Karteikarten
Memory Structure
Karten in diesem Stapel (71)
- 1Memory defintionsMemory is the process involved in retaining, retrieving and using information about stimuli, images, events, ideas and skills after the original information is no longer present.
• Memory is active any time some past experience definitions the way you think or behave now or in the future. - 2Atkinson & Shiffrin (1968) Modal ModelStructural Features
– Sensory Registers
– Short-Term Store
– Long-Term Store
• Control Processes
– Dynamic processes associated with the structural features E.g. rehearsal - 3Sensory storesVisual Sensory Store (Iconic) Sperling (1960) - Decays within about 0.5s – Auditory Sensory Store (Echoic) – Treisman (1964) Duration of information in the store ~2-4s
- 4Short-term memory• 2 sources of active information:
• Internally represented information that is in a state of activation from the sensory stores - Activated information in LTM held in the STM• Decay of activation causes information to leave STM • Rehearsal – retain in STM & more likely transfer to LTM (Rundus 1971) - 5STM Forgetting• Displacement if too much information in STM – Miller (1956) – Magic number 7 +/- 2 – digit span – Additional items will displace current items in STM – Can be improved with chunking – B B C I T V S K Y becomes BBC ITV SKY Decay – if the information is not rehearsed it will decay rapidly
- 6Cowan (2001)– STM capacity may
actually be around 3-4 items - 7Luck & Vogel (1997)used a Change Detection Paradigm
• Two pictures flashed one after the
other, determine whether ‘same’ or
‘different’.
– An easy task if within STM capacity.
– Deterioration began once more
than 4 items in the display - 8Change DetectionUses visual items that cannot be verbally recoded – impossible to rehearse or put them into larger ‘chunks’. • Measures STM capacity independent of control processes/LTM involvement (which you get with word lists). • Individual Differences on this task found which correlate with measures of higher cognitive functioning: – Individuals with larger change detection capacity perform better on complex cognitive tasks (verbal and visual reasoning, reasoning/problem solving etc.) – Suggests visual short-term memory capacity plays a central role in cognitive functioning.
- 9Alvarez & Cavanagh (2004)– Change detection using more complex stimuli • CD capacity:
– coloured squares 4.4
– more complex cubes 1.6
• The greater the amount of information in an image, the fewer items that can be
remembered • A capacity of STM ap - 10Transfer to LTM• Information transfer from STM to LTM
– Rehearsal the important process here
• Depth of rehearsal processing important - 11Craik & Lockhart (1972)• - Levels of Processing Hypothesis
– Maintenance rehearsal – Mere repetition does not aid transfer (STM-LTM) e.g. coins– Elaborative rehearsal – Deeper processing - forming associations, attend to meaning, thinking about new info etc. – More likely to result in transfer of info from STM to LTM - 12(Postman & Phillips 1965)Evidence for Modal Model
(STM-LTM Distinction) Serial Position Curve: P’s asked to recall the list of words in the order in which they hear them.
• Primacy effect • Recency effect -Eliminated by brief delay filled with distractor task - 13Neuropsychological Evidence• Brain damaged patients provide evidence for STM/LTM distinction
– KF - Stroke leading to damage to left parietal lobe• Impaired short-term memory – Digit / letter span - two items or fewer; • Normal long term memory
– Forms & retains new memories his life - 14H.M. & Clive Wearing• Damage to medial
temporal lobe (inc.
hippocampus)
– Impaired LTM
– Intact STM - 15K.F.• Damage to parietal and
temporal lobes
– Intact LTM
– Impaired STM for letters,
words and digits - 16Patients with prefrontal lesions (linked to STM deficit) :• Near normal LTM with structured encoding • Impaired LTM performance:
• When required to initiate strategies to actively encode info • When they require strategic processing during retrieval - 17(Ranganath & Knight, 2003). Hypothesis: Prefrontal cortex is likely necessary for implementing control processes (strategies) that can contribute to both STM and LTM performance.
