AI-Driven Neural Surrogates for In Silico Design of Cognitive-Affective Neuromodulation Targets
Abstract
In neuropsychiatry, the primary goal is often not only to decode brain activity but to change it, for example to lessen a negative affective bias or an overly salient memory. Motivated by control theory, we develop an AI-driven neural-surrogate framework that proposes candidate representational changes and tests their predicted perceptual effects from snapshots of stimulus-evoked fMRI activity, without physical stimulation. The framework combines fMRI decoding, deep generative modeling, and constrained latent-space steering. Valence and memorability are used only as worked examples. Using more than 36,000 image-fMRI observations from four deeply sampled Natural Scenes Dataset participants, subject-specific models recovered coarse generative structure from visually responsive cortex (two-way identification, 0.79-0.88; chance, 0.5). Graded perturbations were reconstructed as images and evaluated with automated scorers and human ratings from 7,200 trials by 18 participants. In the primary VDVAE model, valence shifted from -0.61 to +1.03 SD and memorability from -1.34 to +1.45 SD; a later Versatile Diffusion refinement reduced or altered these effects. Across five perturbation levels, human valence ratings moved in the predicted direction under the linear time-correction model (mean slope, 0.038 SD per unit of alpha; 95 percent CI, 0.003-0.074; positive in 16 of 18 participants). Perceived memorability did not change reliably. Baseline agreement with the automated assessor was suggestive for valence (r = 0.30) and weak for memorability (r = 0.10). Extreme perturbations drifted from the original stimulus, so intended change must be weighed against loss of fidelity. These findings provide a falsifiable upstream method for designing and behaviorally testing candidate representational targets for future neuromodulation in psychiatry, while marking the limits of the present static approximation.
Related
- High-intensity fear engrams extend beyond threat processing to drive cognitive and affective dysfunction
- Individualized AI-driven Neuromodulation Selectively Enhances and Reorganizes Human Tongue Sensorimotor Networks.
- Cognitive load and prefrontal neuromodulation shape exploration, risk-taking, and information seeking in experience-based decision making
- Nanomaterial-genetics enables near-infrared-driven genetically targeted neuromodulation
- CB2 activation during epileptogenesis recruits immunomodulatory monocytes and mitigates cognitive, affective and seizure outcomes
BibTeX
@misc{rothermel2026aidriven,
title = {{AI-Driven Neural Surrogates for In Silico Design of Cognitive-Affective Neuromodulation Targets}},
author = {Marco Rothermel and Madleen Stenger and Soroush Daftarian and Svenja Jule Francke and Bita Shariatpanahi and José C. García Alanis and Mohammad-Ali Nikouei Mahani and Stefan G. Hofmann and Tim Hahn and Hamidreza Jamalabadi},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.27729},
eprint = {2609.27729},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27729},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - AI-Driven Neural Surrogates for In Silico Design of Cognitive-Affective Neuromodulation Targets
AU - Marco Rothermel
AU - Madleen Stenger
AU - Soroush Daftarian
AU - Svenja Jule Francke
AU - Bita Shariatpanahi
AU - José C. García Alanis
AU - Mohammad-Ali Nikouei Mahani
AU - Stefan G. Hofmann
AU - Tim Hahn
AU - Hamidreza Jamalabadi
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2609.27729
UR - https://arxiv.org/abs/2609.27729
AB - In neuropsychiatry, the primary goal is often not only to decode brain activity but to change it, for example to lessen a negative affective bias or an overly salient memory. Motivated by control theory, we develop an AI-driven neural-surrogate framework that proposes candidate representational changes and tests their predicted perceptual effects from snapshots of stimulus-evoked fMRI activity, without physical stimulation. The framework combines fMRI decoding, deep generative modeling, and constrained latent-space steering. Valence and memorability are used only as worked examples. Using more than 36,000 image-fMRI observations from four deeply sampled Natural Scenes Dataset participants, subject-specific models recovered coarse generative structure from visually responsive cortex (two-way identification, 0.79-0.88; chance, 0.5). Graded perturbations were reconstructed as images and evaluated with automated scorers and human ratings from 7,200 trials by 18 participants. In the primary VDVAE model, valence shifted from -0.61 to +1.03 SD and memorability from -1.34 to +1.45 SD; a later Versatile Diffusion refinement reduced or altered these effects. Across five perturbation levels, human valence ratings moved in the predicted direction under the linear time-correction model (mean slope, 0.038 SD per unit of alpha; 95 percent CI, 0.003-0.074; positive in 16 of 18 participants). Perceived memorability did not change reliably. Baseline agreement with the automated assessor was suggestive for valence (r = 0.30) and weak for memorability (r = 0.10). Extreme perturbations drifted from the original stimulus, so intended change must be weighed against loss of fidelity. These findings provide a falsifiable upstream method for designing and behaviorally testing candidate representational targets for future neuromodulation in psychiatry, while marking the limits of the present static approximation.
ER -