Altered functional connectivity of the anterior insula in attention networks of autistic individuals
The anterior insula sits at the junction of the salience and ventral attention networks — the machinery that decides what is worth attending to. It has been implicated in autism for over a decade, with findings that contradict each other: hyperconnectivity in some cohorts, hypoconnectivity in others. One explanation for conflicting results is that a real effect is being averaged away.
The question
Does anterior insula connectivity with the salience and ventral attention networks differ in autism — and does it differ by sex, in a literature that has mostly treated sex as a covariate?
How it is being done
- ROI-to-ROI resting-state functional connectivity between the anterior insula and pre-specified regions of the salience and ventral attention networks.
- Males and females analyzed separately rather than pooled with sex as a covariate — the decision the whole thesis turns on.
- Standard preprocessing across all scans, with ROIs defined a priori and MNI coordinates reported.
- Connectivity matrices regressed against symptom instruments — SRS-2, SCQ and ADI-R — to test whether the differences carry behavioral meaning.
What it shows so far
- Connectivity differed significantly between autistic and non-autistic children, concentrated at the temporoparietal junction and supramarginal gyrus.
- The direction reversed by sex: increased connectivity in autistic females, decreased in autistic males — an effect that cancels toward null when the sexes are pooled.
- Preliminary regressions related resting-state connectivity in autistic females to symptom scores, suggesting connectivity may carry information about severity.
Materials
What comes next
The sex-stratified result argues for replication in a larger cohort with enough autistic girls to power the female analysis on its own — the sample constraint that has shaped this literature from the start.