- 18Some neuropsychological evidence suggests medial temporal lobe damage can impair...STM (as well as LTM)
- Novel visual objects: deficits seen even with 2–10s retention. - 19Alternative hypothesesThere are no large-scale cortical areas that uniquely support STM or LTM but there might be: - Different neural circuits that support STM and LTM within the same cortical area OR - Differences in the dynamics of activation within the same
cortical circuits - 20Modal model: Evaluated - strengthsStrengths • Evidence that the stores
differ in several ways:– Duration– Capacity – Forgetting mechanisms – Effects of brain damage - 21Modal model: Evaluated - weaknessesWeaknesses • STM not only route to LTM
– Unconscious (implicit) learning occurs without ever occupying the STM – Rehearsal may not be as crucial to learning • Neuropsychological evidence
suggests a more complicated picture - 22STM Store• Early discussion around how much information (capacity) can STM hold and
how long (duration) can it hold it for?
– Both highlight STM as a storage mechanism – Research suggest this is too narrow • STM much more dynamic – LTM to STM more active than ‘store’ suggests •better called Working Memory - 23Baddeley & Hitch (1974)defined WM as – ‘a limited capacity system for storage and manipulation of information for complex tasks such as comprehension, learning and reasoning’ • Not just storage of information – much more active – we manipulate information in the STS
- 24Baddeley & Hitch (1974)Continued– it is possible to do more than
one task at once if those tasks are sufficiently different • Examined dual task vs single task performance (doing two things at once) • Proposed that WM must be dynamic and consist of a number of components which can function separately. - 25Working memory model (Baddeley & Hitch, 1974)Central Executive:• Modality free • Resembles attention • Deals with any cognitively demanding tasks• Pulls information from LTM
Co-ordinates the two slave systems and allocates attention to them: 1. Visuo-spatial sketch pad 2. Phonological Loop - 26A Limited-Capacity System• The components of the system are thought to be limited in capacity and relatively independent: – Two tasks using the same component of WM can’t be performed successfully together – Two tasks using different components can be performedas well together or separately
- 27• Baddeley & Hitch (1976) - Two simultaneous tasks (dual task):• Digit span task (repeat list of numbers)
• Verbal reasoning task ( e.g. true or false - B follows A?). • Participants took longer to answer reasoning questions.• Hypothesis - Verbal reasoning task uses the central executive and digit span task uses the phonological loop. - 28Interference Measures Robbins et al. (1996)• Selecting chess
moves involves
central executive and visuo-spatial sketchpad • Not the phonological loop
Effects of secondary tasks on quality of chess-move selection in
stronger and weaker players - 29Other Findings of the WM• WM model has to explain a whole host offindings regarding performance on short termmemory tasks including:
– Effects of articulatory suppression
– Phonological Similarity effect
– Irrelevant speech effect
– Word length effect - 30Murray (1967) -Articulatory Suppression– Digit Span task: Visual presentation of a sequence of digits (1,5,3 …..) – Prevented sub-vocal rehearsal by uttering ‘the’ at audible level during presentation of digits – Memory significantly poorer on digit task in suppression condition – Why does sub-vocal rehearsal interfere in the retention of visual stimulus?
- 31• Larsen, Baddeley, and Andrade (2000)– Recall 25% worse with a phonologically similar list compared to a dissimilar list
Phonological Similarity Effect – Suggests we use speechbased rehearsal processes
within the phonological loop– Again WM model needs toexplain why memory for phonologically similar lists is worse! - 32Salame & Baddeley (1982)Irrelevant Speech Effect • Sequential visual presentation of 9 digits (e.g. 7, 4, 9, 2, 6, 8, 5, 1, 3) with – Condition 1 with spoken words (e.g. bid, pot)– Condition 2 with nonsense syllables(e.g. pid, bot) – Condition 3 with silent control • Told to ignore sounds but digit errors 40% higher in 1 & 2
- 33• Salame & Baddeley (1987)– found unattended noise has no effect on serial recall but word type stimuli do. Why?
