Poster No:
2498
Submission Type:
Abstract Submission
Authors:
David Gruskin1, Gaurav Patel1
Institutions:
1Columbia University, New York, NY
First Author:
Co-Author:
Introduction:
To what extent do our genes shape how we process movies? Studies involving retinotopic and task-free paradigms have established that various aspects of sensory-relevant brain function are under genetic control. However, the extent to which inherited genetic factors underlie individuals' brain responses to complex audiovisual stimuli is still unclear. Additionally, recent work suggests that individual differences in fMRI activity and connectivity patterns are partially driven by lower-level features like functional topography (Haxby et al., 2020), but the extent to which the heritability of these high-dimensional patterns is driven by heritability of stimulus processing vs. functional topography is largely unknown. Here, we address these knowledge gaps by applying multidimensional heritability analyses to movie-watching and resting state 7T fMRI data from the Human Connectome Project (HCP).
Methods:
Analyses were performed on fMRI data from the 176 subjects (including 50 monozygotic and 33 dizygotic twin pairs) who completed two hours of resting state and movie-watching runs across two days as part of the HCP Young Adult 7T release (Van Essen et al., 2013) following denoising with ICA-FIX and global signal regression. We used intersubject correlation (ISC) to measure vertex-level BOLD time course similarity between all pairs of participants and applied a multidimensional heritability analysis (Anderson et al., 2021) controlling for age, sex, and head motion to these ISC values to quantify BOLD time course heritability (h^2). To measure functional connectivity (FC) similarity across participants, we generated resting state and movie-watching FC matrices for each participant and each day of data collection by correlating time courses extracted from 400 Schaefer atlas parcels and then correlating FC profiles for each combination of 17 resting state networks (Schaefer et al., 2018). The resulting FC similarity values were then submitted to the multidimensional heritability analysis to estimate FC profile heritability. To quantify the extent to which this heritability is driven by genetic similarity in functional topography ("where" stimuli are processed) vs. stimulus processing ("how" stimuli are processed), we functionally aligned data from all subjects using both connectivity and response hyperalignment and repeated the ISC and FC heritability analyses on the aligned data. Importantly, we performed this alignment at multiple spatial resolutions to estimate point spread functions of heritable functional topographies.
Results:
Vertex-level BOLD time courses were heritable across most of the cortex (max h^2 = .12, 65% of vertices heritable at FDR-corrected P<.05), especially in auditory and visual regions, and spatial heritability patterns were consistent across days of data collection (r = .94, Pspin<.001). Similarly, movie-watching FC profiles were heritable within and between all networks (avg. h^2 = .34, all network combinations heritable at FDR-corrected P<.05), and FC profile heritability was higher for sensory networks during movie-watching compared to rest. Movie-watching FC profile heritability patterns were also consistent across days of data collection (r = .96, P<.001). Following response and connectivity hyperalignment, activity and connectivity h^2 values were reduced across vertices and networks, especially for sensory regions. Activity h^2 decreased exponentially with hyperalignment resolution, and response hyperalignment at the coarsest spatial resolution decreased activity h^2 by 48% across the cortex.
Conclusions:
Taken together, these results demonstrate the power of using ecologically rich stimuli to study the heritability of sensory processing in cortex. Specifically, these results establish that heritable brain responses to audiovisual stimuli reflect genetic control over multiple independent aspects of brain function, including both "where" vs. "how" information is processed.
Genetics:
Genetics Other 2
Perception, Attention and Motor Behavior:
Perception: Multisensory and Crossmodal 1
Keywords:
FUNCTIONAL MRI
Perception
Other - Heritability; Movie-watching
1|2Indicates the priority used for review
Provide references using author date format
x