Popular in Course
Popular in Psychlogy
This 237 page Class Notes was uploaded by Mrs. Gracie Heathcote on Wednesday October 21, 2015. The Class Notes belongs to PSY 400 at Syracuse University taught by Marc Howard in Fall. Since its upload, it has received 18 views. For similar materials see /class/225615/psy-400-syracuse-university in Psychlogy at Syracuse University.
Reviews for Selected Topics
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: 10/21/15
Foundations of Scientific Psychology plus Terminology Marc W Howard August 27 2003 Overview of today s material 0 The foundations of scientific psychology 1 Psychophysics and the Weber Fechner Law 2 Hermann Ebbinghaus and the scientific study of memory 0 Terms you ll want to know in reading Chapter 2 4 What makes something science a Search to describe objective reality oReject ideas that are contradicted or not supported by reality 0 Occam s razor The scientific method a Hypothesis 0 Predicted data 0 Confirm or deny Classics in the History of Psychology Christopher Green at York University Toronto has assembled quite a resource See the Supplemental Readings links for optional readings by Aristotle Fechner and Ebbinghaus Psychophysics Weber and Fechner o Weber Just noticeable difference 0 Same or different a Depends on stimulus magnitude Fechner s law SzkrlogP where S is the sensation and P is the physical stimulus Weber Fechner law AP AS k P 0 Imagine turning on a light in a room in the daytime o or at night Ebbinghaus 1885 a First experiments a Remote vs adjacent associations 0 Forgetting function Ebbinghaus methods a Serial recall of nonsense syllables o Dependent measure was savings 0 Less time to re Iearn means better memory Isn t all this laboratory stuff kind of artificial o Ecological validity o Replicability o Glenberg s review Ebbinghaus received wisdom on associations Ideas which have been developed simultaneously or in immediate succession in the same mind mutually reproduce each other and do this with greater ease in the direction of the original succession and with a certainty proportional to the frequency with which they were together Asymmetry Remote VS adjacent associations forward remote backyngd ABCDE Transfer study oLearnABCDE o Re learn a shuffled list a Shuffled list differs from original in distance of shuffling 1ACEG 2ADGJ 3AEIM Showed evidence for remote associations Percent Savings N 0 o o O 0 1 2 3 4 5 6 7 Distance Ebbinghaus Forgetting Function Fig 12 in the text Stimulus Terms a stimulus Anything presented to the S a item A stimulus that is to be remembered omask A stimulus that hides or stops another stimulus Perceptual Terms a sensory Having to do w a sensory system rather than higher order cognitive o modality Mode of perception eg visual auditory o tactile Having to do with the sense of touch oprecategorical Based on low level sensory information Verbal terms a word frequency How often a word occurs in written text a semantic Having to do with the meaning of a word Experimental terms a manipulation Change in an independent variable 0 output The S s responses In a recall task these are the words S says Quasitheoretical terms a span How many items of a particular type can be remembered perfectly a capacity How much information memory or a hypothetical memory store can hold Forgetting terms a decay Forgetting caused by the passage of time o interference Forgetting caused by presentation of other material A statistical term a correlation Measure of the relationship between two variables 1 means perfect correlation 0 means not correlated 1 means anticorrelated Dichotomies a central peripheral Having to do with higher order cognitive processes lower level perceptual processes a objective subjective Measurable to an external observer description of S s experience a nominal functional The actual item presented to S the actual stimulus experienced by the stimulus including their reaction to it Assignment 0 First Experiment Report due Wednesday Partial Report a Read Chapter 2 for Wednesday 0 Have a great Labor Day Weekend Interference Theory