A Model of Representational Spaces in Human Cortex

Abstract

Current models of the functional architecture of human cortex emphasize areas that capture coarse-scale features of cortical topography but provide no account for population responses that encode information in fine-scale patterns of activity. Here, we present a linear model of shared representational spaces in human cortex that captures fine-scale distinctions among population responses with response-tuning basis functions that are common across brains and models cortical patterns of neural responses with individual-specific topographic basis functions. We derive a common model space for the whole cortex using a new algorithm, searchlight hyperalignment, and complex, dynamic stimuli that provide a broad sampling of visual, auditory, and social percepts. The model aligns representations across brains in occipital, temporal, parietal, and prefrontal cortices, as shown by between-subject multivariate pattern classification and intersubject correlation of representational geometry, indicating that structural principles for shared neural representations apply across widely divergent domains of information. The model provides a rigorous account for individual variability of well-known coarse-scale topographies, such as retinotopy and category selectivity, and goes further to account for fine-scale patterns that are multiplexed with coarse-scale topographies and carry finer distinctions.

Building a Science of Individual Differences from fMRI

“…The selection of voxels/vertices that are fed into the hyperalignment algorithm is the sole spatial constraint; data from each subject is eventually projected into a common space that can violate topology. The original implementation of the technique relied on pre-selecting a Region-Of-Interest (ROI) [34]; a more recent whole-brain version of hyperalignment uses a searchlight centered on each vertex, then combines the resulting transformation matrices into a large, whole-brain transformation matrix that can be used to project the cortical data from each individual subject to a common representational space [35]. Ongoing work is extending the hyperalignment algorithm to make use of resting-state data instead of movie data, in an effort to maximize applicability to extant datasets.…”

From connectome to cognition: The search for mechanism in human functional brain networks

“…An associated goal is to probe relationships between FC and behavioral measures to capture individual differences in network processing and how these relate to task performance (Schultz & Cole, 2016). Such individual difference approaches might capitalize on recently developed MVPA algorithms that enable “hyperalignment” of representational spaces across subjects (Guntupalli et al, 2016). Relationships between FC and behavior can help verify that network changes across task conditions arise from functionally relevant rather than artifactual sources.…”

Data-driven approaches in the investigation of social perception

“…Moreover, anatomy-based alignment does not capture the idiosyncratic individual variability of coarse topographic features, such as the location, size and conformation of the borders of functional areas such as retinotopically organized early visual areas, motion-sensitive MT or category-selective areas in ventral temporal cortex. A common model of the functional architecture that captures these features-fine-scale patterns of activity and individual variability of coarse-scale features-has been developed using a new algorithm, hyperalignment, and achieves broad general validity by estimating model parameters based on responses to a complex, dynamic, naturalistic stimulus, such as a fulllength movie [47,51,52]. The elements of this model are a common, high-dimensional representational space and individual transformation matrices that project data from idiosyncratic, individual anatomic spaces into the common model space.…”