Back

Convergent and selective representations of pain, appetitive processes, aversive processes, and cognitive control in the insula

Kwon, M.; Bo, K.; Botvinik-Nezer, R.; Kragel, P. A.; Van Oudenhove, L.; Wager, T. D.; The Affective Neuroimaging Consortium,

2025-02-19 neuroscience
10.1101/2025.02.18.638889 bioRxiv
Show abstract

Regions that respond to multiple types of information ("convergence zones") are crucial for the brain to generate coherent experiences and behaviors. The insular cortex, known for its functional diversity, has been hypothesized to be a key convergence hub, yet empirical evidence identifying how and where convergence occurs is incomplete. To address this gap, we analyzed functional convergence across four key task domains--somatic pain, non-somatic appetitive processes, aversive processes, and cognitive control--in a large-scale Bayesian mega-analysis of fMRI data (N=540, systematically sampled from 36 studies). Bayes Factor analyses identified both convergent zones responding to multiple domains and selective zones responding to single domains. Results revealed a hierarchical convergence architecture, with a multi-domain convergence zone in bilateral dorsal anterior insula surrounded by zones showing progressively increasing convergence. Functional decoding and coactivation analyses further support the insulas role as a convergence hub, while cytoarchitectonic and neurotransmitter profiling characterizes the potential neuroanatomical underpinnings subserving convergent and domain-selective function. Overall, our results demonstrate a structured functional topography in the insula that bridges specialized and convergent processing, providing a potential neural basis for how diverse information streams combine into unified subjective experiences.

Published in Nature Communications (predicted rank #1) · training set

Matching journals

The top 4 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.