Eye witness testimony is commonly used as evidence in court, but a body of research demonstrates that it may not be a reliable and accurate source of information. Moreover, research suggests that people are often flawed when trying to accurately recall the details of a crime scene or accident, much like a car crash video on YouTube. Witnesses may absolutely insist on certain details of a an accident, but after reviewing the same YouTube car crash video, will often realize that they are wrong.
Several studies that utilized basic research approaches, and that ultimately contributed to a better understanding of virtually unrecognized bias and prejudice judgments, are Koffka (1935), and Taylor and Fiske (1975). First, Koffka (1935) found that participants who were asked to focus on one of two lights and to express which had moved would report that the one receiving their visual attention had moved. Koffka (1935) demonstrated that because certain objects stand out and capture our attention within our visual field, those objects will consequently be more influential and causal in nature. Moreover, the results of the study suggest that witnesses may only endorse events that are present in their sensory systems. They may overlook events that fall outside of their direct experience, and witnesses therefore may be more strongly inclined to insist on the validity of the individual events they experience.
You can do an experiment on yourself to test how well you recall the details of an event. Watch car crash videos, perhaps Youtube car crash videos, and see how well you can remember the minute details of the event. Write them down after you watch the car crash video, and then go back and replay it in slow motion, and see how much you recalled correctly.
A study by Taylor and Fiske (1975) took this conceptual framework one step further by applying it to social interactions, where participants were intentionally seated and positioned to where their visual attention would be saliently focused on one individual or equally between the two people engaging in a conversation. Taylor and Fiske (1975) concluded that the observers which were focused on a single individual found that person to be more influential in the conversation, and observers visually attentive to both speakers diffused the influence of the conversation equally between both people. Noteworthy, this study explicates the same findings as Koffka (1935), but in the realm of social interactions.
Again, you can create your own experiment that implements the same principles as demonstrated by Taylor and Fiske (1975). Watch one of the many car crash videos on YouTube and focus on just one part of the screen. You can even use a piece of paper to block out the rest of the screen. Watch it, pause it, and then jot down what you can remember from the scene. Then watch the entire video, including the other half of the screen. You will likely find that what you saw when watching half of the the car crash video differs greatly from what actually occurred in the car crash video.