Memory Models

• The video serves as a quick reminder of key A-level psychology points
• A psych boost app is available on IOS and Android for flashcard based active recall
• Tutorial videos and printable resources are on Patreon

Multistore Model of Memory (Atkinson & Shiffrin, 1968)

• Theoretical cognitive model describes how the memory system processes information with three stores
• Sensory register receives raw sensory impressions
• Attention transfers information to short-term memory
  • Coding modality-specific
  • Capacity large
  • Duration very short (250 milliseconds, varies per store)
• Short-term memory (STM) receives information from sensory register via attention or long-term memory via retrieval
• Information maintained in STM through maintenance rehearsal or transferred to long-term memory (LTM) by elaborative rehearsal
  • Coding acoustic
  • Duration approximately 18 seconds.
  • Capacity 7 +/- 2 items (Miller)
  • Loss via displacement or decay
• Long-term memory (LTM) has an unlimited capacity and is coded semantically
  • Very long, potentially permanent duration
  • Forgotten information may be inaccessible
  • Information retrieved back to STM

Evaluations of the Multistore Model

• Glanzer and Kunitz found primacy and recency effects
  • Words at the start and end of a list recalled more easily, suggesting STM and LTM are separate stores
• Sperling found recall of a random row of a 12-letter grid flashed briefly was 75%
  • Suggests a large capacity and short duration of the sensory register
• Baddeley found immediate recall worse with acoustically similar words, delayed recall worse with semantically similar words
  • Suggests STM is coded acoustically, LTM semantically
• Jacobs found STM capacity averages 7 items for letters, 9 for numbers
• Peterson and Peterson found recall of trigrams decreased significantly after 18 seconds without rehearsal
  • Suggests short duration of STM
• Wagner found 75% recall for critical details after 1 year, 45% after 5 years in diary event recall
  • Suggests very large LTM capacity
• Bahrick found 90% recall of school friends' names after 15 years, 80% after 48 years
  • Suggests very long LTM duration
• Cognitive tests of memory like the MSM are often highly artificial which means low mundane realism and low ecological validity
  • Findings may not generalize to real-world memory use
• LTM has different types, not a single store
• Working Memory Model explains STM as an active system with multiple stores
• STM capacity can vary with age and practice

Types of Long-Term Memory

• Long-term memory stores memories over long periods
• Divided into declarative (explicit) and non-declarative (implicit) types
• Declarative memory is consciously accessible and can be expressed in words
• Non-declarative memory unconsciously recalled and hard to verbalize
• Episodic memory stores experiences and events with time and place references (timestamped)
  • Autobiographical
  • Strength influenced by emotion
  • Associated with hippocampus and prefrontal cortex
• Semantic memory stores facts, meanings, and knowledge
  • Strength from processing depth
  • Not timestamped
  • Episodic memories can become semantic over time
  • Associated with the frontal cortex
• Procedural memory stores unconscious skills
  • Often learned in childhood
  • More resistant to forgetting
  • Associated with motor cortex and cerebellum

Evaluations of Long-Term Memory

• Vargha-Khadem found children with hippocampus damage had episodic amnesia but intact semantic memory
  • Suggests separate brain regions for episodic and semantic memory
• Clive Wearing retained semantic (facts about his life) and procedural memory (piano playing) despite anterograde amnesia, however episodic was lost
  • Suggests separate memory stores
• Generalizing from idiographic case studies is problematic due to unique individual factors
• healthy brains studies indicate long-term memories are associated with specific brain areas
• Types of long-term memory may not be truly distinct
  • Episodic memories can become semantic

Working Memory Model (Baddeley & Hitch, 1974)

• An information processing model that replaced the STM store in the MSM
• An active processor with multiple stores
• Central executive
  • Controls attention and filters information
  • Limited capacity (4 items)
  • Can only deal with one strand of information at a time
• Phonological loop processes sound information
  • Acoustic coding
  • Primary acoustic store (inner ear) stores recently heard words
  • Articulatory process (inner voice) stores information via subvocal repetition
  • Capacity of 2 seconds
• Visuo-spatial sketchpad processes visual and spatial information
  • Visual cache passively stores form and color
  • Inner scribe actively stores relationships in 3D space
• Episodic buffer (added in 2000)
  • General store that holds and integrates information from VSS, phonological loop, central executive, and LTM

