Eye Witness Testimony

Página 1

Eye Witness Testimony

ENCODE

RETAIN

RETRIEVE

Legal term for witnesses who give evidence in court, concerning the identity of a suspect

LOFTUS & PALMER 1974 (experiment 1)

Leading questions & memory for a car crash

AIM: to see if leading questions affect recall of car speeds

HOW: Independent Measures Design - Laboratory Situation

FIlm clips of a traffic accident shown to 45 college students

Wrote Descriptions

Answered a number of questions

Different participants asked same question with a different verb.

IV = Verb

DV = Estimated Speed

Question - "About how fast were the cars going when they collided/smashed/bumped/hit/contacted with each other?"

FINDINGS

CONCLUSION

Response Bias Factors - participant not sure of speed so give answer they think researchers want (demand characteristics)

Question causes actual distortion of participants memory of event - word smashed leads participant to recall accident as being more severe than it actually was

LOFTUS & PALMER 1974 (experiment 1)

150 Students shown a film of a multiple car crash which lasted about a minute (action just 4 seconds)

3 groups of 50 asked to give written description, then;

- 1st Group asked how fast when they "smashed" into each other

- 2nd Group asked how fast when they "hit" each other

- 3rd Group not asked anything about speed

1 week later asked "Did you see any broken glass?"

Shows probability of saying "yes" to seeing broken glass is not just related to the speed estimate but by verbs meaning

EVALUATION

Method - Controlled

Usefulness of Research - Show memory is easily distorted - Shows that leading questions can bias the eye witnesses answers

Reliability - not difficult to replicate

Type of Data - Quantitative

Low ecological validity - lab setting - not real life - prepared/expect it - don't normally see fully in front of you

Validity - People who estimated higher speeds, did not remember car going faster but were giving an estimate - why 2nd experiment was conducted

Yuille and Cutshall 1986

Vancouver Shooting Study

Interviewed 13 people who witnessed an armed robbery in Canada

Gave Witness Statements

4-5 months later

Answered some leading & neutral questions

Compar

Compared to original Witness Statements

FINDINGS

Wi

Witness accuracy had not declined over 4-5 months

Misleading questions had no effect on accuracy of recall

Those deeply affected by event were most accurate

Eye Witness Testimony

Loftus et al 1

Loftus et al 2

Loftus Evaluation

Yuille et al