# Children’s state anxiety before MRI scanning and resting state functional connectivity in large scale brain networks

## TL;DR

This study shows that anxiety before an MRI scan can affect brain connectivity patterns, particularly in a group of Latina girls.

## Contribution

The study introduces a constrained network-based statistical method to analyze resting-state functional connectivity in relation to pre-scanning anxiety.

## Key findings

- Reduced resting-state functional connectivity within the default mode network was linked to pre-scanning anxiety in a community sample of Latina girls.
- The association did not replicate in a companion sample of treatment-seeking and healthy youth.
- Sample characteristics and replication are critical for interpreting resting-state connectivity differences.

## Abstract

Most resting-state functional connectivity (rs-FC) research does not consider the participant’s subjective state during magnetic resonance imaging (MRI). Heightened anxiety before an MRI (“pre-scanning state anxiety”) may influence rs-FC and complicate interpretation of individual differences, particularly in underrepresented groups whose scanning experiences may differ from typical research samples.

We assessed associations between pre-scanning state anxiety and rs-FC within and between the default mode network (DMN) and salience network in a trait-anxious community sample of Latina girls (8–13 years) and a companion sample of treatment-seeking and healthy youth (8–18 years) of predominantly non-Latinx background. A constrained network-based statistical approach calculated the average of un-thresholded correlation coefficients from edge-level partial Spearman correlations to produce network-level measures (7 cortical + 1 subcortical). This approach is “constrained” in that analyses operate at the spatial scale of functional networks, rather than individual edges, to increase statistical power. Statistics were compared against a permutation-based null distribution to assess significance (Bonferroni corrected p < 0.00139).

Reduced rs-FC within the DMN (r = − 0.32, p < 0.00139) was associated with pre-scanning state anxiety in the community sample, but did not replicate in our companion sample.

Pre-scanning state anxiety is associated with rs-FC within the DMN, but only among a trait-anxious community sample. Individual differences in MRI scanning experiences may be associated with rs-FC, but sample characteristics and replication should be considered.

The online version contains supplementary material available at 10.1038/s41598-025-34410-8.