PSY 400 Human Memory April 5 2004 Housekeeping 0 Get back your exam 0 Implicit memory due Wednesday a Chapter 6 is closest to the material we ll talk abouttoday Some stats on the exam 0 range of raw scores 25 to 45 out of 50 a score 30 155raw a mean 84 i 95 o 7 A s 5 8 5 5 C s 2 D s Let s go over it Important things from last time o Distinction between explicitimplicit directindirect oLogic results of serial reaction time task experiment a Memory systems and double dissociations What s a double dissociation The implicit learning experiment results 45 55 including last semester Reaction Time ms lgt O1 O1 00 3 Oquot O O O l 4 4 C l 4 C C Did you notice what happened at block 10 Were you aware of the sequence Implicit learning experiment report 0 Empirical results logic of random vs blocked o Methodological issues 0 Theoretical discussion Do you think there s a subconscious memory system driving this result Overview of today s material 0 Some loose ends from implicit memorymemory systems 0 Interference theory Retroactive and Proactive interference Findings from MFR and MMFR Spontaneous recovery Is there unlearning What memory systems are there table 74 1 Procedural 2 Perceptual representation 3 Primary memory 4 Semantic 5 Episodic Procedural a Simple Pavlovian conditioning 0 Motor skills riding a bike 0 Some sequence learning problems Perceptual representation system 0 Argued to be responsible for repetition priming o Priming of possible but not impossible objects Fig 76 o The possible objects must have their own special region a Ratcliff and McKoon 1995 argued this is due to recollection Table 75 Summary of memory systems research 0 This is big business 0 Even if correct at some level it s clearly an oversimplification 0 Procedural distinction well accepted PRS not seprability of episodic semantic hotly debated What causes forgetting over time Decay vs Interference a We forget over time o Is this a consequence of time per se 0 Think of rust Interference Theory 0 Retroactive interference o Proactive interference o Unlearning Melton amp Irwin 1940 the classic retroactive interference study 055 learned serial lists by the method of anticipation a They learned list 1 then list 2 for some number of trials control group zero trials list 2 0 At the end tested on list 1 0 RI is difference between the control group and the experimental groups recall Melton and Irwin s Factor Xquot 0 RI goes up with second list learning a Intrusions from second list go up then down a What accounts for the difference Candidates for Factor X o Unlearning Learning second list causes first list associations to be weakened a Response set learning With additional practice S better able to keep different lists straight Proactive interference Figure 63 0 Recall also Release from Pl experiment 0 Observed similar effects in free recall Major Interference Paradigms PI and RI were extensively studied with paired associate learning where you can localize the interference to a specific pair See Table 61 Why is there Pl in paired associate learning Testing itemspecific interference Goal is to control response competition 0 Modified free recall MFR o Modified modified free recall MMFR Modified free recall a If you learn A B and then A D and i ask A what does it mean if you say D 0 Maybe you know both but you just say D first sometimes a Modified free recall instructed to say first one you think of Briggs 1954 oLearn A B and then A D oWait for some retention interval 000 025 050 075 100 000 025 050 075 100 06 24 33 72 B I h Level of Orlg Learnl ugvel of luterpolated 1mm Perlod 1n Hours Figlre 5 3 Data om the m odjfed ee recall procedure ofBrigg 1954 See text for detaii An Aside about Spontaneous Recovery 0 Responses compete with each other oAssociations decay over time accelerated forgetting o A B should appear to improve as you forget A D like in the Briggs data Modified modified free recall a How do we separate response competition from unlearning oln MMFR you are supposed to say both responses Pl in MMFR 0 Study A B for say 10 trials 0 Study A D for variable