WEEK 6 LEC MEMORY 2: FORGETTING & AMNESIA

Memory is Reconstructive and Forgetting: Key Concepts

  • The lecture emphasizes that memory is not a perfect recording of events but a reconstructive process influenced by current knowledge, beliefs, and context.
  • Eyewitness testimony is especially prone to reconstruction and distortion due to labeling, schemas, expectations, and suggestive questioning.
  • Understanding memory involves exploring sensory memory, working/short-term memory, long-term memory, encoding, storage, retrieval, and the factors that affect each stage.

The Memory Processing Model: Sensory, Working, and Long-Term Memory; Retrieval

  • Sensory memory

    • Senses momentarily register enormous detail from the environment.

  • Working memory (short-term memory)

    • Holds a few items that are noticed and encoded; limited capacity.

  • Long-term memory (LTM)

    • Only some items are encoded into LTM; information can be altered or lost over time.

  • Retrieval from LTM

    • Retrieval depends on interference, retrieval cues, mood, and motives; some information is retrieved and some isn’t.

  • Overall question addressed: How much do we remember, and what determines whether memory is retained or forgotten?

Memory is a Reconstructive Process

  • Retrieval is not a perfect replay of the original event; it involves reconstructing memories.
  • Features of reconstruction
    • Not like a video recording; subject to influence from expectations, beliefs, and others.
    • Memories change over time; even vivid memories can be altered by later information.
    • We filter information and fill gaps, leading to potential distortions.
  • Consequences
    • Memory can be distorted as people fit new information into existing schemas or imagined events into memory.
    • The act of recall itself can introduce errors due to reconstructive processing.

Source Monitoring and Misinformation Effects

  • Source monitoring error: attributing a memory to the wrong source (experienced, heard, read, or imagined).
  • Misinformation effect: information encountered after an event can alter memory of the event.
  • Illustrative example (generic): People may misattribute the origin of a memory after exposure to post-event information.

Classic Experiments in Memory Reconstruction

  • Carmichael, Hogan & Walter (1932)
    • Participants were shown a figure and given a label that varied by group.
    • The label influenced how participants later drew the figure; demonstrates labeling can bias memory recall.
  • Loftus & Palmer (1974)
    • Experiment on how leading questions influence memory for a car crash.
    • Critical manipulation: verb used to describe impact in the question.
    • Reported speeds (estimates) by participants varied with verb:
    • Smashed into each other? ~ (66 \text{ km/h})
    • Hit into each other? ~ (55 \text{ km/h})
    • Contacted each other? ~ (51 \text{ km/h})
    • Demonstrates how memory is shaped by wording and expectations about events.
  • Question wording and memory distortion
    • Follow-up question about glass can bias answers: e.g., asking about broken glass after a crash increases reported incidence.
    • Reported breakdown (illustrative percentages):
    • Control group: ~12% said yes to seeing broken glass
    • Smashed condition: ~32% said yes to seeing broken glass
    • For broken headlight wording variations:
      • Did you see the broken headlight? ~20%
      • Did you see a broken headlight? ~6%
    • These results show how suggestive questioning inflates memory distortions.
  • Memory reconstruction and the false memory literature
    • People fill in gaps with plausible guesses; memories are built and revised to fit schemas.
    • Imagining events can create false memories; memory reconstruction can lead to distortions.
  • Lost in the Mall Experiment (referenced) as an example of implanted false memories in everyday contexts.

Eyewitness Testimony: Practical Implications and Conditions

  • Eyewitness recall is not an exact replica of events; recall is constructed from multiple sources and can be shaped by schemas.
  • Factors that influence eyewitness accuracy (summarized from slides):
    • Visual conditions: time to observe the perpetrator, lighting, visibility, and disguise.
    • Duration: Short timeframes between witnessing and identification reduce accuracy.
    • Ethnicity bias: Greater inaccuracy when observing people of different ethnic backgrounds from one’s own.
    • Post-event information: Discussing with other witnesses can reduce accuracy.
    • Stress and weapon focus: Under stress or when weapons are present, attention may be disproportionately focused on the weapon, reducing recall of other details.
  • Practical takeaways: The reliability of eyewitness testimony can be improved with careful procedures, but overall it remains vulnerable to distortion.

The Wording and Conditions That Reduce or Exacerbate Eyewitness Error

  • Conditions that increase reliability
    • Good lighting and clear visibility of the perpetrator.
    • No disguises; straightforward views.
    • Short time between witnessing and identifying.
  • Conditions that reduce reliability
    • Viewing individuals of different ethnic backgrounds can reduce accuracy.
    • If witnesses discuss the event with others, their memories can become more similar and possibly distorted.
    • High-stress situations and weapon focus can impair overall recall.

How Do We Forget? Theories of Forgetting

  • Four broad categories of forgetting mechanisms:

    • Encoding failures: information never gets into long-term storage because it was not effectively encoded.
    • Decay: memories fade over time if not used; neural pathways deteriorate if not reinforced.
    • Interference: new or existing information competes with or disrupts memory; includes retroactive and proactive interference.
    • Retrieval failure: information is stored but cannot be retrieved without appropriate cues.

  • Encoding and transfer to LTM

    • The process from sensory input to long-term memory involves attention and encoding success.
    • Encoding failure reduces transfer to STM/ LTM.

  • Encoding failure and the transfer to STM/LTM (illustrated by schematic processes):

    • Sensory input → Sensory Memory → Encoding → Working Memory → Long-Term Memory
    • Forgetting can occur at any stage due to encoding failures, decay, or interference.

Forgetting as Decay

  • Decay theory: memories fade gradually if not used; over time neural pathways deteriorate.
  • Visual cue: “Use it or lose it.”
  • Example from the literature: HM case demonstrates severe memory impairment after surgical removal of hippocampal regions (details below).

