Cog Sci Midterm 2

I <3 u, you're going to kill it

processes in acquiring and retaining info

encoding, storage, retrieval

long term memory

declarative memory

LTM recall

goal directed, search for info

free recall

recall info in any order (ex. grocery list)

serial recall

recall info in a specific order (ex. phone number)

cued recall

recall on the basis of retrieval cues (ex. remembering names by what seat they sit in)

LTM recognition

identifying whether presented item in already in memory, familiarity, goal directed or not, stimulus or cue driven

ebbinghaus nonsense syllables

-syllables until perfect recall then measured rate of info loss
-important because can't be encoded semantically

ebbinghaus forgetting curve

-within minutes memory is good,
-falls off 20-60 min
-plateaus after few days

ebbinghaus evidence for stm

remember accurately after a short time then ltm takes over and memory remains constant

information processing model (atkinson & shifrin)

-separate systems with unique duration and capacity
-sensory memory (msec-sec)
-stm (15-30sec)
-ltm (years)

memory control processes

-attention
-rehearsal
-encoding
-retrieval

stm characteristics

limited storage capacity, duration of storage (minutes) which is lost through interference and decay

stm capacity (Miller)

7 +/- 2 (5-9)

chunking

-grouping info into smaller parts,
-can vary from single letters to whole words,
-reduces memory load (ex. btwlol vs gerfnt)

Ericsson semantic chunking

-SF had initial span of 7 digits and after training increased to 79,
-used chunking to encode numbers as running records because he was a runner,
-used super chunks/hierarchies

limitations of Ericsson

-3-4 digit chunks
-no improvement if chunks couldn't relate to existing knowledge
-specific to digits
-not "bigger" stm but learned encoding strategy

visual chunking (Chase & Simon)

chess experts had better recall when pieces were in game positions but did the same as novices when pieces were in random spots

STM vs WM

-storage vs processing
-static vs active (mental workbench)

mental workbench example

solving a math problem in your head

3 Part WM Model (Baddeley)

central executive, phonological loop, visuospatial sketchpad, episodic buffer

Working Memory

system for temporary storage + manipulation of info for essential cognitive tasks
-limited capacity
-multiple parts

central executive

-controls recourses and allocated to subsystems
-initiates retrieval and decision process
-transfer to LTM via encoding

phonological loop (verbal-auditory processing)

passive storage but active rehearsal

phonological store (inner ear)

holds small amount of speech info, differs from auditory sensory memory because WM requires attention

articulatory loop (inner voice)

rehearsal process to prevent decay

acoustic confusion (evidence for phonological loop)

sound based errors in verbal retrieval
-ex. BFVMNP errors for same sounding letters
-ex. MAP/CAP more confusable than HUGE/WIDE

articulatory suppression (evidence for pl)

occupying verbal resources impairs verbal WM

verbal WM span

can remember as many words as you can say in about 2 sec so capacity should be influenced by word length/speaking rate

word length effects

-longer words reduce span
-longer words take longer to say so reduces recall

cross linguistic word length effects (Ellis & Hennelly)

tested English/Welsh bilinguals, smaller effective digit span in Welsh because digits take longer to say

visual/spatial info in VSSP

mental imagery, tracking location, route planning
ex. how many windows does your house have

chess board piece recall VSSP

requires allocation of resources to VSSP, PL shouldn't be involved

modality specific interference paradigm

see and reproduce chessboard, interference task
-articulatory loop suppression (PL)
-tap keys in a pattern (VSSP)
-produce random list of letters (CE)
-control did the best

episodic buffer

multimodal integration and storage, leaves CE responsible for control processing

LTM capacity and duration

unlimited

acquisition (encoding)

WM \>> LTM

retention

storage

retrieval or forgetting

LTM \>> STM

paired associate learning

remember words as pairs by rehearsing
-study phase "car-shirt"
-test phase "car-?"

rehearsal encoding into LTM

deliberate recycling of STM contents increases probability of info being transferred to LTM

serial position curve

free recall, primacy (first items get rehearsed more), recency (remember items still in STM)

rote rehearsal

repetition

organization encoding into LTM

enhanced by putting material into larger meaningful structures

organization example

remember one list with animals and utensils better than random list

clustering

evidence that grouping is present in memory in retrieval, semantic cues can trigger recall clusters
-ex. free recall of animals will say pets, then farm animals etc
-ex. zoo animals

imagery encoding (visuospatial knowledge) in LTM Engelkamp

paired associate learning
-heard auditory pairs
-told to learn items as well as possible or by forming images with nouns
-free or cued recall
-cued recall stronger overall, imagery condition better at both

depth of processing encoding into LTM

degree of semantic involvement, more powerful than simple rehearsal

depth of processing Hyde & Jenkins

-study word list for memory test
-rate words' pleasantness
-count number of letters in each word
-intentional learning and pleasantness (incidental but semantic) groups performed same
-even when you aren't trying to learn, paying attention to meaning is effective encoding

depth of processing Craik & Lockhart

can process info at incidental, shallow level (maintenance rehearsal using PL) but intentional deep processing is better

depth of processing Craik & Tulving

SHARK
-semantic "is the word a type of fish"
-rhyme "does it rhyme with park"
-orthographic "does it start with S"
-semantic best

attention to meaning in encoding

better memory regardless of intent but doesn't explain why deeper is better

elaboration in encoding to LTM

connecting new info to previous knowledge

Elaboration (Anderson & Bower)

better remembered "the doctor hated the lawyer... because of the malpractice suit" than just the first half

encoding specificity

better recall if conditions at retrieval are similar to those present during encoding, contextual info provides cues

encoding specificity Tulving & Osler

best remembered related cue and no cue than alternate cue
-empty stomach vs hurt stomach

encoding specificity Godden & Baddeley

deep sea divers learned lists and were tested on the beach or underwater, best was dry-dry and wet-wet conditions

context dependent memory

-external (odor, room, color)
-internal (state-exercise, mood-angry/happy)
-stronger for recall than recognition

transfer-appropriate processing

we encode perceptual and cognitive processes active during learning so activating similar processes at retrieval will show better memory

transfer-appropriate processing Morris

CAT
-semantic "does it have a tail"
-rhyme "does it rhyme with hat"
-recognition- did you see cat before
-cued recall- did it rhyme with hay
-best in standard semantic
-deeper processing by itself doesn't necessarily lead to better memory performance

Mnemonics

ROY G BIV

LTM retrieval

searching through memory

external cues

specific recall prompt (test questions), context clues (priming)

internal cues

self generated (where did I leave that?)

prime

stimulus presented earlier in time

target

stimulus that follows prime, influence of prime based on existing connection in memory

same cue can bring out different memory, implicit memory retrieval

list of unscrambling tulip and petal likely to see ELAF as leaf, list of insect and dog likely to see it as flea

lexical decision task (Meyer)

determine if word is real or not and shown word pairs either in succession or not (bread/doctor, nurse/doctor), RT longest for nonword, unrelated prime, related prime
-memory accessibility, retrieval faster for having just processed related info

availability

information can be in LTM

accessibility

may or may not be retrieved in WM but priming makes it better

Brown-Peterson paradigm trigrams

3 letter trigrams, count back by 3s for 3-18s to prevent attention and rehearsal (decay and interference), loss of info is rapid, near 0 at 18s, evidence for RI

retroactive interference

later material interferes with memory of previous material

Keppel & Underwood early vs late trigram

little forgetting at first trial but more with each trial, "builds up," PI

proactive interference

previous items confusable with current memory set, caused in part by similarity of older material to new idents, decrease by distinctiveness

Wickens PI

fruit, meat, profession for first 3 trials then either swtiched or stayed fruit, reduced recall over first 3 trials (PI) but recall rebounded trial 4, size of increase related to similarity (professions recall highest, meat, fruit)

serial position curve simple rehearsal

primacy- most rehearsal, recency -most recent in STM

serial position curve interference

primacy- RI only
recency- PI only
middle- RI + PI

Minami & Dallenbach interference or decay

condition A learn maze, immobilize for 24h (decay), retest 25% forgetting, condition B learn, active exploration (interference plus decay), retest 75% forgetting

consolidation

integrating info into LTM

declarative memory consolidation

slow wave sleep

non-declarative memory

REM sleep

walker sleep consolidation

group 1- train monring, awake, test, sleep, test morning
group 2- train night, sleep, test morning, awake, test night
results- gains in speed and accuract after sleep period regardless of when, consolidation mediates in hippocampus activity during REM

inhibition

suppression of non target info, ex. who's birthday is?

retrieval induced inhibition

prevents irrelevant memories from being retrieved instead of target memory

anderson inhibition study

vehicle or pet category with 4 items, category and half of items practiced, others not at all, better recall for practiced items in practiced categories, lowest for unpracticed items in practiced categories

what does anderson study show

inhibition- retrieval of certain practiced items inhibited retrieval of unpracticed

retrieval induced enhancement

repeated retrieval (testing) can enhance learning and memory

roediger retrieval induced enhancement

read a story, study-study group read again, study-test group recalled story, recall test after 5 min 2 days 1 week,
results- after 5 min study-study higher but after a week study-test higher (long term advantage)

autobiographical memory

episodic memories for specific events

3 levels of detail autobiographical memory (conway)

specific event (min/hours), general event (weeks), macro level (years)
ex. going to college

pillemer autobiographical memory

better recall in college alumni for start of freshman year and end of senior year, milestones and transitions remembered best

reminiscence bump

when asked to remember their lives people give more detail about specific times (adolescence and older adulthood)

shrauf & rubin reminiscence bump

US immigrants found a shift in RB, earlier immigrants showed earlier bump and later is later, report more memories from that milestone

flashbulb memory

rich detailed memories encoded after an emotionally salient event

brown & kulik flashbulb memory (special)

"now print" mechanism, emotionally charged, consequentiality (consequences of event reverberates)

neisser flashbulb memory (not special)

FB memory influenced by ordinary memory mechanisms
-frequent narrative rehearsal
-decay, inaccurate details
-plausible distortions

narrative rehearsal leveling

detail loss

narrative rehearsal assimilation

change to fit preconceptions

narrative rehearsal sharpening

embellishment

talarico falshbulb vs everyday memories

day after 9/11 asked about memories of the event and an ordinary event from 9/10, follow ups 7/42/224 days later, no difference in consistency but maintained confidence in FBM and not ordinary, beliefs/consequentiality of FB feel different