Gunslingers, poker players, and chickens 1: Decision making under physical performance pressure in elite athletes

The cognitive skills required during sport are highly demanding; accurate decisions based on the processing of dynamic environments are made in a fraction of a second (Walsh, 2014). Optimal decision-making abilities are crucial for success in sporting competition (Bar-Eli et al., 2011; Kaya, 2014). Moreover, for the elite athlete, decision making is required under conditions of intense mental and physical pressure (Anshel and Wells, 2000), yet much of the work in this area has largely ignored the highly stressful context in which athletes operate. A number of studies have shown that conditions of elevated pressure influence athletes' decision quality (Kinrade et al., 2015; Smith et al., 2016), response times (Hepler, 2015; Smith et al., 2016) and risk taking (Pighin et al., 2015). However, almost all of this work has been undertaken in nonelite athletes and participants who do not routinely operate under conditions of high stress. Thus, there is very little known about the influence of pressure on decision making in elite athletes.

This study investigated the influence of physical performance pressure on decision making in a sample of world-class elite athletes. This allowed an examination of whether findings from the previous work in nonelite athletes extend to those who routinely operate under conditions of high stress. How this work could be applied to improve insight and understanding of decision making among sport professionals is examined. We sought to introduce a categorization of decision making useful to practitioners in sport: gunslingers, poker players, and chickens.

Twenty-three elite athletes who compete and have frequent success at an international level (including six Olympic medal winners) performed tasks relating to three categories of decision making under conditions of low and high physical pressure. Decision making under risk was measured with performance on the Cambridge Gambling Task (CGT; Rogers et al., 1999), decision making under uncertainty with the Balloon Analogue Risk Task (BART; Lejuez et al., 2002), and fast reactive responses and interference with the Stroop Task (Stroop, 1935). Performance pressures of physical exhaustion was induced via an exercise protocol consisting of intervals of maximal exertion undertaken on a watt bike.

At a group level, under physical pressure elite athletes were faster to respond to control trials on the Stroop Task and to simple probabilistic choices on the CGT. Physical pressure was also found to increase risk taking for decisions where probability outcomes were explicit (on the CGT), but did not affect risk taking when probability outcomes were unknown (on the BART). There were no significant correlations in the degree to which individuals' responses changed under pressure across the three tasks, suggesting that elite athletes did not show consistent responses to physical pressure across measures of decision making. When assessing the applicability of results based on group averages to individual athletes, none of the sample showed an "average" response (within 1 SD of the mean) to pressure across all three decision-making tasks.

There are three points of conclusion. First, an immediate scientific point that highlights a failure of transfer of work reported from nonelite athletes to elite athletes in the area of decision making under pressure. Second, a practical conclusion with respect to the application of this work to the elite sporting environment, which highlights the limitations of statistical approaches based on group averages and thus the beneficial use of individualized profiling in feedback sessions. Third, the application of this work in a sports setting is described, in particular the development and implementation of a decision-making taxonomy as a framework to conceptualize and communicate psychological skills among elite sporting professionals.