosses a certain threshold that fires an action potential

Problems with template theory

-no perfect match in memory -sometimes we see stimuli that aren't a perfect match -so this one has a curved back so maybe more like pig but oriented differently from cat -theres no perfect match in memory in this cartoon and that's the case with everything you experience -very few things are perfect matches to things we have experienced before -too many views possible (would require an implausibly large database of templates stored in LTM) -the template theory is faced with this challenge that there is an impossible large amount of templates we would need to have and how do we deal with these different alterations

Top-down processing and bottom-up processing

-object recognition is powerfully influenced by the stimulus itself- that is by the features that are in view -processes that are directly shaped by the stimulus are sometimes called "data driven" but are more commonly termed bottom-up processes -the effect of context, however, reminds us that recognition is also influenced by your knowledge and expectations -this sort of influence - relying on your knowledge- is sometimes called "concept-driven" and processes shaped by knowledge are commonly called top-down processes

Change blindness

-observer's inability to detect changes in scenes they're looking directly at -why is it hard to notice the change (initially)? -when motion detection is disrupted, it is very difficult to observe changes to unattended image locations (When motion is disrupted by the flicker, it is very difficult to observe changes to unattended image locations) -brain makes reasonable assumption that things do not change unexpectedly (in the absence of motion cues) -Things don't usually disappear and reappear so we sort of assume that in general the environment will be coherent -When things do change we normally experience motion - so you could be looking at a field and a bird flies across it you notice that change -In this case you are showcasing a weakness in our perception that we don't notice a change when our attention isn't highly focused on one location -And this shows that our perception of the world is a lot sparser than our subjective experience of seeing suggests -humans' interpretation of the visual field is much sparser than the subjective experience of "seeing" suggests -we have trouble detecting changes in our environment - changes that we don't attend to -What these illustrate is that even though when you see the picture it just seems like a coherent image of all the different features but really the only features we process well are whats in our narrow focus of attention - so your spatial attention is darting around you have sort of an illusion of completeness but really the way your perceptional system you have fine grained detail and good processing of a narrow area and then your attention shifts - maybe this time you did notice the box but you didn't hold your attention there long enough and you move over here and moved over here and eventually you get back here and so when your attention is specifically at the box the change

Processes

-operations that transform one representation into another -they act on representations and are what allows us to transform representations -allows us to make sense of the representations -this should be intuitive that there is some world out there that we believe exists independent of our minds but the only way we access it is through these mental processes -we have our senses and have all these cognitive apparati - our attention, perceptual analysis, language, and memory and those will take the information out there and try to organize them in some way and we can engage different kinds of processes that operate on the information and transform it -how we change one representation to the other -ex. mental rotation task - you have some sort of object and you mentally rotate it in your brain - mentally rotating them (that act) is a process

Orientation-specific visual fields

-other cells fire at their maximum only when a stimulus containing an edge of just the right orientation appears within their receptive field -some of these cells fire at their maximum rate when a horizontal edge is presented; others, when a vertical edge is in view; still others fire at their maximum to orientations in between horizontal and vertical -note, that in each case, these orientations merely define the cells "preference", because these cells are not oblivious to edges of other orientations -if a cell's preference is for, say, horizontal edges, then the cell will still respond to other orientations but will respond less strongly than it does for horizontals -specifically, the further the edge is from the cell's preferred orientation (say, a vertical edge for a cell that prefers horizontal), the weaker the firing will be, and edges sharply different from the cell's preferred orientation will elicit virtually no response -One of the things you can do is see maybe it has orientation selectivity maybe it responds more to vertical bars then horizontal bars or diagonal bars so you can map out which orientation it prefer -Some cells will respond more to like this one likes the vertical bars, the ones just off vertical they will fire a little less and the horizontal bars they wont fire barely at all -Any neuron in the brain we record from an individual system we can give the animal different kinds of stimuli and see, does it like things that are static or moving? Or does the neuron not care about that aspect of the visual world -You can change orientation, contrast, special frequency and start to figure out what kinds of features the neuron is extracting

How do we make sense of the visual world?

-our perception of the visual world is organized in ways that the stimulus input is not -the organization must be contributed by the perceiver; this is why the perceptual whole is often different from the sum of its parts -some years later, Bruner voiced related claims and coined the phrase "beyond the information give" to describe some of the ways that our perception of a stimulus differs from (and goes beyond) the stimulus itself -people resolve ambiguity in everyday situation -to do this, we rely on a few basic principles -ex. fruit bowl: its almost certain that you perceive segments B and E as being united, forming a complete apply, but not that this information isn't provided by the stimulus; instead its your interpretation -its also likely that you perceive the banana as entirely banana-shaped and, thus, continuing downward out of your view into the bowl, where it eventually terminates with the sort of point that's normal for a banana -similarly, surely you perceive the horizontal stripes in the background as continuous and merely hidden from view by the pitcher -even with this ordinary scene, therefore your perception goes "beyond the information given" and so the unity of the two apple slices and the continuity of the stripes is in the eye of the beholder not in the stimulus itself

Downside of the word superiority effect

-over-regulation -one downside to this organization is that it leads to errors of over-regulation -here, the presented stimulus is "CQRN" but is likely to be misread as "CORN" -However, the network's biases usually help achieve correct perception -but there is one potentially adverse consequence of this although I would argue that it is adaptive which is when you see something like this (CQRN) most people will say CORN and this is known as overregulation -because the CO detector is so much more likely - has a higher baseline state then the CQ given somewhat ambiguous or degraded input the system will make the guess and from a reasonable guess most of the time you are going to see the word CORN with an O and not with Q and so you might misread this as CORN -the networks biases are usually there to help achieve correct perception but there are interesting cases where it can go wrong because the assumptions you are making here is that this would be a real English word when in fact its not -the basic idea is that the network is biased, inevitably favoring frequent letter combinations over infrequent ones -in effect, the network operates on the basis "when in doubt, assume that the input falls into the frequent pattern" -the reason of course is simply that the detectors for the frequent pattern are well primed- and therefore easier to trigger -the bias will pull the network toward errors if the input happens to have an unusual spelling pattern, but these inputs are less common in your experience -hence, the network's bias necessarily helps perception more often than it hurts

The word superiority effect

-people are more accurate in recognizing a letter in the context of a word than when the letter is presented in isolation, or in the context of a non-word -If I just flash a letter on screen quickly and ask whether it was a d or a K youll get it right some percentage accuracy but if I flash a word on the screen youll do better in detecting that there was a D or a K -Going to do better in determining whether there is a d or a k in one with a legit word (load) rather than nonsensical word (lxvd) -This example is referring to the latter part of this definition in the context of a non-word but this also works with a single letter -Do better in recognizing if d or k in the word load rather than when just the letter d is flashed on the screen -That's because the context - the letters that are nearby here are going to help you to know what letter strings you just saw -if d is flashed too quickly for you to recognize all of the features you don't have much else to go off of but here you have a higher probability if your L detector captures that there is an L and your O detector captures that there is an O it helps prime this D detector to think D is a likely ending of this LOA___