- 34Baddeley, Thomson & Buchanan (1975)Word Length Effect–Repeat sequences of words (controlled for frequency)– Short (sum, wit)– Long (aluminium),– Retrieval/span worse for longer words– Why? - because they take longer to rehearse? – Baddeley et al – we can remember the number of items
we can pronounce in 1.5-2 secs–Effect is eliminated by articulatory suppression - 35Ellis & Hennelly (1980)Word Length Effect • Application
– Digit span is greater in languages where digits faster to articulate– Chinese 265ms: digit span 9.9
– English 321ms: digit span 6.6
– Welsh 385ms: digit span 5.8 - 36Two Components of phonological loop• Phonological store
–Auditory presentation (speech) has
direct access
• Articulatory Process
–Visual presentation indirect access to
phonologica - 37• Phonological Similarity effect– Confusion in phonological store
- 38• Irrelevant Speech Effect– Visual presentation requires verbal recoding
– Interference from concurrent speech - 39• Articulatory Suppression– Prevents verbal re-coding of visually presented
material into phonological store - 40• Word Length Effect– Phonological loop with capacity of approx. 1.5
secs - 41Henson, Burgess & Firth (2000) –neuroimaging • Phonological store
– Left inferior parietal cortex activity associated with phonological store
• Articulatory process – Left inferior frontal cortex activity associated with verbal rehearsal - 42– Vallar & Baddeley (1984)damage results in intact phonological
store but damaged articulation process - 43Uses in Everyday Cognition1. Role in learning to read, language
comprehension and vocabulary
acquisition
2. Developmental dyslexia
• Striking feature - impaired memory
span - 44• Majerus (2006) –correlation
between phonological loop
capacity and vocabulary size - 45Visuo-Spatial Sketchpad• Temporary storage/ manipulation of spatial and visual
information - 46• Logie (1995) - visuo spatial sketchpadconsists of:
– Visual cache stores visual form & colour info – Inner scribe processes spatial & movement info
• VSSP research is less well-developed than phonological
loop - 47Shepherd & Metzler (1971)• Look at two pictures, decide as rapidly as possible if they are same or different. • When one shape rotated 40 degrees it took 2 seconds to say they were the same, 140 degree rotation took 4 seconds. • Longer decision greater differences in orientation •mentally rotate images to answer question - VSSP
- 48Klauer & Zhao (2004)good on spaital and visual but not two of the same e.g. spatial spatial
- 49Farah, Hammond, Levine & Calvanio (1988)LH:– Good on spatial tasks
– Brookes’ spatial matrix task,– Shepherd’s mental rotation task• Poor on imagery tasks– e.g. judging the relative sizes of animals– dog vs. sheep
– Consistent with separate visual and spatial systems - 50Visuo-Spatial Sketchpad: pros and consStrengths• Supported by research showing independence in spatial and visual tasks • Support from studies of brain-damaged patients (LH) Limitations • Many tasks require both comonents– Not yet clear how information is combinedand integrated
- 51Central Executive• Imagine you are driving along in your car and you are: 1. Being given directions by your passenger friend • Requires the phonological loop to understand
instructions and the VSSP to visualise the streets ahead of you 2. The news is on the car radio - 52Baddeley - CE -pure attentional system’ •Divided attention – between two messages • Switching attention – e.g. random generation of digits – switch between generating and checking randomness • Monitors and co-ordinates slave systems, allocates limited resources • Lack of info CE greatest weakness of model - most important component
- 53Frontal Lobe Damage• Leads to problems with WM and control of attention • Behaviour such as Perseveration – repeating the same action even though it is not achieving the goal of that behaviour. – E.g. when asked to follow a rule (e.g. choose the smallest object), patient with FL damage will be fine until the rule is changed (e.g. pick the largest object). – They will persevere with the old rule even when given feedback that their responding is now incorrect. • Perseveration represents a breakdown the CE’s ability to control attention.
- 54Problems with initial modelEvidence WM can hold much more than predicted based on capacities of phonological loop and VSSP alone.
