Manning JR, Zhu X, Willke T, Ranganath R, Stachenfeld K, Hasson U, Blei DM, Norman KA (2017). A probabilistic approach to discovering dynamic full-brain functional connectivity patterns...

View the Full Publication Here.

Manning JR, Zhu X, Willke T, Ranganath R, Stachenfeld K, Hasson U, Blei DM, Norman KA (2017). A probabilistic approach to discovering dynamic full-brain functional connectivity patterns. bioRxiv: 106690.

Abstract

Recent work indicates that the covariance structure of functional magnetic resonance imaging (fMRI) data – commonly described as functional connectivity – can change as a function of the participant’s cognitive state (for review see [32]). Here we present a technique, termed hierarchical topographic factor analysis (HTFA), for efficiently discovering full-brain networks in large multi-subject neuroimaging datasets. HTFA approximates each subject’s network by first re-representing each brain image in terms of the activations of a set of localized nodes, and then computing the covariance of the activation time series of these nodes. The number of nodes, along with their locations, sizes, and activations (over time) are learned from the data. Because the number of nodes is typically substantially smaller than the number of fMRI voxels, HTFA can be orders of magnitude more efficient than traditional voxel-based functional connectivity approaches. In one case study, we show that HTFA recovers the known connectivity patterns underlying a synthetic dataset. In a second case study, we illustrate how HTFA may be used to discover dynamic full-brain activity and connectivity patterns in real fMRI data, collected as participants listened to a story. In a third case study, we carried out a similar series of analyses on fMRI data collected as participants viewed an episode of a television show. In these latter case studies, we found that both the HTFA-derived activity and connectivity patterns may be used to reliably decode which moments in the story or show the participants were experiencing. Further, we found that these two classes of patterns contained partially non-overlapping information, such that classifiers trained on combinations of activity-based and dynamic connectivitybased features performed better than classifiers trained on activity or connectivity patterns alone.

Manning JR, Hulbert JC, Williams J, Piloto L, Sahakyan L, Norman KA (2016). A neural signature of contextually mediated intentional forgetting. Psychonomic Bulletin & Review.

View the Full Publication Here.

Abstract

The mental context in which we experience an event plays a fundamental role in how we organize our memories of an event (e.g. in relation to other events) and, in turn, how we retrieve those memories later. Because we use contextual representations to retrieve information pertaining to our past, processes that alter our representations of context can enhance or diminish our capacity to retrieve particular memories. We designed a functional magnetic resonance imaging (fMRI) experiment to test the hypothesis that people can intentionally forget previously experienced events by changing their mental representations of contextual information associated with those events. We had human participants study two lists of words, manipulating whether they were told to forget (or remember) the first list prior to studying the second list. We used pattern classifiers to track neural patterns that reflected contextual information associated with the first list and found that, consistent with the notion of contextual change, the activation of the first-list contextual representation was lower following a forget instruction than a remember instruction. Further, the magnitude of this neural signature of contextual change was negatively correlated with participants’ abilities to later recall items from the first list.