Subjective Uncontrollability over Aversive Events Reduces Working Memory Performance and Related Large-Scale Network Interactions

Abstract

Lack of control over significant events may induce a state of learned helplessness that is characterized by cognitive, motivational, and affective deficits. Although highly relevant in the pathogenesis of several mental disorders, the extent of the cognitive deficits induced by experiences of uncontrollability and the neural mechanisms underlying such deficits in humans remain poorly understood. Using functional magnetic resonance imaging (fMRI), we tested here whether uncontrollability over aversive events impairs subsequent working memory performance and, if so, which neural processes are involved in such deficits. We assessed working memory and the involved neurocircuitry in the MRI scanner before and after participants underwent a task in which they could either learn to avoid electric shocks or had no instrumental control over shocks. Our results show that subjective, but not objective, uncontrollability over aversive events impaired working memory performance. This impact of subjective uncontrollability was linked to altered prefrontal and parahippocampal activities and connectivity as well as decreased crosstalk between frontoparietal executive and salience networks. Our findings show that the perceived uncontrollability over aversive events, rather than the aversive events themselves or the actual, objective control over them, disrupts subsequent working memory processes, most likely through altered crosstalk between prefrontal, temporal, and parietal areas.

(Lack of) Effects of noradrenergic stimulation on human working memory performance

“…We performed identical analyses using the highest n -back level as the dependent variable. In line with previous studies (Bogdanov and Schwabe 2016 ; Wanke and Schwabe 2020 ), we observed a significant training effect in working memory performance, as shown by significant increases in the accumulated n -back level ( F (1,112) = 23.469, p < 0.001, η 2 = 0.173; M pre = 24.388; M post = 26.802) and highest n -back level ( F (1,112) = 13.263, p < 0.001, η 2 = 0.106; M pre = 3.942; M post = 4.273) from baseline to post-manipulation. These analyses did, however, not reveal any significant interaction effects of session and noradrenergic manipulation, session and controllability manipulation, or session, noradrenergic manipulation, and controllability manipulation on working memory performance, as indicated by a lack of significant interaction effects for the accumulated n -back level (all F s ≤ 0.381, p s ≥ 0.392, η 2 s ≤ 0.017) and the highest n -back level (all F s ≤ 0.702, p s ≥ 0.498, η 2 s ≤ 0.016; see Fig.…”

“…Additional analyses were performed to take into account findings from a previous study showing that objective and subjectively experienced controllability may substantially differ and that subjective perceptions of control rather than objective group assignments affect working memory performance. More specifically, a recent study from our lab showed that low subjective uncontrollability was associated with reduced working memory performance, whereas no differences in performance were observed between groups based on objective (un)controllability (Wanke and Schwabe 2020 ). Thus, we additionally performed all analyses with the factor subjective controllability instead of experimental controllability.…”

A Translational Paradigm to Study the Effects of Uncontrollable Stress in Humans

“…EC performed worst in the working memory task, but we had hypothesised YC would show the lowest scores. As working memory has previously been found to be impaired by acute stress [24] and lack of control [26], this contradictory result may reflect problems with our manipulation of stressor controllability. Perhaps, EC experienced greater strain due to higher task demands compared to YC during the stress induction.…”

“…However, it remains unclear how these findings translate to humans, as only a few studies have examined the neural underpinnings. Wanke and Schwabe [26] examined the effects of stressor controllability on working memory performance using an fMRI design to elucidate differential effects on a neural level. They report that perceived control, in contrast to objective control, altered prefrontal activation during the memory task.…”

How representative are neuroimaging samples? Large-scale evidence for trait anxiety differences between fMRI and behaviour-only research participants.

“…Therefore, we set out to gather a large dataset of existing trait anxiety scores from labs across multiple institutions who routinely collect trait anxiety measures in their behavioural and MRI studies 1,25,[41][42][43][44][45][46][47][48][49][50][51][52][53][54] . In order to control for possible confounds and examine interaction effects, we additionally collected the following variables: gender, age, whether and how participants were screened for affective/psychiatric disorders, and whether the study involved the presence of a stressor and/or pharmacological manipulation (see Methods for details).…”