Interference may have caused forgetting (not just decay)
Doesn't reflect real-life memory tasks
Jacobs (1887) - STM Capacity (Digit Span)
Aim: To investigate the capacity of short-term memory
Procedure: Participants heard sequences of digits/letters. Had to repeat them back in correct order. Sequence length increased until unable to recall correctly.
Findings: Mean digit span: 9.3 items. Mean letter span: 7.3 items.
Conclusion: STM has limited capacity (7±2 items) - Miller later confirmed this
Aim: To investigate encoding in short-term and long-term memory
Procedure: Four groups shown word lists: acoustically similar (cat, cab, can), acoustically dissimilar, semantically similar (great, large, big), semantically dissimilar. Recalled immediately (STM) or after 20 minutes (LTM).
Findings: STM: worse recall for acoustically similar words. LTM: worse recall for semantically similar words.
Godden & Baddeley (1975): Divers learned words underwater/on land. Recall 40% lower when context changed (learned underwater, recalled on land or vice versa).
Carter & Cassaday (1998): Antihistamines created different internal state. Recall worse when state at learning ≠ state at recall.
✓ Strengths
Research support (Godden & Baddeley, Carter & Cassaday)
Real-world application - exam technique (study in similar conditions)
Comprehensive explanation - covers various types of cues
Explains retrieval blocks and "tip of tongue" phenomenon
✗ Limitations
Context effects may be overstated - Baddeley found smaller effects with recognition
Doesn't explain all forgetting
Some studies failed to replicate context effects
Learning in multiple contexts may be better (prevents context dependence)
👁️ Eyewitness Testimony (EWT)
Misleading Information - Leading Questions
Questions phrased in a way that suggests a certain answer
Response Bias: Wording influences answer but doesn't change memory
Loftus & Palmer (1974): 45 students watched car crash videos. Asked "How fast were the cars going when they _____ each other?" Verb varied: smashed (40.5 mph), collided (39.3), bumped (38.1), hit (34), contacted (31.8). Week later: "Did you see broken glass?" (there was none). "Smashed" group: 32% said yes vs. 14% in "hit" group.
Conclusion: Leading questions can alter memory (not just response bias)
✓ Strengths
Controlled lab experiment - high internal validity
Applications in legal system - improving interview techniques
Replicated many times
✗ Limitations
Artificial task - watching videos vs. real accidents (lacks emotional impact)
Student sample - age bias, may not generalize
Individual differences in susceptibility
Demand characteristics - students may guess aim
Misleading Information - Post-Event Discussion
Witnesses discussing events can contaminate each other's memories
Memory Contamination: Co-witnesses mix information from others into their own memory
Memory Conformity: Witnesses go along with others for social approval
Gabbert et al. (2003): Pairs watched different videos of same crime. Then discussed. 71% mistakenly recalled items from other person's video. Control group (no discussion): 0% errors.
✓ Strengths
Research support (Gabbert et al.)
Important real-world application - separating witnesses
Explains why witness accounts often converge
✗ Limitations
Lab studies lack emotional impact of real crimes
Difficult to separate conformity from actual memory contamination
Individual differences - some more susceptible
Anxiety Effects on EWT
Yerkes-Dodson Law: Inverted-U relationship. Moderate anxiety improves accuracy, very high/low anxiety impairs it.
Weapon Focus Effect: Anxiety about weapon reduces accuracy for other details
Loftus et al. (1987): Participants heard discussion in lab. Low anxiety: man emerged with pen, hands greasy. High anxiety: man emerged with knife, hands bloody. Identification: 49% (pen) vs. 33% (knife).
Yuille & Cutshall (1986): Real crime (gun shop robbery). 13 witnesses interviewed after 4-5 months. Very accurate recall (88%). Those with highest stress most accurate.
✓ Strengths
Practical applications - legal system awareness
Weapons focus supported by eye-tracking studies
Real-world research (Yuille & Cutshall) shows high accuracy possible
✗ Limitations
Contradictory findings - some studies show anxiety improves memory
Ethical issues - inducing high anxiety in participants
Lab studies lack realism of real crimes
Weapon focus may be due to surprise, not anxiety
Individual differences - some cope better with anxiety
Cognitive Interview (Fisher & Geiselman, 1992)
Report Everything: Include all details, even seemingly trivial
Reinstate Context: Mentally recreate environment and emotions
Reverse Order: Recall events in different chronological order
Change Perspective: Recall from different viewpoints
Enhanced CI: + rapport building, minimize distractions, open questions, witness controls flow
✓ Strengths
Research support - Kohnken et al. meta-analysis: 34% increase in correct information
Real-world effectiveness - used by many police forces
Based on psychological principles (encoding specificity, multiple retrieval paths)
Reduces leading questions
✗ Limitations
Time-consuming - not suitable for all situations
Requires special training - expensive
Also increases incorrect information (Kohnken: 61% increase in errors)
Some components more effective than others (context reinstatement best)
Not suitable for all witnesses (e.g., children, learning difficulties)