Testing hypothesized neural substrates with lesions

Correlative neuroimaging generates anatomical hypotheses causally testable by lesion studies. Regionally weighted lesion volume-deficit regression - comparing slopes of behavioural dependence on damage inside vs outside a hypothesized neural substrate - and compartmental volume comparisons inheriting the same logic have been widely employed for this task, but without analytical examination or empirical validation of the underlying methodology. Here we show analytically that this approach does not provide a valid test of whether the hypothesized compartment identifies the true substrate. To illustrate the empirical consequences, we conduct semi-synthetic simulations using 4119 ischaemic lesions, modelling behavioural deficits dependent on a global lesion attribute, on widely investigated distributed cortical networks and white matter tracts, or on a right frontal substrate of fluid intelligence previously identified by multivariate methods. We apply regionally weighted regression with each hypothesized compartment as the target over 5000 iterations spanning multiple realistic noise levels. Where the hypothesized target is not the true substrate, we find spuriously significant dependence despite definitionally absent anatomical specificity; adjustment for lesion volume attenuates or exacerbates the observed bias dependent on the chosen target. Conversely, where the target is the true substrate, lesion volume correction universally reduces already poor sensitivity. Our analysis potentially explains common discrepancies across localisation studies, and highlights the necessity of multivariate approaches that explicitly model lesion and substrate anatomy at finely resolved spatial scales.

Authors

Institutions

Publication Details

Journal
bioRxiv (Cold Spring Harbor Laboratory)
Published
2026-11-19
DOI
https://doi.org/10.64898/2026.11.18.767704
Primary Topic
Functional Brain Connectivity Studies
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Testing hypothesized neural substrates with lesions

Patrik Bey, Parashkev Nachev, Lisa Cipolotti, Edgar Chan et al.
bioRxiv (Cold Spring Harbor Laboratory)
Functional Brain Connectivity Studies
preprint

Testing hypothesized neural substrates with lesions

Patrik Bey, Parashkev Nachev, Lisa Cipolotti, Edgar Chan, James K Ruffle, Amy Nelson
preprint en

Abstract

Correlative neuroimaging generates anatomical hypotheses causally testable by lesion studies. Regionally weighted lesion volume-deficit regression - comparing slopes of behavioural dependence on damage inside vs outside a hypothesized neural substrate - and compartmental volume comparisons inheriting the same logic have been widely employed for this task, but without analytical examination or empirical validation of the underlying methodology. Here we show analytically that this approach does not provide a valid test of whether the hypothesized compartment identifies the true substrate. To illustrate the empirical consequences, we conduct semi-synthetic simulations using 4119 ischaemic lesions, modelling behavioural deficits dependent on a global lesion attribute, on widely investigated distributed cortical networks and white matter tracts, or on a right frontal substrate of fluid intelligence previously identified by multivariate methods. We apply regionally weighted regression with each hypothesized compartment as the target over 5000 iterations spanning multiple realistic noise levels. Where the hypothesized target is not the true substrate, we find spuriously significant dependence despite definitionally absent anatomical specificity; adjustment for lesion volume attenuates or exacerbates the observed bias dependent on the chosen target. Conversely, where the target is the true substrate, lesion volume correction universally reduces already poor sensitivity. Our analysis potentially explains common discrepancies across localisation studies, and highlights the necessity of multivariate approaches that explicitly model lesion and substrate anatomy at finely resolved spatial scales.

bioRxiv (Cold Spring Harbor Laboratory)
University College London (GB)
Functional Brain Connectivity Studies
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon