Reconstructive Memory

memory

an active processing system which involves encoding, sorting, and retrieving information for future use

reconstructive memory

theory that suggests that memory is not a perfect recording of events, but rather an active process of reconstruction influenced by prior knowledge, expectations and schemas

leading questions

questions that contain suggestive information which can alter a person’s memory of an event

schema

mental representations based on prior experiences that help us understand the world

Loftus and Palmer (1974) - Aim

To investigate whether the use of leading questions would effect estimation of speed

Loftus and Palmer - Method

45 volunteers from a university; true lab, independent measures, convenience sampling

Loftus and Palmer - Procedure

• Divided into 5 groups and watched seven films of traffic accidents

• After viewing, each group was asked a leading question: "How fast were the car going when they ___ /into each other?"

• Verb changed for each group: hit, smashed, collided, bumped and contacted

Loftus and Palmer - Results

• Mean estimates of speed were highest for smashed and lowest for contacted

• Smashed: 40.8mph

• Collided: 39.3mph

• Bumped: 38.1mph

• Hit: 34.0mph

• Contacted: 31.8mph

Loftus and Palmer - Conclusion

Suggest that the results could be due to the participant’s labels being distorted by the verb used to describe the car crash

Loftus and Palmer - Link

• Demonstrates reconstructive memory by showing that information provided after an event can influence how that event is subsequently remembered, meaning memory is not necessarily an exact reproduction of the original experience

• If memory was exact, changing the verb should not have systematically influenced estimates of speed

• "smashed" may have activated schema associated with more severe collisions, whereas "contacted" may have activated expectations of a less sever collision

• Demonstrates that retrieval can involve combining information from the original event with subsequent information and existing knowledge or schemas, resulting in a reconstructed representation of the event

Loftus and Palmer - Strengths

• High internal validity

• Quantitative data - objective comparison

Loftus and Palmer - Limitations

• Demand characteristics or responses bias

• Sampling bias

Yuille and Cutshall (1986) - Aim

To investigate the accuracy in recall of eyewitnesses to a real crime, in response to leading questions and over time

Yuille and Cutshall - Method

13 eyewitnesses aged 15-32 (3 females, 10 males); field study with semi-structured interviews, purposive sampling

Yuille and Cutshall - Procedure

• Interviewed 4-5 months after incident

• Asked to give their own account and then asked questions after

• Half of the group were asked if they saw a broken headlight and the other if they saw the broken headlight (there was no broken headlight)

• Half were asked about a yellow panel vs the yellow panel (panel was blue)

• Asked to rate stress and if they had any emotional problems

• Scoring procedure to turn qualitative data into quantitative data

• Researchers used systems of 'action details' and 'description details' to collate information

Yuille and Cutshall - Results

• Misleading questions had very little effect on their recall

• 10 witnesses said that there was no broken headlight and no yellow panel (correct)

Yuille and Cutshall - Conclusion

Eyewitnesses were not inaccurate, contrary to findings of previous research (were true lab). Misleading questions had little effect on the eyewitnesses.

Yuille and Cutshall - Link

• Challenges the idea that memory is reconstructive, despite exposure to misleading questions, most eyewitnesses accurately recalled details suggesting that emotionally charged conditions memory may be less prone to reconstructive

Yuille and Cutshall - Strengths

• High ecological validity

• Scoring procedure produced quantitative data, objective comparison

Yuille and Cutshall - Limitations

• Limited generalisability

• Some weak points in scoring procedure, reducing reliability