Forgetting as a Function of Time: Ebbinghaus Forgetting Curve

  • Ebbinghaus' forgetting curve (conceptual): memory retention declines over time after learning.
  • Typical representation (schematic):
    • Memory Retention (%) vs. Time since learning.
    • Initial retention is high (near 100%), followed by rapid early decline, then a slower decline over time.
  • A common schematic featured in lecture slides shows exponential-like decay with time points such as immediate recall, 20 minutes, 1 hour, and 9 hours to illustrate rapid early forgetting.
  • Mathematical representation (model):
    • A common formalization is the exponential forgetting function: R(t) = R_0 e^{-k t}
    • Where:
    • (R(t)) is memory retention at time (t),
    • (R_0) is initial retention (often 1 or 100%),
    • (k) is the forgetting rate constant ((k>0)).
    • This captures the idea that forgetting is faster shortly after learning and slows down over time.

Forgetting as Interference

  • Interference theory: memory can be impaired by other similar information.
  • Retroactive interference: new information learned today interferes with information already stored from the past.
  • Proactive interference: existing knowledge interferes with new learning.
  • Practical implication: When learning similar material close in time, performance can degrade due to interference.

Retrieval Failure and Retrieval Cues

  • Forgetting can occur when retrieval is unsuccessful, even if the information is stored.
  • Retrieval cue theories propose that successful retrieval depends on available cues; absent cues lead to forgetting.
  • The process involves recognizing or recalling information with the aid of cues; without cues, retrieval may fail.

Amnesia: Types, Mechanisms, and Case Studies

  • Amnesia is defined as the loss of memory.
  • Retrograde amnesia: inability to remember events that occurred before a brain trauma.
  • Anterograde amnesia: inability to remember events that occur after brain trauma; inability to transfer new information into LTM.
    • In typical amnesia, memory for past events (some retrograde) may be retained partially, but new memory formation is impaired.
  • HM case study (Henry Molaison)
    • 1926–2008; underwent surgical removal of large portions of temporal lobes in 1953, including both hippocampi.
    • Resulted in virtually complete anterograde amnesia; some retrograde amnesia for about 11 years prior to surgery.
    • IQ remained above average (around 112).
    • No explicit (declarative) memory, but preserved implicit (procedural) memory, demonstrated by intact performance on tasks like mirror drawing.
  • Implication: Distinct memory systems exist; some memories can be stored and retrieved implicitly even when explicit memory formation is impaired.

Memory Subsystems and Long-Term Memory Organization

  • Long-Term Memory (LTM) is often divided into explicit (declarative) and implicit (non-declarative) memory:

    • Explicit memory (conscious recall):
    • Semantic memory: memory for facts about the world.
    • Episodic memory: memory for personal experiences and events.
    • Implicit memory (without conscious recall):
    • Procedural memory: memory for knowing how to do things (skills, procedures).
    • Priming: facilitated processing of a stimulus due to prior exposure.
    • Conditioning and Habituation: learned associations and responses.

  • Neural substrates (general):

    • Hippocampus: crucial for forming new memories; also implicated in spatial memory and certain aspects of working memory.
    • Neocortex: storage for long-term memories; distributed across cortical areas depending on memory type.
    • Modern views suggest memory storage is more distributed and complex than early, hippocampus-centric models.

  • HM and memory subsystems highlight that memory is not monolithic; different systems contribute to different kinds of memory performance.

The Anatomy of Memory: Highlights from Neurological Findings

  • Hippocampus
    • Involved in spatial memory and in the formation of new memories (critical for encoding into LTM).
  • Neocortex
    • Storage site for long-term memories, with different cortical areas supporting different types of information (e.g., semantic vs. perceptual details).
  • The broader memory network is more intricate than a single region; memory is supported by interactions across brain regions.

Practical and Ethical Implications of Memory Research

  • Because memory is reconstructive and highly susceptible to suggestion, the use of leading questions, false information, or repeated interviewing can distort eyewitness accounts.
  • In legal settings, this has important implications for the reliability of testimony and the design of interviewing protocols to minimize misinformation effects and source monitoring errors.
  • Clinically, understanding amnesia and memory subsystems informs diagnosis and rehabilitation strategies for patients with memory disorders.
  • Philosophical considerations: memory does not perfectly preserve reality; our sense of personal identity and continuity can be shaped by the reliability of our memories.

Summary of Core Takeaways

  • Memory is reconstructive: recall involves constructing memories, not replaying them verbatim.
  • Forgetting can occur at encoding, storage (decay), interference, or retrieval stages.
  • Eyewitness testimony is particularly vulnerable to distortion due to source monitoring errors and misinformation effects.
  • Classic experiments (Carmichael et al., Loftus & Palmer) demonstrate how labels, wording, and post-event information bias memory reports.
  • The forgetting curve illustrates rapid early forgetting followed by a slower decline over time; a common mathematical representation is R(t) = R_0 e^{-k t}.
  • Interference (retroactive and proactive) and retrieval cues are central to explaining why we forget.
  • Amnesia demonstrates distinct memory systems: explicit (semantic/episodic) vs. implicit (procedural, priming, conditioning, habituation).
  • The HM case underscores the role of the hippocampus in forming new explicit memories, while implicit memory can remain intact.

References and Key Names Mentioned (from lecture content)

  • Carmichael, Hogan & Walter (1932) – labeling effect on memory reconstructions.
  • Loftus & Palmer (1974) – eyewitness memory influenced by verb choice (smashed, hit, contacted).
  • Lost in the Mall Experiment – demonstrates implanted false memories.
  • HM (Henry Molaison) – classic amnesia patient with bilateral hippocampal removal; preserved implicit memory.