number of trials o MMFR shows PI 0 Tm On Secund Lust Why is there Pl in MMFR Do you unlearn AB when you learn AD 0 Maybe there s still response competition 0 Independent associations Logic of independence of associations If there s unlearning then the better you learn A D the worse you should remember A B In particular if there s unlearning PBlD lt PB Also PBampD lt PBPD Some MMFR data Abra 1969 Immed MMFR 24 hr MMFR Resp D f D f B 43 11 54 42 35 77 E3 38 08 46 14 09 23 81 19 56 44 Evidence for independent associations Abra 1969 ImmaiMNWR 24hrMMFR PKI l P113 PKI l PKI3 53 54 75 77 1BampD1B D1BampD1B D 43 44 42 43 Factors that could yield artifactual independent associations 0 Maybe positive correlation from some property of A o Artifact of averaging over 55 or items 0 Maybe the associations are independent if so how do you explain the decrease in A B recall Connection of PA learning to other paradigms 0 Most of forgetting is RI 0 What does this A BA D thing have to do with say free recall a You can think of buildup of PI in free recall as an A B A D experiment o but then what is A What is the cue in uncued recall a In Brown Peterson you re asked to recall the most recent trigram o In free recall you re asked to recall the most recent list 0 There s no explicit cue but something must be prompting recall 0 Let s revisit interference in Brown Peterson Revisiting Keppel and Underwood 0 Perfect performance 5 09 on first trial even g 0 with long delay E 02 E 07 n 065 lt 3 Seconds E 06 39 9 Seconds n O 5 39 E 18 Seconds I 0 Fig 64 Revisiting the Buildup of Pl 0 Remind us about the buildup of PI paradigm o Retrieval effect Gardiner et al 1972 Release from Pl paradigm 0 Word trigrams from a category 0 Eg CAR BIKE SHIP 0 Some 55 get a category shift after a few trials 0 Eg shift from methods of transportation to vegetables Release from Pl results 1 09 I 08 07 06 05 04 O 3 u C onttol Proportion C orrect 02 01 39 Exper39nental 0 I l 2 3 4 Trial Pl in BrownPeterson as a Retrieval Phenomenon Figure 66 o Subtle change in category garden flowers vs wildflowers oTell S either before learning critical trigram before test on the critical trigram or not at all Semantic Distance in Release from Pl Degree of release depends continuously on semantic distance Proportion Correct l l 1 2 3 4 Trial F39guxe4 5Propoxtbn conectjn eexecalllasa mctbn oft he numberoftc39ails W J39ckens 1972 Forthe zst neett ls J39temsconslsted oftypesofftujt IIIanng39yen on the burlh tbalwetese lected from the semantic category ildirated above More evidence that forgetting in BrownPeterson due to interference o Turvey Brick and Osborn 1970 varied delay intervals table 65 0 Found that PI was comparable for different delays 0 When a final delay of fixed value was given recall depended on prior delays Assignment 0 Developmental next topic Ch 14 o Implicit learning report due Wednesday 0 Next experiment is false memory The Modal Model II The Atkinson amp Shiffrin Model PSY 400 Human Memory Spring 2004 February 2 2004 Important stuff from last time oDistinction between primary and secondary memory 0 Brown Peterson task Keppel and Underwood Release from PI 0 Waugh Norman result decay vs RI in probe digit task 0 Concept of rehearsal in maintaining info in STS Some notes about the experiment reports 0 Everyone got a nine or a ten great job 0 Theory was usually the problem 0 Might go to 100 point scale in the future 0 No points for length per se 0 caret Overview of today s material a The Atkinson and Shiffrin model 1 Control processes 2 Continuous paired associate learning 3 Detailed description of performance 0 The two store model in free recall 1 Rehearsal as an empirical phenomenon 2 The recency effect in free recall 3 The primacy effect in free recall The state of primary memory research in 1968 oTheoretical distinction between primary and secondary memory olmportance of rehearsal in maintaining information understood 0 Role of interference Atkinson and Shiffrin 1968 o Watershed moment in cognitive psychology still cited 100 times a year 0 Explosion of modeling and experiment