The importance of context

-perception is not automatic -extensive processing is needed to build a coherent percept -Top-down processing occurs involves making inferences based on past experiences and/or surrounding information -conceptually-driven as opposed to data-driven -Perception is not this automatic process where information comes into our eyes or any of our senses and then just perculates up through these different levels of analysis where we extract information that we are getting from the world -we are always making guesses or assumptions about what is likely to be experienced in a particular context and that makes perception much more efficient then it would be if it was just this purely bottom up (Meaning incoming information is coming in and getting analyzed at the V1 level and then going onto V2 and then V4) there is bottom up signaling - information does come up this hierarchy but there is also at every level this sort of top down signals where higher levels of processing are constraining your information about what is happening at the lower level -You need extensive processing and top down processing is going to allow you to take your beliefs, expectations, past experiences surrounding contextual details all of these sources that could guide your perception and potentially influence or modulate what is happening at a lower level -We refer to this as being conceptually driven rather than data driven

Law of Pragnanz

-we tend to perceive any given visual array in a way that most simply organizes the different elements into a stable and coherent form -whenever faced with visual stimuli we assume that the simplest explanation is the most likely - useful because allows us to make quick assumptions

What illusions teach us about perception

-perceptions are indirectly related to the objective world -perceptions are guesses (predictive hypotheses - often are wrong) -phenomena of illusions can be used to discover principles of perception -Physiological adaptations: what happens when feature-selective neurons get tired out? -Bottom-up vs. top-down processing: effects of expectation/context -Cognitive rules and assumptions (heuristics): learned from our experience living in a 3D world -perceptions are indirectly related to the objective world they usually give us a good sense of what is out there but they can be wrong because they are guesses -we are using predictions based on lots of factors to make these assumptions about what we are seeing and they are often going to be if not completely wrong somewhat wrong and the illusions are really so cool to experience because they show us this disconnect between what we are seeing and what actually is there in the world -theres a few kinds of principles that we can discover - one is physiological adaptations - just by tiring out or fatiguing a population of neurons in the brain you can create certain kinds of illusions -bottom up vs. top down processing we will look at a little bit in terms of the effects of expectation/context -and then woven into this are just heuristics or these rules of them - things that we expect of the world that lead us to make certain kinds of assumptions

Flow of sensory input

-photoreceptors: take in the signal those are located in the back of our eyes in the retina and they consist of rods and cones --> bipolar cells --> ganglion cells (their axons end up going to the optic tract) --> optic tract --> LGN (thalamus: relay station for all sensory information: LGN is a subpart of that dedicated to vision) --> occipital lobe (V1) --> V5 -Rods: sensitive to low levels of light, they can distinguish different intensities of light (are color-blind) not as sensitive as cones in general -Cones: sensitive to color differences -there are three different types of cones: short, medium, long wavelength - each having its own pattern of sensitivities to different wavelengths -cones have another crucial function: they enable you to discern find detail - the ability to see detail is referred to as acuity - this explains why you point our eyes toward a target whenever you wish to perceive it in detail - what you are actually doing is positioning your eyes so that the image of the target falls onto the fovea where there are a lot of cones, the very center of the retina

Biederman Experiment

-pictures of geometric objects -priming methodology: -faster to recognize something the second time around (facilitated processing based on prior experience) -had subjects name objects twice; some rotated Hypothesis: -priming occurs because structural description in long-term memory is "active" -structural description does not change as object is rotated Prediction: -if structural description is the same when object rotates, then priming should be unaffected by rotating object between 1st and 2nd trial Measure priming: -reaction time difference between Block 1 and 2 -for identical and rotated versions of objects Biederman's Results: -relatively little effect of rotation on priming -mostly viewpoint independent! -Experiment: pictures of geometric objects -And used a priming methodology - the idea that we are faster at processing info we have already processed before -Having experienced something and having brain make judgement on it now our neural ensembles will be more efficient when seeing it -facilitated processing based on prior experience -they had subjects name objects -every object they saw twice - so second time they are doing the same task -sometimes they were the same orientation and sometimes rotated -the hypothesis is that priming is going to occur because structural description in long-term memory is active -when you see this lamp it is activating the relationship of the geons and next time you see it it is activating the same set of geons the same structural description -structural description just means which geons and how they are connected -the structural description does not change as object is rotated -prediction: if structural description is the same when object rotates, then priming should be unaffected by rotating object between 1st and 2nd trials -they measured priming by measuring the reaction time difference between Block 1 and 2 -they had identical trials so saw the same thing in block 1 and 2 and rotated versions so different orientations in block 2 -1st block you see objects like this and then second block you either see it the exact same degree of rotation (0 degrees) or you see it with some other rotation -this should be fast due to priming (0 degrees) and if processing is viewpoint independent it should be just as fast and if its viewpoint dependent and theory doesn't hold then it should be not as fast -heres the first time you see it - mean reaction time was at 825 seconds and round 2 it gets much faster -first time they say lamp and it takes them 800 milliseconds and second time they say lamp it takes them around 600 milliseconds -the question is what happens when it is rotated -there is a little bit of slowing not exactly as fast when rotated -in general though there is relatively little effect of rotation on priming -the answer is mostly-viewpoint independent -these data are interpreted as showing strong evidence for viewpoint independence

Short-term memory

-psychologists reserve the term short-term memory to refer to information that you hold actively in mind on a moment-to-moment basis -generally lasts less than 20 seconds -Short term memory is sort of whats in the contents of your mental workspace -Your current attentional focus or whatever information is actively available to you right now -And it generally lasts less than 20 seconds (but hard to put a number on it because it depends on the complexity of the information and how many different pieces of information there are) -Its really what you are holding in mind on a moment-to-moment basis -Finding nemo ex. -the only thing that dory can do well is use short term memory -When things are currently available to her in her moment to moment mental workspace she can keep a coherent conversation -As soon as she gets distracted it becomes apparent that none of that got laid down into long term memory -So dory fails to form new long term memories her short term memory is just fine -"not only is my short term memory horrible, but my short term memory is horrible" -to say that shows that you forgot what you just said a few seconds ago (more of a deficit in short term memory) -short term memory is pretty short -a central idea regarding short-term memory is that information is only available for a very brief period if not rehearsed -so you can with certain kinds of information particularly verbal information - like words, numbers - you can just keep rehearsing it as long as you have the motivation and focus to do so and keep it in mind -you can actually even store it in long term memory and be able to retrieve it later if you keep rehearsing it so you can keep certain kinds of things active -but if you don't rehearse its just something you are thinking about for a second and then your attention is turned somewhere else that information will be lost from short term memory -and you would need to have a long term memory stored in order to retrieve it

