Sports performance at the highest level requires a wealth of cognitive functions such as attention, decision making, and working memory to be functioning at optimal levels in stressful and demanding environments. Whilst a substantial research base exists focusing on psychological skills for performance (e.g., imagery) or therapeutic techniques for emotion regulation (e.g., cognitive behavioral therapy), there is a scarcity of research examining whether the enhancement of core cognitive abilities leads to improved performance in sport. Cognitive training is a highly researched method of enhancing cognitive skills through repetitive and targeted exercises. In this article, we outline the potential use of cognitive training (CT) in athlete populations with a view to supporting athletic performance. We propose how such an intervention could be used in the future, drawing on evidence from other fields where this technique is more fruitfully researched, and provide recommendations for both researchers and practitioners working in the field.
The Role of Cognition in Sport
The role of cognition and neuroscience in understanding, predicting, and potentially improving elite sports performance is an area that has received increased interest in recent years (Yarrow et al., 2009; Walsh, 2014; Katwala, 2016). This notion is validated by studies showing that athletes perform faster and more accurately on specific cognitive tasks (Mann et al., 2007; Voss et al., 2010). Such findings have been supplemented by studies showing that baseline cognitive ability is able to predict future sporting achievement (Vestberg et al., 2012, 2017; Mangine et al., 2014).
Given the above evidence, the aim of this paper is to introduce some of the considerations in this potentially booming field of practice, incorporating knowledge of cognitive training (CT) in other cohorts. We highlight that further research is needed before we can reliably inform coaches, athletes, and support staff of any potential benefits from this technique. Well planned studies which incorporate collaborative interdisciplinary knowledge are needed to progress this field most rapidly.
A Brief Introduction to Cognitive Training
Computerized CT is a flourishing field of research [and commercial business (George and Whitehouse, 2011)] within the scope of cognitive enhancement, with applications being studied extensively in many different cohorts. The central focus of CT is to target specific cognitive functions, through repetitive computerized exercises. Complexity and response time demands change frequently during and across sessions, in accordance with changes in individual performance as to avoid over- or under-stimulation.
Cognitive training has shown efficacy in terms of post-training performance on cognitive testing, assumed to represent an improved capacity in the specific domain (i.e., near transfer), though relevant to this discussion, also on aspects of motor functions such as gait (Smith-Ray et al., 2015; Walton et al., 2018). Improvements in cognition have been shown in those with neurodegenerative disease, along with other psychiatric and neurological disorders (Keshavan et al., 2014; Lampit et al., 2014b; Leung et al., 2015; Hallock et al., 2016; Hill et al., 2016; Motter et al., 2016).
Despite many positive findings for CT on cognition, it must be acknowledged that there is a strong and healthy debate surrounding overall efficacy, justifiably, given the claims from some commercial companies often outweigh the underlying scientific evidence (e.g., see the well documented exchange between researchers) and extensive review by Simons et al. (2016). Additionally, the CT field has struggled in general from high levels of methodological heterogeneity amongst studies, a poor ability to define improvement in a functional capacity, and small sample sizes (Walton et al., 2014). In the current context, it is also worth noting that CT has predominantly shown most promise in populations characterized by deficits in cognition, in that it has primarily been used to raise what may have previously decreased, or reduce further losses. As illustrated above, elite athletes may actually have superior functioning within specific domains, and thus it is currently unknown whether CT can enhance cognitive performance in this sample.
Enhancing Cognition for Elite Performance
Anecdotal evidence suggests that exercises which resemble CT are already being implemented in sports environments. Indeed, there are many companies now selling software aimed to deliver this very product (e.g., NeuroTracker, Axon Sports). As researchers and advocates of CT, it is encouraging to see the enthusiastic uptake of the technology in new settings. However, it appears the bulk of existing evidence regarding CT’s efficacy, on which athletes and coaches must currently rely, comes from direct claims delivered by some of the commercial companies themselves (or their sponsored athletes), which often do not appear backed up by peer-reviewed accessible science. The early stages of CT research more generally were once in a similar state, however, the field now sees hundreds of publications per year (Walton et al., 2014), progressively fine-tuning facets of design. Nevertheless, given that CT is not a ‘one size fits all’ intervention, our knowledge of what does, doesn’t, or could work for these specific sporting purposes lags significantly behind other cohorts (Harris et al., 2018). This must change before these interventions are to be wholeheartedly endorsed and promoted.
