Aims of the study

Loftus and Palmer (1974) conducted research to explore how the wording of questions can influence people's memories of events. Specifically, they wanted to test whether leading questions could affect the accuracy of eyewitness accounts, focusing on speed estimates in car crashes.

Experiment 1: Investigating speed estimates

Participants

45 students from the University of Washington took part, divided into five groups of nine participants each.

Method

This was a laboratory experiment using an independent measures design, where different groups experience different conditions. The independent variable (IV) - the factor manipulated by researchers - was the verb used in the critical question: "hit", "smashed", "collided", "bumped", or "contacted". The dependent variable (DV) - the factor measured - was the participants' estimate of the cars' speed in miles per hour.

Procedure

1. Participants watched seven short films of traffic accidents, presented in a random order.
2. After each film, they wrote a general account of what they had seen.
3. They then answered specific questions, including the critical one: "About how fast were the cars going when they [verb] each other?"
4. Each group heard a different verb in this question.

Results

The verb "smashed" led to the highest average speed estimate, while "contacted" produced the lowest. Other verbs gave estimates in between. These findings show that participants are generally poor at estimating speeds accurately, but more importantly, the wording of the question systematically influences their judgements, with harsher verbs leading to higher estimates.

Experiment 2: Investigating memory distortion

Method

This was also a laboratory experiment with an independent measures design. The IV was the verb in the speed question: "smashed", "hit", or no question about speed (control group). The DV was whether participants reported seeing broken glass (yes or no) in response to a later question.

Procedure

1. Participants viewed a film showing a multiple-car accident.
2. They answered questions, including the critical speed question with either "smashed" or "hit" (or no speed question for the control group).
3. Later, participants were asked additional questions, including: "Did you see any broken glass?"

Results

Participants who heard "smashed" gave higher speed estimates than those who heard "hit". For the broken glass question, more participants in the "smashed" group reported seeing broken glass compared to the "hit" and control groups. This suggests that the verb "smashed" not only inflated speed estimates but also increased the chance of falsely remembering broken glass.

Key conclusions and implications

Loftus and Palmer concluded that leading questions can distort eyewitness memory by integrating post-event information with the original memory, creating a single, altered recollection. This indicates that information from the original memory is merged with information after the fact, producing one distorted memory.