Loftus & Palmer (1974). Reconstruction of Automobile Destruction — An Example of the Interaction between Language and Memory.
Arhe Vaninetti
Research Methods and Statistics, PSY301, Pacific University, Oregon
March 6, 2023
During the era of history, the issue of false memories and languages’ effects on interpretations were being heavily researched due to various court-based cultural factors. The understanding of memory, whether short term or long, was reviewed largely due to weak understanding. The notion of implanting memories, false eyewitness memories, false accusations, and confabulated memories were Loftus & Palmer’s (1974) influences in creating a study on memory distortion in eyewitness testimony. This study was set up not only to be about memory confabulation, but specifically in the context of a legal issue.
In the study, the researchers solicited 45 student participants who were divided into six groups. Within each of the groups, the participants were shown different video footage of car accidents that lasted between 5 and 30 seconds. The participants were then asked, “About how fast were the cars going when they hit each other?” In each group, the word hit was replaced with smashed, collided, bumped, and contacted. Loftus & Palmer (1974) stated that the experiment in its entirety lasted 1.5 hours.
Within the results section of the study, Loftus & Palmer (1974) include an analysis of variance on the verbs (i.e., hit, smashed, etc.) and with the subjects and video footage as random effects. This produced a large F ratio, and a P-value of <0.005, indicating that the results were statistically significant.
Within the discussion of the study, the authors illustrate that the phrasing of a question can influence the answer given, including in higher stakes activities like reviewing footage of a crash. They indicate that this may be due to the biasing of the verbs (i.e., smashed vs. hit), or because the question changes the participant’s minds’ eye view of the footage. Either way, the reason for the change in answer after a biased question should be researched to discover which situation is more likely—or if a combination of both is best suited.
In the second experiment, 150 student participants were solicited and placed into groups of various sizes (similar to experiment I). The participants watched video footage of a multi-car crash and were then tasked with answering a questionnaire that included a short answer response and various checkbox questions about the accident. One of the questions was the same question from experiment I, “About how fast were the cars going when they hit each other?” In 50 of the participants, they received the hit question, 50 received the smashed question, and 50 received no question about speed at all. One week later, the participants returned and answered the same questionnaire again, but with the additional yes/no question of, “Did you see any broken glass?” The researchers noted in the study that there was no broken glass in the footage.
In the results of experiment II, Loftus & Palmer (1974) found that the participants who had received the smashed question were more likely to write that the cars were going a faster speed than the participants who received the hit question (10.46 MPH vs. 8.00 MPH, respectively). They also found that the probability of saying yes to the broken glass question was 0.32 with smashed question participants, as compared to 0.14 with hit question participants. In the discussion section of experiment II, the researchers state that the effects of framed questions not only have immediate effects, but effects that last at least a week. Furthermore, Loftus & Palmer propose that factual memory and confabulated memory combine to create a memory that we cannot distinguish between truth and fiction. The addition of the broken glass suggestion created a confabulation in the participants’ memory.
An interesting addition that could be made to this article would be to utilize participants with aphasia (i.e., those who cannot view images in their mind’s eye). This could create the differentiation between the confabulated mental imagery and the perceived truth.