Memory

Coding, capacity, duration

AO1

Coding STM & LTM: Baddeley (1966) Acoustic STM & Semantic LTM

Capacity STM: Jacobs (1887) 9.3 numbers, 7.3 letters and Miller (1956) 7+/- 2 items

Duration: STM Peterson & Peterson (1959) Trigrams 80% at 3s / 3% at 18s LTM Bahrick (1975) 70% 48 years

AO3

+/- Internal & ecological validity

- Size of the chunk not specified

- Individual differences (age)

- Miller overestimated capacity - closer to 4 (Cowan, 2001)

​

Working Memory

Model

AO1

Central executive - supervisory role, attention, decisions

Slave systems:

PL - phonological store (inner ear) / articulatory process (inner voice): holds 2 secs worth of info

VSS - visual cache (store) / inner scribe (spatial tasks): 4 objects

Episodic buffer - temporary,

allows for integration

AO3

+ Evidence from KF (VSS intact, PL damaged)

+ Dual task studies - Baddeley, 1975

- CE least understood

- Evidence lacks mundane realism

​

Eye witness testimony:

Misleading info

AO1

Leading questions

Loftus & Palmer (1974) contacted 31.8mph, smashed 40.5mph

Response bias or substitution?

Broken glass follow up = substitution more likely

Post event discussion

Gabbert (2003) 71% reported info not seen & 60% wrongly accused

Memory contamination or memory conformity

AO3

+ Real world application to criminal justice system

- Studies: low ecologically validity

- Higher recall in field experiment, Yuille & Cutshall (1986)

- Individual differences - age bias

Multi Store Model

of Memory

AO1

First model of memory

Atkinson & Shiffrin (1968)

Linear, sequential model

SS - modality specific, less 1/2 second and high capacity

STM - acoustic, 18-30 secs, 7+/-2

LTM - Semantic, infinite unlimited

Attention, maintenance rehearsal

AO3

+ HM - more than one store

+ Very influential model

- Over emphasis of rehearsal (elaborative needed)

- KF more than one STM

​

Forgetting:

Interference

AO1

Proactive - old memory affects new memory

Retroactive - new memory affects old memory

Effects of similarity: more similar = more interference

McGeoch & McDonald (1931) - synonyms 12% and numbers 37%

Effects of time: shorter time = more interference

AO3

+ Baddeley & Hitch - Rugby study - Interference not time

- Studies lack ecological validity

+ Individual differences - greater WM, less susceptible

- Memories need to be very similar

Eye witness testimony:

Anxiety

AO1

Anxiety: negative effect

Johnson & Scott (1976) Waiting room, pen knife study: 49% low anxiety, 33% high anxiety

Anxiety: positive effect

Yuille & Cutshall (1986)

High stress 88%, low stress 75%

Yerkes-Dodson (1908) curve: inverted U theory can explain contradictory findings

AO3

+ Pickel (1998) Measuring surprise rather than anxiety?

- Y&C: confounding variables

- Individual differences: stables v neurotics

- U theory is too reductionist in only anxiety = arousal

​

​

Types of

LTM

AO1

Tulving (1972)

Episodic - personal events, time stamped, conscious recall

Semantic - knowledge, not time stamped, conscious recall

Procedural - motor skills, not time stamped, unconscious recall

AO3

+ HM - more than one LTM

+ Brain scans  - episodic (right frontal) semantic (left frontal)

+ Real life application for helping with dementia

- Cohen & Squire: Only two stores declarative v non-declarative

Forgetting:

Retrieval Failure

AO1

Forgetting due to absence of cues

Encoding specificity principle

Context dependent forgetting

Godden & Baddeley (1975)

State dependent forgetting

Carter & Cassaday (1998)

AO3

+ Real world application

- Studies: ecologically validity

- Cannot test ESP as do not know if cue was encoded or not

- Results not replicated for a recognition task

​

​

​

​

Eye witness testimony:

Cognitive interview

AO1

Fisher & Geiselman (1992)

Report everything (cues)

Reinstate context (context dependent forgetting)

Reverse the other (schemas)

Change perspective (schemas)

Enhanced CI - social elements

AO3

+ Kohnken (1999) meta analysis - 41% more accurate

- Report everything and reinstate context most important

- Time consuming and training

- Individual differences: Wright and Holliday (2007) more effective when respondents are older