done by Atkinson s lab in the mid 19605 o Brought it all together in a coherent theory The Atkinson and Shiffrin 1968 Model 0 STS is a m Externii mew SENSORY REGISTER r capacity buffer 0 Control processes 39 E E i H L SHORTTERM STORE j l i operate on STS r4 m L o Forgetting in LTS by i l I I i A L r n I II V I p LONG TERM STORE j v I decay interference r n I I I 39 nc quot L p J Fla 1 Structure of the memory system Automatic vs controlled processing 0 Automatic processing happens independently of volition on the part of the S a Controlled processing is a result of strategic decisions on the part of the S Sensory registers VS STS oType of information sensory vs modality independent avl o This means semantic information available in STS not so much sensory registers 0 Time course STS typically longer 0 Type of processing automatic vs controlled Controlled Processing in STS 0 Control processes F maintain information in ll STS SHORTTERM STORE o Rehearsal is main one o The longer in STS the more transfer to LTS l nzmv LDSS or Fla 2 The mhanrsal buffer and its relation to the memory ayahem A t 39 d 39 t t t 5 H E E a H E M I m us 1m 5 a in 1 Fla 3 A sample sequence 11 um fur Exparimenb 1 A Mathematical modeling of STS o Number letter paired associates tests a New item would either enter the buffer or not a If stimulus already in buffer certain to enter a If stimulus not in buffer some probability of it displacing a random item Stimulus already in buffer Before After 17J 17J 34L 34 0 34L 29K 29K 48M 48M Stimulus not already in buffer 17J 4444 34L 4444 29K 4 48NI 4 gt A prediction 0 An S R pair replaces an S R pair in buffer with the same S a An S R pair can displace an S R pair if the S isn t in buffer a Chance of displacement are higher if number of S is higher than buffer capacity Mathematical modeling of LTS 0 Amount in LTS related to time in STS o Exponential trace decay over time Describes Performance 0 Level of performance at long delay depends on time in buffer 0 Chances of entering buffer greater if fewer stimuli o What is an alternative explanation Allsame vs Alldifferent in cont PAL Same 12 X 37 G 37 L 37 0 37 D 12 M Different 12 X 37 G 62 L 29 0 54 D 12 M Another prediction What happens if you get the same stimulus over and over a An incoming stimulus can only displace items in the buffer if the incoming stimulus is not in the buffer a Forgetting should be slower for same A detailed description of performance All same All different PROBABILITY of A connacr RESPONSE moawmn or A connzcr RESPONSE A N a n 39m 4 m a And predicts effect of overt covert rehearsal o In the overt condition subjects repeat each pair a Atkinson and Shiffrin argued this led to automatic entry into buffer a Model predicts certain patterns of results Forgetting with overt covert rehearsal m o mman V or A comm nEsmNsE m and details with overtcovert rehearsal PnoeAavurv or A CORRECT RESPONSE The twostore model in free recall 1 Rehearsal as an empirical phenomenon 2 The recency effect in free recall 3 The primacy effect in free recall Serial Position Experiment The free recall task 0 Get a series of words 0 Recall them in any order 0 Let s give it a try BLESSING SUMMER CANOE FLAVOR DUTY RATTLE GARMENT MOTION OVEN FEVER HAMMER TENNIS CONGRESS POWDER LEVEL NEEDLE PAPER DAUGHTER TRAFFIC How is this different from what we did in coglab AampS model also applied to free recall oThe recency effect in immediate and delayed recaH o The primacy effect in free recall a Directly measuring rehearsal The recency effect in immediate and delayed free recall Howard and Kahana 1999 The AampS model described this as two stores 0 Immediate recall uses STS LTS 0 Delayed recall uses only LTS Serial Position 0 Recency effect due to STS Many variables affect one end of the SPC but not the other 0 Modality better recency for auditory o Amnesia Baddeley and Warrington 1970 0 Word frequency better recall of early items 0 Semantic similarity of list better early 0 Presentation rate better early Presentation rate and recency 1 Probability of Recall 02 5 10 15 20 Serial