Date of Award

Summer 8-15-2016

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



The manner in which the human brain recognizes certain stimuli as novel or familiar is a matter of ongoing investigation. The overarching goal of this dissertation is to improve our understanding of how this may be accomplished. More specifically, work contained herein focuses on a recently described "parietal memory network" (PMN; Gilmore et al., 2015) that shows opposite patterns of activity when perceiving novel or familiar stimuli: deactivating in response to novelty, and activating in response to familiarity. Critically, our understanding of this network is based on explicit memory tasks, in which subjects are deliberately instructed to learn or remember information to perform the experimental task. The aim of this dissertation is to determine if the same opposing patterns of activity are present in task conditions in which no explicit orientation to stimulus history is required (i.e., implicit memory conditions). In Chapter 1, I review evidence that links activity within the PMN to encoding and retrieval processes, and describe how the perception of novelty and familiarity may explain various observations from prior literature. In this chapter I discuss several techniques that utilize functional magnetic resonance imaging (fMRI) to measure activations within the brain. These include single experiments utilizing blood oxygen level dependent (BOLD) activity as a means of associating specific behavioral phenomena with specific neural correlates; meta-analyses of many such fMRI studies; and the use of BOLD correlations in the absence of explicit task conditions to estimate the functional network structure of the human brain. It is from the work reviewed in this chapter that this dissertation's empirical questions were derived. In Chapter 2, I discuss experimental data collected under implicit memory task conditions. This was designed to assess the degree to which activity predicted by explicit memory tasks is recapitulated under implicit conditions. Subjects observed stimuli multiple times, making simple semantic judgments during each presentation. The BOLD responses within each subject were measured for each presentation of each stimulus. PMN regions demonstrated two of three predicted patterns of activity: they deactivated relative to a resting baseline when initially processing a stimulus, and they increased in activity across multiple item presentations. Predicted above-baseline activations during final presentations were not observed. This suggests that existing hypotheses describing PMN functions should be revised in a way that suggests a more prominent role for attention in producing familiarity-related activations. In Chapter 3, the task data from Chapter 2 are compared to an individual with superior memory abilities. This individual (ND) is a "memory athlete" who has trained extensively in the use of mental imagery as a tool for rapid learning. When comparing him to the control group characterized in Chapter 2, we found no appreciable differences in neural activity in the implicit memory task. These findings are consistent with those observed in prior literature that suggest memory athletes do not possess unusual memory skills outside of the tasks they specifically train (Maguire et al., 2002; Ramon et al., 2016). In Chapter 4, resting-state functional connectivity (RSFC) MRI data are examined to estimate the functional network organization of all participants examined in Chapters 2 and 3. The PMN and several control networks were localized using this independent approach, and the activity within regions of each network was compared using data from the implicit memory task. Results suggest that the implicit memory task produces very similar activity in PMN and adjacent default mode network regions, and suggests that the task itself is not a practical means of localizing the PMN within single subjects. Chapter 5 serves as a summary of the results from Chapters 2-4. In this chapter I place key findings of the dissertation in a broader context and suggest future directions that might be taken to better understand the PMN. An updated hypothesis of PMN function is proposed to better account for possible attentional affects on network activity, and several future directions are considered.


English (en)

Chair and Committee

Kathleen B. McDermott

Committee Members

Ian G. Dobbins, Nico U. Dosenbach, Steven E. Petersen, Henry L. Roediger,


Permanent URL: