Can cognitive functions predict athlete's overall effectiveness and contribution to the team in small-sided soccer games? A Machine Learning approach

• Grant number: 21/15134-9
• Support Opportunities: Scholarships in Brazil - Master
• Start date: August 01, 2023
• End date: August 31, 2024
• Field of knowledge: Health Sciences - Physical Education
• Principal Investigator: Paulo Roberto Pereira Santiago
• Grantee: Rafael Luiz Martins Monteiro
• Host Institution: Faculdade de Medicina de Ribeirão Preto (FMRP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil

Abstract

Soccer is a team sport that requires quick and accurate decision-making. The brain and its various cognitive functions are primarily responsible for the decision-making process. Understanding how each one relates to the performance of athletes on the field is essential. Thus, the present study aims to predict the overall effectiveness and players contribution to the team in small-sided games by analyzing the cognitive functions: sustained, selective and distributed attention, working memory, executive control, impulsivity control and multiple object tracking through the application of machine learning in addition to describing the association between cognitive variables and field performance. 46 soccer athletes working in the under 17 category of professional clubs will participate in the study. To evaluate the athletes' cognitive functions, the following tests will be used: d2, stroop, counting span, Go/No Go, trail making and multiple object tracking. For the evaluation of the athletes' performance on the field, a protocol with reduced games will be used that will provide the general effectiveness of the players and the contributions of the same to the team. For the prediction of performance in the field through the analysis of cognitive functions, classificatory supervised machine learning codes will be used. Statistical analyzes will be performed using measures of central tendency and variability and Pearson's correlation will be used to verify how the results of the cognitive tests are correlated with the performance variables in reduced games. In all cases the significance level will be p d 0.05. It is expected that it is possible to predict performance in small-sided games through the analysis of cognitive tests and that there is a positive correlation between the performance variables of athletes in small games and cognitive tests.