Position Murdock 1962 The primacy effect in free recall Probability of Recall O O O O O O O O O OI Lll ilgt39 ICDIJICDIQD L 10 20 30 40 Serial Position Murdock 1962 Advantage for recall of first several items in the list AampS model predicts primacy o No displacement from buffer until buffer full 0 Capacity typically 2 4 a First items receive more time in buffer hence better recall Evidence for AampS account of primacy 0 Early items are rehearsed more fig 39 0 Conditions under which rehearsal is discouraged affect primacy 1 Incidental learning 2 Concurrent task Incidental learning a Subject doesn t know there ll be a memory test a Presumably no need to engage in rehearsal a Marshall and Werder 1972 found essentially no primacy effect with incidental learning Incidental learning results 1 Intentional HG 39 08 7 InCIdental O D E 06 7 O 3 g 04 7 0 9 a 02 i 0 5 10 15 Serial Position Concurrent task 0 Subjects know there will be a test 0 But they have to do other processing on each word 0 Should prevent control processes Concurrent task and primacy 1 Murdock 1962 no 08 MESSquot 39 M62 15 task g E067 0 Howard and 04 7 Kahana 1999 E concreteabstract 0392 judgement 0 i i i 5 10 15 Serial Position The basis of rehearsal 0 Subjects rehearse o In overt rehearsal Ss rehearse out loud oYou can then analyze recalls as a function of rehearsal Rehearsal predicts recall 0 Nominal vs functional serial position Figure 34 Tan amp Ward 2000 Brodie and Murdock 1977 a Word frequency effects Ward et al 2003 o Spacing effects Rundus 1971 0 von Restorff effects Rundus 1971 a Do you really need two stores maybe rehearsal isjust recall Summary 0 A85 model relied on control processes in STS 0 A85 model describes continuous paired associate learning in extreme detail oAampS model describes basic and important characteristics of free recall Assignment 0 Serial Position experiment report due Monday a Chapter 6 by Monday 0 Supplemental readings 1 Kahana Howard Zaromb and Wingfield 2002 for Wednesday 2 Howard and Kahana 1999 optional 3 Chapter 3 Kahana Foundations of Human Memory optional PSY 400 Memory amp the Brain Marc W Howard January 13 2003 Am i in the right place a PSY 400 Memory amp the Brain 0 SECMOOQ 39838 c HBC 323 Mon 3 420 Who the heck is this guy 0 Marc Howard Assistant Professor Dept of Psych a Training and research expertise a Teaching experience a httpmemorysyredu Should i really take this course 0 This is a science course 0 We will eventually cover relatively advanced topics in Cognitive Psychology and Neurobiology 0 Although this requires little or no background in either of these topics it will be much easier for those who have some Syllabus Readings 0 Cognitive Neuroscience of Memory by Howard Eichenbaum 0 Supplemental readings httpmemorysyredu a You are responsible for the material covered in lecture Syllabus Grading I want everyone to do well work hard and respect the material a Exams 2 plus a final 0 Class participation 0 Extra Credit assignments Extra Credit assignments Write a brief 2 4 pages report on a paper from the literature 0 Up to one per week 0 Up to two points per assignment 0 Up to 10 points on your final grade Extra Credit Opportunity Eectrophysioogica Correlates of Human Spatial Navigation Michael J Kahana Volen National Center for Complex Systems Brandeis University 1030 AM Friday Jan 31 location TBA Syllabus Overview of content a Neurons o Synapses o LTP o Non NMDA dependent plasticity o Kahana prep 0 Cortical plasticity o Non declarative Memory 0 Anatomy of the MTL o Single unit studies of the MTL o Pharmacology and Physiology of the MTL o MTL Amnesia o Declarative memory 0 Semantic and Episodic Memory 0 Memory Space o Recency and Contiguity in Human Memory 0 The Temporal Context Model Preliminaries o Read Chs 1 2 for Wednesday 0 Things you might need to know first What is 0 Ion a Current 0 Voltage 0 Concentration gradient Ions 0 Electric charge 0 Electrons and Protons o Atoms and ions Voltage 0 Electrical Potential 0 Imagine a ski slope Current 0 Movement of charge 0 What s it moving through 0 V IR 0 Resistance