Abstract
The events of an accident that students watched in a YouTube clip were followed by some questions to these students. The question about how fast the cars were going before they smashed into each other resulted into higher estimates of speed than when the verb was replaced with hit. On conducting another test on the same participants a week later, the probability that the participants who received the verb smashed recalled seeing glass after the accident was higher than that of the participants who received the verb hit. These disparities in the answers are consistent with the view that the nature and structure of the question asked would determine the expected answer. This implies that the participants reconstruct their memory of the events following the accident when different questions are used, which suggest the severity with which they consider the accident. The duration of the accident was also estimated differently in accordance with the magnitude the participants equated to the accident.
Remembering the details of complex events that people witness usually vary. Even more, people who have just heard information from second and third parties often fail to remember the exact details of the events they were told. For instance in the case of an accident involving vehicles, eyewitnesses would give different information. More specifically when they are asked about the quantification of numerical variables such as the estimated time the accident took, the speed that these cars involved in the accident were travelling, or how much time elapsed between the surrounding of a horn and the moment of collision Loftus and Palmer (1974).
According to Whipple (1909) and Bird (1927), most people usually have a difficulty in reporting numerical variables like distance, time and speed. This has led to the differences and inaccuracy in the estimations of how long, fast or far an event took through the testimonies of the eyewitnesses. According to previously established research, eyewitnesses have always overestimated the duration of complex and disturbing events such as accidents. This could be because these complex events cause psychological torture to the eyewitnesses and they would always wish the events end quickly. However, these research results such as those of Block (1974); Marshall (1969); and Ornstein (1969) have indicated that judging the speeds of occurrences of complex events like accidents is usually difficult thereby resulting in notably large variations in the testimonies of one witness to another. In Marshall’s study (1969), he featured a test that was run on Air Force personnel who had prior information about the intended research and questioning of the speed, they yielded variable results. The car was moving only 12 mph, but the variations in the estimations by the Air Force personnel were between 10 and 50. These variations in speed usually come about because measuring speed requires more than the visual activity of the human eye, but also machines and technology that could measure the velocity of moving objects. As opposed to speed, time can be counted or estimated considering how long an eyewitness presumed he had been standing at the scene of the accident or watching the same in a video clip.
Just as lightly discussed in the above paragraph, there are certain variables that influence these estimates, thereby leading to the inaccuracies and variations in the eyewitness testimonies. This research investigates one such variable, which demonstrates that the wording of questions about an event a person witnessed can influence their description of an event as well as later recall or memory the same event. In the occurrence of an event, the questions that the eyewitnesses are asked usually vary. Some questions are usually leading and suggestive than others. The Supreme Court (1973) gave legal rules on the indications of when leading questions are allowed. A leading question refers to the questions that are structured in content and form to suggest to the witness the desired answers (Loftus, 1975).
In this study, which replicates the Loftus and palmer’s (1974) study, the participants were shown a YouTube clip of a car driving erratically, followed by an accident that took approximately 2 minutes in duration. The participants were then allowed to complete a short online questionnaire. The participants were randomly allocated samples and divided into two groups with the first half answering the first five questions and the other half answering the second question. The first five questions that were asked in the questionnaires included describing the accident in the participants own words, the number of cars involved in the accident with specific answer choices, and the time of the occurrence of the accident (day or night). They were also asked about the road conditions at the time of the occurrence of the accident with options to choose from them, and the estimates in kilometers per hour the speed at which the cars were going when they smashed or hit each other. In order to identify if the participants could memory the events of the accident, they were asked in an online to complete the questionnaire asking if they saw any broken glass in the YouTube video that they had watched with yes and no choices. The verbs hit and smashed were used differently in order to identify the magnitude of the accident in the view of the eyewitnesses. Hit would appear more gently than smashed.
Method
In this assignment experiment, one hundred and fifty eight students participated as the eyewitnesses. These students were instructed by the lecturer to watch a video clip of a car driving erratically, followed by an accident. The length of the film was approximately two minutes. After watching the YouTube film, the participants were requested to answer online questionnaires in two phases. The first phase of the questionnaire was to be filled immediately after watching the clip and the second section of the questionnaire was to follow a week later. The first requirement of the questionnaire asked them to describe what they saw in the accident in their own narration. This was followed by some specific questions about the accident. Half of the participant sample was asked about the first five questions of the questionnaire, and the second half of the sample asked the subsequent questions. The most important question of the one that interrogated the participants about the speed that the vehicles moved before they ‘hit’ or ‘smashed’ into each other. The two verbs were used separately to identify the magnitude with which the participants rated the accident. Additionally, the question that was asked one week later was very important since it sought to underline if the participants could still recall the events of the accident even a week after watching the YouTube clip.