Neural evidence for early selections

-relevant evidence comes, for example, from studies that record the electrical activity of the brain in the milliseconds after a stimulus has arrived -these studies confirm that the brain activity for attended inputs is distinguishable from that for unattended inputs just 80 ms or so after the stimulus presentation - a time interval in which early sensory processing is still under way -apparently, in these cases, the attended input is privileged from the start -There is actually a lot of neural evidence for early selection - someone doing dichotic listening task and inputs are coming in the left and right ear - in this case sort of beats at different times (sound segments) -And you are only asked to attend to one of them - doesn't matter if you are attending to the right ear ignore the left or attend to the left and ignore the right -You can plot the electrical wave form here measured - this is measured at the scalp with EEG -Event related potential -EEG waveform of electrical activity and the time scale here is much faster than fMRI -So this is 0 when the sound occurs to 200 milliseconds so just a fifth of a second -So whether you are attending or not (attending in red - unattending in blue) -And when you are attending to a sound - the signal will start to get a boost relative to the unattended sound as early as 80 milliseconds so very soon after you hear the stimulus -This shows that before you have time to extract a lot of information about stimulus or make decisions etc your brain is already bias processing up for relevant information letting it through - processing it more strongly and reducing the amount of neural activity for unattended information -You can see this in the visual system too - fMRI study that illustrates task where subject would be looking at this fixation cross and trying to pay attention to something on the left side of space or the right side of space -So in this case the checker board is flickering and they just have to detect some very faint change in contrast but they are either told to attend to the left side or the right side -The visual input is the same that they are seeing a lot of stimulation or visual input on both sides the difference is just whether you are attending or not -When you are attending to the left side of space you are going to get more activity in the right -This image here is flipped so confusing - in this region here which is LGN of the thalamus the first stage of visual processing happening beyond the retinal system where the signals come here in the thalamus and then they are ultimately going to end up at the visual cortex area (V1) -You can see even at the thalamus this relay station attention makes a difference -When you are attending to the left vs the right you are going to get more activity - if attending to the left the right LGN will show more activity then when you are attending to the right -Everything is crossed in the way its processing - the left side of the brain is processing everything in the right visual field and vice versa -You see this even more so in the visual cortex -These areas here have an even bigger affect of attention - attention can affect processing at early stages and its surprising how early it can be

Magnetic resonance imagine (MRI)

-relies on the magnetic properties of the atoms that make up the brain tissue, and it yields fabulously detailed pictures of the brain -MRI tells us about the shape and size of brain structures, but they tell us nothing about the activity levels within these structures. -the dominant way that cognitive neurosciences are studying brain function -Put people in donut shaped magnet - very powerful -The subject basically lies on back - scanners are designed to scan any part of body -Put person so head is in the center here - they are experiencing strong magnetic field - going to line up all the ions in their body with the magnetic field and that is happening in virtue of lying in there - the scanner is always on and taking ions either with or against the magnetic field -The trick is you send a pulse of radio waves tuned to frequency of these hydrogen proteins and that is going to cause them to tip out of alignment -In the scanner ions line up and then we pulse them and tip them out of equilibrium and watch to see how long it takes them to come back -its the decay period in which ions go back to their state and through that we can create these images of the brain - can look at anatomy -MRI can look at anatomy of any part of body -Structural MRI's are useful for clinical imaging (diagnosis) - was there a stroke? is there bleeding in the brain? is there an area of atrophy? -the results of a MRI scan are relatively stable, changing only if the person's brain structure changes (because of an injury, perhaps, or the growth of a tumor)

The Baddeley and Hitch (1974) model of working memory

-replaced the concept of a unitary "short-term store" with a multi-component "working memory" system -they said short term memory is not just one thing (like information goes from senses to short term memory to long term memory - that is too simplistic) working memory is a system and it has this storage component (actually multiple storage components) that then interact and feed into this processing component which is what they call the central executive -There are these two systems they call slave systems: the phonological loop system that holds auditory verbal linguistic information in mind (this is something like rehearsing words or numbers using auditory representations)a -And then there is a visual-spatial buffer - this is going to be holding visual and spatial information in mind for brief periods of time -And these two slave systems hold information and send it on to the central executive which then can maintain the things that are most important (so if you have more information than you can handle which ones do you want to rehearse - do you need to do anything with them? Do you need to organize them) -So the central executive is kind of coordinating the process in the case of the phonological loop - the phonologocial loop just cycles through so if you remembering a number 5, 4,7,9,4,0,6,5,4,7,9,4,0,6 it just keeps looping the more numbers you have the more likely that youll mess up and by the time you get to the end of it you might lose the first one but this can continue to loop -But then if I told you to do something with that information like reorganize them into numerical order that would require the central executive to kind of manipulate information here

Search times and set size

-search times can be influenced by set size -just the number of items in the display -graph: -Typically this is what would be the hypothetical graph if you did this experiment and varied the set size (the number of items on display and there is only one target present and you are just looking at the response time you can get two different patterns) -You could get two different patterns -One kind of search the number of items in display doesn't effect response time (disjunctive) - and another search the number of stimuli on display does affect response time (conjuctive) -Because this search is in parallel the number of distractors in the disjunctive search doesn't affect response time -since the conjunctive search is in serial the number of distractors in the conjunctive search does affect response time

Taxonomy of memory

-short-term memory (a.k.a. working memory) -maintains and manipulates information relevant to one's current goals -NOTE: sometimes the term STM is reserved for simple information maintenance, whereas WM is used to imply maintenance plus manipulation -however we will use the terms STM and WM interchangeably -some psychologists make a distinction between short term memory - when a distinction is made short term memory is referring to pure maintenance so you flash a picture on the screen or present a set of numbers or letters or words and then several seconds later you test them - was this the picture you just saw etc -so simple maintenance -whereas working memory is maintence plus some degree of manipulation -so if I gave you a sequence of letters and asked you to repeat them back in alphabetical order that's not just holding onto the information that's working on it in this active sense -or if I give you a picture of some sort of colored dots and ask you to imagine what the mirrored image would be so you can mentally rotate or flip it in some way - so whenever you do that that's working memory -but for this course you can just think of them synonymously - not with this distinction Long term memory: -holds the stored record of prior experience -the vast repository that contains all of you knowledge and all of your beliefs - most of which you happen not to be thinking about (i.e. not working on) at this moment -LTM has a few different types... -Declarative memory = memories that can be verbally expressed (memories you can tell someone about) -Declarative memories have two different flavors that are dissociable in the way there supported by the brain -episodic memory: memory for specific life events or episodes ex. remembering your high school graduation, remembering wedding etc -just an episode that has a time and place (so spatial context: where it happened and temporal context: when it happened) -semantic memory (memory for facts and general knowledge: contextual) ex. knowing the capital of California -And sometimes we have both - sometimes you have memory for the fact and sometimes you have memory for where you learned the fact -Your like oh I know the answer to that and I know it because I was sitting in this room and experienced it in this way so sometimes when you retrieve something you have both semantic and episodic memory its not either or -They are just different expressions of remembering but both declarative -Nondeclarative memory = memories that cannot be verbally expressed -main ex. of this is procedural memory: memory for skills and habits; demonstrated by doing (knowing how to tie your shoes knowing hot ride a bike etc.)

