Memory

According to the multi-store model (Atkinson and Shiffrin, 1968), the predominant coding in short-term memory and in long-term memory respectively is best described as:

• AVisual in STM and acoustic in LTM
• BAcoustic in STM and semantic in LTM✓
• CSemantic in STM and acoustic in LTM
• DSemantic in both STM and LTM

Answer: Baddeley's research showed STM is largely coded acoustically (acoustically similar words are confused in immediate recall) while LTM is largely coded semantically (semantically similar words are confused after a delay).

Which pair correctly states the capacity and duration of short-term memory as described on the AQA specification?

• ACapacity 7±2 items; duration approximately 18–30 seconds without rehearsal✓
• BCapacity potentially unlimited; duration up to a lifetime, even without rehearsal
• CCapacity 7±2 items; duration approximately 2 hours (even without rehearsal)
• DCapacity 1–2 items; duration about 4 seconds

Answer: Miller estimated STM capacity at 7±2 chunks, and Peterson and Peterson (1959) found STM duration is roughly 18–30 seconds when rehearsal is prevented. Unlimited capacity and lifelong duration describe LTM.

Tulving distinguished types of long-term memory. A memory for a specific event such as your first day at school, including the time, place and emotions involved, is an example of:

• AProcedural memory
• BSemantic memory
• CEpisodic memory✓
• DWorking memory

Answer: Episodic memory stores personal events ('episodes') and is time-stamped and consciously recalled. Semantic memory holds general knowledge and facts, while procedural memory holds skills such as riding a bike, recalled without conscious effort.

Knowing how to ride a bicycle without being able to easily put the actions into words is an example of which type of long-term memory?

• AEpisodic memory, which is declarative and time-stamped
• BSemantic memory, which is declarative knowledge of facts
• CSensory memory, which is usually modality-specific
• DProcedural memory, which is largely non-declarative✓

Answer: Procedural memory stores motor skills and actions that are performed automatically and are difficult to explain verbally; it is non-declarative. Episodic and semantic memory are declarative (can be consciously stated).

In Baddeley and Hitch's (1974) working memory model, which component is the limited-capacity attentional system that allocates resources and directs the slave systems?

• AThe phonological loop
• BThe central executive✓
• CThe visuo-spatial sketchpad
• DThe episodic buffer

Answer: The central executive is the supervisory, limited-capacity attentional component that controls and coordinates the slave systems (the phonological loop and visuo-spatial sketchpad). It does not store information itself for long.

Within the working memory model, the phonological loop is subdivided into the phonological store and the articulatory process. The articulatory process is best described as:

• AThe 'inner voice' that allows maintenance rehearsal by sub-vocal repetition✓
• BThe 'inner ear' that holds spoken words briefly
• CThe 'inner scribe' that records spatial arrangement
• DThe store that integrates visual and verbal information with time sequencing

Answer: The articulatory process (the 'inner voice') allows sub-vocal rehearsal to keep verbal information active, with a capacity of about two seconds' worth of speech. The phonological store is the 'inner ear'; the inner scribe and the episodic buffer belong to other components.

McGeoch and McDonald's (1931) study, in which recall was worst when newly learned material was similar to the original list, provides evidence for which explanation of forgetting?

• ARetrieval failure caused by the absence of appropriate cues at recall
• BInterference, specifically that similarity increases interference✓
• CTrace decay, in which the memory trace fades away with the passage of time
• DRepression, in which anxiety-provoking memories are pushed into the unconscious

Answer: Participants who learned a second list of synonyms showed the poorest recall of the first list, showing that greater similarity between materials increases interference. This supports interference theory rather than cue-dependent forgetting.

Godden and Baddeley's (1975) deep-sea diver study, where divers recalled best when the learning and recall environments matched, is most often used to illustrate:

• AState-dependent forgetting, where the internal cue is the learner's mood
• BProactive interference from earlier and similar learning
• CContext-dependent forgetting and the encoding specificity principle✓
• DThe serial position effect (primacy and recency)

Answer: Divers who learned and recalled in the same environment (both on land or both underwater) recalled more than those whose environments mismatched, demonstrating context-dependent retrieval failure consistent with Tulving's encoding specificity principle. State-dependent cues are internal, not environmental.

Loftus and Palmer's (1974) first experiment showed that changing the verb in the question 'About how fast were the cars going when they ___ each other?' affected speed estimates. Which verb produced the HIGHEST mean estimate?

