AQA Psychology A-level: Topic 2 - Memory
Part 1 — Coding, Capacity, and Duration of Memory
Coding: Refers to the format or sensory "type" of information stored in memory. * Short-Term Memory (STM): Coding is primarily acoustic. * Long-Term Memory (LTM): Coding is primarily semantic. * Research Evidence: Baddeley (1966) found that participants made more mistakes when recalling acoustically-similar words immediately after learning (STM). In contrast, more mistakes were made when recalling semantically-similar words after a 20-minute interval (LTM recall).
Capacity: Refers to the total volume or amount of data a memory store can hold at any given time. * Short-Term Memory (STM): Miller suggested the capacity is items. He noted that information often comes in groups of 7 (e.g., 7 days of the week, 7 notes on a musical scale), suggesting an evolutionary or biological predisposition. * Chunking: The method of grouping information together to increase the amount held in the STM. * Research Evidence: Jacobs demonstrated that the mean letter span was 7.3 and the mean digit span was 9.3. * Long-Term Memory (LTM): The capacity is thought to be virtually unlimited.
Duration: Refers to the length of time information can be held in a memory store. * Short-Term Memory (STM): Lasts approximately 18-30 seconds. * Research Evidence (STM): Petersen et al. (1959) studied 24 undergraduates. They used consonant syllables and varying retention intervals. To prevent mental rehearsal, participants had to count backward from a 3-digit number. Accuracy decreased as retention intervals increased. * Long-Term Memory (LTM): Duration is potentially unlimited. * Research Evidence (LTM): Bahrick et al. (1975) studied 396 participants on their ability to recognize former graduating classmates. High school photo recognition was 90% after 15 years, and decreased to 70% after 46 years.
Critical Evaluation of Coding, Capacity, and Duration: * Standardization Issues: Historical research, such as the work by Jacobs, often lacked the rigorous standardization and scientific methods found in modern laboratory experiments. Factors like noisy rooms or difficult word lists may have acted as confounding variables. * Mundane Realism: Bahrick et al.'s (1975) study is praised for high mundane realism because it used meaningful, personal stimuli (high school faces) rather than artificial stimuli, giving it high ecological validity. Conversely, Petersen et al. and Miller's studies used artificial stimuli (nonsense syllables/random lists), which may not reflect real-life memory. * Capacity Re-evaluations: Recent research suggests Miller may have overestimated STM capacity; it may be closer to 4 chunks rather than the original 5-9 items estimate.
Part 2 — The Multi-Store Model of Memory (MSM)
Structure of the MSM: This model describes memory as a flow of information through three distinct stores: the Sensory Register, Short-Term Memory (STM), and Long-Term Memory (LTM).
The Sensory Register: * Includes a sub-store for each of the five senses (e.g., the echoic store for auditory data). * Capacity: Very large. * Duration: Extremely short, less than half a second ( seconds). * Transfer Mechanism: Information passes from the sensory register to the STM only if attention is paid to it.
Short-Term Memory (STM): * Coding: Acoustic (Baddeley). * Capacity: items (Miller). * Duration: 18-30 seconds (Petersen). * Maintenance Rehearsal: Occurs when we repeat info to ourselves. Prolonged rehearsal allows transfer to LTM. Lack of rehearsal leads to forgetting.
Long-Term Memory (LTM): * Coding: Semantic. * Capacity: Unlimited. * Duration: Unlimited (at least 46+ years per Bahrick et al.). * Retrieval: The process of transferring information from LTM back into the STM for use.
Critical Evaluation of the MSM: * Unitary Store Limitation: The model treats LTM as a single store, but Tulving et al. proposed three distinct types (procedural, semantic, episodic). The MSM fails to distinguish between unconscious recall (procedural) and conscious recall (semantic/episodic). * Rehearsal Types: Craik and Watkins (1973) argue that maintenance rehearsal (repetition) is less effective for LTM transfer than elaborative rehearsal, which involves linking new data to existing knowledge. * Support for Separate Stores: The model correctly identifies qualitative differences between STM and LTM in terms of coding and duration, supported by Baddeley and Miller. * STM Unitary Store Issue: Shallice and Warrington's study of patient KF showed he had poor auditory recall but accurate visual recall, suggesting STM is not one unitary store but has sub-components.
