The Relationship Between Language, Memory And Thought

Introduction

Language and memory have historically been studied apart as unique cognitive abilities. However, new evidence suggests that these two cognitive abilities are more intertwined than believed. The research done is based upon Barlett’s schema theory, which suggests that memories can be influenced by the previous knowledge of a person. This implies that we use our previous knowledge or schemas to guess at things we are unsure of such as seeing a small, blue flying object in the air and guessing that it is a bluebird based on previous knowledge. The interpretation (although uncheckable) would be recorded in memory as a bluebird. Loftus and Palmer (1974) experimented on the two abilities of language and memory. The researchers proposed that a leading question, by its form or content, suggests to the participant what answer is desired or leads him/her to the desired answer. In their classic study, the researchers showed participants films of traffic accidents and then presented them with questions about the accidents. The participants were specifically interrogated about the speed of the vehicles with the switching of a verb which triggered schemas about the speed of the vehicles. Verbs such as smashed, hit, collided, bumped, and contacted were used in the original study. The verbs involved different specifications of a moment which shows how smashed ended up having a mean speed estimate 9.0 miles per hour greater than that of the verb contacted.

The aim of this study is to determine whether the answers to a complex event such as a car crash can be manipulated based on how the questions are phrased more specifically on which verb is used to describe the complex event. It is interesting to investigate as it supports the distortion of eyewitness testimony accounts as the account would become distorted by the suggestive cues and schemas provided in the question.

The null hypothesis is that the way a question is worded and the language used will not have an effect on the memory or answer of the participant who is being questioned.

The research hypothesis is that if the questions phrased use a highly suggestive verb such as smashed will produce significantly higher speeds for one singular impact of a set speed. This will be accomplished by changing the verb used to either hit or smashed and asking the participants what they estimate the speed to be after showing a singular collision.

The independent variable is the verb used in the question; the dependent variable is the estimated average speed recalled by the participant.

Exploration

This experiment used a between-groups design where two sets of participants were used, one for each condition. This design will minimize the transfer of information and will ensure that participants do not notice the switch in the verb to sway their answers. The sample was made up of high school students in grades from Junior to Senior between the ages of 16 to 18. All participants were not International Baccalaureate Psychology students nor students who had already participated in a study of the same aim or design. In this case, it also meant that the participants were familiar with driving and students who were older who would understand the importance of the study and would be more motivated to give a genuine answer. Both groups were given a questionnaire that had multiple questions and the target question randomized in the order and each group had the verb switched from either hit or smashed.

In order to create a list of questions, we went to multiple law sites regarding car crashes and consulted witness testimony questions to focus on the car crash. In addition to these witness testimony questions, we added in the question targeting the change in verbs by asking “How fast was the white car going when it hit/smashed the brown car.” Questions were chosen to be objective and contained language which was friendly to all members of our study.

A lab experiment was conducted, in the natural environment of an empty classroom to avoid other conflicts and distractions to direct focus to the task at hand. Participants were randomly allocated a letter from a list that was randomized for each condition and confidentiality. We as the experimenters gave an introduction from a pre-written script and a final warning about graphic content. The consent forms were collected due to the graphic nature of the experiment involving a car crash. In each condition, participants were shown a video of a car crash and then given an “answer sheet” to fill in which included the target question and the same filler questions. The questionnaire was administered to test how accurately the participants could recall the details of a complex event and if their answers were affected by the verb used in the estimating speed question. There was no talking during the passing out of the answer sheet as well as until all the answer sheets were collected.

Analysis

In order to provide a richer interpretation of our data, I have calculated the averages of both conditions. As can be seen from the data above, it appears that on average the group that used the verb smashed in the question estimating the speed yielded a higher average speed by 9.50 miles per hour. Both groups had the same filler questions and only the changing of the verb was targeted as the independent variable.

Discussion

The results of this experiment indicate that the form of a question (in this case, changes in a single word) can markedly and systematically affect a witness’s answer to that question. The actual speed of the vehicles controlled little variance in the final report. The interpretation is that the question form (specifically one word) causes a change in a subject’s memory of the accident. The verb may have changed the memory through suggestiveness and the participant “sees” the accident as being more severe or high impactful than it was. There are two kinds of information that contribute memories: information gained during an event and information gained after. Small changes such as a verb in a question after the fact can lead to distortion of memories which is why human memory is susceptible to change and decay. As can be seen, by the results stated above, we were able to support the findings of Loftus and Palmer (1974). They found a smaller increase in the estimated speed over all five conditions but specifically a 6.5 miles per hour increase from hit to smashed. This could be because our participants were younger and may have not identified with some of the words and linked them to their own driving experience or other destructive concepts. It could also have to do with the memory recall of younger students was better as well as only one film being shown rather than a series of films from the original study.

The results are supported by the theory proposed by the original researchers on how the leading question, by its form or content, suggests to the participant what answer is desired or leads him/her to the desired answer. Which in our case, is more specifically the schema that goes along with the verb hit or smashed that triggers a higher impact (and faster speed) collision rather than a lower impact (and lower speed) collision.

One of the strengths of our study was that we did a laboratory study that allows us to control many aspects of our environment and the experience of the participant, which reduces variables and increases the validity of the study. Along with these controls, we were also able to use a script and standardized method for the experiment which guaranteed that there was no variation between the two groups.

A limitation of the research is that it lacked mundane realism. Participants viewed a video clip rather than being present at the real life accident which warrants more emotional impact as witnessing such a traumatic event firsthand. The emotional impact of a real life collision could be enough to cause the participant to pay less attention to the details and be less motivated to be accurate in their judgments. A study conducted by Yuille and Cutshall (1986) conflicts the finding of the study done by Loftus and Palmer (1974). They found that misleading information did not alter the memory of people who had witnessed a real armed robbery. This implies that misleading information such as the filler questions asking about irrelevant details to the aim of the study and would have had a greater influence in the lab.

Another limitation is the use of students are participants. Although all the students participating were drivers, they, especially highschool students, are not a good representation of the general populous. They are most likely less experienced drivers and therefore have less confidence in their ability to estimate speeds. This may have influenced them to be more swayed by the verb in question. Using a specific age group such as students could mean that a person of a different age group may be more or less alert to the crash and therefore give a different answer or pay attention to more details.