Memory

Models, studies & evaluation — A-Level Psychology

Atkinson & Shiffrin (1968) — Multi-Store Model

Memory consists of three distinct stores, each with different encoding, capacity, and duration characteristics.

Sensory Register

Duration: <1 sec
Capacity: Very large
Encoding: Modality-specific

Attention

Short-Term Memory

Duration: 18-30 sec
Capacity: 7±2 items
Encoding: Acoustic

Rehearsal

Long-Term Memory

Duration: Lifetime
Capacity: Unlimited
Encoding: Semantic
Click a memory store above to explore its characteristics, supporting evidence, and evaluation.

Evaluation

+ Supported by case studies like HM — damage to hippocampus impaired LTM but not STM, supporting separate stores.
+ Serial position effect: primacy (LTM) and recency (STM) effects support two distinct stores.
Oversimplified — LTM is not a single store (episodic, semantic, procedural). MSM doesn't account for this.
Rehearsal is not always needed for LTM — flashbulb memories form without rehearsal. Elaborative processing matters more.

Baddeley & Hitch (1974) — Working Memory Model

Replaced the unitary STM with a multi-component system that actively processes information rather than passively storing it.

Central Executive

Allocates attention
Directs subsystems
Limited capacity

Phonological Loop

Phonological store
(inner ear)
Articulatory process
(inner voice)

Episodic Buffer

Integrates info
Links to LTM
Added 2000

Visuo-Spatial Sketchpad

Visual cache
(what things look like)
Inner scribe
(spatial relations)

Key Evidence

Baddeley et al. (1975) — Word Length Effect

Evidence for the phonological loop

Participants recalled fewer long words than short words in serial recall. Supports the idea that the articulatory process rehearses items in a time-limited loop — longer words take longer to rehearse, so fewer fit.

Dual-Task Studies

Evidence for separate components

Two tasks using the same component (e.g., two visual tasks) interfere more than two tasks using different components (e.g., one visual + one verbal). This supports the idea of separate subsystems with independent resources.

+ Better than MSM — explains why we can multitask (separate subsystems) but struggle with similar simultaneous tasks.
+ Supported by brain imaging — different brain areas active for different WMM components.
Central executive is vague — what exactly does it do? May need to be further subdivided.
Only explains STM/working memory — doesn't account for LTM processes.

Factors Affecting Accuracy of Memory

Loftus & Palmer (1974)

Leading Questions & Eyewitness Testimony

Participants watched car crash videos. Those asked "How fast were the cars going when they smashed into each other?" gave higher speed estimates than those asked about "contacted." A week later, the "smashed" group were more likely to report (falsely) seeing broken glass.

Conclusion: Post-event information can alter and distort memory. Leading questions create a response bias and can actually change the memory trace itself.

Anxiety — Weapon Focus Effect (Johnson & Scott, 1976)

Anxiety & Memory

Participants who witnessed a person carrying a bloodied knife had poorer recall of the person's face than those who saw someone carrying a pen. Anxiety narrows attention to the source of threat (the weapon), impairing memory for peripheral details.

However: Christianson & Hubinette (1993) found real-life witnesses to bank robberies had accurate memories. Real-life anxiety may enhance rather than impair memory.

Cognitive Interview (Fisher & Geiselman, 1992)

Improving Eyewitness Testimony

Four techniques: (1) mental reinstatement of context, (2) report everything, (3) recall in different order, (4) recall from different perspective. Designed to improve accuracy by accessing multiple retrieval cues.

Exam tip: When discussing eyewitness testimony, always include: the study (researcher, method, findings), what it tells us about memory accuracy, practical applications (e.g., police interview techniques), and evaluation (lab vs real life, ecological validity).