Study Design.

A) Stress and control weeks were counterbalanced for order, followed by stress and control MRI scans that were also counterbalanced between subjects. B) Typical scan day for a participant. C) Three tasks performed during MRI task blocks including a standard 2- Back, facial oddball, and associative retrieval task.

Residualization-based affective reactivity score calculation. The averaged effects of examination weeks on (A) subjective stress levels and (B) positive affect, with individual colored lines shown for the modeled random subject-level effects. (C) These random effects (RE) were extracted from each model and then regressed onto each other. Positive affect changes were then residualized with respect to the exposure measure and inverted, resulting in affective reactivity scores. Thus, blue dots above the line indicate lower affective reactivity to stress exposure, while red dots under the regression lines indicate higher affective reactivity scores.

Physiological changes in response to SECPT.

A) Salivary cortisol response showing increased cortisol in response to the SECPT compared to the control procedure. B) Average heart rate (beats per minute) and C) average heart rate variability (RMSSD, ms) per scanner run separated by whether participants had the control or stress scan first. Participants who had the stress session first exhibited increased heart rate and decreased heart rate variability (RMSSD) during the stress session. Error bars = SEM.

Main Task Effects.

Main task effects across all task runs for each of the contrasts for the SN- oddball task (A, MNIxyz=41 91 35 mm), ECN-2back task (B, MNIxyz=27 33 29 mm), and DMN-associative retrieval task (C, MNIxyz=41 91 35 mm). Increased activity seen within expected regions of interest as well as some overlap in the SN and ECN contrasts, as well as an unexpected decrease in the DMN contrast. Results are whole-brain corrected with a cluster forming threshold of Z>3.1 and whole-brain corrected cluster significance level of P<0.05.

Effects of stress on large scale networks over time and real-life stress reactivity.

A) Stress-Control differences in each of the networks in both the early and late phases of the stress response showing overall suppression of DMN under stress, and an increase in SN activity in early phase of stress reactivity. B) Decreased SN reactivity to stress is related to increased resilience in daily life. Error bars = SEM.

Oddball Task Measures.

D-prime (A), Hits (B), and False alarms (C) differed significantly between the stress and control sessions, with lower d-prime under stress driven by both fewer hits, and more false alarms. SN (salience network) activity was also related to false alarms differently in the stress and control session (D).