Jacobian-Based Interpretation of Nonlinear Neural Encoding Model

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Gao, Xiaohui, Yang, Haoran, Cheng, Yue, Zuo, Mengfei, Liu, Yiheng, Li, Peiyang, Hu, Xintao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912790668640256
author Gao, Xiaohui
Yang, Haoran
Cheng, Yue
Zuo, Mengfei
Liu, Yiheng
Li, Peiyang
Hu, Xintao
author_facet Gao, Xiaohui
Yang, Haoran
Cheng, Yue
Zuo, Mengfei
Liu, Yiheng
Li, Peiyang
Hu, Xintao
contents In recent years, the alignment between artificial neural network (ANN) embeddings and blood oxygenation level dependent (BOLD) responses in functional magnetic resonance imaging (fMRI) via neural encoding models has significantly advanced research on neural representation mechanisms and interpretability in the brain. However, these approaches remain limited in characterizing the brain's inherently nonlinear response properties. To address this, we propose the Jacobian-based Nonlinearity Evaluation (JNE), an interpretability metric for nonlinear neural encoding models. JNE quantifies nonlinearity by statistically measuring the dispersion of local linear mappings (Jacobians) from model representations to predicted BOLD responses, thereby approximating the nonlinearity of BOLD signals. Centered on proposing JNE as a novel interpretability metric, we validated its effectiveness through controlled simulation experiments on various activation functions and network architectures, and further verified it on real fMRI data, demonstrating a hierarchical progression of nonlinear characteristics from primary to higher-order visual cortices, consistent with established cortical organization. We further extended JNE with Sample-Specificity (JNE-SS), revealing stimulus-selective nonlinear response patterns in functionally specialized brain regions. As the first interpretability metric for quantifying nonlinear responses, JNE provides new insights into brain information processing. Code available at https://github.com/Gaitxh/JNE.
format Preprint
id arxiv_https___arxiv_org_abs_2510_13688
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Jacobian-Based Interpretation of Nonlinear Neural Encoding Model
Gao, Xiaohui
Yang, Haoran
Cheng, Yue
Zuo, Mengfei
Liu, Yiheng
Li, Peiyang
Hu, Xintao
Neurons and Cognition
In recent years, the alignment between artificial neural network (ANN) embeddings and blood oxygenation level dependent (BOLD) responses in functional magnetic resonance imaging (fMRI) via neural encoding models has significantly advanced research on neural representation mechanisms and interpretability in the brain. However, these approaches remain limited in characterizing the brain's inherently nonlinear response properties. To address this, we propose the Jacobian-based Nonlinearity Evaluation (JNE), an interpretability metric for nonlinear neural encoding models. JNE quantifies nonlinearity by statistically measuring the dispersion of local linear mappings (Jacobians) from model representations to predicted BOLD responses, thereby approximating the nonlinearity of BOLD signals. Centered on proposing JNE as a novel interpretability metric, we validated its effectiveness through controlled simulation experiments on various activation functions and network architectures, and further verified it on real fMRI data, demonstrating a hierarchical progression of nonlinear characteristics from primary to higher-order visual cortices, consistent with established cortical organization. We further extended JNE with Sample-Specificity (JNE-SS), revealing stimulus-selective nonlinear response patterns in functionally specialized brain regions. As the first interpretability metric for quantifying nonlinear responses, JNE provides new insights into brain information processing. Code available at https://github.com/Gaitxh/JNE.
title Jacobian-Based Interpretation of Nonlinear Neural Encoding Model
topic Neurons and Cognition
url https://arxiv.org/abs/2510.13688