PSYC2361 Episode 2: Memory Reliability and Eyewitness Evidence

Memory Reliability and the Common Sense Fallacy

• Despite advances in forensic science (e.g., DNA analysis), many criminal convictions still rely solely on eyewitness testimony. Consequently, understanding the reliability and operations of eyewitness memory is crucial to the justice system.
• There is a prevalent belief among some criminal justice professionals, such as judges, that eyewitness memory is "common sense." However, behavioral science research frequently contradicts this assumption.

Empirical Evidence on Juror Beliefs vs. Reality

• Brigham and Bothwell Study (Florida State University): John Brigham and Robert Bothwell conducted research where prospective jurors were given descriptions of real studies involving simulated crimes and subsequent lineups.
• Juror Predictions: Prospective jurors estimated that identification accuracy would fall between $51\%$ and $71\%$.
• Actual Accuracy: The actual accuracy of participants in those studies was significantly lower, ranging between $13\%$ and $32\%$.
• Conclusion: Common sense beliefs regarding memory accuracy are generally over-optimistic and do not align with empirical data. These inaccuracies are problematic because they dictate the amount of credibility professionals and jurors assign to witnesses.

Case Study: The Memory of George W. Bush and 9/11

• To investigate if high-stakes real-world situations improve memory accuracy, researchers examined George W. Bush's recollection of how he heard about the terrorist attacks on September 11, 2001.
• Account 1 (December 4, 2001): During a speech in Florida, Bush told a boy named Jordan that he was sitting in a classroom and saw the first plane hit the building on a TV outside the room. He stated, "I thought it was a bad pilot," before going into the classroom and later being informed of the second plane by Andy Card.
• The Problem with Account 1: Footage of the first plane hitting the North Tower did not emerge until September 12. It was impossible for him to have seen it live on TV before entering the classroom.
• Public Reaction: Anti-Bush websites (e.g., bushwatch.com) and mainstream media (e.g., The Guardian) used this discrepancy to accuse the President of lying or being involved in a conspiracy, based on the assumption that one cannot misremember such a monumental event.
• Account 2 (December 20, 2001): Interviewed by the Washington Post, Bush’s story changed to reflect the fact that he was informed of the first crash before entering the classroom but did not see it on video until later.
• Account 3 (January 5, 2002): Bush reverted to his first account, again claiming he saw the first plane hit the building on TV before entering the classroom and commented on the "terrible pilot."
• Analysis: These errors illustrate that even for the most significant events, memory is not a perfect recording.

Explanations for Memory Distortion

Theoretical Explanations by Daniel Greenberg (Duke University)

1. "Wrong Time-Slice" Error: People often retrieve a memory of an event but attribute it to the wrong time. Visual imagery, especially from TV footage, is dominant and easily modified. Bush likely saw the footage repeatedly later and inadvertently shifted that memory back in time to the morning of the attack.
2. Narrative Construction: Humans reconstruct gaps in stories based on "scripts" (general ideas of what usually happens). Greenberg argues Bush saw video of the aftermath later and "added cause to effect" to create a logical narrative. His verbal hesitations (e.g., "the TV was obviously on") suggest he was reasoning out or reconstructing the memory as he spoke.

Confidence and Accuracy: The Pezdek Study

• Kathy Pezdek Study: Seven weeks after 9/11, Pezdek surveyed undergraduates in Hawaii, California, and New York.
• The Question: "On September 11, did you see the videotape on television of the first plane striking the first tower?"
• Results:
  • Hawaii: $84\%$ gave the wrong answer.
  • California: $61\%$ gave the wrong answer.
  • New York: $76\%$ made the same mistake as Bush.
• Critical Finding on Confidence: Participants who correctly answered "no" were less confident than those who incorrectly answered "yes." In this case, higher confidence was associated with a higher likelihood of being wrong. This is vital because confidence is the primary cue used by legal professionals to judge witness accuracy.

The Three Stages of Memory

Memory is a creative, constructive process rather than a video playback. Errors can occur at any of these three stages:

1. Encoding: The initial acquisition of information. If info is never encoded, it never becomes a memory.
2. Storage: Keeping the memory intact over time. Information may be lost (forgotten) or modified.
3. Retrieval: Reporting the memory at a later date. The memory may exist but be inaccessible or biased by the way it is requested.

The Encoding Stage: Subjective Reconstruction

• Subjective Reality: Perception is a subjective reconstruction created by the brain.
• Change Blindness: In a demonstration where an entire car is removed from an image with a grey flicker in between, many people fail to notice the change. The brain focuses on narrow elements and "fills in" the rest.
• Expectation and Influence: The Fischer (1968) image (Old woman / Young woman / Old man) shows that different people interpret the same visual stimulus differently based on expectation.

The Weapon Focus Effect

• Definition: A phenomenon where a witness’s memory for a suspect’s face is impaired because their attention is focused on a weapon.
• Explaining the Effect:
  • Threat Hypothesis: The witness is so scared they fixate on the source of the danger.
  • Unusualness Hypothesis: The weapon is unexpected in the context and draws attention away from peripheral details.
• Meta-Analysis (Jonathan Fawcett): Empirical support exists for both explanations. While some argue this only happens in labs, Fawcett found the effect persists in the real world once variables like "duration of exposure" (armed suspects may stay longer) are controlled.

