Effects of psychological fatigue on college athletes’ error-related negativity based on artificial intelligence computing method

Abstract

Psychological fatigue includes mental fatigue and burnout. In order to investigate the impact of psychological fatigue on athletes' response monitoring, event-related potentials technique is typically used, and the most critical indicator is error-related negativity. Two experiments were carried out to explore cause-effect relationships of psychological fatigue and response monitoring. The event-related potentials data processing was based on Artificial Intelligence computing methods, including wavelet transform, grayscale transformation and other algorithms. The first experiment was done to explain mental fatigue and response monitoring. 15 participants operated continuously 60 min Flanker task, and then operated 15 min task accompanied by light music. From the results of behavioral performances, the fatigue period compared with the fatigue-free period showed significant differences, including reaction time (p = 0.029) and correct rate (p = 0.046). From the results of error-related negativity, the amplitude of the fatigue-free period was bigger than that of the music adjustment period, the latter was bigger than that of the fatigue period (p < 0.001). The second experiment was conducted to explore burnout and response monitoring. Twenty-four participants were separated averagely into two groups. One group was burnout group, the other group was no burnout group. They both operated 15 min task. From the results of behavioral performances, no burnout group was better than burnout group. From the results of error-related negativity, no burnout group’s amplitude was bigger than that of burnout group, but the difference was marginally significant. The conclusions are as follows: Artificial intelligence technology is feasible for processing event-related potentials data. Psychological fatigue weakens athletes’ response monitoring ability, and the effect of mental fatigue is significant. In future researches, the following topics should be concerned, including the mediating or moderating effects of third variables, different ways of recovering from mental fatigue, computer data simulator and date accuracy, brain-computer interfaces and error-related negativity, etc.

Stability of Performance Monitoring with Prolonged Task Performance: A Study of Error-Related Negativity and Error Positivity

“…Study 1, error-trial Pe uniquely decreased over the task battery and individual tasks, whereas during Study 2, error-trial Pe only decreased during individual tasks. Present findings from both task batteries contradict other studies that showed Pe becomes larger (Clayson et al, 2012) or does not change with prolonged task performance (Li et al, 2022;Xiao et al, 2015). Pe relates to conscious error processing (Steinhauser & Yeung, 2012;Ullsperger et al, 2010;Wessel, 2012;Wessel et al, 2012), and, it is possible that reduced task engagement could relate to concomitant reductions in conscious processing of errors.…”

“…This preregistered study aimed to determine how prolonged task performance affects ERN amplitude both within a single task and across a battery of tasks, using data from two different studies. We predicted nonlinear decreases in ERN amplitude both during individual tasks and across the task battery due to an initial decrease in ERN amplitude followed by a floor Decreases in ERN amplitude over this time frame are consistent with findings that ERN amplitude decreases with extended task engagement, lasting up to an hour (Clayson et al, 2012;Kato et al, 2009;Kato et al, 2012;Li et al, 2022;cf. Moore et al, 2017).…”

“…Studies of the relationship between ERN and prolonged task performance generally observe decreases in amplitude over the course of a single task. For example, ERN amplitude decreases with prolonged task performance for as long as an hour (Clayson et al, 2012;Kato et al, 2009;Kato et al, 2012;Li et al, 2022;Ma et al, 2017;cf. Moore et al, 2017) or two (Boksem et al, 2006;Lorist et al, 2005).…”

“…Pe is studied less frequently than ERN in the context of prolonged task performance, and the findings are mixed. Whereas one study showed that correct-and error-trial Pe became larger over the course of a one-hour task (Clayson et al, 2012), another study showed no change in Pe amplitude over one-hour tasks (Li et al, 2022;Xiao et al, 2015). Therefore, relationships between Pe and prolonged task performance remain unclear.…”

Higher education worries and response in the era of artificial intelligence

“…The fourth industrial revolution has been opened, and AI is the most important technology in the 21st century, it has triggered a chain reaction in all fields of society, including education. The rapid progress of technology and the lagging adaptation of higher education constitute a contradiction [10]. The development of education and technology is not synchronized.…”