Introductory Cognitive Psychology
Introductory Cognitive Psychology PSYC 2145
Popular in Course
Popular in Psychlogy
This 19 page Class Notes was uploaded by Megane Sauer on Thursday October 29, 2015. The Class Notes belongs to PSYC 2145 at University of Colorado at Boulder taught by Staff in Fall. Since its upload, it has received 7 views. For similar materials see /class/232026/psyc-2145-university-of-colorado-at-boulder in Psychlogy at University of Colorado at Boulder.
Reviews for Introductory Cognitive Psychology
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: 10/29/15
What is Cognitive Psychology History What People Actually Did It is a Science Until late 1800 s O Observe 0639M T t Theorize Test Theorize Just a bunch of philosophers sitting around thinking 0 Aristotle mind is in the heart 0 Nativist vs Empiricist of course its both Devil is in details History German Psychophysicists 180039s History Wundt starting 1879 Observe ntrospec on Introspectlon Test Theorize Theorize Introspection Fechner Weber Helmholtz The full method Q What is smallest Change in stimulus detectable Emphasized actual experiments and depts journals students ApproaCh present stimuli39 see if Change demcmdl But introspection method strongly determined results But only applied to lowlevel sensory processing not cognition one Step above Sitting around thinking Structuralism find atoms of cognition History James also late 1800 s Obgvve t Theorize William James didn t actually run many experiments The Tao of Psychology But somehow he had a lot of damn good insights introspectionl Structures Functions Wundt James Functlonallsm evolutionary function of cognition representations processes brain areas cognitive operations hippocampus memory neurons activation Behaviorism 1913 195039s Seeds of the Cognitive Revolution Language quot Test Th em l ze Behaviorists Watson Pavlov Thorndike Skinner o Rejected introspection and focus on processing 0 Focused on observables learned responses to stimuli Associationism same as Berkeley etc o Generative not obviously determined by stimuli o Rulegovemed must be internal rules grammar 0 Complex Cannot be learned through simple associations Memory 0 Internal organization according to schemas categories Artificial Intelligence 0 Abstract symbolic processing rigorously studied on computers E Cognitive Psychology Observe Computation I Stimuli H Processing H Response I Test Theorize Internal processes rigorously formalized as computation Opened up a rich set of experimental paradigms amp much data 1 0 What Cognitive Psychology Looks Like Processing categor rehearsal ization Visual recognition d m Register I AUditory Shortterm Lon term 8 imuli Register Memory Merger i Response J recoding g I reorganiza ion J manipu ation Past Eat Relation Subject Time Relation Subject Relation Slow Children Agent Objw Bread Cold Figure 57 A propositionalnetwork representation of the sentence Children who are slow eat bread that is cold Structure buffers propositions Function rehearsal recoding 11 Key Features of Cognitive Psychology 0 Enumerate possible information processing strategies 0 Derive signature predictions for each 0 Run the appropriate experiment 12 The Fundamental Problem of Inference If P then Q does not imply if Q then P If drinking beer then must be 21 If 21 then drinking beer not necessarily If serialexhaustive then linearslope amp YN same If linearslope 8 YN same then serialexhaustive not necessarily What else could it be 0 Linear slope 2 item rehearsal interferes with parallel search 0 Initial search is parallel confirmation is serial exhaustive he Fundamental Problem with Cognitive Psychology We don t have independent access to the internal processes Therefore we can only test via behavioral outcomes But we are always limited by our imaginations and preconceived 1 r I dUULLI I 0 systems Metaphors vs The Real Thing When you don t know what something really is you use metaphor Metaphors for mind 0 Hydraulics Descartes Freud blockage o Switchboard Helmholtz etc a Computer symbol processor 1950slow death now Metaphors support but also constrain our imaginations Cognitive Neuroscience The New Frontier Cognitive psychology is undergoing a new revolution by