Eyewitness Psych - Midterm

Michael Green

Spent 27 years in prison largely on eyewitness testimony and was later exonerated via DNA evidence.

Fallibility of memory

Memory is reconstructive, susceptible to distortions, decay, interference, and suggestion.

Importance of procedure

Some variables are under control (e.g. how lineups are constructed), others are not.

Trade-off of realism vs control

Laboratory studies provide experimental control but may lack ecological validity; field studies carry realism but less control.

Memory Models

Includes sensory, short-term, and long-term memory and how they apply to face recognition.

Signal Detection Theory (SDT)

Helps model decisions as discriminating signal (target) vs noise (non-target) under uncertainty.

Hits, false alarms, misses, and correct rejections

Concepts in lineup tasks modeled by Signal Detection Theory.

Dual-Process / Recognition Models

Distinguishes between familiarity and recollection-type processes in recognizing faces.

Face Processing & Person Memory Specifics

Face recognition has special properties like holistic processing and configural information.

Distance, lighting, exposure time

Variables that affect accuracy in eyewitness identification.

Confidence-accuracy relationships

Explains under what conditions confident responses are trustworthy.

Lineup decisions

Can be biased or misled by noise or suggestive cues.

Signal detection curves

Graphical representations used in Signal Detection Theory.

Criterion shifts

Changes in decision thresholds that affect identification outcomes.

ROC (Receiver Operating Characteristic) analyses

A tool used to evaluate the performance of a binary classifier system.

Dual-process distinctions

Differentiates between processes of familiarity and recollection in memory.

Face-specific cognitive phenomena

Includes effects like inversion effect and own-race bias.

Noise in memory

Refers to memory variability that can affect decision thresholds.

Bias in decision thresholds

Can lead to inaccuracies in eyewitness identification.

Estimator variables

Factors beyond the control of the justice system that help estimate how reliable an identification might be.

Exposure duration / viewing time

How long the witness saw the perpetrator; more time results in more accurate identifications.

Physical separation, lighting quality, obstruction

Conditions that degrade accuracy, such as poor lighting, oblique angles, and distance.

Distance & viewing conditions

Factors like lighting, angle, and visibility that affect identification accuracy.

Stress / arousal

Emotional/stress level during witnessing; high arousal can narrow attention and reduce memory for other features.

Weapon focus effect

The phenomenon where the presence of a weapon draws attention and impairs encoding of other features.

Retention interval / delay

Time between the event and identification; accuracy decays with time due to forgetting or interference.

Perpetrator distinctiveness

How distinctive the face, clothing, or appearance is; more distinctive features facilitate recognition.

Own-race / cross-race bias

The tendency to recognize faces of one's own racial group more accurately, leading to misidentification of cross-race individuals.

Witness characteristics

Factors such as age, attention, intoxication, victim vs bystander status, and cognitive ability that can undermine recognition.

Other moderating factors

Visual obstructions, disguises, disguise changes, and environmental context that can affect identification accuracy.

Critical Insights & Caveats

Some factors show strong, replicable effects while others are weaker or context-dependent.

Interactions between estimator variables

The possibility that different estimator variables may interact, such as stress affecting exposure time.

Laboratory vs real-world

Field and archival data tend to support lab results, though effect sizes often attenuate.

Degree of factors

It's not just presence or absence but the degree of factors (e.g., more vs less lighting) that matters.

Study Strategies

Creating flashcards with definitions, typical empirical effects, and caveats for each estimator variable.

Real-world examples

Using scenarios like low light at night or cross-race interactions to anchor abstract findings.

Possible Test / Essay Questions

Questions to explore topics such as the weapon focus effect, retention interval, and own-race bias.

System variables

Factors that can be adjusted to improve accuracy and reduce error in identification procedures.

Lineup composition / foil selection

The non-suspect 'fillers' must be plausible and match general description.

Simultaneous vs sequential presentation

Simultaneous presents all options at once, while sequential presents one at a time.

Estimator- vs absolute-judgment instructions

Estimator instructions tell the witness the suspect may or may not be present, while absolute implies 'someone is in the lineup'.

Double-blind administration

The administrator does not know who the suspect is to prevent inadvertent cues or influence.

Feedback / confirming statements

Whether feedback is given after selection can inflate confidence and bias memory.

Multiple lineups or repeated identifications

Repeated lineup procedures can contaminate memory and encourage guessing.

Delay between event and lineup

Delays may degrade memory; introducing fillers too late may bias choices.

Instructions & warnings

Pre-lineup instructions aim to reduce guessing and pressure.

Interviewer techniques & questioning

How questions are phrased can mislead or implant false memory.

Double-blind lineup

Recommended procedure where administrators are unaware of suspect identity to prevent cueing.

No feedback

Post-identification feedback increases witnesses' confidence in erroneous identifications.

Sequential lineups

Encourage more conservative responses but may reduce correct identifications.

Proper filler selection

Poor fillers can lead to suspect standing out or lineup unfairness.

Unbiased instructions

Instructions that indicate the suspect may or may not be present help reduce forced choice pressure.

Best practices

Courts and legislatures can mandate procedures like double-blind and careful instructions.

Judicial scrutiny

Judges are expected to scrutinize identification procedures and provide juries with reliability instructions.

Expert testimony

May need to explain which system procedures were used (or not) in a given case.