Part 3 — Types of Long-Term Memory
Episodic Memory: Personal memories of life events, including context (when and where), details of people, and places. Example: A wedding or first meeting a partner.
Semantic Memory: General knowledge of the world, meanings of words, concepts, and themes. Example: Understanding how one concept helps explain another (general facts).
Procedural Memory: Memories for learned skills or actions. Example: Driving a car or swimming. These processes are recalled unconsciously.
Critical Evaluation of LTM Types: * Neuroscanning Evidence: Petersen et al. found semantic memories activated the left prefrontal cortex, while episodic memories activated the right prefrontal cortex, proving different neurological bases. * Practical Applications: Belleville et al. noted that mild cognitive impairments primarily affect episodic memory. Understanding these distinctions allows for targeted treatments. * Alternative Classifications: Cohen and Squire distinguish memory by Declarative (Episodic and Semantic—conscious recall) and Non-declarative (Procedural—unconscious recall). * Case Studies: Patients HM and Clive Wearing had severely impaired episodic memory but intact procedural and semantic memory. Clive Wearing could still play piano (procedural) and understand music (semantic) but couldn't remember his wife visiting minutes earlier (episodic).
Part 4 — The Working Memory Model (WMM)
Components of the WMM: * Central Executive (CE): An 'attentional process' with very limited capacity. It coordinates the slave systems and allocates tasks. * Phonological Loop (PL): Processes auditory data. Comprised of the Articulatory Process (stores spoken words) and the Phonological Store (inner ear). * Visuo-spatial Sketchpad (VSS): Processes visual and spatial data (the 'inner eye'). Capacity is roughly 4-5 chunks (Baddeley). It is divided into the Inner Scribe and Visual Cache. * Episodic Buffer: Added later to integrate visual, spatial, and auditory data into a single coherent picture. It acts as the storage component for the CE and links STM to LTM.
Critical Evaluation of the WMM: * Vague Central Executive: The CE is the most important but least understood part. Critics say it is more complex than a single 'process' and might contain its own sub-components. * Support from Patient KF: Shallice and Warrington found KF could remember visual info but not auditory, supporting the existence of separate VSS and PL components in working memory. * Dual-Task Performance: Studies show that undertaking two tasks simultaneously (one visual, one verbal) causes performance drop because slave systems compete for the CE's limited resources. * Neuroscientific Support: Braver et al. used scans to show that as tasks for the CE became more difficult (higher cognitive load), activation in the prefrontal cortex increased significantly.
Part 5 — Explanations for Forgetting: Interference
Types of Interference: 1. Proactive Interference: Older memories interfere with the retrieval of newer memories. 2. Retroactive Interference: Newer memories interfere with the retrieval of older memories.
Research Evidence: McGeoch and McDonald (1931) had groups learn a list of words and then a second list. The second list varied (synonyms, antonyms, unrelated words, numbers, syllables). * Result: The synonyms group (most similar to the original) had the worst recall, with 3.1 fewer items remembered than the control group. * Conclusion: Similarity correlates positively with the extent of forgetting.
Critical Evaluation of Interference: * Lab Validation: Highly controlled conditions in lab studies increase internal validity. * Artificiality: Use of random word lists results in low mundane realism; real-life learning involves meaningful information. * Real-World Application: Baddeley and Hitch studied rugby players. They found that forgetting was determined by the number of games played (interference) rather than time passed from the start of the season. * Time Limitation: Lab studies often test recall after 1-2 hours, which doesn't reflect the days or weeks usually involved in real-life forgetting.
Part 6 — Explanations for Forgetting: Retrieval Failure
Theory: Forgetting occurs due to an absence of 'cues' that were present during encoding. This is the Encoding Specificity Principle (ESP) proposed by Tulving.
Context-Dependent Forgetting: * Occurs when the external environment at the time of recall differs from encoding. * Study: Godden and Baddeley (1975) tested deep-water divers in four conditions: learning on land/recalling on land, learning land/recalling underwater, learning underwater/recalling land, learning underwater/recalling underwater. Matching environments resulted in significantly higher recall.