iilLui pui suing 39 cognitive 0 Uses neural data to inform understanding of processes 0 Theories need to make predictions at neural level 0 Everyone needs to go back to school to learn neuroscience This is a very exciting time in cognitive psychology Summary What is Cognitive Psychology Understanding behavior in terms of internal cognitive processes Paradigms for theorizing about cognitive processes 0 Traditional symbol processor computer information processing 0 Connectionist or neural network based on brain 0 Abstract 39 more de Liiprive than Paradigms for testing theories 0 Behavioral data RT accuracy eye tracking etc o Neural data neuroimaging brain lesions etc Neurons output integration synapses inputs Detector axon cell body membrane potential Neuron Neurons are detectors eg smoke detector Looking for patterns in their inputs Target pattern is determined by weights Electrophysiology Single Cell Recording I Lnymu wigquot 1m Advantages 0 Direct measure of activity 0 Spatial individual neurons 0 Temporal millisecond Disadvantages 0 Only in animals usually 0 Only shows correlations 0 Need parallel recording The Brain Cortex O 3 language o 39n 06906 Brain Lesions Infer function of damaged brain region via behavioral effects Key problem If P then Q does not imply if Q then P If drinking beer then must be 21 If 21 then drinking beer not necessarily If broca szsyntax then broca s lesion 2 bad syntax If bad syntax then broca szsyntaxl not necessarily Brain Lesions Summary Advantages 0 Shows causal role of brain area a Can make highly accurate lesions in animals Disadvantages a Brain may not be modular damage complex effects PET Positron Emission Tomography Positron Advantages o Nonfinvasive shows function of normal brain a Gives big picture view of where cognition happens Disadvantages 0 Indirect measure blood ow a Correlational a Brain may not be modular pure insertion 0 Has worse spatial temporal resolution than fMRl Emission Tomo a h o Lesions in humans are very messy gr P y M w 1 a 39 acti itv PET Summary MRI amp fMRI Measures differences in magnetic propertie Structural grey vs white water content good for lesion loc Functional oxygenated blood is magnetic fMRl Summary Advantages o Noninvasive shows function of normal brain 0 Gives big picture view of where cognition happens Disadvantages 0 Indirect measure blood ow 0 Correlational 0 Brain may not be modular pure insertion 0 Has better spatial temporal resolution than PET Converging Methods Lesions to MT also produce motion processing impairments Using multiple methods reduces limitations of each EEG Electrical Activity in the Scalp mm REIWEWWWNWWWWWWWWWW i D wmwmwmm I 39me I Defies WWM quotW I cum wWL WU I l l so pw 1 sec 28 ERP Event Related Electrical Activit Avera ed y g E E6 2le 1 l Ampli er Stimulus onset 39 1quot x 100 nIaIs v k R Q 1 Signal averaging v f ERP s f 1sz I I a 1 Sum mus unset Suuml generator 700 msec E ERP Summary Advantages o Non invasive shows function of normal brain 0 Gives big picture view of when where cognition happens o Is a direct measure of neural activity 0 Has excellent temporal resolution relative to fMRI PET Disadvantages o Correlational 0 Brain may not be modular pure insertion 0 Spatial localization is terrible Overall Summary Brain Map olumn Layer Neuron Log size mml Dendme Synapse 74 1 4 o 1 2 a 4 39 Mlllisecund Semnd Ilnue Huur Day Lug time sec Can You See This Inverse optics problem our eyes register 2D we see 3D Let s Get Relative Contrast ONCell OFFCell ON area OFF area OFF area ON area Actron poremials Surround ugm Ltg39hl IN I l T 05 1o 0 Tvme Tbrne b Center 1 5pm or ugm in center 2 Spot n Iigm m urrouna S This is what the cortex l sees V1 likes Lines Edges Summary Biological Tools for Vision 0 Relative encoding of brightness centersurround o Edges lines are particularly informative V1 transforms dots into edges and bars The Three Indeterminacies Shape Likelihood Principle 0 Shape 3D from 2D o Shading Light source re ectance shadow from luminance 0 Depth Size lt gt distance Is this a cube Viewed edgeon Highly unlikely Brain assumes sim lest inter retation Occam s razor The bra1n makes assumpt1ons to solve these p p 37 Shape Intersecting Lines 2 Clues Perkins 71 Y