Implicit processing of unattended hemifield

-shows that even information that is in the unattended hemiphal (the side of visual space) can be processed its just not consciously processed -in this experiment patient was showed a house like this and then a house like this -and were asked questions ex. What do you see? A House And now? A house Are they the same or different? The same Anything wrong with either? No Which one would you live in? They are the same Choose one The top one -patient picks the house that is not on fire 9/11 times -this illustrates that even though the patient was unaware that there was a fire drawn on one of the houses because it wasn't in their attended hemiphal still the information getting in the visual system influenced them unconsciously to think maybe that wasn't the better choice of house

Gestalt principles of organization

-similarity: things that are similar are perceived to be more related than things that are dissimilar -Things that tend to be grouped together that are things that will be similar looking (circles that are similar so we view it as a triangle surrounded by a backdrop of squares) -proximity: things that are close to one another are perceived to be more related than things that are spaced farther apart -we tend to perceive groups, linking dots that are close together -We view this object as a square and here we view it as three verticle bars -We assume the proximity here is that these two columns should be grouped together and these two columns should be grouped together rather than any other way -good continuation: elements arranged on a line or curve are perceived to be more related than elements not on the line or curve -we assume that the curve lines is one object and the straight line is another object because objects in world have proper continuation -closure: incomplete objects will tend to be perceived as wholes -we assume that objects in the world based on our experience of how things really are in the world tend to be closed -they have defined boundaries -this triangle here that doesn't exist that is only implied -we assume that the circles should be closed and the triangle should be complete and what are the chances you just have these pac man figures arranged to make the triangle figure -simplicity: we tend to interpret a form in the simplest way possible ex. we would see the form on the left as two intersecting rectangles rather than as a single 12-sided irregular polygon

Object recognition in the brain: the dorsal and ventral streams

-starts at the primary visual cortex (V1, V2, V3) then goes to from V4 to IT along the inferior temporal lobe is the ventral stream -Its responsible for inferring what it is we are experiencing - we call it the what pathway: Recognizing objects -If you move visual information in parallel (starting at V1, V2, V3) will get processed in the occipital cortex and travel along temporal lobe to figure out what you are experiencing (ventral stream) -And travel (starting at the primary visual cortex or V1, V2, V3) along the dorsal occipital to the parietal lobe to figure out where the information is - to extract spatial details to help localize it to approach it and reach for it and to know how it can be distinguished in space form other objects (dorsal stream) -So this information seems to us consciously to be highly integrated - that's wrong - the spatial information and object identity information is not happening in the same area -parallel processing remains in evidence when we move beyond the occipital cortex (primary visual cortex) -this pathway, often called the what system (ventral stream) plays a major role in the identification of visual objects, telling you whether the object is a cat, an apple, or whatever -at the same time, activation from the occipital lobe (primary visual cortex) is also passed along a second pathway, leading to the parietal cortex, in what is often called the where system (dorsal stream) -this system seems to perform the function of guiding your action, based on your perception of where an object is located - above or below you, to your right or to your left -information from the primary visual cortex at the back of the head is transmitted to the infertotemporal cortex (IT) or the what system or the ventral stream, and to the posterior parietal cortex - the where system (the dorsal stream)

The Generation Effect

-subjects learn pairs of words, either read them: rapid-fast, lamp-light -or generate the second item in the past: rapid-f___, lamp-l___ -LATER: given a cued recall test, and people do much better with generation than mere reading -Another way to engage in deep coding that is very effective is the generation effect -Subjects can do a task where they learn pairs of words and they later will be tested - the word rapid when it is presented is paired with fast - the word lamp is paired with light and later they will get a word and they have to pair it with its associate -So this is paired-associate learning -In one condition they learned the word pairs like that and then in another condition they had to generate the word pairs themselves: they don't see the word fast they have to generate it so they see the letter f___ and they have to pick the word that goes there and people will say fast cuz it's the most likely completion -The act of generating it - thinking about what would go there and coming up with a solution is going to strengthen the memory - so when you are forming this memory even though you aren't seeing the word on the screen the generation effect will ensure this deep encoding -And later when you are given a cued recall test (you get one cue thats the first word like rapid and you have to come up with the associate) if you compare subjects who learned this way rather than just reading them subjects will do much better if they generated the associate themselves -so the generation effect is very related to the testing effect

Mental imagery can be sufficient to trigger context reinstatement

-subjects learn words in a distinct room; tested in 1 of 3 conditions: 1) same context 2) different context 3) different context, but imagine encoding context -Result: imagining the encoding context almost as good as being there -so this seems like it could undermine a lot of things because we are always learning in one context a pilot might be learning in a flight simulator and then they are tested in the real thing -and just this mismatch of context could be a real problem -but it turns out that if you are tested and you are not in context its not the end of the world because all you need to do is imagine yourself in the context -in this experiment there was three different conditions so subjects either learned and tested in the same context (they remembered 18 words on average), learned and tested in a different context (they remembered 12 words), and if you learn in one context and are tested in a different context but right before the test you are told to imagine yourself in that room and then take the test that is enough that imagining yourself back in this classroom might help you remember the information you are looking for so it just provides these roots to the memory

Experiment showing contextual effects on memory

-subjects learned a list of words -and they either learned it on land and were tested on land or they learned them underwater and were tested underwater or there was another condition where they learned underwater and were tested on land or learned on land and tested underwater -the data showed this striking effect - people who studied on land and tested on land do the best overall cuz we are used to that and you don't have all the distractions of the scuba gear and being underwater -but that wasn't what they were interested in - they were interested in when you look at people who studied on land when they are tested underwater they have this big performance hit -but people who learn underwater do best when they are tested underwater -if you study underwater and you come up on land and are tested you don't do as well -so you do better when the context of study matches the context of test

