Psychology revision: memory

The multi-store model

Atkinson and Shiffrin (1968) proposed three stores: the sensory register, short-term memory (STM) and long-term memory (LTM). Attention moves information from the sensory register into STM, and rehearsal moves it from STM into LTM.

• STM capacity: about 7 plus or minus 2 items (Miller, 1956). Cowan (2001) argued it is closer to 4 chunks.
• STM duration: about 18 to 30 seconds without rehearsal (Peterson and Peterson, 1959).
• Encoding: mainly acoustic in STM and mainly semantic in LTM (Baddeley, 1966).
• LTM duration can be lifelong: Bahrick et al. (1975) found people recognised classmates' faces with about 70% accuracy 48 years after leaving school.
• Evaluation: case studies such as KF (Shallice and Warrington, 1970) suggest STM is not a single store, which the working memory model explains.

The working memory model

Baddeley and Hitch (1974) replaced the single STM store with a central executive that directs attention, and slave systems: the phonological loop (sounds and words) and the visuo-spatial sketchpad (visual and spatial information). Baddeley added the episodic buffer in 2000, which links information across the systems and to LTM. Dual-task studies support separate systems: people can do a verbal and a visual task together more easily than two visual tasks.

Types of long-term memory

Tulving (1985) divided LTM into episodic memory (personal events), semantic memory (facts and knowledge) and procedural memory (skills, such as riding a bike). Episodic and semantic memories are explicit; procedural memory is implicit. The case of Clive Wearing, whose episodic memory was devastated but who could still play the piano, supports separate stores.

Explanations for forgetting

• Proactive interference: old learning disrupts new learning.
• Retroactive interference: new learning disrupts old learning. Interference is strongest when the material is similar (McGeoch and McDonald, 1931).
• Retrieval failure: forgetting happens when the cues present at learning are missing at recall (Tulving's encoding specificity principle). Godden and Baddeley (1975) found divers recalled words best in the environment where they learned them (context-dependent forgetting). Carter and Cassaday (1998) found the same for internal states (state-dependent forgetting).

Eyewitness testimony

• Leading questions: in Loftus and Palmer (1974), participants who heard 'smashed' estimated a mean speed of 40.5 mph, compared with 31.8 mph for 'contacted'.
• Post-event discussion: Gabbert et al. (2003) found that 71% of witnesses who discussed an event reported details they had not seen themselves.
• Anxiety: Johnson and Scott (1976) found a weapon reduced identification accuracy, but Yuille and Cutshall (1986) found witnesses to a real shooting were highly accurate months later.
• The cognitive interview (Fisher and Geiselman, 1992) uses four techniques: report everything, reinstate the context, reverse the order and change perspective.

Questions

Related topics