State-Dependent Forgetting: * Occurs when internal physiological or emotional states at recall differ from encoding. * Study: Carter and Cassaday (1998) used antihistamines to create a drowsy state. Recall in matching states was 40% higher than non-matching states.
Critical Evaluation of Retrieval Failure: * High Consensus: Eysenck argues this is the primary reason for LTM forgetting. * Ecological Validity Issues: Baddeley notes that contexts as extreme as land vs. water are rare in real life, making the theory hard to generalize. * Recall vs. Recognition: Godden and Baddeley found that when testing for word recognition rather than recall, context effects disappeared, suggesting the theory isn't universal. * Circular Reasoning: The ESP relies on assumptions that cannot be independently tested; if a cue works, it's assumed it was encoded, and if it doesn't work, it's assumed it wasn't encoded.
Part 7 — Factors Affecting Eyewitness Testimony (EWT): Misleading Information
Leading Questions: Questions phrased to suggest a specific answer. * Study: Loftus and Palmer (1974). Participants watched a car crash. The critical verb in the question ("How fast were they going when they [X] each other?") changed speed estimates. "Smashed" resulted in an estimate 8.7 mph higher than "contacted". * Mechanism: Response bias (pressure to answer a certain way) or substitution explanation (actual memory change). Those who heard "smashed" were more likely to falsely recall broken glass 2 weeks later.
Post-Event Discussion (PED): Co-witnesses discussing events can distort memory (memory contamination). * Study: Gabbert et al. (2003) matched-pairs design. Pairs saw the same crime from different angles. After discussion, 71% of participants recalled aspects they hadn't seen, compared to 0% in the control group. * Mechanism: Memory conformity (accepting others' views).
Critical Evaluation of Misleading Information: * Own Age Bias: Anastasi and Rhodes found people are better at identifying targets from their own age group. EWT studies often use young targets, disadvantaging older participants. * Demand Characteristics: Zaragosa and McCloskey argue participants exhibit the 'Please-U' effect, providing answers they think the researcher wants. * Mundane Realism: Video clips lack the emotional stress and anxiety of real-life crimes.
Part 8 — Factors Affecting Eyewitness Testimony (EWT): Anxiety
Negative Effects of Anxiety: * Weapon Focus Effect: Anxiety narrows focus to the source of the threat. * Study: Johnson and Scott (1976). Low-anxiety (greasy pen) vs. high-anxiety (bloody knife/argument). High-anxiety recall was 16% lower.
Positive Effects of Anxiety: * Fight or Flight: Heightened arousal improves attention to the surroundings. * Study: Yuille and Cutshall (1986). Field study of a real gun store shooting in Canada. Witnesses with 'high' self-reported anxiety had 11% higher recall accuracy 5 months later.
Critical Evaluation of Anxiety factors: * Yerkes-Dodson Law: Proposes an 'inverted-U' relationship between arousal and performance; moderate anxiety is optimal. * Weapon Focus vs. Surprise: Pickel found decreased EWT accuracy when a person held a raw chicken in a hair salon, suggesting 'unusualness' (surprise) rather than anxiety (threat) causes the effect. * Ethics: Exposing participants to trauma/bloody weapons breaches BPS guidelines regarding psychological harm. * Field Study Reliability: Lack of control over variables like TV reports or PED in the Yuille and Cutshall study.
Part 9 — Improving Eyewitness Testimony: Cognitive Interviews
The Four Stages of CI: 1. Report Everything: Include every detail to prevent cue-dependency issues. 2. Reinstate the Context: Return to the scene mentally (weather, mood) to trigger context cues. 3. Change the Perspective: Imagine the scene as the victim or another witness to minimize schema influence. 4. Reverse the Order: Prevents schemas and makes lying more difficult.
Enhanced Cognitive Interview (ECI): Developed by Fisher et al. (1987). Focuses on interviewer-witness social dynamics like eye contact and building rapport to reduce anxiety.
Critical Evaluation of Cognitive Interviews: * Practicality: CI is time-consuming and requires extensive officer training, which police forces often lack. * Quantity vs. Quality: Kohnken et al. (1999) found CI increased correct information by 81% but also increased incorrect information by 61%. * Component Success: Milne and Bull (2002) found that using just 'Report Everything' and 'Reinstate the Context' provides the best results, suggesting the full interview might not be necessary.