We thank Dr. Ali for his concise annotation of our efforts to validate a tool that evaluates mental workload in surgery [1, 2]. Unlike other safety critical domains, the field of surgery has been slow to acknowledge the impact of intraoperative stress on surgical performance, but recently a sea change has been triggered by authorities in the field of surgical education [3]. We agree with Ali that stress is not by default detrimental to performance. Our aim was to develop a diagnostic tool that identifies the factors that contribute to disrupted performance, should it occur. Indeed, studies of the effects of acute stress on operating performance have shown considerable variability, ranging from no effect to either facilitative or debilitative effects [3–5]. The Yerkes-Dodson law emerged from the earliest attempts to explain the relationship between physiological arousal and performance, but it has been criticized for treating stress as a unitary construct, influenced solely by physiological factors [6]. More recently, Catastrophe Theory has been invoked to model the relationship, using both physiological and psychological (cognitive anxiety) components of stress [7]. The model proposes that physiological arousal displays a mild inverted-U relationship with performance when cognitive anxiety is low, but that catastrophic declines in performance can occur if both physiological arousal and cognitive anxiety are high. Recent surgical literature has elucidated the complexity of stress as a construct and signaled the need for a multidimensional approach to the examination of stress in surgery [3].

The importance of investigating both subjective evaluations of stress and objective measures of performance under intraoperative conditions is evident when the workload measures from our study [2] are compared with the performance and stress measures that we used in a previous study [5]. In the previous study, we investigated laparoscopic peg transfer performance under time pressure, multitasking, and evaluation apprehension conditions using a multidimensional approach. That is, we included objective measures of motor performance (Imperial College Surgical Assessment Device; ICSAD) and both subjective (state anxiety) and objective (heart rate) measures of stress. Time pressure was perceived as most stressful, yet performance was maintained at baseline levels. In contrast, multitasking was not perceived as stressful, yet performance was disrupted [5]. Mental workload is an extremely useful construct when the intention is to examine the mechanisms underpinning such effects, as performance is governed by interacting demands of the task, circumstances in which the task is completed, and the skills, propensities, and perceptions of the individual performing that task [2]. Consequently, while multitasking was not perceived as stressful in our earlier study, it was rated subjectively as causing the highest task complexity, mental demands, and distraction on our SURG-TLX measure [2]. Based on these findings, it is not enough simply to reveal that a particular intraoperative condition caused a surgeon objectively to perform worse (as suggested by Ali); without clues to why performance was affected, prevention of potential catastrophes in the future is a “stab in the dark.”

Ali makes a valid point that individuals react differently to stress. For example, a surgeon who feels that he or she has sufficient resources to meet the demands of a situation may view stress in a positive, “challenge” state, but a surgeon who feels dispossessed of the resources required to meet the demands of the situation may view stress in a negative, “threat” state [8]. We have previously shown that individual differences in trait anxiety [9] or the propensity for reinvestment [10, 11] have a significant influence on how efficiently an individual performs a motor task when under evaluative pressure, and recently we have begun to examine these questions in the surgical performance context.