and Conductance Diffusion 0 Concentration gradient 0 Imagine a room with snow blowing around and another room without What are we and how do we remember 0 The brain a Circuits 0 Cells The Brain mew mW m m M m Emu Gluilll39 Mum Ith can Pm mull cm Circuits A VII Ec layers 5an Cells The mirror effect and Recent advances in the study of recognition memory PSY 400 Human Memory March 22 2004 Housekeeping oRememberKnow Experiment report due Wednesday 0 Are there questions about that Words of wisdom 0 Experiment reports are 40 of your grade 0 There have been for reports There will probably be about another 3 or so a Not handing in two or three experiment reports is about like missing an exam 0 You get points for handing in late experiment reports so do that Important things from last time o Episodicsemantic distinction o Recollection and Familiarity 0 Signature of recollection and familiarity in ROC curves What is the difference between episodic and semantic memory How do recollection and familiarity map on to this Confidence levels in yesno recognition memory oYesno recognition asks you to rate your confidence on a 2 point scale 0 Rather than responding yes or no you give a more graded response 1 2 3 4 5 Sure No Sure Yes Confidence judgements and ROC curves 0 There are several criteria eg gt 5 gt 4 etc a We can calculate a hit rate and a false alarm rate for each of the criteria a Plotting these as a function of each other yields an ROC curve ROC curves 0 Several different criteria for 1 yes 08 06 oROC is hits vs falseEs alarms 04 02 t o for several different 00W i criteria PM How do recollection and familiarity contribute to the ROC curve Overview of Today s Material 0 Mirror effect 0 Recent advances in the study of recognition memory 1 The effects of scopolamine on recollection and familiarity 2 Continuous response technique 3 Contiguity effects in recollection The mirror effect 0 Consider two classes of stimuli one of which is easy to remember the other of which is not a The hit rate will be higher for the easy class a The false alarm rate will also be lower for the easy class eg table 98 o This is a pretty general finding Table 97 Why is this called a mirror effect a See figure 96 o This seems paradoxical there s a difference in responses to unstudied items a If you re less likely to say yes to new easy items why aren t you also less likely to say yes to old easy items 0 Signal detection theory is incomplete Imagine how this would look in a SDT framework Scopolamine and recognition memory oScopolamine is a drug that has amnestic properties a It is a cholinergic antagonist that binds to muscarinic acetylcholine receptors 0 Acetylcholine is important in memory Methods The task a 300 complex photographs presented during study 0 Blocks of 30 pictures one item each 2 s o Indooroutdoor orienting task a 15 min delay prior to recognition testing 0 Subjects rated confidence on a 5 point scale 1 2 3 4 5 Sure No Sure Yes ls recognition of pictures different from recognition of words Old or new How are pictures different from words Methods Pharmacological concerns 0 IM injection of 4 mg scop 0 Presented pictures for recognition 90 min after injection 0 Recognition testing began approximately 230 after injection New X24 68ns Old X24 38 p lt 001 Sure No Sure Yes Sure No Sure Yes 0 0 Scopolamine Proportion Responses 39 n lo Lo 2A Proportion Responses 0 o o 0 N A A 3 m o on NEW OLD Sure No Sure Yes Sure No Sure Yes 0 0 Control Proportion Responses 39 n lo Lo 2A Proportion Responses 4 3 2 1 3 m Behavioral Results OLD There is a recognition deficit with scopolamine N w Old 0 4 0 4 0 3 0 3 E E m m a a o 02 o 02 2 m x 0 1 0 1 0 1 1 1 1 0 1 1 1 1 0 01 02 03 04 0 01 02 03 04 Control Control The Yonelinas High Threshold Model a Probes are recollected with probability R o If recollected the S says yes to every possible criterion 0 If not recollected decision depends on a signal detection process The Yongjinas High Threshold Model Probability Density P yes old P lyeSWnew Od Strength R1 R NWOJQSMS z ROC curves 0 ROC is hits vs false alarms is a z