Evaluations of the Working Memory Model

• Baddeley found performance better when tasks used different processing systems
  • Suggests VSS and phonological loop are separate
• KF had impaired verbal STM but intact visual STM after brain injury
  • Suggests separate processes located in separate brain regions
• Integration of spatial and verbal information increases activity in the prefrontal cortex.
  • Episodic buffer likely exists in prefrontal cortex
• Baddeley found recall of monosyllabic words was better than that of polysyllabic words.
  • Capacity of phonological loop is time-based (2 seconds)
• The working memory model seems more realistic than the STM component of MSM
• Memory tasks often lack mundane realism
• Central executive's function lacks full explanation, needs development
• Model relies on inferences about cognitive processes

Explanations for Forgetting: Interference Theory

• Forgetting occurs due to confusion/disruption of LTMs by other information
• Proactive interference: old information disrupts new (works forward)
• Retroactive interference: new information disrupts old (works backward)
• Similarity: interference more likely with similar information
• Time sensitivity: interference less likely with a larger gap between learning

Explanations for Forgetting: Retrieval Failure

• Forgetting due to absence of appropriate cues
• Information is available but inaccessible
• Encoding specificity principle: cues encoded with memories
• Context-dependent cues: external environment (e.g., sight, sound, smell)
• State-dependent cues: internal environment (e.g., emotions, drug states)
• Category/organizational cues: relate to organization or category of memories

Evaluations of Forgetting

• Smith found people who moved more often recalled fewer street names
  • Suggests retroactive interference
• Greenberg and Underwood found recall decreased with more previously learned word pairs
  • Suggests proactive interference
• Interference may not explain many common examples of forgetting
• Godden and Baddeley found recall best when learning and recall environments matched
  • Suggests context-dependent cues promote recall
• Overton found recall best when internal state matched during learning and recall
  • Suggests state-dependent cues promote recall
• Tulving and Pearlstone found recall improved with category cues
  • Suggests organizational cues aid recall
• Interference and cue-dependent forgetting may only explain temporary memory loss
• Research has practical applications for revision strategies and cognitive interviews

Factors Affecting Eyewitness Testimony: Misleading Information

• Bartlett argued memory is reconstructive and prone to errors
• Schemas influence memory recall
• Leading questions imply answers, influence recall
• Memory Conformity and response bias
• Discussion after the event alters accuracy of another witness
• Witnesses may go along with other accounts for social approval

Factors Affecting Eyewitness Testimony: Anxiety

• Anxiety is arousal with concern, tension, and physiological changes
• High anxiety may decrease recall due to weapon focus effect
• High anxiety may increase recall due to alertness
• Yerkes-Dodson Law: eyewitness accuracy increases with anxiety to a point, then decreases

Evaluations of Eyewitness Testimony

• Loftus and Palmer: verb used in questioning affected speed estimation of car crash
  • Suggests leading questions influence recall
• Gabbert found that discussing events led to inaccurate information in eyewitness testimony
• Bodner found if warned participants changed less
• Lack of emotional impact in lab-based research may lower validity
• Johnson and Scott found weapon focus effect decreased accuracy
  • Support of weapon focus and anxiety altering eyewitness testimony
• Yuille and Cutshall found high stress witnesses of a real shooting had accurate recall, and resisted misleading information
• Research led to the cognitive interview technique
• Lab research may have demand characteristics
• Research may break ethical guidelines

Improving Eyewitness Testimony: Cognitive Interview

• Eyewitness testimony can be inaccurate
• Fisher defined the flawed standard interview and suggested the cognitive interview as improvement
• Techniques of the cognitive interview:
  • Context reinstatement: mentally return to the crime scene
  • Report everything: include all details, even irrelevant ones
  • Recall from changed perspective: consider other viewpoints
  • Recall in reverse order: switch timeline to check accuracy

Evaluations of Cognitive Interview

• Fisher found the interview produces 47% more information in real interviews after their training and 63% more information than the standard interview detectives
  • Suggests a cognitive interview effectively enhances the memory of Witnesses in the real world
• A meta-analysis found increased correct information but also increased incorrect information
  • Accuracy rate similar to standard interview
  • Suggests that the cognitive interview may be of limited practical use due to increased errors
• Each aspect produces similar recall levels, but context reinstatement and report everything produce more correct recall
• The cognitive interview is time-consuming and requires training
• The cognitive interview may not be adopted due to limited Financial Resources
• It is not effective in improving recognition of suspects
• Less effective with young children due to egocentrism.

Test your knowledge of memorymodels, including the multi-store model and working memory. Questions cover coding methods, duration, capacity of short-term memory. Also includes long-term memory and interference.