Exploring the Neural Mechanisms Underlying Pattern Separation in Preschool-Aged Children: an fMRI Paradigm

• In Event: Breaking Barriers to Better Understand Early Cognitive Development: Novel Task-Based fMRI Paradigms in Young Children

Abstract

Pattern separation is a computational process by which patterns of neuronal activation underlying similar memories are made distinct to counteract interference during retrieval (Norman & O’Reilly, 2003). This process is essential to distinguish similar memories but cannot be measured directly; however, mnemonic similarity tasks (MST) where participants discriminate similar stimuli can serve as an index (e.g., Lacy et al., 2011). During these tasks, participants encode pictures and then must discriminate between those they saw during encoding (Targets); new pictures they did not seen before (Foils); and pictures that are similar, but not the same as the previously seen pictures (Lures). Previous research in young children has shown that hippocampal structure shows dramatic changes, and these structural changes relate to age-related improvements in children’s ability to form precise memories through pattern separation. However, we have yet to establish 1) whether there are functional hippocampal changes during early childhood and 2) whether functional changes relate to pattern separation development. To address these gaps, we developed a novel, passive-viewing, block design, fMRI task in which participants encode 60 pictures outside of the scanner, passively view 30 in the scanner (10 Targets, 10 Lures, 10 Foils) in two short runs, and complete an active retrieval of 30 pictures (yes/no judgement about whether they saw the pictures during encoding) after the scan. We also used Framewise Integrated Real-time MRI Monitoring (FIRMM) software to monitor motion during the scan and provide the participants with feedback during the scan. We investigated: 1) whether anterior and posterior hippocampus show differential activation to Targets, Lures, and Foils during passive viewing and 2) whether the magnitude of the neural response relates to behavioral performance outside of the scanner. The behavioral marker calculated was the Lure Discrimination Index (LDI), an index of how well participants distinguish targets and lures. We recruited a sample of eight 3-4-year-old children who completed the out-of-scanner encoding, in-scanner passive retrieval, and out-of-scanner active retrieval. We found marginally significant differential activation for Targets relative to Foils in left posterior hippocampus (t = 2.48, pcorrected = .084312). There were no significant differences in activation for Targets relative to Lures or Lures relative to Foils. Additionally, multiple regression analyses revealed that magnitude of neural response did not predict LDI performance. This was true for all contrasts. These results may be due to a lack of power. Therefore, we plan to continue recruiting participants and update the results with a larger sample size by the time of the meeting as well as investigate other metrics beyond mean activation, such as regional heterogeneity of hippocampal functioning during the task.

Authors

• Jade Dunstan, University of Maryland - College Park
  Presenting Author

• Tracy Riggins, University of Maryland - College Park