Unique and persistent individual patterns of brain activity across different memory retrieval tasks

Abstract

Fourteen subjects were scanned in two fMRI sessions separated by several months. During each session, subjects performed an episodic retrieval task, a semantic retrieval task, and a working memory task. We found that 1) despite extensive intersubject variability in the pattern of activity across the whole brain, individual activity patterns were stable over time, 2) activity patterns of the same individual performing different tasks were more similar than activity patterns of different individuals performing the same task, and 3) that individual differences in decision criterion on a recognition test predicted the degree of similarity between any two individuals' patterns of brain activity, but individual differences in memory accuracy or similarity in structural anatomy did not. These results imply that the exclusive use of group maps may be ineffective in profiling the pattern of activations for a given task. This may be particularly true for a task like episodic retrieval, which is relatively strategic and can involve widely-distributed specialized processes that are peripheral to the actual retrieval of stored information. Further, these processes may be differentially engaged depending on individual differences in cognitive processing and/or physiology.

Optimizing preprocessing and analysis pipelines for single‐subject fMRI. I. Standard temporal motion and physiological noise correction methods

“…This increased heterogeneity of between-subject reproducibility is expected, given that individual subject brain responses are inherently variable, potentially more so for young, healthy participants [McIntosh et al, 2008]. Furthermore, our results indicate that stable group responses are at least partly a result of using fixed preprocessing pipelines, and that inter-subject differences, along with subject-dependent pipeline optimization, need to be more carefully considered than is usual in the current literature [Miller et al, 2009]. Individual subject optimization increases between-subject signal variance relative to the fixed-pipeline set, particularly for nominally task-activated regions with higher mean Zscore.…”

Encoding and recognition after traumatic brain injury: Neuropsychological and functional magnetic resonance imaging findings

“…Each of the following paragraphs reports the results of the between-group contrasts. Since there was variation even among individuals in the same group, initial consideration of the data from the group comparison was conducted using a threshold of p = 0.01 uncorrected, with a cluster size > 10 voxels, (see Miller, et al, 2009 for a discussion of individual differences in fMRI). More restrictive thresholding ( p < 0.001, uncorrected, cluster size > 10 voxels) was then implemented.…”

Neural Prediction of Communication-Relevant Outcomes

“…Neuroimaging data can have high test-retest reliability (Miller et al, 2009), depending on a number of factors (for a review, see : Berkman, Cunningham, & Lieberman, in press). …”