intersection Arrow T intersection lowerconcave upper corner occlusion corner 38 Shading Bags of Tricks ll lg ht check in ahaiircn39vii39 is the same graft the dark r inertia o Hts tuftta the s hardcrew E H 39lt lE EII39IquotI Assume constant shading light source 2 upper right Assume shadows are fuzzy can tell from squares Assume color from relationship to neighbors light 2 dark 39 Shading Contrast CenterSurround 40 Depth The Third Dimension There are many depth cues Monocular o Occlusion linear perspective relative height texture atmosphere Binocular 0 Convergence eyes go inwards for near objects 0 Stereopsis retinal location further apart for near objects retinal disparity Pencil test Monocular Depth Assumptions Occlusion linear perspective relative height texture atmosphere Binocular Depth Stereopsis n I Perceived Objects I t I I u Actual Objects o o o o Tightly spaced 2 closer loosely spaced 2 further The Three Indeterminacies Assumption Summary 0 Shape 3D from 2D likelihood assumption line junctions o Shading Light source re ectance shadow from luminance Constant shading shadows fuzzy objects sharp color 2 relative 0 Depth Size lt gt distance Occlusion linear perspective relative height texture atmosphere Convergence retinal disparity depth Higher Level Vision What is vision for o Recognizing objects faces places etc o Navigating in space Two Streams Ventral what Dorsal wherehow PO Template Matching Input Template Input Template Input 0 0 THE 0 0 00 o n O o oo Template Ill Problems with Template Matching hut MMwgzzMM 4 I39ll F 13919 l M ve M 0 Too many templates needed for all possible objects amp variations Feature demons Feature Matching Pandemonium 1 Vertical line 2 Horizontal line 3 Upright sideways line 4 Upleft sideways line Cognitive demons 5 T 1 2 6 V 3 4 7 A 2 3 4 8 K 1 3 4 Problems with Feature Matching Geons Are These the Features Geons o What are the features 1 2 o Combinatorics how many combinations 3 4 25 Objects 5 6quot 5 A 5 E Figure 931 Examples of geons and their presence in objects Five simple geons are shown together with several common objects that can be analyzed into con gurations of geons From Biederman 1987 Problems with Geons Alternative Local Views DO geons really describe all shapes Instead of object centered representatlon of ObJGCt features parts could store viewe rcente red representation of what objects look 0 Combinatorics how many combinations and how are geon hke from dlfferent perspeCthes 7 relatlonshlps dlstortlons etc encoded Templates agam Hello but Geons are easy to recognize from basic features line junctions These are fancy templates where you rotate Views to align them parallel lines curved lines etc This is main strength of theory with stored representations Q 45 90 Q 45 Brick t J Horn A o Claw Lemony Conek a b V Noodlem V Cylinderv q Soapd g L Fry 4 y Wedge amp Reaclmn Time ms Tarr s Local View Experiments 900 Viewpoint Dependence Results Tan etal 1998 Proportions me am rates 700 0 45 90 results fmml an simihxexpnmems Reaction time and errors increase as even basic geons are rotated away from studied orientation Problems with Local View s Chickenandegg problem need to recognize before you know how to rotate o Views are clearly not literal images but what are they 0 How are they matched Summary of Object Recognition Objectcentered recognition is orientationindependent o Templates 0 Featuresparts lines geons Viewercentered recognition is orientationdependent 0 Local Views rotation to closest View Two Streams Ventral quotwhatquot Dorsal where how 57 Where Ungerleider amp Mishkin nonmatching to sample vs landmark task A 39 r7 a 0 a B A Ventral pathway lesions impaired object comparison B Dorsal pathway lesions impaired location perception 58 and How Which central circle is larger But is your reaching system fooled No Goodale amp Milner Also not fooled by slope of hill misjudgment skiers anyone Motor control system how uses perception differently 59 BottomUp vs TopDown Sometimes the bottomup stimulus is ambiguous need to use context to make sense 60 TRACE Model amp Word Superiority Letters are better recognized in words than nonwords But you need to recognize the letters before the words You can do both bottomup and topdown