Harris et al. (2018) reviewed the evidence for real-world transfer of effects using commercially available CT interventions. These authors found only one study (Romeas et al., 2016) to have been completed within a sporting context, illustrating the lack of evidence for CT in athletes. This study employed 3-dimensional multiple object tracking (3D-MOT), a task which challenges users to keep track of multiple moving objects in a dynamic and changing visual field. Intuitively, this skill has implications for sports performance where athletes must be able to accurately process, for example, multiple teammates, the opposition, obstacles and targets all at once. Athletes have been shown to excel in this task, with Faubert (2013) showing that professional athletes across multiple sports have a higher baseline ability to perform this task, but also faster learning curves than non-elite athletes, and non-athletes. Romeas et al. (2016) examined the training in 19 male soccer players over three groups (3D-MOT, passive and active control). The experimental group trained twice weekly for 5 weeks, while the active control watched 3D soccer videos accompanied by short interviews based on decision making, thus reinforcing the expectation of training benefit to the athletes. Following training, the intervention group improved by 15% in a measure of on-field passing decision-making, in addition to subjective confidence levels in decision-making accuracy. There were not improvements in shooting or passing accuracy, which again reflects the potential constraints on transferring of CT benefits to related-but-different tasks.
There were limitations to this work, not least that the intervention group only included seven athletes (two dropped out). It must also be acknowledged that this study was conducted by researchers who are, ostensibly, heavily invested in the tool; providing further evidence that navigating the realm of combining scientifically rigorous studies with financially lucrative tools will be inherently difficult (Rabipour and Raz, 2012; Simons et al., 2016). This potential conflict-of-interest has previously been a common criticism of CT, where some companies who have enormous financial incentives to show positive results have been involved in the research studies which seek to objectively determine efficacy. While we certainly suggest no wrongdoing whatsoever, and the author’s conflicts of interest were clearly provided, we would like to highlight that separating proof of efficacy research studies from those invested in the outcomes is always preferential (Ahn et al., 2017).
Separately, and not reviewed by Harris et al. (2018), Hirao and Masaki (2018) used the Simon Task to make those trained more able to shoot toward the opposite direction of a goal-keepers initial lateral movement. Twenty-nine lacrosse players were split into two groups, either conducting Stimulus-Response Compatibility Training, or an active control. In line with the authors’ hypothesis, the intervention group shot to the opposite side of the goalie’s movement more often than the active control post-training, though this did not lead to more goals being scored, potentially due to poor shooting velocity. Additionally, though there was a significant difference between groups at post-test, the treatment group did not show a significant improvement from baseline. It is also worth noting that the treatment group performed the cognitive task less accurately at follow up, and significantly worse than the control group following training. Therefore, while this study is interesting and has some well-designed elements, we cannot obtain a full picture of the training efficacy and theoretical underpinnings for the improvements found in this work.
The work of Romeas et al. (2016) and Hirao and Masaki (2018) are certainly exciting, and a positive step in the right direction for investigating the potential efficacy of CT in sport via peer-reviewed controlled trials. However, given the known difficulty of achieving far transfer following CT, it is surprising that the only known studies have both provided positive effects. Replication is required before such results can be relied upon, and of particular importance, publication of null results in similar studies is encouraged so as to minimize creating a biased literature.
Of note, there are studies which have examined other techniques of training which also incorporate some cognitive-perceptual ability (see review by Hadlow et al., 2018). By contrast, these studies have been more focused around aspects including: (a) video-based training that is highly specific to the outcome (e.g., quickly predicting the direction of a batsman’s strike from video (Hopwood et al., 2011); (b) computer-based putting training (Fery and Ponserre, 2001); or (c) making decisions faster than ‘real-time’ on sporting scenarios (Lorains et al., 2013; Farahani et al., 2017). While this work is very interesting and likely has great potential for investigating the role of cognition in increasing performance, we do not consider this to be CT per se, but rather an alternative method of sport-specific practice that involves computerized tools. By definition, CT should target specific cognitive functions that are not simply reflective of the desired outcome. Given this, when discussing CT for sport, we are specifically interested in the act of improving core cognitive processes which in turn fundamentally underlie sports performance.
We suggest that there is a significant gap in our knowledge-base regarding how CT can be implemented to improve athletes’ performance. Given the link between cognition and sporting ability, there is a clear rationale for further investigating whether CT could benefit athletes. However, the current evidence-base means that we cannot know whether this tool is effective, and given the difficulties achieving far transfer in other cohorts, we caution around investing too heavily in such methods at this point in time. We do, however, recognize there is merit to investigating further, and research that would develop this understanding will require the assistance of coaching staff and athletes to establish high quality studies, with the ultimate aim of better understanding how these methods could help athletes maximise every potential for their performance.
Read more HERE