Biology subjects

Sadhukha, S.

Publications and source records attributed to Sadhukha, S..

2 recordsLinked to original sources

bioRxiv2025

Structured Neural Variability from Repeated Naturalistic Video Watching Experiences

The brain continuously integrates perception and interpretation processes to navigate dynamic environments. It is often assumed that these processes are invariant to repeated experiences. Here we investigated how neural responses to naturalistic stimuli evolve across multiple viewings using functional magnetic resonance imaging. Twenty participants watched 24 short videos across five separate scanning sessions. Between-subject correlations (BSC) decreased progressively across all brain regions with repeated viewing, indicating increasing individual variability. Within-subject correlations (WSC) consistently exceeded BSC, demonstrating that individuals are more similar to themselves than to others. However, the structure of within-subject variability differed fundamentally between sensory and interpretive regions. Early visual cortex exhibited strong repetition effects, with brain activity becoming increasingly dissimilar with each successive viewing, accompanied by reduced coupling to low-level visual features and decreased anti-correlation with default mode regions. In contrast, ventromedial prefrontal cortex (vmPFC) showed temporal proximity effects, where sessions closer in time exhibited more similar activity patterns regardless of viewing number. This temporal structure in vmPFC aligned with fluctuations in participants video preferences across sessions. These findings suggest that repeated experiences are never processed identically: sensory systems adapt efficiently to familiar input while interpretive systems continuously reshape meaning based on current internal states.

neuroscience↗

bioRxiv2020

Custom-molded headcases have limited efficacy in reducing head motion for fMRI

Effectively minimizing head motion continues to be a challenge for the collection of functional magnetic resonance imaging (fMRI) data. The use of individual-specific custom molded headcases is a promising solution to this issue, but there has been limited work to date. In the present work, we examine the efficacy of headcases in a larger group of participants engaged in naturalistic scanning paradigms including: long movie-watching scans ([~]20-45min) and a recall task that involved talking aloud inside the MRI. Unlike previous work, we find that headcases do not reliably reduce motion during movie viewing compared to alternative methods such as foam pillows or foam pillows plus medical tape. Surprisingly, we also find that motion is worse when participants talk aloud while wearing headcases. These differences appear to be driven by large, brief rotations of the head as well as translations in the z-plane as participants speak. Smaller, constant head movements appear equivalent with or without headcases. The largest reductions in head motion are observable when participants were situated with both foam pillows and medical tape. Altogether, this work suggests that in a healthy adult population, custom-molded headcases may provide limited efficacy in reducing head motion beyond existing tools available to researchers. We hope this work can help improve the quality of custom headcases, motivate the investigation of additional solutions, and provide additional information about head motion in naturalistic contexts.

neuroscience↗