PSYC325 Final Exam: Lectures 17 and 18

B.F. Skinner was not a fan of cognitivist psychology and said...

'The real question is not whether machines think but whether men do' Cognitive science (as far as Skinner was concerned) is the creation science of psychology, as it struggles to maintain the position of a mind or self All this cognitive science stuff that allowed us to have minds was effectively pseudoscience to Skinner Went to his grave believing this - 1990 paper finished the day before he died

John et al (2012) found that...

- A lot of psychologists are doing things with the experiments they do which go against the logic of hypothesis and significance testing itself - anonymous survey sent to 5964 psychologists at major US universities with incentivised truth telling - No one has ever genuinely had a null hypothesis

What is Bayesian statistics?

- Associated with 18th century mathematicians/ clergymen Thomas Bayes - With Bayesian statistics, you have a prior estimate of likelihood, and you gather data to alter that likelihood - Your prior estimate is loaded with your previous beliefs

What did statistics have to do with biology?

- Post-Darwinian biologists had a new appreciation for variation, and the role that might play in biology - In order to understand variation at the biological level - biologists started thinking in terms of variation and so started thinking in terms of statistics which might help us understand this variation - Whereas previously they had just tried to get the most representative specimen of a species - To understand variation, you needed to employ statistics - Galton was Darwinian in the sense that he was Darwin's cousin

Fischer statistics involves...

- Ronald Fischer; in 1925s Statistical Methods for research Workers, he adapted an extended 'Student's' 1908 t-test into ANOVA - had to use an anonymous name instead of his own so he chose student, not actually intended for students - Provided a rationale for using t-tests and ANOVAs in a carefully designed experiment they can determine whether we should reject the null hypothesis (that two sample groups are just from the same population) - Fischer was bitterly opposed to the idea that hypothesis testing allowed decisions to be made on the meaning of data - he did not believe that a p value of less than 0.05 showed that two things were different except in the context of a very carefully controlled experiment where there was no other explanation

What relationship did Karl Pearson and Ronald Fischer have?

- They hated each other with a passion - Statistics has a metatheory and there are different points of view about how best to use statistics in a scientific psychology

Explain Pavlov's context and how he influenced psychology...

Around the same time as Thorndike Pavlov came from a physiological background - found evidence of environmental stimuli causing a behavioural response which could be measured by physiological means implies there is no need to involve the mind Behavioural response because animal's behaviour changes - but measured with physiological means Pavlov wanted to avoid metaphysical ideas of psychology - psychology in terms of philosophy of mind Pavlov and Thorndike set up some of the ways in which behaviourism would be used in order to understand about the human mind - both were interested in humans but used cats/dogs

Rats and Stats in the 70's involved...

Behaviour research explaining how rat's behaviour Behaviourism and statistics

How is behaviourism 'the way of no ideas'?

Cartesian Dualism --> no one really agreed with Descartes - subjective experience vs scientific determinist world Radical Behaviourism --> environmental consequences cause us to behave in certain ways - if there is a ghost in the machine i.e. if we have a mind, it has nothing to do with the actual behaviour Methodological behaviourists --> cannot study the consciousness/ghost scientifically - we'll just focus on what we can study Behaviourism takes cartesian mind body problem - solves it by suggesting we get rid of the mind - mind doesn't actually influence our behaviour at all - solution to the way of ideas is to have no ideas There's a fundamental thing about behaviourism in that its entirely stupid - we have a mind that controls behaviour - this should be straightforward and true - this makes us fundamentally different to animals Didn't think that cognition mattered - thoughts matter exactly the same as someone speaking to you

Statistics can be split into...

Correlational (spearman's and Pearson's tests) vs inferential (t-tests and f-tests) statistics in psychology - Psychologists enthusiastically took up (Pearson's) correlational statistics pretty early - e.g. Spearman's g (1906) and factor analysis - Galton was also a psychologist interested in the mind and was implementing stats into his work - As a result of this you get the rise of the IQ test

Psychology slowly takes up statistics. Provide an overview of this...

Cowles (1989) identifies one of the earliest psychology experiments using null hypothesis significance testing as being from 1940, and a slow rise to acceptance until 1970's - one of the first psychologists in Australia to use a t-test or an f-test was Gordon Hammer in about 1949. Plenty of famous experiments in the intervening period don't use null hypothesis significance testing; e.g. Miller (1956); plus or minus seven idea, Milgram (1963); shocking people - neither of these experiments show statistics First medical RCT (randomised control trial) - was done in 1946 - First meta-analysis done in 1980

Why was the history stats surprisingly dodgy?