transformed o Z ROC ROC oSDT predicts straight line z ROCs zHits I L 125 O 15 05 zFAs 05 How would a z ROC with recollection look YHT and ROC curves P hit 2 P faR1 R lt1gt c ltIgt g c o YHT makes z ROCs with JP two parameters R and d oYHT predicts non linear O I Q z ROCs 1 2 25 15 05 05 15 o R causes the non linearity FA 2 S Sure No Sure Yes Sure No Sure Yes Scopolamine Proportion Responses Sure No Sure Yes Sure No Sure Yes Controls Proportion Responses NEW YHT does not fit the scop data OLD YHT does not fit the scop data 2 Control Scopolamine zhts zHts 0 Sure No Sure Yes 0 Sure No Sure Yes Sure No Sure Yes Q Sure No Sure Yes Q Proportion Responses 90090 Amwbm 06 Proportion Responses 90090 Amwbm 06 Proportion Responses 0 o 099 Amwbw 06 0 Sure No Sure Yes Sure No Sure Yes Proportion Responses opoop O Nubu 06 Control Proportion Responses opoop O Nubu 06 Proportion Responses 0 o 009 OANmbm 06 Proportion Responses 0 o 099 Amwbw 06 0 Sure No Sure Yes Q Scopol Proportion Responses o o 009 Ammgtm 06 0 Sure No Sure Yes H amlne Qualitative pattern observed across 55 Qualitative pattern observed across Ss Control Scopolamine 3 H ts H ts zHns A H l 4 5 05 o 125 15 05 05 15 715 705 05 15 r 71 zFAs zFAs zFAs zFAs H ts H ts zHIIs A H 1 4 r 72 4 1 2 1 zFAs zFAs zFAs zFAs An expanded twoprocess model of recognition 0 Instead of all or none some or none recollection Probability Density 0 R yields a distribution llll 1234 0 New parameters LLR Strength and JR Expanded model does fit the scop data NEW OLD 05 05 II In 3 3 g 04 S 04 9 a o 3 03 g 03 g n n C x c o 2 02 2 02 1 E o 01 o 01 1 I a 1 I 0 I 0 Sure No Sure Yes Sure No Sure Yes 05 05 0 II o d 0 g 04 g 04 l o o 39g 0 03 03 2 E 1 O x n g 02 9 02 O t 1 a a a o 01 o 01 n n 0 I o Sure No Sure Yes Sure No Sure Yes Expanded model does fit the scop data Control 2 Scopolamine L zhts K zhts 0 MN q HR 0R Parameter values Response distributions YHT VR Control Scopolamine Control Scopolamine 246 246 246 181 153 149 153 159 126 152 126 125 032 017 32 40 493 408 07 19 Parameter values Z ROCS YHT VR Control Scopolamine Control Scopolamine d 062 052 60 33 R 035 022 36 35 MB 25 18 0R 5 39 Modeling lessons 0 Scopolamine affects quality not quantity of R I 0 Sure No Sure Yes 0 Sure No Sure Yes Control 05 NEW 05 OLD a a o Scopolamine affects F 2 E 03 503 t 8 IE 02 IE 02 00 3 EN EN E E 0 0 Conclusions from Scopolamine study a Scopolamine affects recognition memory a Recollection and familiarity were both affected by scopolamine perhaps they re not so separate after all a Recollection is not all or none It s some or none Motivation for the continuousresponse technique 0 In yesno recognition you have two choices 0 With multiple confidence levels you have say 5 choices a 5 is better than two so why not have an infinite number of choices How many heights are there The continuous response technique 0 Continuous response variable 0 Continuous ROC oMore constrained for modeling NO YES Comparable discriminability ROC with Mouse and Keyboard MJMMouse MJMKey 08 i 07 i 06 i 05 Hit Rate 04 i 03 i 02 i 01 0 01 02 03 04 05 06 07 08 09 1 False Alarm Rate What type of continuous response distributions would you expect to see if SDT holds What about the Yonelinas model Signal Detection Theory JMMouse Response Distribution 2000 mouseaux 1600 ouse39 o Shifted old item distribution Frequency 0 Subjects made discrete responses 0 i i i i i i i i i 0 01 02 03 04 05 06 07 08 09 1 Response Individual Subjects 1 an 7 us m an r Iuor Rem WM i an 7 Resume Individual Subjects JM MauseeBeOLDeResprmSY JMMauseeseReSPROC 1am e use 15D e DE MD e me 2 e us FrequenDV M R at m me as U2 us JMMausee Nsvweeuowsr FrequenDV An interesting possiblity Maybe there isn t a continuous strength at all The Malmberg model a Malmberg argues that there are a small number of discrete memory states c There is noise in the mapping between these memory states and the responses leading to a smooth looking ROC curve 0 Here we appear to see discrete memory states Continuous Response Technique Summary 0 Subjects show comparable discriminability with continuous