Feature Integration Theory

-suggests that we have distinct feature maps -e.g. orientation, shape, color, size -each map has info about the location of its feature (loaded without attention) -Suggests that we have distinct feature maps: in the visual system -there are different kinds of systems we can be sensitive to -Information about shape, orientation (angle of line/verticle/horizontal) color, size -E.g. orientation, shape, color, size -Each map has info about the location of its feature (where it occurs in space) -If you have map where red occurs it's a spatial map in your brain that indicates all the places in your environment - in space - in visual space where color red is present -You would have a different map for the color blue and a different map for things that are vertical -So you have these maps and these maps are processed without attention -As visual system is receiving information about surroundings you are automatically processing it through the different stages and neuron's that care about certain kinds of features -They are telling the next area up this features is present within my receptive field and those are assembled into a map so each individual neuron in early visual cortex might only be sensitive to one kind of line orientation and one part of space - but as that information feeds forward and gets processed you end up getting a spatial map -in an isolated feature search (disjunctive search, you only have to consult your feature map - EX. If I put this display up and ask which one of these displays the black oval it'll be easy you'll notice this one where as in a combined feature search you need to bind multiple maps - this is what makes it take longer -in a combined feature search we must bind information from 2 (or more) maps -binding the 2 maps requires attention, which takes longer -Now you have to find the presence of two features -Its not enough to identify if there is an oval or an off diagonal line - you want the co-occurence of those two features - you need your orientation map and your color/shape map to be aligned in space - need to look do these two features occur at this location (no-only this the vertical line does) -As you shift your attention through space you are sort of binding these maps together and that takes longer

Graph representation of double dissociation

-suggests that working memory is not a unitary function -And this is just a scematic of what the double dissocation looks like so if you take these two patients cases of what can happen in the brain it shows that you can get damage that impairs one and spares the other and you can get damage that spares the first and impairs the second -With this patient impaired auditory performance good spatial with this one good auditory performance impaired spatial

The neuron

-the basic parts of a neuron are: -dendrites, which detect incoming signals from neuron to neuron (input) -cell body, which contains the nucleus and cellular machinery (integrates everything together) -the axon, which transmits signals to other neurons (axon terminals contain vesicles with neurotransmitters) -Synapse: gaps between presynaptic and postsynaptic neurons (those gaps are what have the neurotransmitters which allows the neurons to communicate with one another) -Not all neurons look like this but this is a nice textbook style neuron where a nice axon seperates the cell body, there is the nucleus, this is the part of the cell where metabolic machinery is and then the axon allows the neuron to send signal to some other areas - it could be several millimeters away or could be several centimeters or inches (some of these axons are pretty long) and the axon will have terminals -The dendrites are going to be the input area for the neuron where its listening in to see how other neurons are communicating with it -What neurotransmitters might arrive at these dendrites to tell the neurons something about what's happening in another part of the brain -And then the cell body that contains the nucleus and the axon which is going to transmit signal to other neurons -the main flow of information through the brain - from the sense organs inward, from one part of the brain to the others, and then from the brain outward - is made possible by the neurons -neurons have three major parts: the cell body is the portion of the cell that contains the neuron's nucleus and all the elements needed for the normal metabolic activities of the cell -the dendrites are usually the input side of the neuron, receiving signals from many other neurons. -in most neurons, the dendrites are heavily branched, like a thick and tangled bush -the axon is the output side of the neuron; it sends neural impulses to other neurons. -axons can vary enormously in length; the giraffe, for example, has neurons with axons that run the full length of its neck

Tissue types: Gray and white matter

-the gray matter of the brain contains the cell bodies of the neurons -cerebral cortex is gray matter (dark) -the white matter contains the axons and fiber bundles (Light) -they are what connects between the grey matter -axons are covered with myelin sheaths (act as insulators so signal does not lose strength - acts as a coating), which insulates and increases the speed of conduction -white matter is the parts of brain - if you cut skull lengthwise -you can see areas that look whiter - and those are areas of bundles of axons - these axons are covered in these myelin sheath -it is like insulating wire - so the electrical signals traveling through don't lose strength when traveling through -where there is a lot of myelin the tissue looks lighter and when there isn't a lot of myelin the tissue looks darker -so the grey matter is where cell bodies are heavily packed in - dense arrangements of cell bodies -gray matter: cell bodies -white matter: connections between them

Distrupted knowledge

-the network's knowledge is not locally represented anywhere; it is not stored in a particular location or built into a specific process -thus, we cannot look just at the level of priming in the CO-detector and conclude that this detector represents a frequent bigram, nor can we look at the CF-detector to conclude that it represents a rare bigram -Instead, we need to look at the relationship between their levels of priming, and we also need to look at how this relationship will lead to one detector being more influential than the other -the knowledge about bigram frequencies, in other words, is distributed knowledge - that is, it is represented in a fashion that's distributed across the network and detectable only if we consider how the entire network functions -What the network "knows" about spelling or what it "expects" or "infers" about the patterns it sees is not locally represented in any single detector, but rather is a property of the network as a whole this is an example of distributed knowledge -the information that the feature net knows about the world in this case about visual inputs and reading is not locally represented - each of these detectors is only doing a small piece of the processing but the networks knowledge is really a property of the knowledge as a whole and this is what is called distributed knowledge and most of information processing in the brain is kind of distributed -you have all these nodes looking out for specific features but the networks knowledge of the english lexicon and all the properties of that - the bigrams likely come from this sort of set of connections and the fact that this CL bigram has a certain activity level has to do with all the words that you've ever seen and how often the C and the L get stimulated simultaneously and so forth - each of these pathways has a certain connection strength and you cant look at any of these individual units here and know what the person is perceiving

What is cognitive psychology?

-the scientific study of the acquisition, retention, and use of knowledge -the scientific study of knowledge and mental processes -broadly speaking it is the scientific inquiry into the nature of knowledge and thought primarily in humans but not exclusively -it involves topics such as perception, memory, attention, problem solving, decision-making, reasoning, language and other things too like creativity, mental imagery -learning and memory -attention -language -decision-making -information processing -similar to how we do things everyday

Subcortical parts of forebrain

-the subcortical (within the cortex) parts of the forebrain include: -thalamus (relay station for nearly all the sensory information going to the cortex) -hypothalamus (a structure that plays a crucial role in controlling motivated behaviors such as eating, drinking, and sexual activity) -limbic system: more for emotional responses (amygdala) as well as for memory (hippocampus)

Qualities of mental representations

-they are formed in the mind -Here I am using mind in the loose sense they are really formed in the brain and mind and brain are sort of interdependent (not trying to put these as different concepts) -The sum total of all of our mental experiences are happening in our brains so that's really what I mean here -They make it possible to think about objects and events even in their absence -Homer here imagining this donut - he may have never experienced a pink sprinkled donut but he can imagine it and that mental representation is sort of a symbolic entity that is happening in his mind that represents something that could happen in the world or perhaps did exist in the world that he is remembering from a past experience -Representations are going to be reliably activated by objects/events in the world - if you see this red flower in the world then that is going to activate a mental representation of the flower - as im looking at each of you im getting information about your face your location and its activating a mental representation so theres this correspondance