• AContacted
• BHit
• CBumped
• DSmashed✓

Answer: 'Smashed' produced the highest mean estimate (about 40.8 mph) and 'contacted' the lowest (about 31.8 mph). This demonstrates how leading questions can distort eyewitness memory through response-bias or substitution.

Johnson and Scott's (1976) 'weapon focus' study is often cited as evidence that anxiety can REDUCE the accuracy of eyewitness testimony. The proposed mechanism is that:

• AHigh anxiety always enhances the encoding of all details (the Yerkes-Dodson optimum)
• BAttention narrows onto the weapon, impairing recall of the perpetrator's face✓
• CAnxiety triggers post-event discussion that contaminates memory
• DAnxiety causes source confusion between schemas

Answer: Witnesses who saw a man holding a knife were less accurate at identifying him than those who saw a pen, suggesting anxiety/arousal narrows attention onto the weapon at the expense of other details. The Yerkes-Dodson law predicts an inverted-U, not that anxiety always helps.

What are the components of the working memory model?

• ACentral executive, sensory register, short-term and long-term store
• BEpisodic memory, semantic memory and procedural memory
• CCentral executive, phonological loop, sketchpad and buffer✓
• DEncoding, storage and retrieval operating in a fixed sequence

Answer: Baddeley and Hitch proposed a central executive directing a phonological loop and visuo-spatial sketchpad, with the episodic buffer added in 2000. 'Sensory register, short-term and long-term store' belongs to the multi-store model, a different account.

What did Peterson and Peterson (1959) find about short-term memory duration?

• AAbout 30 minutes without rehearsal
• BAbout 18 seconds without rehearsal✓
• CAbout 2 seconds without rehearsal
• DEffectively unlimited without rehearsal

Answer: Recall of trigrams fell to around 10% after 18 seconds when counting backwards prevented rehearsal. The artificial stimuli mean it may underestimate duration for meaningful material.

What is the capacity of short-term memory according to Miller?

• AFive items, plus or minus two
• BSeven items, plus or minus two✓
• CNine items, plus or minus two
• DEffectively unlimited in capacity

Answer: Miller's magic number seven describes chunks rather than individual items, so chunking meaningfully increases how much is held. Later work by Cowan suggests four chunks may be nearer the truth.

What are the three types of long-term memory Tulving proposed?

• AEpisodic, semantic and sensory memory
• BProcedural, sensory and iconic memory
• CEpisodic, semantic and procedural✓
• DSemantic, iconic and echoic memory

Answer: Episodic stores personal events, semantic stores facts and meanings, and procedural stores skills. Only episodic and semantic are declarative; procedural memory is difficult to put into words, which is why it survives some amnesias.

What did Loftus and Palmer (1974) demonstrate about eyewitness testimony?

• AAnxiety always improves the accuracy of witness recall
• BWitnesses recall violent events with near-perfect accuracy
• CLeading questions can distort what a witness recalls✓
• DPost-event discussion has no effect on witness accounts

Answer: Speed estimates rose with the severity of the verb used, and a week later those asked about cars that 'smashed' were more likely to report broken glass that was never there. The lab setting limits how far it generalises.

What is proactive interference?

• AOlder memories disrupting the recall of newer ones✓
• BNewer memories disrupting the recall of older ones
• CAbsence of cues preventing retrieval of a memory
• DAnxiety at encoding preventing a memory forming

Answer: Proactive interference works forwards in time — an old phone number making the new one hard to recall. Retroactive interference works backwards, and both are worse when the material is similar.

What is the encoding specificity principle?

• ARecall improves when retrieval cues differ from those at encoding
• BRecall improves when retrieval cues match those at encoding✓
• CRecall depends only on how long the material was rehearsed
• DRecall depends only on how meaningful the material was

Answer: Tulving argued that a cue present when the memory was laid down aids retrieval. Godden and Baddeley showed it with divers recalling word lists better in the environment where they learned them.

What is the cognitive interview designed to improve?

• AThe accuracy and quantity of eyewitness recall✓
• BThe speed at which a witness gives a statement
• CThe confidence a witness feels in their account
• DThe ability of police to detect deliberate lying

Answer: Its four techniques — report everything, reinstate context, reverse the order and change perspective — draw on encoding specificity and on the idea that memories have several retrieval routes. It takes longer than a standard interview, which limits use.

Why does the multi-store model struggle to explain the case of KF?

• AHis long-term memory was poor but his short-term memory was intact
• BHis verbal short-term recall was poor but visual recall was intact✓
• CHe could form new memories but could not recall any of the old ones
• DHe showed no impairment in either short- or long-term memory

Answer: A single unitary short-term store cannot account for one modality being damaged while another is spared. That dissociation supports the separate phonological loop and visuo-spatial sketchpad of working memory.