Generally, the mathematicians who figured out the theory of stats - Pascal, Fermat, Laplace - Were mostly trying to help their gambler friends get better at making bets - they didn't see it as something that could be useful if applied to science - gambling by using stats - The first scientists who applied statistics to scientific problem solving were 'advocates of eugenics who were interested in biometrics as a way of identifying 'poor stock' - people who shouldn't pass on their genes - E.g. Francis Galton, perhaps the first scientist to do correlations, donated money in his will to University College London to fund the Galton Chair of Eugenics; 1911-33; Karl Pearson (Pearson's R) - 1933-43; Ronald Fischer (null hypothesis, significance testing, p values)

Explain the relationship between rats (behaviourism) and stats

However, behaviourists - being good logical positivists - thought correlational and inferential statistics were beside the point - they weren't really science because they didn't fit into hypothetico-deductive laws vs statistical inferences (Karl Hempel) - If you actually read behaviourist experiments they tend not to report t-tests or f-tests they just report a behaviour increasing over time in graphical form

What are the tenants of behaviourism?

Humans, like other animals, are the product of evolution - not that much different to what James believed as a Darwinian However, unlike James - We adapt our behaviour to our environment - not only do we adapt to an environment, but we have a whole set of inherent processes by which this happens - inherent in the nature of being an animal that you have such patterns of reinforcement Humans like other animals are tabula rasa (blank slate) - humans don't come into the world with preconceived ideas - we are written on by experience - because we're all tabula rasa we're not any different to animals we're just more complicated Behaviour can be studied objectively and quantitatively, from a verificationist point of view - Verificationism, also known as the verification principle or the verifiability criterion of meaning, is the philosophical doctrine that only statements that are empirically verifiable (i.e. verifiable through the senses) are cognitively meaningful, or else they are truths of logic (tautologies). Behaviour is explained by logical positivism - this is the kind of philosophy of science specifically that the behaviourists were going for. Verificationist - logical positivism Behaviourism is explained deterministically by environmental factors, not the mind - things that cause behaviour are in our environment not our mind, not that we have a desire to achieve a goal and that determines behaviour - BUT something in the future i.e. the goal, can't cause something in the present to happen. So the behaviourists argued that the way to deal with issue of future focused goals not able to cause current behaviour, that it is only environmental factors to the extent that we think it is our mind causing us to do things. Because of all this... The dominant topic of psychology should be the learning of behaviour

Explain the Open Science Collaboration (2015) project

In 2015, what was then the biggest psychology project ever conducted was published in the prestigious journal of science It featured 100 separate experiments, and 269 co-authors It aimed to replicate a study/experiment from every journal article in three randomly chosen 2008 journal issues from: psychological science, Journal of Personality and Social Psychology, Journal of Experimental Psychology; Learning, Memory and Cognition Their aim was to see how much psychologists bad statistical practice effects results - replicate research and see if they find the same results Found that only 39 of 100 replication attempts were successful - assume 3 of 5 times the research is probably wrong Most psychology research is probably wrong - we don't understand the metatheory of statistic

How is the success of behaviourism measured?

In terms of predicting and controlling behaviour, what behaviourists wanted psychology to be about, in animals at least, were very successful e.g. Cats doing tricks

E. Pearson and Jerzy Neyman brought about...

Karl Pearson retired in 1933, but his son Egon Pearson and his collaborator Jerzy Neyman had already taken up arms against Fischer Pearson and Neyman revised the null hypothesis significance testing into basically what it is today... for better or worse. And introduced confidence intervals and the concept of type 1 and 2 errors.

Psychology and measurement have issues because...

One of the fundamental issues here is measurement - makes sense to measure how tall people are b/c can measure in cm, however the mind is harder to measure - harder to measure someone's need for autonomy as seven units long - doesn't really make sense in the same way - this presents an issue for stats is psychology - Francis Galton argues against what Wundt did - a lot of early psychology didn't use measurement in the same way as today

Explain Pavlov's Conditioned and conditional stimuli...

