Jordan Clark and Dr. Brock Kirwan, Psychology Department
The goal of this project was to increase our understanding of how human memory works. Specifically, we wanted to investigate what happens in the brain when we make memory mistakes, and to see if there are regions of the brain whose relative activation levels could be predictive as to whether a memory mistake will be perpetuated or corrected. In response to the first part of that research question, we hypothesized that when a memory mistake is made, the memory trace from the mistake is encoded separately from the original memory, rather than overwriting the original memory. For the second part of the research question, we predicted that there would indeed be differences in brain activation levels when memory mistakes are perpetuated as compared to when the mistake is corrected, particularly within the temporal lobe cortex. Our project was carried out by utilizing fMRI techniques at the BYU MRI Facility.
To study these research questions, we recruited 35 participants (20 male, 15 female, mean age = 23.1), each of whom completed our behavioral research task. The task consisted of three phases (Study, Test 1, and Test 2), all of which were performed while the participant was inside the MRI scanner. During the Study phase of the task, participants were shown a series of 200 images of common objects, and they were instructed to respond whether the object in each image would typically be found indoors or outdoors. The accuracy of the indoor/outdoor responses given during the Study phase was irrelevant to the results of the study; this was merely used to ensure that participants remained alert and focused on the stimuli being presented. Test 1 was again a series of 200 images of the same objects as in the Study phase, however, half of the images had been altered slightly (e.g. the color of the object changed, the angle at which the object was oriented changed, etc.) and half of the images remained identical to those presented in the Study phase. With each image in this phase, participants were instructed to respond whether the image was exactly the same as an image they had seen during the Study phase, or if the image was similar to, but not exactly the same as, an image that had been presented previously. Test 2 was a second memory task for the participant, but during this phase, both the original image presented during the Study phase as well as the similar-but-slightly-altered image were presented simultaneously to the participant and they were instructed to choose which of the two images was the original one presented to them during the Study phase.
The behavioral data (the accuracy of the participants’ responses during Test 1 and Test 2) show that when participants correctly identified an image during Test 1, on average they were much more likely to also correctly identify that image during Test 2. This make sense considering that when we access memories, the neural pathways to those memories are strengthened, thus making it more likely for us to continue correctly identifying the information contained within that memory trace. When a mistake was made during Test 1, however, participants were equally likely to correct that mistake during Test 2 as they were to perpetuate the mistake and again incorrectly identify the image when it was presented during Test 2. This answers our first research question by suggesting that at some times, a memory mistake will overwrite the original memory, thus causing perpetuation of the mistake because the original memory cannot be accessed. At other times, a memory mistake will coexist with the original memory, which allows for access to the original memory and thus the ability to correct ourselves after making a mistake. Why both of these situations occur in the brain, and what causes one to occur rather than the other at any given time, remains unknown.
The functional data (the regions of brain activity obtained by the MRI scanner) give us the answer to our second research question. In cases where a memory mistake was made during Test 1, five distinct regions of the brain were found to have significantly different levels of activation relative to a baseline that may be predictive of Test 2 performance. The Right Superior Frontal Sulcus and the Right Inferior Frontal Gyrus (see Fig. 1, images 1 and 2) are the first two of these regions. When these two regions showed higher levels of activation during the Study phase, and a mistake was made during Test 1, the participant was more likely to be able to correct their mistake during Test 2. The other three regions that showed significantly different levels of activation were the Right Temporal Pole, the Right Anterior Insula, and the Right Inferior Parietal Lobule (see Fig. 1, images 3-5). When these three regions showed higher levels of activation during Test 1, and a mistake was made during Test 1, the participant was more likely to perpetuate their mistake during Test 2.
Currently, the significance of the results of this study are similar to many other findings in the neuroscience field in that the practicality of the findings is yet to be seen. We do not yet have the technology to make widespread, public use of much of the information obtained from neuroscience research. For now, this is knowledge simply for the sake of knowing it. However, it is important to continue to explore the brain and learn how it works because eventually there will be practical uses found for these various pieces of knowledge that have already been discovered.
The results from this study were presented at the Mary Lou Fulton Poster Conference on the campus of BYU in April 2017. We intend to submit the findings to peer-reviewed journals for publication, though we are still in the process of writing the findings. We hope the manuscript will be ready for submission to journals within a couple of months.