Testing as Learning graph

-this experiment has two different conditions there is a study study condition and a study test condition -so subjects were given some information - I think they had to read a passage - reading comprehension test -and then they either got to read it again - or they only read it once and then they are tested on it -in study study condition they studied twice and then they are tested and in study test condition they study once then they are tested on it then 5 minutes or 2 days later or 1 week later they are tested again -so its really study study test and study test test -and when its only a 5 min retention interval its better to study it twice - this is what cramming does -if you study study study and rush into test everything will be fresh in your mind and studying is actually pretty good at keeping that information accessible because it was so recent -you do get an advantage studying twice for the short duration -but if you are tested 2 days later or 1week later you get a huge advantage for the study test so cramming is very good for remembering information minutes later but once you get to the next day the people who crammed are going to forget a lot of what they were cramming whereas the people who were studying and testing themselves are going to do better later on -itll give you a more durable long term memory

Example of Ambiguity (face mask)

-this image looks like a face at first with a mask but how many can see the other interpretation -there are two faces in there - you are trying to use your top down processes to constrain how could this be two faces -now that you experience it both ways hopefully this pushed those of you who were still struggling to the edge -it's the case of this image which is ambiguous and the visual system interprets it in the more likely way where a mask would contain a face

Short term memory study

-this is a classic study - where they showed accuracy on a task where subjects had to recall some letters or numbers and you could see here there is like 100% accuracy 0% accuracy and this is the recall interval - so how long it is between when the stimuli were spoken to the subject and when they had to repeat them back -if you just get the sequence of letters and numbers and subjects can rehearse them performance should theoretically be at this level (around 90% accuracy) -if you are allowed to rehearse it doesn't matter if you are waiting 8 seconds or 18 seconds the information will be there -but if you give those letters and have subjects count backwards - count backwards from 100 by 3's for ex - just that act of doing that counting in your head or allowed - its going to tie up your working memory system -it prevents rehearsal - and therefore the information that the subject was presented with (the letters or numbers) will just fade over time -without rehearsal by 18 seconds or so they are remembering only 20% so this shows without active processing or effort to keep it in this available mental workspace the information will fade -and the only way that you can retrieve it at that point is if you had stored a long term memory

Making sense of an ambiguous visual world

-this is a complex figure here (look at figure) - wanted to reiterate the confluence of top down signals that are coming from higher levels and influencing lower levels (conceptually different processing and bottom up) -so you have the environment like the real world you are trying to figure out what is out there -you have sensation your sensory organs that are trying to process information from the environment (and in the case of vision you are trying to look at the depth of things, other organizational principles like the gestalt principles, what is grouped together, and organize this into units) -your left retina is taking in, in this example some portion of this and you are trying to figure out which portions of this go together and group them and then analyze them into parts and figure out oh there is sort of a picture on the wall here so these are all guesses you are making from pieces of the image that you are interpreting and your top down processing is influencing that -if you expect there to be art on the walls if you realize this is a room and that a painting or photograph is likely that could increase your chances of interpreting that as art as opposed to something else that might be on the wall

The history of functional brain mapping

-this really started with Galen (129-200 AD) who came up with a theory of brain organization which is notable for trying to attribute certain cognitive functions to specific parts of the brain -his theory was based on the ventricles of the brain -he thought that this anterior ventricle controls perception, the middle ventricle controls cognition, and the posterior ventricle controls memory -the ventricles are just a fluid filled space in the brain -they have some important metabolic function but don't contain neurons and there is no information processing in the ventricles -a lot of things wrong with this theory - but was the first attempt to map cognitive processes to different brain locations

Visual imagery as a key part of mental disorders

-visual imagery is also affected by a range of disorders -PTSD (the imagery of negative experiences - these vivid memories of traumatic events - these are being retrieved from memory but they are very vivid in detail and much of the therapy patients have is to reduce that) -with depression it's the opposite - their ability to imagine times in the past where they had positive experiences or to envision a positive view of the future - an optimistic perspective - is hard for them to do

Memory is a reconstructive process

-we take whatever comes back to mind and fill in the gaps -retrieivng a past event engages the same brain mechanisms as imagining a future event -patients with hippocampal damage show an impaired ability to envision the future • But other details that don't come to mind we like to fill in the gaps and elaborate and sometimes we fill them in with plausible accounts with what could have happened • We are filling in or reconstructing with other likely things which can lead us to have false recollections • When you scan peoples brains and they are asked to recall autobiographical memories (past experiences of their life) these parts of the brian light up • But if you give people the same scan and ask them to invision something in the future (going on future vacation) its basically the same areas - so reconstructing details from the past and envisioning potential versions of the future really have a lot of overlap not much difference going on in the brain • Its heavily reconstructed and involves these mental imagery processes and semantic activation and other things that probably get some things right but a lot of the details might be different from what you experience • When patients have damage to hippocampus it doesn't only impair their ability to go back in time but also impairs ability to envision the future • Compared to controls amnesic patients have poor scores when imagining future scenarios • They aren't giving as many experiental details and the details aren't as spatially coherent

Decoding much finer-grained mental images from fMRI activity

-what if you could decode whether some subject was imagining the letter X or the letter O based on the regions of the brain that perceive the letter X or the letter O -if you do this in the visual cortex it cant be about a proposition its about the visual processing -train a classifier model to learn whats the brain pattern when you see the letter X or the letter O and then... -ask you to imagine X or O -so here they are training their classifier in perceptual regions - lateral/occipital cortex to learn whats the brain pattern -you test it when they are imaging X's and O's -in this experiment it was right 62% of the time significantly better than guessing -and shows when you are reconstructing this pattern in mental imagery it does closely mirror the picture like pattern -activity patterns in left lateral occipital region could predict what the subject was imagining 62% of the time

Illusory Motion (fictions)

-when you look at this stimulus which isn't moving at all this perception of motion is due to tiny eye movements known as microsaccades so if you hold your fixation in any one place -The image where you are fixating is static and you can only sort of hold that for a few seconds and then you kind of lose it and start experiencing the illusion again -But you have these very small eye movements that are almost imperceptible that are causing this to give you illusory motion

Adjusting the beam of attention

-you can think of attention as having a spotlight with different properties -you can hold the spotlight you can have it be more narrow or more wide you can shift it around and so forth -And this figure from the book illustrates that you can adjust the beam of attention you can either focus more globally (see skull) or zoom in more narrowly and see women in dressing room

Differences between left hemisphere of brain and right hemisphere of brain?