Storage and Retrieval Challenges

Forgetting and Misinformation

• Forgetting Major Events: Geoff Maycock and Julia Lester (UK) found that drivers forget approximately $30\%$ of the car crashes they are involved in per year. This suggests people forget entire events, not just minor details.
• The Misinformation Effect: Stored memories can be altered by introducing false information after the event.
• Harmut Blank Study (2013): Participants watched a CCTV robbery. After discussing it with a "plant" who gave misinformation (e.g., identifying the wrong man with the gun), they incorporated that false data into their own reports a week later, even when warned to ignore the other witness.
• Charles Morgan (Yale) Study: Military personnel in a high-stress prisoner-of-war simulation identified the wrong interrogator after being shown a misleading photograph.

Retrieval Biases: Loftus and Palmer

• Leading Questions: Elizabeth Loftus and John Palmer showed participants films of car crashes.
• The Word Effect: Participants asked how fast cars went when they "SMASHED" gave higher speed estimates than those asked how fast they went when they "HIT" each other.
• False Memories: One week later, the "SMASHED" group was more likely to falsely remember seeing broken glass.

Forensic Hypnosis

• Definition: An altered state of consciousness intended to heighten focus and concentration.
• Anecdotal Successes:
  • Chowchilla Kidnapping (1976): A bus driver was able to recall a van license plate (missing only one digit) under hypnosis after his normal recall failed.
  • San Francisco Kidnapping: A victim recalled unique rust spots and a specific mechanic transaction involving a credit card, leading to an arrest.
• Scientific Critique: Laboratory studies generally show no memory improvement under hypnosis.
• Suggestibility and Thresholds: Marilyn Smith suggests hypnosis increases suggestibility, lowering the threshold for what someone accepts as a memory. This increases "Correct Hits" but also increases "False Positives."

The Contingency Table of Memory Recall

• Context Reinstatement: Research by Krafka and Penrod (University of Wisconsin-Madison) showed that simply reinstating the context (e.g., asking a shop assistant to remember the environment of a transaction) improved accuracy without the risks associated with hypnosis.

The Cognitive Interview (CI)

Developed by Edward Fisher (UCLA) and Ronald Geiselman (FIU) to replace standard, often interruptive, police interviewing techniques.

Theoretical Foundations

1. Encoding Specificity Principle (Endel Tulving): Memory retrieval is most effective when the cues used during retrieval overlap with the information stored during encoding.
2. Multiple Retrieval Paths: A single memory can be accessed via several different cues.

The Four Original Mnemonics

1. Report Everything: Encourage the witness to report every detail, even if trivial or partial. This requires breaking social rules about conversation dominance.
2. Mental Reinstatement of Context: Asking the witness to mentally recreate the physiological, cognitive, and emotional state they were in during the event. Looking at a blank wall or closing eyes helps concentration.
3. Change Perspective: Recalling the event from the viewpoint of another witness. (Note: Critics like Kebbell & Wagstaff warn this may lead to speculation).
4. Change Order: Recalling events in reverse chronological order. This disrupts the use of "scripts" or "schemas" which often cause people to report what usually happens rather than what actually happened.

The Enhanced Cognitive Interview (ECI)

Fisher and Geiselman (1992) layered communication skills into the CI, creating a 9-phase process:

1. Greet and Rapport: Introduce names, use neutral questions, and build empathy.
2. Environment/Transfer of Control: Ensure a peaceful environment; explicitly tell the witness they are in control of the interview.
3. Free Report: Use context reinstatement to prompt an uninterrupted account.
4. Questioning: Follow the witness's own retrieval order; use open-ended questions; avoid "topic hopping."
5. Varied Retrieval: Use reverse order or change perspective; prompt using the five senses (e.g., smell in arson cases).
6. Investigation Specifics: Revert to standard questioning for specific needs (this has lower evidential value).
7. Summary: Check the interviewer's understanding against the witness's words.
8. Closure: Return to neutral topics and explain the next steps.
9. Evaluation: Discuss performance with colleagues for professional development.

Efficacy and Modifications

• Study Results: Lab studies showed a $25-35\%$ increase in information. A field study with Florida detectives found a $47\%$ increase in corroborated facts.
• Limitations: ECI is time-consuming, difficult to train, and requires a cooperative witness.
• Category Clustering Recall (CCR): Proposed by Paulo et al. (2016) as an alternative to the "Change Order" mnemonic. Witnesses organize recall into categories: Objects, Locations, Actions, and Sounds/Conversations.
• CCR Study (Paulo et al., 2016): A study using a clip from the drama Inspector Max found that CCR was superior to the traditional "Change Order" mnemonic in eliciting correct details, likely because semantic categorization is more natural for the human brain.

Questions & Discussion

Question from Jordan (to President Bush): "How did you feel when you heard about the terrorist attack?" Bush's Response (Account 1): "I was sitting in a schoolhouse in Florida… and I was sitting outside the classroom waiting to go in, and I saw an airplane hit the tower—the first one. There was a TV on. And I used to fly myself, and I said, 'Well, there's a terrible pilot.' And I said, 'It must have been a horrible accident.' But I was whisked off there… I didn't have much time to think about it."

Question to President Bush (January 5, 2002): "What was the first thing that went through your head when you heard that a plane crashed into the first building?" Bush's Response (Account 3): "I was sitting there, and my Chief of Staff—well, first of all, when we walked into the classroom, I had seen this plane fly into the first building. There was a TV set on. And you know, I thought it was pilot error and I was amazed that anybody could make such a terrible mistake. And something was wrong with the plane, or—anyway, I'm sitting there, listening to the briefing, and Andy Card comes in."

Standard Interviewer Prompt (Traditional): Asking specific, closed questions and interrupting frequently to get details relevant to the investigation leads to less information and more potential for leading the witness.