Pavlov actually talked about conditional stimuli; stimuli that were conditional in causing an effect - the idea of conditioning wasn't something he wanted to push necessarily but rather came from a mistranslation

Definitions of psychology in terms of rats and stats...

Psychology is the science of mental life, both of its phenomena and their conditions, the phenomena are such things as we call feelings, desires, cognitions, reasoning, decisions and the like; and superficially considered, there variety and complexity is such as to leave a chaotic impression on the observer - William James, 1890 - Psychology is about mental life - is the science of mental life - the things that happen in our minds In contrast... Watson argues that psychology is a purely objective experimental of natural science. Its theoretical goal is the prediction and control of behaviour. - just focused on behaviour, not interested in the mind. The first year definition of psychology comes from a combination of these two perspectives

Chomsky suggests... about language learning. This goes against Skinner's perspective of verbal behaviour.

Reinforcement cannot solely account for our learning of language as kids e.g. a child might say 'mummy goed home', despite never hearing an adult say 'goed' - child is taking information about the world and applying it to other stuff and this has to involve some sort of mental process. - Unless you can explain how 'colourless green ideas sleep furiously' is different to 'friendly young dogs seem harmless' - there has to be more than just reinforcement in language - We cannot explain language without having a mind - Descartes also thought we cannot have language without a mind - expression of the mind and justification of others existence

Explain the rise of behaviourism

Remember the functionalist/structuralist debates of introspective psychology - between function of consciousness (William James, James Angell) and structure of consciousness (Wundt) Functionalists in this debate evolved in behaviourists - key differences between introspective psychologists and behaviourists, was that introspective psychologists were initially very interested in introspective results e.g. interested in consciousness BUT... At the turn of the century, functionalists became increasingly less worried about introspective results and more focused on objective results (e.g. Angell 1903) Research gives them license to believe this is the best way to go about psychology - to start focusing on behaviour - research of Edward Thorndike

What is Thorndike's Law of Effect?

Research on cats learning to escape a 'puzzle box' - reinforcement - positive reinforcement is important to animal learning - satisfaction of escaping the box The law of effect is a psychological principle advanced by Edward Thorndike in 1898 on the matter of behavioural conditioning (not then formulated as such) which states that "responses that produce a satisfying effect in a particular situation become more likely to occur again in that situation, and responses that produce a discomforting effect become less likely to occur again in that situation."

What is the behaviourists relationship with the received view?

S.S. Stephens in 1935 introduced Bridgeman's operational definitions into psychology - instead of defining airy fairy ideas about thought or cognition, we could specifically define it as a certain kind of behavioural output of that which was more real than the thing itself Verificationism, hypothetico-deductive method Tolman in the 1930s travelled to Vienna to study with the logical positivists - Behaviourists justified the use of operational definitions by arguing that they were literally all that needed to be studied - we didn't need to know anything about the mind behind those constructs that are operationally defined because that had no effect on behaviour By 1950, theorists like Hull and Tolman were criticised for not being logical positivist enough (despite their claims)

Why Skinner not a stimulus response psychologist?

Skinner argued that a stimulus doesn't elicit a response, but enables an organism's discrimination - trying to get rid of drives. - Where S-R psychologists see organisms seeking to reduce unpleasant stimuli associated with e.g. hunger (drives) - Skinner just sees reinforcement/punishment Behaviours as not just movements in space by ways of interacting with reinforcers Did not think this was scientific enough - drive theories and goals are not as scientific - primary drives are unscientific Skinner is a radical behaviourist

Verbal Behaviours in behaviourism refer to...

Skinners tacts; verbal behaviour under the stimulus control of something in the environment We have been trained via reinforcement to response in particular ways to particular situations in the environment around us Skinner argues that when someone says 'I am looking for my glasses' what they really mean in terms of stimulus response is 'when I have behaved this way in the past, I have found my glasses and have then stopped behaving in this way' - there's something in the environment we want to stop happening so we look for our glasses and this influences future experiences of losing glasses and thus looking for them Something in your training reinforces your behaviour - we learn these behaviours based on past experiences

What is the basic assumption in behaviourism?

Stimulus response (S-R) - The idea that behaviours are essentially reflexes - responses - caused by environmental stimuli e.g. a baby grasping your finger if you put near it - caused by reinforcement Not caused by belief-desire folk psychology - caused by environmental stimuli The language of independent, dependent, and intervening variables originally came from the behaviourist Tolman - Indep. variable was the stimulus - Dep. variable was the response

How was Jame Angell's Functionalism related to J.B. Watson... and what was Watson famous for...