response 0 Subjects appear to use a few discrete states 0 This is consistent with the Malmberg approach Continuous Response Directions 0 Deconvolve error a See if instructing different causes more SDT friendly distributions Recognition vs recall 0 Recognition is often thought to be a test of the item alone item information oFree recall is usually thought to involve associations between items 0 But they should both be episodic memory so why the difference Associations in free recall 0 Brief reminder about the CRP a General properties of episodic associations What is free recall Measuring Episodic Association Association when A is called to mind B tends also to spontaneously come to mind AB SENCE HOLLOW PUPIL Recalled RIVER DARLING RIVER CAMPAIGN DUTY CAMPAIGN HELMET SISTER P U P I L VELVET Conditional Response Probability CRP Kahana 1996 Measuring Episodic Association Association when A is called to mind B tends also to spontaneously come to mind 3 2 1 9 i g 3 Recalled vggk bbb RIVER 02gt 06 9 0 7 0 lQltgtOz 4 akaC CAMPAIGN PUPIL Conditional Response Probability CRP Kahana 1996 Measuring Episodic Association Association when A is called to mind B tends also to spontaneously come to mind 2 3 2 1 9 i g 3 Recalled gtslacks RIVER 02gt 06 9 0 7 0 lQltgtOz 4 akaC CAMPAIGN PUPIL Conditional Response Probability CRP Kahana 1996 Measuring Episodic Association Association when A is called to mind B tends also to spontaneously come to mind 3 2 1 9 i g 3 Recalled z a a a s waggt052 RIVER lo 42 lt9 4 CAMPAIGN PUPIL Conditional Response Probability CRP Kahana 1996 Measuring Episodic Association Association when A is called to mind B tends also to spontaneously come to mind 1 2 3 Recalled A p 2 6 lt5 amp 4 RIVER x 0 74gt 0 Q 5 C 4 af xO 476x99 CAMPAIGN PUPIL Conditional Response Probability CRP Kahana 1996 Contiguity and Asymmetry in the CRP CRP 4 2 o 2 4 4 2 o 2 4 4 2 o 2 4 4 2 o 2 4 4 2 o 2 4 4 2 o 2 4 Lag 4 3 2 1 O Conditional Response Probability 0 0 0 0 J igt CD 03 L Our CRP Does the CRP reflect mental time travel 2 Delayed Cont Distractor Measuring associations in recognition memory a A hallmark of recall is better memory for stuffthat happened near in time after you recall something 0 Unlike free recall the order of testing is fixed by the experimenter oSo let s compare cases in which we present consecutive old items Relative lag in recognition What will happen a Will 7 have any effect at all o If so should it depend on whether you recollect the first item or not 0 Why does this happen Associations in recognition memory 08 a 06 6 n f 04 02 Adjacent 0 Remote 0 0 02 04 06 08 1 Local False Alarm Rate Association and recollection 0 Separate on pnor response 0 Open symbols r ghest con dence recoHchveresponse 0 Filled symbols con dence 1 5 presumably not recol lected 08 07 Hit Rate 04 Ha 03l l l l l l l l l ll 1086420 2 4 6 810 Relative Lag Why is this 0 Let s say i only open my eyes for 7 s at a time while studying the pictures a This could make an effect like the one we just saw o If this is why we see the effect then it shouldn t matter if you actually test the items in order or not o It does matter If recall and recognition are so similar why is the recognition CRP symmetric Conclusions Temporal retrieval effects in recognition 0 Recollection of an item results in better memory for old items that follow if they re near in the list to the first item 0 Maybe recollection really does result in mental time travel 0 Put another way maybe recollection retrieves context context changes gradually over time and we make recognition judgements by comparing an item s context to the current context Conclusions 0 This distinction between R and F is meaningful o The relationship between R and F is much more complex than currently appreciated oRecollection retrieves context that changes gradually over time Assignment 0 Sherman et 3 optional reading would help a RememberKnow due Wednesday 0 Exam review Wednesday bring questions 0 Next chapter is Implicit Memory probably not till week from Monday
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'