Results
The results of this experiment was presented in different tables according to the different questions that were asked in the experiment. The first table below indicates the frequencies of ‘yes’ and ‘no’ responses to the question asked one week later about seeing the broken glasses. In this table, the research sought to find the differences in the proportions of the people who said that they saw the glasses versus those who didn’t, according to the verb (‘hit’ or ‘smashed’) used in the survey immediately after viewing the YouTube clip.
Frequencies of ‘Yes’ and ‘No’ responses to the question, ‘Did you see any broken glass?’
Verb condition
Smashed Hit Total
Response Yes Observed freq 32 27 59
No Observed freq 49 50 99
Total 81 77 158
An independent chi-square test on these responses was beyond the .025 level, x2(2) =7.76. From the above table, it is evident that the number of the participants that responded to have seen the glasses after the accident when the verb smashes was used than those who saw them when the verb hit was used. However, those who never observed the glasses when the verb fit was used were more than when the verb smashed was used. The following tables indicate the t-tests conducted on the other questions asked in the questionnaires.
Group Statistics
Experimental group N Mean Std. Deviation Std. Error Mean
Speed 2 Smashed 81 81.6975 11.20705 1.24523
Hit 77 70.5779 11.27966 1.28544
Discussion
This experiment provides additional information that demonstrates a commonly known phenomenon, that is, the nature of a question always determines the expected response. When an individual is asked leading questions, or questions that suggest possible answers to him, it is obvious that the person asking the question has predetermined desired answers he expects the respondent to give (Loftus, 1975). In this experiment, we realized that the question that asked about the speed of the cars before they smashed into each other led to higher speed estimates than when the verb hit was used, despite maintaining the question. Even further, we noticed that the verb used in the question (smashed or hit) later influenced the answers asked a week after the occurrence of the accident.
After watching the video, the eyewitnesses formed some representations of the accident. This gave rise to the first question of the questionnaire that required the participants to narrate the events of the accident. However, when the interviewer asks the participants about the speed of the cars and the duration of the accident, the participants are supplied with external information thereby the differences in using the smash and hit verbs. The participant combines this information and integrates in his memory, consequently influencing his answer to the question that was asked a week later. Since the presence of glass at the scene of the accident would mean a severe accident, the participants would interpret the verbs smashed and hit differently to recall the events of the accident.
Loftus and Palmer (1974) conducted a similar experiment and yielded equal results. They conducted two experiments where the subjects were requested to view films of automobile accidents and later answered questions about the events occurring in the films. In their experiment, forty five students participated in groups of various sizes, and there were seven different automobile accident films that they watched. They used verbs such as smashed, collided, bumped, and contacted in the place of hit. The number of students who reported seeing the glasses after the accident when the verb smashed was used was more than those who reported the same when the verb was replaced with contacted.
When the participants were asked about the speed of the cars before the accident using the verb ‘smashed’, the participants would consider the experimenter labeling the accident as a smash and not a hit. Therefore, this would influence their responses a week after the experiment since they would consider the accident more serious. This explains how external information can change the eyewitness’s testimony relating to the events that they witnessed, and when they are asked about the same after some time. Additionally, the nature and structure of the question is important in gathering of the required information. Leading questions are equally important as open-ended questions since every situation requires a different means of collecting the required information from the eyewitnesses.
References:
Bird, C. (1927). The influence of the press upon the accuracy of report. Journal of Abnormal and Social Psychology, 123-129. doi: 10.1037/h0075870
Block, R. A. (1974). Memory and the experience of duration in retrospect. Memory & Cognition, 153-160. doi:10.1037/0278-7393.1.1.31.
Loftus E. F. (1975). Leading Questions and the Eyewitness Report. Cognitive Psychology 7, 560-572. doi:10.1016/0010-0285(75)90023-7.
Loftus, E. F. & Palmer, J. C. (1974). Reconstruction of automobile destruction: An example of the interaction between language and memory. Journal of Verbal Learning and Verbal Behavior, 585-589. doi:10.1016/S0022-5371(74)80011-3
Marshall, J. (1969). Law and psychology in conflict. New York: Anchor Books.
Ornstein, R. E. (1969). On the experience of time. Harmonds-worth. Middlesex. England: Penguin Publishers
Supreme Court Reporter, (1973). Rules of Evidence for United State Courts and Magistrates.
