ePoster

EFFECT OF EARLY VISUAL EXPERIENCE ON SPONTANEOUS NEURAL VARIABILITY ACROSS THE VISUAL HIERARCHY

Gargi Majumdarand 5 co-authors

University of Hamburg

FENS Forum 2026 (2026)
Barcelona, Spain
Board PS06-09PM-343

Presentation

Date TBA

Board: PS06-09PM-343

Poster preview

EFFECT OF EARLY VISUAL EXPERIENCE ON SPONTANEOUS NEURAL VARIABILITY ACROSS THE VISUAL HIERARCHY poster preview

Event Information

Poster Board

PS06-09PM-343

Abstract

Perception relies on intrinsic cortical priors that help inference under uncertainty. Interactions between spontaneous activity and sensory input during early development are thought to shape these intrinsic dynamics. Visual deprivation from birth provides a critical opportunity to test how early input shapes neural priors.

We examined intrinsic neural variability and neural timescales in individuals born with dense bilateral congenital cataracts (CC) who later underwent sight-restoration surgery. Resting-state fMRI data were acquired during eyes-open (EO) and eyes-closed (EC) conditions from presurgery (CC-Pre) and postsurgery (CC-Post) CC individuals, along with age-matched sighted controls (SC) (Fig.1a). Variance-based BOLD fluctuations were quantified using the standard deviation of the BOLD signal across the visual hierarchy (V1–V4). A hierarchical Bayesian model integrated data across scanners, cohorts, regions, and conditions to obtain robust group-level estimates (Fig.1b).

All groups exhibited a systematic increase in BOLD variability from V1 to V4 (Fig.1c,d). However, CC-Pre individuals showed significantly higher variability across the visual hierarchy compared to sighted controls. Crucially, restoration of visual input (CC-Post individuals) was associated with reduced variability after surgery, but the variability in CC- Post remained higher than in SC individuals across the hierarchy, especially in EO condition.

To assess whether altered variability was accompanied by changes in neural timescales, BOLD autocorrelation decays were computed (Fig.1e,f). While SC individuals showed slower autocorrelation decay from V1 to V4, CC individuals exhibited faster decay, suggesting shorter intrinsic timescales. Together, these findings demonstrate the crucial role of early visual input in shaping the spatiotemporal organization of intrinsic cortical dynamics.


(a) Bar plot showing age distributions for the three participant groups: CC-Pre (presurgery CC), CC-Post (CC postsurgery), and SC (normally sighted control) individuals for 1.5T and 3T datasets. (b) Graphical plate diagram of the Bayesian hierarchical regression model, with neural variability (SDBOLD) as the dependent variable modeled using a Gamma likelihood. Regression coefficients (β) are assigned Student-t priors, with subject-level random slopes ( γ 𝛾 ) and subject- and scanner-level random intercepts (σ). (c-d) ROI-level posterior estimates of neural variability for the (c) eyes-open (EO) and (d) eyes-closed (EC) resting-state conditions across visual areas V1–V4, for CC-Pre, CC-Post, and SC groups. Neural variability increases from V1 to V4 for all groups in both conditions. However, CC-Pre individuals exhibit significantly higher SDBOLD across all ROIs, followed by CC-Post and SC individuals. These differences were more pronounced for EO compared to EC condition. (e) Neural timescale analysis for EO and EC conditions comparing CC-Post and SC groups; CC-Post individuals showed faster autocorrelation decay, indicating shorter intrinsic timescales across all ROIs in both the EO and EC conditions.

Recommended posters

Cookies

We use essential cookies to run the site. Analytics cookies are optional and help us improve World Wide. Learn more.