-you probably have heard the idea that the left hemisphere and right have different attributes - left more analytical and right more holistic and creative -there is a little truth to this in the sense that the two hemispheres when you study them in patients who have damaged to one or the other do seem to have bias' in the way in which they process information -if you take an image like this which looks like a giant H but the details are little T's and show it to a patient with damage (they have a local feature and a global feature) and you look at patients with in general damage to right or left hemisphere these are extreme examples of what they would draw -person with damage to right hemisphere is getting that it is composed of z's but not capturing shape of M well -patient with left damage misses detail but correctly captures the global form here -if your left hemisphere is damaged you are going to miss that analytical aspect - attention to detail so will be biased to process global form -and if your right hemisphere is damaged you may miss the holistic aspect - and using your left hemisphere will focus more on the fine grained details -with us we can toggle between both forms -this leads people to think am I more left or right brained -this article highlights a point that despite what youve been told there isn't evidence that some people are more right or left brained -there is a division of labor but most people use both hemispheres to do everything they do

Mental Imagery and Ambiguous Figures

Chambers and Reisberg (1985) -showed ambiguous figures for 5 seconds and asked for first interpretation -removed picture, asked people to form a mental image -then asked is there something else that it could have been? -Results: people were unable to discover a second interpretation from the image -then drew the figure from memory and could then find the other interpretation Conclusion: a propositional code may override the imaginal code in some circumstances -What this experiment shows is that peoples visual memory for it was good enough that they could draw it but the ability to look at the mental image and discover the other interpretation is not very good to do that you must be looking at the picture -So there must be something different between pictures and images and the images are not sufficient to be able to discover these ambiguous interpretations -So for ambiguous illusions you need to actually look at it to see the ambiguity - there propositional code over rides analogous code Another ambiguity example: -pick one of these animals and memorize what it looks like -rotate it in your mind by 90 degrees clockwise and memorize what it would look like -none of the subjects could identify the new animal produced by the mental rotation -So they thought if you could mentally rotate this just like all the anolog theoriests say we can do why cant we see the interpretation? -argued that images are intrinsically bound to an structural interpretation -evidence for the propositional view

Two types of search

Disjunctive search (isolated feature): ex. look for the dial with pointer that is not vertical - when all other pointers are vertical -single feature enough to find target (not shared with distractors) -done in parallel -not attention demanding ("Preattentive) -is looking for an isolated feature and that is enough to find the target without a lot of attention -This is done in parallel meaning the information about these features is processed in different parts of the brains - shapes, colors, and such and we can find the feature we want without having to search stimulus by stimulus -So its done in parallel meaning you can look at the entire image at once you don't have to scan your attention stimulus by stimulus to find it -So this is not attention demanding in fact its called preattentive Conjunctive search (conjunctive search): ex. look for the dial not vertical AND with a warning light on -A combination of features need to find the target -done by checking each item (serial) -demands attention -Whereas combining feature search where you need to find multiple features is done in serial so one object at a time -You have to scan your attention you have to pick a starting point and sort of scan through and check each one - does it meet the characteristics -So this demands attention

Summary: the hippocampus and memory

Encoding: -the hippocampus plays a critical role in the formation of new declarative memories (both episodic and semantic) -facilitates binding of many different types of event details into a coherent episodic memory -damage results in anterograde amnesia -but, non-declarative (implicit) knowledge can still be acquired RetrievaL: -by indexing the inter-relationship between event elements, the hippocampus can contribute the recollection of episodic memories (not typically involved in semantic memory retrieval) -as time passes, episodic memories become gradually consolidated in cortex and depend less on hippocampal indexing -damage results in temporally graded retrograde amnesia

Hindbrain, Midbrain, and Forebrain

Forebrain: (largest region of brain) contains the cortex: giant thin convoluted sheet of tissue that covers our brain, 3mm thick -variety of subcortical structures that are beneath the cortex Midbrain: movement coordination, pain regulation, auditory pathways (circuits that relay auditory information from the ears to the areas in the forebrain where this information is processed and interpreted) Hindbrain: (sits directly on top the spinal cord) Controls heart rhythms and breathings -regulates alertness -also has the cerebellum which controls breathing and heart rate and key life functions

Evidence for Levels of Processing

Incidental learning -deep processing leads to learning, even in absence of an intention to learn -learning that just happens as you go - not trying to memorize information but as you go about life you are just encountering things and some percentage of that will end up getting remembered and the thing that determines the likelihood of later recalling it is the depth of processing -Even in an absence of intention to learn things that you think about deeply and process in this way - in a semantic way - will be better remembered Intentional learning -usually ensures deep processing will occur -not necessarily any better than incidental learning with deep processing (Its not necessarily any better than incidental learning and in some cases if you are doing intentional learning wrong it can be worse) -Intentional learning is what you do when you are trying to learn - when you go into a lab and you are given a memory test and you know you are going to be tested so now you are trying to use strategies to intentionally do as well as you can -Its what you do when you are studying for a test but the reason it works so well is not because you put your brain into learning mode its because the goal of learning itself encourages deep processing -Memory is a byproduct of processing (things that you process deeply are going to be better remembered)

More examples of the inverse optics problem

Magnet-Like slopes: -Heres a nice example of an illusion - so this is the actual object they built its something you can build in the real world and that's uphill -the actual object is like this - it doesn't have that kind of symmetry you see -the illusion only works from a particular vantage point -this illustrates the problem of inverse optics and this is the real 3D world and your interpretation of it was kind of a best guess given the information you have -seemed more likely that it was symmetrical and that all of these were at right angles from each other and going towards the center Charlie Brown Street Art: -they appear to be coming off pavement in 3D and in order to create this you also need to take advantage of inverse optics -if in the real world there was a person standing here the image cast on the retina has certain qualities depending on the visual angle so to create this impression in the 2D version painted directly on the sidewalk you need to draw it like this -the heads need to be much bigger than the bodies and then it appears like this -so your visual system is adjusting for that - you are expecting through your lifetime of experience in the 3D world that things that are larger here - that cast a bigger angle on the retina - will be in this case closer to you

The cocktail party effect

Problem #1 for early selection theories: -people notice when their own name is spoken at a noisy party, even when they're not paying attention -If you are not paying attention to what is going on in the background but suddenly someone says your name you are going to be more likely to hear that then if they are just saying something else that is irrelevant Experimental stimulation of cocktail party effect: -Right ear is relevant left ear is irrelevant -He is paying attention to mary had a little lamb and then every once and awhile in his unattended ear they would say a sentence with his name in it -And then at the end of the trial you ask them did you notice anything from the ear you weren't paying attention to - any words that stood out to you? -Most people don't notice it: w/o own name 6% -But if it was your own name 33% noticed - so it's a major increase -The filtering is still very good even when its your own name but why is it if you are blocking information before perceptual analysis how is it that your own name would have an advantage -Doesnt have to be your own name; anything that your interested in can capture your attention -So the filter theory cant really accommodate for that

Key distinctions between STM and LTM

STM: -active contents of consciousness -active nodes in LTM (if I give you a set of numbers to rehearse (like 5,3,7,9,8) each of those numbers have some representation in long term memory - you are aware of those numbers they are spoken in the English language and they have meaning for you - each one of those is a node in your LTM and your just kind of transiently activating them as you cycle through) -fast access to contents (because its occupying our consciousness its something that we can report back very quickly - sort of what is it you were just rehearsing and you just spit it back out) -limited capacity -fast forgetting -transiently increased neural firing relative to baseline (so whatever is currently occupying your thoughts is going to be actively represented by neural firing) LTM: -not currently in consciousness but can be retrieved (if I retrieve a long term memory and I remember what I had for breakfast yesterday morning now that's in working memory - I remember that and can think back to where I was and so now my working memory is occupied with whatever I retrieved from long term memory) -inactive until cued -slower access -unlimited capacity (seemingly) -slower forgetting -plastic changes in synaptic connection in strength (so plastic changes meaning your brain has been changed because of your experiences and then later when you encounter something your ability to retrieve it is because of that change in synaptic strength)

