Lecture Notes Cognitive Psych 3/23 - 3/25
Lecture Notes Cognitive Psych 3/23 - 3/25 PSY 0422
Popular in Cognitive Psychology
Popular in Psychlogy
This 5 page Class Notes was uploaded by Lindsay Notetaker on Friday March 27, 2015. The Class Notes belongs to PSY 0422 at University of Pittsburgh taught by in Spring2015. Since its upload, it has received 111 views. For similar materials see Cognitive Psychology in Psychlogy at University of Pittsburgh.
Reviews for Lecture Notes Cognitive Psych 3/23 - 3/25
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 03/27/15
323 Lecture 18 Language bridging to concepts Mean Fixation Duration ms N G G H 1 UI Using eyetracking to study garden path sentences Garde off the O O n path sentence after the musician played the piano was quickly taken stage People slow down when they read the inconsistent information later in the sentence This supports the immediacy of processing idea that we construct representations of text immediately while reading each word of the sentence Surfacelevel ambiguity Nongarden path sentence after the musician bowed the piano was quickly taken off the stage Garden Path l Control D3 D2 Dl D D1 D2 Position in Sentence Syntax and Semantics Summary We try to understand the syntactic structure of a sentence on a wordbyword basis building propositional representations in memory Shows how language and memory come together This involves making some guesses Sometimes we are fooled when the best guess or most common choice is wrong Pragmatics 4th level phonology syntax semantics pragmatics Making sure that people understand what is said and what is meant People follow rules in conversation to help the listener know what is meant Conversational Maxims People are assumed to cooperate in a conversation and therefore follow rules 0 Relevance be relevant 0 Quantity be informative 0 Quality say what s true 0 Manner and tone be clear 0 Relations with partner infer and respond to partner s knowledge and beliefs 0 Rule violations signal intentional violations Indirect Speech Acts Example Parent Where did you go Teenager Out Parent What did you do Teenager Nothing Review Sounds of separate words blur together much more than we think they do and we use our past experience with words to separate them e g implicit rules Our brains appear to parse semantics meaning separately from syntax grammar Meaning is the memory units we activate and grammar is the way we connect the meanings together in working memory Understanding a text involves building propositional representations and language ambiguities give us insight into how we process text Pragmatics is the complex way we use the situation rules of conversation to get at real intended meaning Language and Thought SapirWhorf Hypothesis Strong version language determines thought 0 Can t think things that you don t have words for 0 Too strong language is exible I Kiriwina has word mokita which means the truth that everyone knows but no one talks about I Ojibwe has word inaabate which means the smoke that curls there 0 Can express in English though no speci c word 0 Examples in slides Weak version language favors some thought processes over others 0 Will tend to think in ways your language suggests Testing the hypothesis color English has many color terms Other languages have as few as two 0 Dani of Indonesia mili dark mola light If the SapirWhorf hypothesis is correct then expect that perception of memory for color should be different for Dani speakers and English speakers How Test two groups using color chips o Naming colors different e g amount of agreement but memory for colors very similar 0 Make same kind of errors 0 Do best for the same prototypical colors Results do NOT support the strong version of the SapirWhorf hypothesis 0 But color may be a bad choice strong biological component Weaker hypothesis Two related points Processing costs some things easier to say in some languages Thinking for saying if language requires some distinctions to be made then people speaking that language need to pay attention to those aspects of the world Summary Language and Thought Language and thought are often closely related 0 Our thought processes get re ected in our language 0 We use linguistic codes to help us think But language is exible enough not to constrain thought as strong versions of the SapirWhorf hypothesis would suggest However language can make information more or less accessible and having verbal labels can help in some tasks Importance Functions of concepts Concept mental representation of a class Category examples picked out by that concept Classification decision that an item is in the class Reasoning and cognitive economy don t need information for every instance Prediction go beyond given information Understanding explanation Building blocks for complex concepts Communication 325 Lecture 19 Concepts Conceptual representations and classification How do we classify Theories Classical view Probabilistic view Exemplar view Classical View All instances of a concept share common properties that are singly necessary and jointly suff1cient If an instance has all these properties it is a member of the category If not it is not a member Unitary Description true of all members of the category single representation of a category in our mind that s true of every member within it Classical View Benefits eff1cient storage of concepts clear category boundaries decisions Problems failure to specify def1ning features fails to capture typicality effects many unclear cases Probabilistic View Unitary representation that includes features usually true of instances of that concept Example bird feathers beak wings can y two legged lays eggs Prototype summary representation including all the typical properties 0 Used to represent the concept classify new instances and reason 0 The prototype is the central tendency of the category Prototype and family resemblance Similar to an extended family Structured by family resemblance Prototype consists of features that occur with many of the instances 0 Best example or most ideal member of the category 0 Do you think of the most ideal version of a category Family resemblance and typicality In artificial categories family resemblance predicts 0 Ease of learning how easy we learn the category or new object o Categorization RT after learning 0 Typicality ratings 0 Output order eg furniture usually generated rst Context Sensitivity Example At Thanksgiving time someone says The bird is in the oven what s the bird Turkey more typical of an answer Probabilistic view Accounts for typicality unitary representation Problems 0 Lack of sensitivity 0 Not sensitive to many factors that people are sensitive to I Frequency of instances Variability Correlation of feature values Exemplar view Representation consists of separate disjunctive descriptions of some of its exemplars No unitary description Explicitly disjunctive can use different information for different classifications DOG collies or spaniels or Fido Exemplar view Different ways of making concrete 0 Most similar exemplars 0 Set of most similar exemplars o All exemplars weighted by similarity Can predict many prototypelike effects Can predict in uences of frequency variability correlation of features Realworld evidence physicians f1res think about the specific f1re they ve seen in the past and determine what to do in a new situation of a fire based on that past situation Exemplar view explains a lot but People have abstractions germ positive feedback loop Do people have abstractions in disjunctions If so how Why do these exemplars go together Beyond similarity Theories Similarity not enough Need for explanatory knowledge DEMO mutilated items Item I Raccoon paint it black paint a white stripe on its back 0 Operate put super smelly yucky stuff in its tail Item 2 Coffee pot punch holes in it pull off spout hang from tree fill with bird seed Why going beyond the similarity of observable features Concepts are organized by theories To decide if something is a member 0 Not match of features rather 0 Ask if it ts Role of similarity o A good heuristic guess 0 Deep and surface are correlated 0 However if similarity and knowledge con ict go with knowledge Summary of Conceptual Representations Classical Probabilistic Exemplar But similaritybased may not be enough need to take into account knowledge
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'