fMRI Encoding Models

fMRI: Functional Magnetic Resonance Imaging

Latest papers

8 papers2 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

When Model Complexity Matters in Cross-Subject Naturalistic fMRI Prediction
Nov 2, 2026Junhao Zhao, Jacob Kang, Colin Kurniawan +2fMRI Encoding Models Neuroimaging
Department of Electrical and Computer Engineering, University of Maryland, College Park, MD, USA · Fischell Department of Bioengineering, University of Maryland, College Park, MD, USA · Department of Computer Science, University of Maryland, College Park, MD, USA +2
Alignment of visual brain encoding models with individual perceptual similarity judgments
Oct 15, 2026Daniel Chong, Haomiao Chen, Keith W Jamison +4fMRI Encoding Models Neural Encoding
Cornell University · Cornell Tech · Weill Cornell Medicine +2
MultiFlow: Vision-Driven Multimodal fMRI Encoding via Stimulus-Informed Source Flow Matching
Oct 11, 2026Bui Van Hung, Hoang-Son Vo, Mai Duc Minh Huy +2fMRI Encoding Models Neuroimaging
Chonnam National University, Gwangju, South Korea · Ho Chi Minh University of Science, Ho Chi Minh City, Vietnam · AISEED, Inc., South Korea
Cross-attention encoding models reveal dynamic spatiotemporal routing across human higher visual cortex
Sep 28, 2026Iishaan Inabathini, Margaret M. HendersonfMRI Encoding Models Cross-Attention
Carnegie Mellon University
Coarse-to-fine Hierarchical Architecture with Sequential Mamba for Brain Reconstruction
Jun 3, 2026Hoang-Son Vo, Van-Hung Bui, Minh-Huy Mai-Duc +2Visual Neuroscience Hierarchical Representation Learning
Chonnam National University, Gwangju, Republic of Korea · Vietnam National University - Ho Chi Minh City, University of Science, Vietnam · Institute for Cybersecurity and Digital Technologies, Russia
MIRAGE: Adaptive Multimodal Gating for Whole-Brain fMRI Encoding
May 28, 2026Abdulkadir Gokce, Badr AlKhamissi, Martin SchrimpfNeural Encoding fMRI Encoding Models
NeuroAI Lab, EPFL
Nonlocal operator learning for fMRI encoding and decoding tasks
May 19, 2026Andreas Kramer, Saugat Acharya, Alice Giola +1fMRI Encoding Models Neural Decoding
Department of Computer Science, Idaho State University 921 S. 8th Ave Mail Stop 8060, Pocatello, ID 83209-8023 · Department of Mathematics and Statistics, Idaho State University Physical Science Complex, 921 S. 8th Ave., Stop 8085, Pocatello, ID 83209
Independent-Component-Based Encoding Models of Brain Activity During Story Comprehension
Apr 27, 2026Kamya Hari, Taha Binhuraib, Jin Li +2fMRI Encoding Models Independent Component Analysis
School of Electrical and Computer Engineering, Georgia Institute of Technology · School of Psychological and Brain Sciences, Georgia Institute of Technology · Department of Linguistics, Stanford University

Nov 2, 2026J/K move · Enter open · S save

When Model Complexity Matters in Cross-Subject Naturalistic fMRI Prediction

Junhao Zhao, Jacob Kang, Colin Kurniawan, David Simberg+1

Department of Electrical and Computer Engineering, University of Maryland, College Park, MD, USA · Fischell Department of Bioengineering, University of Maryland, College Park, MD, USA · Department of Computer Science, University of Maryland, College Park, MD, USA+2

fMRI Encoding Models Neuroimaging Functional Magnetic Resonance Imaging

Open paper AskPDF arXiv

Increasingly complex neural models predict blood-oxygen-level-dependent (BOLD) responses to naturalistic stimuli, but the predictive value of this added complexity remains unclear. We ask when pretrained stimulus representations help and how much downstream complexity is needed, with and without recent measured BOLD history. We compare linear diagnostic references and nonlinear predictors across HCP movie fMRI (154 subjects), BOLD Moments (10 subjects), and Algonauts 2025 (4 subjects). Under a shared offline protocol, models are fitted on training subjects and evaluated on held-out subjects viewing largely familiar stimuli, without refitting. Pretrained visual features generally improve prediction over raw pixels; for linear autoregression, this advantage becomes more pronounced over longer rollouts (recursive predictions without additional measured BOLD). Measured BOLD history provides substantial gains that generally diminish {as rollouts lengthen}. Nonlinear stimulus-only predictors can improve on linear readouts, whereas linear autoregression provides a strong history-conditioned baseline. Controls examine context length, regularization, fitting procedures, and model capacity. Regional response consistency across subjects is associated with held-out prediction accuracy, while less consistent regions tend to gain more from measured history. In paired simulations with fixed linear neural dynamics, slower hemodynamic responses generally sustain high temporal prediction correlations over longer rollouts, illustrating how the observation process can contribute to predictive persistence. Together, these findings support evaluating stimulus representations and downstream complexity separately, using history-conditioned linear prediction as a diagnostic reference under matched inputs and measurement availability.