William James and memory

William James (1980): postulated two forms of memory Primary memory: -the immediate contents of consciousness -effortlessly available and fleeting -limited capacity -he thought there was primary memory: which is kind of like short term memory -whatever is immediately available to you right now -its effortlessly available - because it just happens to be whatever is flutering around in your mind -but its fleeting -so if your attention shifts you might lose that thought -and its limited capacity - there is only so much you can keep track of at once -if I showed you a complex display with blue red, green, purple dots and asked you to remember that for 10 seconds and something changed you might not be able to do it because there is too many stimuli to keep track of Secondary memory: (LTM) -memories of the past -permanent but available with effort -unlimited capacity -its just the sum total of all the knowledge you've experienced in your life that you've stored in some more permanent way -so everything you know about yourself and your friends and all the places you've been that's all secondary memory -its just this sort of permanent store - permanent in the sense that it lingers -james acknowledged that this would be hard to retrieve or might be forgotten but its much more permanent then primary memory -and it takes effort to retrieve -some memories are easy and come back to us readily some we really have to think about (might be able to remember what you did for your last b-day but two or three years ago you have to think how am I going to reconstruct what I did three or two birthdays ago) -unlimited capacity

Encoding and Episodic Memory

we remember different kinds of things: -specific events or episodes that have this time or place (episodic memories) -can often later recollect the source (context) of the information -so with specific events you have the opportunity to not just know that it happened but you actually can reconstruct or travel back in time in your mind and pull out all these details about that event -So that is recollection the source of the experience and not just what happened but where did it happen - who were you with - how were you feeling -If I gave you a word and said was this on the list you could say yes or no - that's not recollection but just recognition or you might have a sense of familiarity - yeah I think the word table was on the list -But then if I asked do you recollect it - im asking do you remember the experience of sitting there in a chair in the lab in franz hall and this word appeared on the screen and a thought occurred to you -You can go back to this moment in your life and recollect it - that's recollection -but sometimes we later only experience a sense of familiarity -Sometimes an experience seems familiar to us but we cant really place it - you encounter someone on campus who looks familiar but you don't know how - you cant place where you know them from -This happens a lot where you have familiarity without recollection How do we learn? -maintenance rehearsal (a.k.a. item-specific rehearsal): simply focus on the to-be-remembered items themselves, with little thought about what the items mean or how they are related to each other -You can engage in maintenance rehearsal or item specific rehearsal where you simply focus on information you are trying to maintain - if you are studying for chemistry you might have these molecular structures and you are just sort of looking at them and saying oh I hope I remember that this one is water or something -Or for vocab you might just look at the word and think about it in your head -elaborative rehearsal: thinking about what the to-be-remembered items mean and/or how they're related to each other and to other things you know -Or you could do elaborative rehearsal - so thinking beyond the item itself and trying to elaborate in some way that is going to be more meaningful -So associating it with other pieces of information connecting it to something in your life - just doing something with that item that is going to make it more memorable for you -So just thinkinga bout what it means in a deeper way -Elaborative rehearsal is going to be much better

Working memory vs. long term memory in amnesisa

• HM had impaired long term memory but had in tact working memory • If you gave him serial position like work learning task - set of words then test memory after • You see controls here show us regular serial position curve have recency affect • Amnesic patients show serial position curve as well but you can see they are completely matched with controls - in fact here these patients numerically exceed the controls • They show completely normal recency effect because recency effect doesn't depend on hippocampal at all its just sort of dumping whatever items are in the working memory and now you get the opportunity to just tell experimenter words on the list and those are the freshest • They do show a small primacy effect and this is because in most of the patients the obliteration of the hippocampus isn't 100% complete • But primacy is impaired • The primacy is impaired by hippocampal damage as well as the middle list part but the recency is spared - so working memory is intact for hippocampal amnesia for the most part

Non declarative memory and amnesia

• Skill learning tasks - one task that is challenging is the mirror traving task • You get a star with the border here and you have to put a pencil on the top and trace your way around without pencil going around border • Everytime you go out you get points off • Its easy to do if you were doing it with a piece of paper in front of you but if you were doing it relying on a mirror its hard because you have to do the opposite of what you see and so it takes learning to perfect this • It's a skill learning task • And one that doesn't ddepend on hippocampus • HM showed learning on first day and then day two he has no clue he did this task yesterday but once you tell him all that he does better - on second day his starting performance is better then first day and third day there is almost no errors on the first trial even though he doesn't remember doing this task • This is just one of many demonstrations that nondeclarative memory seems that you don't need a hippocampus to do • Mirror tracing depends on other regions - cerebellum important for motor corrdination and timing • Put cerebellum up to illustrate this double dissocation - you have brain region A and brain region B and two different cognitive processes or regions • So if damage to brain region A (hippocampus) affects one thing declarative memory but spares another thing procedural memory - and if that's all you knew up until the last slide that would be a single dissociation • What makes it a compelling dissocation - is to find a case like this patients who have cerebellum damage who cant do the mirror tracing task - but they do find that all the other tasks HM is bad at • They have no deficit in declarative memory but a deficit in procedural memory • So now that you have those cases you have a double dissociation • Seems like one area will affect one kind of memory and other area affects another so the systems seem to be largely independent

the DRM paradigm

• words from actual list people do well • Unrelated words people did well not recalling it • But then that theme word (could have list that has bed, pillow, night, snore and people would think sleep was on the list) • Can create list of semantically related words and leave out the most common word they are all related to and each of these words as you encounter it will activate that concept • Reason you had that false memory is that word chair even though it wasn't on the list was something you were thinking about - every one of the words were about chair so you keep activting that chair concept - you had a memory for it because you kept thinking about it that's why you had this false memory • Shows that we really only remember the gist of much of our life - we don't recall the factually accurate details or true to life (veridical) details

Anterograde amnesia

•What does this teach us about encoding and retrieval about memory? •There are two different kinds of amnesia •Everything I have been talking about thus far is what you would call anterograde amnesia - if you imagine a timeline of your life and this is the brain imagery •This is the time your hippocampus gets compromised •Everything that happens after that is going to be what we call anterograde amnesia •The inability to remember events you experience (episodic memory) and facts you encounter (semantic memory) after the brain injury •Both episodic memory and forming new semantic memory after the brain injury •So you cant learn new declarative memories following injury