PhD Defence Charelle Bottenheft | Impact of multiple stressors on cognitive performance: Towards a predictive model

Impact of multiple stressors on cognitive performance: Towards a predictive model

Charelle Bottenheft is a PhD student in the Department of Human Media Interaction. Promotors are prof.dr. J.B.F. van Erp from the Faculty of Electrical Engineering, Mathematics and Computer Science and prof.dr. E.L. Groen from TNO & University of Cranfield, Aerospace Transport.

Professionals in various high-risk and demanding domains often perform complex cognitive tasks under mentally and physically challenging conditions inherent to their work environment. These conditions, referred to as stressors, can significantly influence cognitive functioning. While existing research has extensively examined the effects of individual stressors, professionals are often exposed to multiple stressors simultaneously. The aim of the thesis is to develop and validate a theoretical model to understand and predict the combined effects of multiple stressors in a general way.

Cognitive Performance Model

To address our multiple stressor approach and its impact on cognitive performance, we propose to integrate two models: a combined Stressor Load model into a Cognitive Performance model (Chapter 1). The Cognitive Performance model assumes that cognitive performance follows an inverted U-function of Task Demand, predicting that performance become suboptimal when Task Demand is too low (underload) or too high (overload). To explicitly analyze the effects of multiple stressors, we define Task Demand as the sum of Task Load and combined Stressor Load. Task Load is strictly used to denote the inherent demand of the task, reflecting how a task can influence an individual’s cognitive load. Our hypothesis is that combined Stressor Load reflects the impact of internal or external stressors that may interact and influence cognitive performance, depending on the extent to which underlying mechanisms overlap or not. We propose that this interaction is determined by two types of mechanisms: competition for physiological resources and competition for cognitive resources. Depending on this overlap, interactions can be additive, antagonistic, or synergistic.

Our Cognitive Performance Model uses the inverted U-curve to illustrate how combined Stressor Load shifts Task Demand and affects cognitive performance. Depending on the magnitude of this shift, performance can improve (moving from underload to optimal load), decline (from optimal load to overload), or remain stable within the optimal range.

Research Questions

Four laboratory and simulator studies were conducted to address the following research questions: 

RQ 1: What is the effect of multiple stressors on cognitive performance?

RQ 1A: How does a combination of stressors, each with assumed distinct underlying mechanisms, affect cognitive functioning?

RQ 1B: How does a combination of stressors, with assumed shared underlying physiological mechanisms, affect cognitive functioning?

RQ 2: How effective is the Cognitive Performance model in predicting the effect of multiple stressors on cognitive task performance?

RQ 2A: Is cognitive performance an inverted U-function of Task Demand?

RQ 2B: Can Task Demand be defined as the sum of Task Load and combined Stressor Load? 

Results

To address the first research question, three studies examined combinations of stressors with either distinct or shared mechanisms. Chapter 2 investigates the influence of a metabolic (skipping a meal) combined with a sensory stressor (noise bursts) on cognitive task performance. These stressors were assumed to affect the human body through different mechanisms, addressing RQ 1A. Contrary to our expectations, cognitive performance on the working memory task improved in the presence of noise bursts, likely due to increased arousal. Skipping breakfast did not induce hypoglycemia and was therefore ineffective as a metabolic stressor, which may explain the absence of a combined effect.

Chapter 3 examines the combined effects of heat load and hypobaric hypoxia on cognitive performance. These stressors were expected to affect cognitive performance by a shared physiological mechanism, answering RQ 1B. Hypobaric hypoxia alone did not affect overall cognitive performance, likely due to the relatively mild level of hypoxia, which allowed participants to compensate or adapt. On the other hand, heat load caused a decline in performance, probably due to thermal discomfort drawing cognitive resources away rather than reduced oxygen supply. No combined effects were found across tasks, but an unexpected additive effect appeared on one multitasking subtask, suggesting independent mechanisms.

Chapter 4 examines the combined effects of total sleep deprivation and acute social stress on cognitive performance. This study also explores their expected shared underlying physiological mechanism, also addressing RQ 1B. Both sleep deprivation and the acute social stressor caused a cognitive performance decline. The combined condition of sleep deprivation and acute social stress showed an antagonistic interaction, as the stressors had opposite effects on cognitive performance. This suggests they may share a common mechanism, with social stress improving performance in sleep-deprived participants. Two possible explanations for a shared mechanisms emerged: an arousal effect, reflected in increased autonomic responses, and a similar pattern to that observed in cognitive performance was observed in immune responses (IL-22 levels in blood).

Overall, the studies showed limited evidence for the interactions between stressors because stressors were not always effective in producing an effect on cognitive performance. In addition, interaction effects were not always in the predicted direction. Therefore, we can neither accept nor reject the combined Stressor Load model, and are unable to provide a definitive answer to RQ 1A and 1B.

To address the second research question, the fourth experimental study examines if task performance follows the assumed inverted U-shape as function of Task Demand by varying Task Load within an individual. In addition, the effect of a stressor was tested by varying Stressor Load at two levels (with or without the presence of heat load) (Chapter 5). We used a Tetris game to measure performance under conditions of high load (i.e. the right side of the performance curve). Results showed that performance decline towards higher Task Load levels could be estimated with an inverted sigmoid for most participants. However, the Stressor Load, i.e., heat load, did not produce the expected negative effect on Tetris performance, so we were unable to demonstrate a shift along the performance curve.

Conclusions

Overall, our Cognitive Performance model could not predict the effects of multiple stressors on cognitive performance, addressing RQ 2 (Chapter 6). Our findings suggest that our Cognitive Performance Model may be too simplistic to fully capture cognitive performance under combined stressor conditions. Humans are adaptive and capable of adjusting to varying levels of (combined) Stressor Load and Task Load. Besides this, factors such as fatigue or inattentiveness from time-on-task effects or sleep deprivation influence the model, resulting in a suboptimal mental state. Therefore we propose adjusting the model by introducing mental state as an internal parameters alongside Task Load and (combined) Stressor Load, through which stressors can alter the shape of the performance curve rather than merely shifting the position along the Task Demand axis.