The 1903 experiment of Angell's were a blindfolded observer pointed to where sound was coming from - this was research that was starting to move away from introspective psychology and towards more objective means of measurement. The blindfolded observer was almost certainly J.B Watson - who was one of Angell's PhD students at the time Watson was most famous for the little albert experiment - but his most influential thing was 'psychology as the behaviourist views' (1913) basically a manifesto for psychology should do it this way - argued for behaviourism rather than cognitivism and that psychology should discard consciousness - didn't want psychology to be introspective and focused on consciousness - suspected our minds had nothing to do with our behaviour Psychology 'started' in 1879 with Wundt but took less than 40 yrs for psychology to say well are we really focused on the mind or the behaviour

Why rats and stats?

The answer: while psychiatry was fixated on psychoanalysis, psychology was conditioned to respond to behaviourism From J.B. Watson's 'Psychology as a The Behaviourist Views it' (1913) to 1960 (when there are finally more cognitivists key words in the literature) Skinner publishing screeds against cognitivism in 1990

What did behaviourist research on animals result in?

The kinds of animals that behaviourist research was done on were actually specific kinds of animals - Pigeons, rats, and dogs (typical test subjects) are all omnivorous invasive species with a lot of behavioural flexibility Other species are not like this; e.g. squirrels have more limited behavioural routines As a result of all this, psychology needed a metatheory capable of accounting for mental states, as language cannot be accounted for without assuming mind, and a metatheory with less emphasis on the tabula rasa - Welcome to cognitive psychology (cognitivism)

Radical behaviourism infers that...

The mind has nothing to do with behaviour - thought is simply inner verbal behaviour and no different to saying out loud- Lashley, Skinner, Hull - we don't have thinking in our heads, thoughts were inner verbal behaviours, no different to saying things out loud

The finding that behaviour is partly genetic, meant what for behaviourism?

The tabula rasa (blank slate) evidence of behaviourism is not justified because genetics influence behaviour (nurture side of debate less valid now) - Nim Chimpsky - Behaviourists in the 1970s tried to raise a chimpanzee to sign American Sign Language - to prove Chomsky wrong... did not learn grammar - found that chimpanzees were pretty good at learning individual signs but were crap at learning grammar e.g. could not teach a chimp the difference between 'my aunt killed my uncle' and 'my uncle killed my aunt'

Behaviourists aimed to have a verifications science, this meant...

the kind of science that is logical positivist, found out things and you verified them by doing experiments - but behaviourists like Hull and Tolman still used unverifiable mental states surreptitiously Skinner attempted to get around this in 'Verbal Behaviour' (book) attempted to account for langue in purely behavioural terms - language is the Everest of behaviourism - language is specific to humans and tried to be accounted for by behaviourism, account for language without talking about the mind - language without the mind is too difficult to talk to

Methodological behaviourism infers that...

to be scientific psychology has to focus on objective behaviour, rather than the mind - Tolman exemplifies this Still thought we might have had minds and those minds might affect our behaviour But figured to be scientific must look at what can be studied and that is the environmental impacts on our behaviour - Behaviour and reinforcement can be studied objectively and quantitively

What is Chomsky's view of verbal behaviour (in terms of behaviourism)

we can't really know what we are responding to as stimuli because when it comes down to it Skinner is avoiding the question of our discrimination abilities e.g. doesn't focus on how we learn to discriminate that thing as a white board instead of something else. Just assumes that it happens, also very hard to do this without considering the perceiving to be mental. Shows that there is still mind there that we cannot get rid of in quite the way that Skinner wants us to. (goes against Skinners radical behaviourist perspective)

Describe B.F. Skinner's behaviourist view...

wrote a utopian novel called Waldon two - in which society runs around behaviourist principles - was a radical behaviourist Skinner's explanation of learning and behaviour - operant condition - is most fully developed and influential version of radical behaviourism Positive and negative reinforcement/punishment - Schedules of reinforcement - how these change behaviour - Brought a new methodological and philosophical rigour to behaviourist explanations - Theory as 'a summary of the ways in which observable variables correlate' - very logical positivist point of view