Seth Spencer and Brock Kirwan, Department of Psychology

Introduction

Memory is an essential component for day to day living. Recognition memory in the brain has been associated with specific neural structures such as the medial temporal lobe (MTL), including the hippocampus and the adjacent MTL cortex [1]. The hippocampus is known for its ability to encode and retrieve memories through the distinct processes of pattern separation and pattern completion [2]. Pattern separation is the process of separating two similar stimuli to create distinct memories (i.e. discerning which of 2 similar toothbrushes belongs to you). Pattern completion is the retrieval of an old representation given a noisy or degraded cue (i.e. concluding a toothbrush is yours when presented with one that is not, but is similar). Many previous studies have been performed which use either a test-retest or a continuous recognition paradigm to study memory. This study took a novel approach of a study-test-second test, in order to better understand how false memories are represented in the brain. We are interested in understanding whether or not false memories rewrite the original memory.

Methodology

This project is a behavioral task in which we acquired data of neuronal activation. Methods have been developed based on previous work by Dr. C. Brock Kirwan. Ten healthy young-adults have been recruited. Participants were excluded if they have already participated in one of our previous studies, if they were MRI-incompatible, if they were left handed, non-native English speakers, have a history of a head injury, have a history of physiological, psychological, or neurological disorders, or any history of alcohol or drug abuse.

This study used a three-block, study-test-test design in a highly specific memory task. The study phase consisted of participants being exposed to 200 visual stimuli. Incidental encoding was achieved as participants were required to respond if each stimuli was indoor or outdoor. Test phases were performed concurrent with scanning at the BYU MRI Research facility. A high-resolution multiband fMRI protocol was used to assess neuronal activity. For the first test phase participants were again shown a series of visual stimuli. Each stimulus was either exactly the same as one in the study phase or highly similar to one in the study phase. Participants were asked to respond which of the two groups (old or similar) they felt the stimuli best fit in. This resulted in four categories of interest: an old called similar (pattern completion; FA, false alarm), a similar called similar (pattern separation; CR, correct rejection), a hit (old called old), and a miss (all other responses). For the second test phase, participants were presented with two visual stimuli simultaneously. One stimuli was the original (from the study phase) and the second was a similar stimuli (from the first test phase). Participants were asked to respond which stimulus they had seen during the original study phase through a forced-dichotomous confidence scale. Each stimulus required an expression of confidence for which was the correct image. Instructions were given such that one end of the button pad represented strong confidence the left image was the repeat, and the opposite end represented strong confidence the right image was the repeat. The buttons in between represented varying levels of confidence. Responses which were incorrect were labeled as false alarm (FA), and those which were correct were labeled as hit.

Results

Stimuli that had been pattern completed (false memories) from the first test phase were of particular interest. Our analysis of test one FA items concluded that performance during the second test between false memories correctly identified (FA-hit), and false memories incorrectly identified (FA-FA) was essentially at chance. Analysis also showed that confidence levels were indistinguishable in these areas.

Analysis of functional data produced from the high resolution multiband fMRI scans revealed significant differences in the left inferior frontal and supramarginal gyri. Differential activation in these areas during the first test FA responses were predictive of correction (FA-Hit) or of continued pattern completion (FA-FA).

Conclusion

The results so far present some exciting findings that implicate the neural regions of the left inferior frontal and supramarginal gyri as resistant to pattern completion. However, a greater number of participants will need to be run to evaluate the extent of how memories are re-written (FA-FA) or persist (FA-Hit). It is planned for more participants to be run through this study to clarify the results.

Discussion

A greater understanding of false memory will enable other researchers who are working to better understand and improve the conditions of those with poor memory, dementia, or other conditions. Indeed it may be that some memories are still available, and only need activation or inhibition of specific neural areas to be recalled. Knowing how to activate these areas would be very valuable to maintaining and preserving quality of life for those who have memory struggles.

Sources

1. Squire, L. R., Stark, C. E., & Clark, R. E. (2004). The medial temporal lobe. Annual Review of Neuroscience, 27, 279-306. doi: 10.1146/ annurev.neuro.27.070203.144130 2.

2. Mcclelland, J. L., Mcnaughton, B. L., & Oreilly, R. C. (1995). Why There Are Complementary Learning-Systems in the Hippocampus and Neocortex – Insights from the Successes and Failures of Connectionist Models of Learning and Memory. Psychological Review, 102(3), 419-457. doi: 10.1037/0033-295x.102.3.419