Loftus & Palmer (1974)
Reconstruction of automobile destruction.
Background: Various early studies of memory such as Bartlett 1932, showed how memories aren't accurate records of our experiences. It seems that we attempt to fit our past events into our existing representations of the world, making memory more coherent, clear, or make more sense for us.
Aim/Method: The aim of Loftus & Palmer's laboratory experiment was to investigate how information can influence a witness' memory when supplied after the incident. The study uses two laboratory experiments which both use an independant measures design.
Participants: 45 students from the University of Washington.
Procedure: Students were shown seven short clips of cars colliding. They were then split into five groups (9 participants in each) and were all asked - 'About how fast were the cars going when they _____ each other?' The groups were then given a different verb to fill in the blank, which were - 'smashed, collided, bumped, hit or contacted'. This therefore shows the independant variable (IV) was the verb that was used. The dependant variable (DV) was the estimate of the speed given by the participants.
Controls: All participants:
- were of a similar age
- were asked the same questions apart from the change of verb in the critical question
- watched the same video clips
- were tested in the same environment (laboratory)
The critical question was hidden among other questions regarding the collision to reduce demand characteristics (preventing the participants in guessing what the researcher is really interested in).
- IV: wording of the critical question
- DV: estimate of how fast the cars were going
Results: How the question was phased influenced the participants speed estimates. When the verb 'smashed' was used, participants estimated the cars were travelling much faster than when the verb 'contacted' was used.
Verb Mean Estimate of Speed (mph)
- Smashed 40.8
- Collided 39.3
- Bumped …