Ava struggles to recall the 11-digit phone number 07945813226 after hearing it once. When she groups it as 0794 581 3226, she recalls it easily.

Which concept explains the improvement, and which feature of short-term memory does it relate to?

• AChunking, which increases the effective capacity of short-term memory✓
• BChunking, which extends the duration of short-term memory to minutes
• CElaborative rehearsal, which codes the number semantically in STM
• DMaintenance rehearsal, which increases the capacity of long-term memory

Answer: Eleven separate digits exceed Miller's span of about 7 ± 2 items, but three chunks fall well within it, so chunking increases how much STM can effectively hold: its capacity. Chunking does not lengthen STM's duration, grouping digits by position does not involve processing meaning, and maintenance rehearsal concerns keeping items in STM rather than expanding LTM.

After brain surgery, Mr J cannot form new memories of events: each day he says he has never met his therapist before. Yet over several weeks his speed on a new motor puzzle steadily improves.

What does this case most directly suggest about long-term memory?

• AIt is a single store, as the improvement uses the same system as recalling events
• BIt relies on STM capacity, as the puzzle skill is held in the visuo-spatial sketchpad
• CIt codes everything semantically, as the skill improves through grasping its meaning
• DIt has separate stores, as procedural memory can form while episodic memory cannot✓

Answer: Mr J cannot lay down new episodic memories, yet he acquires a new motor skill without remembering practising it. One kind of LTM can form while another cannot, which supports Tulving's view of separate episodic, semantic and procedural stores and challenges the multi-store model's single LTM. A skill retained over weeks is not held in STM, and procedural learning does not depend on understanding meaning.

Mrs Okafor has moved to a new school this term. She keeps calling pupils in her new class by the names of pupils she taught last year.

Which explanation of forgetting best fits this?

• ARetroactive interference, as new learning disrupts recall of old names
• BRetrieval failure, as the context at learning differs from that at recall
• CProactive interference, as old learning disrupts recall of new names✓
• DDecay, as the traces of the new names have faded through lack of use

Answer: Older information (last year's names) is getting in the way of newer information (this year's names), which is proactive interference. Retroactive interference runs the other way, with new learning disrupting old memories. She learns and uses the new names in the same school, so there is no change of context, and the problem is old names intruding, not new ones fading.

Ellie cannot remember the name of a film actor. When a friend says, 'It begins with K,' the name comes back to her immediately.

Which explanation of forgetting does this illustrate?

• AProactive interference, as an older memory blocked storage of the actor's name
• BRetrieval failure from absent cues, as the name was stored but not accessible✓
• CDisplacement from STM, as newer information pushed the name out of the store
• DDecay of the memory trace, as the name had faded until the cue rebuilt it

Answer: The name returns as soon as a cue is supplied, so it must have been available in LTM all along but inaccessible without the cue: retrieval failure. If the memory had decayed or been displaced, a single letter could not restore it, and interference would not be relieved by a cue to the target itself.

A detective asks a bank cashier to describe a robbery as it would have been seen by the customer standing nearest the door.

Which cognitive interview technique is being used, and why?

• AReport everything, to bring out details the witness thinks trivial
• BContext reinstatement, to recreate the witness's mood at the time
• CChange order, to reduce reliance on the expected order of events
• DChange perspective, to disrupt the effect of schemas on recall✓

Answer: Asking the witness to recall the event from another person's viewpoint is the 'change perspective' technique; it reduces the influence of expectations and schemas and offers new routes to the memory. Report everything asks for every detail however minor, context reinstatement asks the witness to mentally return to the scene, and change order means recalling the events in a different sequence, none of which the detective asked for.

Which of the following is the most valid criticism of Loftus and Palmer's (1974) research into leading questions?

• AParticipants took part in each verb condition, so order effects could explain the different estimates
• BParticipants watched film clips in a lab, so they lacked the stress and stakes of seeing a real crash✓
• CParticipants were trained police officers, so their estimates may not reflect those of ordinary witnesses
• DParticipants' speed estimates were unaffected by the verb, so the study cannot support leading questions

Answer: Watching film clips of car accidents in a laboratory is very different from witnessing a real crash, with its emotional arousal and real consequences, so the findings may lack ecological validity. The study used independent groups (each participant heard one verb), the participants were students, and the verb DID affect estimates, with 'smashed' producing the highest mean.