• E.g. people can remember long sentences (15-20 words)LTM (knowledge) • Binding problem• Integration of information across modalities/sources does occur - 55Baddeley (2000)•added episodic buffer to WM model
• ‘a limited capacity temporary storage system that is capable of integrating
information from a variety of sources’
– Episodic – holds integrated information about episodes or events combining visual, auditory and other sources of
information – Buffer – acts as a buffer between other components of the WM system and links WM with processes such as perception and LTM - 56Baddeley (2012)states the EB is still ‘work in progress’ and
the main take home message is that it increases storage
capacity and communicates with LTM - 57Coding in LTMBaddeley presented word lists:Acoustically Similar: cat, hat, satSemantically Similar: big, large, longAcoustically Dissimilar– Semantically Dissimilar Semantic similarity - impaired LTM memory – Acoustic similarity - no effect on LTM (does effect STM)• Coding primarily in terms of meaning
- 58Nickerson (1965)• Visual LTM: 600 pictures of various
scenes & events
• Tested recognition 1 day or 1 year later
• Interspersed new and old displays
– 1 day - 92% correct
– 1 year - 63% correct - 59Retrieval from LTM• Cue-dependency – Retrieval requires correct internal or external cue• Context-dependency – retrieval better if environment at retrieval and encoding match - State-dependent memory• Dual process theory– Recollection – slower, requires attention• Hippocampus – Familiarity - faster • Perirhinal cortex
- 60• Spiers, Maguire & Burgess (2001)– Review 147 patients - damage hippocampus
• Impaired episodic memory, little semantic deficit
• No STM impairments - 61• Vargha-Khadem et al. (1997)• Bilateral hippocampal damage - impaired episodic memory, normal
semantic memory - 62– Verfaellie et al. (2000)- Hippocampal patient PS
– anterograde amnesia limited to episodic memory
• Episodic dependent on hippocampus - 63Prince et al. (2007):– Left hippocampus
• Active during episodic encoding but NOT semantic
retrieval
• Consistent with neuropsychological evidence - 64Explicit vs. Implicit memoryExplicit Memory Involves conscious recollection• Impaired in people with amnesia Implicit Memory • Typically reflected through changes in behaviour – Procedural memory- e.g. how to ride a bike • Doesn’t involve conscious recollection Preserved in amnesiacs
- 65Tulving, Schacter & Stark (1982)Given a list of multi-syllabled rare words (‘toboggan’) followed by an Implicit memory test - 1hr or week later:
– Fragment Completion Task - e.g. _O_O_GA_ ‒ Participants unaware 50% solutions in list 1 • Repetition-priming effect - more correct solutions for items in list - 66Amnesic Syndrome• Anterograde amnesia - Impaired remembering new info acquired after onset of amnesia • Retrograde amnesia - Difficulty remembering events prior to
amnesia (temporally graded) • STM - often only slightly impaired (e.g. normal digit span) • Some remaining implicit memory - 67Cohen & Squire (1980)• Reading mirror-reversed script
• Text reflected in the mirror - Amnesics show: • General improvements – practice:
increased speed • Specific improvements – re-reading same
text - 68HM• Deficits in episodic & semantic memory BUT: – Motor skills acquired at normal rate: • Mirror tracing - improvement with
practice (procedural memory)• Patients with amnesia - reasonably good learning of sensorimotor and perceptual skills • Skill learning (procedural memory)- intact or nearly so! - 69Graf, Squire & Mandler (1984)• Part 1 - Rate how much liked words on a word list; • Part 2 – 4 Memory tests:
– 3 tests of explicit memory - free-recall, recognition memory & cuedrecall– 1 test of implicit memory - fragment completion• Three letters e.g. STR_ _ (e.g. STRAP, STRIP)• Amnesic patients worse on explicit but not implicit memory - 70Explicitepisodic and semantic
- 71Implicitprocedural
priming
conditioning
Informationen
- Beschreibung
- U24127: Cognitive Psychology Karteikarten am Memory Structure, erstellt von Jasmine Stow am 30/11/2018.
- Erstellt von
- Jasmine Stow
- Aufrufe
- 0
- Bewertung