arXiv q-bio.NC: NeuroVolve targets individualized brain responses with generated images

arXiv q-bio.NC: NeuroVolve targets individualized brain responses with generated images

The authors of NeuroVolve present a data-driven framework for generating images intended to produce user-defined patterns of brain activation in individuals. They report that it recovers established selectivity across visual cortical regions and, with subject-specific brain encoding models, generates personalized stimuli reflecting individual differences in neural preference. The framework also targets responses across multiple regions, characterizing how those regions jointly represent visual content during naturalistic perception. In a prospective fMRI experiment, stimuli targeting the amygdala produced measured activation consistent with predictions in novel participants. The authors describe NeuroVolve as a less constrained tool for probing individual-level neural selectivity across the brain than approaches limited to features defined before experiments. The abstract does not provide sample sizes or detailed results, limiting assessment of the scope of these findings.

Publication: 53f0b431-697b-4b34-8fb1-3dfc1d5610ed

attribution — pass: The claims are attributed to the paper's authors as reported findings. The title prefix 'arXiv q-bio.NC' matches the supplied publisher metadata, although the page lists cs.CV as the primary category and q-bio.NC as a cross-list. This does not misattribute the work.

Authors:Haomiao Chen, Keith W Jamison, Daniel Chong, Yunrui Zhang, Mert R. Sabuncu, Amy Kuceyeski

summary — uncertain: The summary follows the abstract, but readers without specialist background are left without explanations of the central terms 'selectivity', 'brain encoding models', 'visual cortical regions', fMRI and amygdala. It also omits the source's point that the amygdala is a classically non-sensory region, which is why the authors present that result as generalization beyond visual cortex. Smallest repairs: briefly gloss the core terms, or add term-library tooltips. Also restore the phrase 'a classically non-sensory region' and the beyond-visual-cortex framing in the amygdala sentence.

We further demonstrate generalization beyond visual cortex: in a prospective fMRI experiment, NeuroVolve-generated stimuli targeting the amygdala, a classically non-sensory region

date — pass: evidence.originalPublished is 2025-11-29. This matches the v1 submission date, not the 1 Nov 2026 revision or the feed date.

[Submitted on 29 Nov 2025 (v1), last revised 1 Nov 2026 (this version, v2)]

support — pass: Each claim is supported by the abstract and stays framed as an author report. The 'less constrained' comparison follows from the abstract's contrast with handcrafted stimuli and its closing description of the tool. The limitation about missing sample sizes and quantitative results is accurate for the abstract. The question is open, not leading. The reading adds no unsupported benefits or forecasts.

establish NeuroVolve as a less constrained, data-driven tool for probing individual-level neural selectivity across the brain

media — pass: The selected media is the verified direct link to the original arXiv abstract page. This is the most direct way for readers to access the source.