at the same time 61 Cocktail Party in the Lab Dichotic Listening ignored inputs Attended lhputs President Lincoln often read by the light of the re The horses ga oped across the field Speech output Headphones Also ask at end what they heard from other ear President Lincolln often read by the Mg ht of the fire What if you really liked horses Would it matter if you also found Lincoln boring What if your name was presented on the ignored side 63 Early Filtering Evidence Cherry S s didn t notice if language changed or backwards but did notice a pure tone gender changes Physical not semantic What about fire or FIRE or FIRE Signal averager N1 amplitude effect with attention 0 100 200 msec Attended I I I I I I I I I d 7 Loudspeaker 62 Where is Attention Happening Early Late Eelection Selection m Executive g k unctions 4 s ssss p g E 5 Decisions i 3 at g 34E 39 memoryetc sensory Response inputs If early what do you expect What about late 64 Early Filtering Problems Moray People do notice their own name 35 I SAW THE GIRLSong was wishing me that bird JUMPING IN THE STREET dogs six eas eight scratch two dogs scratch eas 39 13Ame GIRL JUMPING Shadowing follows semantics Must be processing semantics of unattended ear so no early filter 65 Late Filtering Everything is processed to semantics Evidence indirect measures of semantic processing 0 GSR galvanic skin response lie or shock detector Words were paired with shock does GSR go up if unattended o priming taxi gt fare not fair Nothing held up that well on further scrutiny 66 Is it Early or Late Early people can occasionally switch between ears Late Yeah but what about those crossover results Reality according to me some others topdown and bottomup graded interactions if it jives it lives if it don t it won t This makes for a movable filter Also Triesman s graded attenuation theory 67 Visual Attention What can do we pay attention to 0 Locations in space Without moving your eyes 0 Objects 68 Valid Trial Invalid Trial CueD D 80 Target I D 60 Spatial Attention Posner Task 120 100 Lesioned Intact 0 Neutral Brightening of box draws covert attention to space 1 Valid Target is more rapidly detected if appears in attended location and more slowly detected in unattended location 2 Invalid 69 Self portrait Spatial Attention Unilateral Neglect Mt lliL l TlEil39lquotS COPY Patient bisections Test paper Vertical lines with horizontal lines on it copying line bisection tasks In all cases patients with parietal temporal lesions seem to forget about 12 of space but they still see it 7O Object Based Attention Baylis amp Driver Which corner is higher either on white or black shapes Same stimuli used for all subjects Faster for judgments on same object contiguous region of same color vs different objects 71 Reaction time milliseconds E Visual Search Where s Waldo FIGURE 312 Results from Treisman and Gelade Reaction time to detect a target as a function of number of distraotors and whether the distraotors contain separately all the features of the target Adapted from Ti eisman amp Gelade 1980 HTinIZ 0quot TinIY l I l 5 15 30 Display size T in LY Unique features in T popout parallel search T in IZ and are both present only conjunction is unique to T so you have to scan serial search 72 Visual Attention Summary 0 Attention can go to both spatial locations and objects 0 Brain damage can result in neglect of parts of space 0 Visual search requires spatial attention scanning for conjunctions but not unique features Learning and Attention Automaticity Some processes seem automatic others require more attention Does it matter how much experience you ve had Driving reading skiing typing Stroop Task Reading is Automatic Name the ink color Magenta Blue Red Reaction Time msec Stroop Data Dunbar amp MacLeod 84 0 0o Control Conflict Congrueni Condition Word reading is more automatic interferes with color naming Other Examples of Automaticity Stroop with shapes 6 red M green Initially color naming interferes with shape naming with enough training it reverses Shiffrin amp Schneider 1977 numbers vs letters consistent compared to mixed inconsistent also a training version Flanker Task SSS S SSS vs SSS H SSS
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'