مطالعه مقدماتی در ارزیابی ویژگی‌های سیگنال مغزی به منظور ارائه شاخص مرتبط با توجه بصری در ورزشکاران نیمه‌حرفه‌ای تنیس روی میز

هدف: در ورزش‌های مبتنی بر عملکرد شناختی بالا مانند تنیس روی میز، آموزش نوروفیدبک موثر نیازمند شاخص‌های زیستی دقیق و متناسب با عملکردهای شناختی خاص، مانند توجه بصری است. با این حال، اغلب پروتکل‌های نوروفیدبک فعلی بر ویژگی‌های کلی eeg تکیه دارند که لزوماً بازتاب‌دهنده صحیحی از عملکردهای ذهنی مورد نیاز در این ورزش‌های سرعت‌محور نیستند. پژوهش حاضر با هدف رفع این خلا، به بررسی ویژگی‌های سیگنال eeg مرتبط با توجه بصری پرداخته است. مواد و روش ها:  براین اساس  در این مطالعه، سیگنال eeg از ناحیه cz در دو حالت استراحت و تصمیم‌گیری از 8 ورزشکار مرد نیمه‌حرفه‌ای تنیس روی میز (18 تا 25 سال) ثبت شد. ویژگی‌های آماری متوسط و واریانس، میانگین دامنه و توان تبدیل فوریه ، میانگین قدرمطلق و انحراف معیار ضرایب تبدیل ویولت، و ویژگی‌های غیرخطی آنتروپی، بُعد فرکتال، بُعد همبستگی و نمای لیاپانوف استخراج گردید. سپس همبستگی این ویژگی‌ها با نمرات آزمون tova از جمله زمان واکنش و خطاهای توجه محاسبه شد و آزمون تی برای بررسی تغییرات قبل و بعد آزمون به کار رفت. یافته ها: نتایج نشان داد ویژگی‌های فرکانسی و زمان-فرکانسی بیشترین همبستگی را دارند به گونه‌ای که متوسط همبستگی ویژگی‌های فرکانسی و زمان فرکانسی با شاخص زمان واکنش به ترتیب 84/0 و  89/0 بدست آمد. علاوه بر این ویژگی‌های فرکانسی و زمان-فرکانسی نیز بیشترین تغییرات را در قبل و بعد از آزمون (p ≤0.02 ) داشت.نتیجه گیری:  این یافته‌ها نشان می‌دهند که می‌توان از این ویژگی‌ها به‌عنوان نشانگرهای زیستی موثر برای پایش و ارتقاء فرآیندهای توجهی در ورزشکاران استفاده کرد. بنابراین، استفاده از این ویژگی‌ها در طراحی پروتکل‌های نوروفیدبک به‌منظور بهبود واکنش سریع شناختی در ورزشکاران مستعد، به‌ویژه در رشته‌هایی مانند تنیس روی میز، کاربردی و موثر خواهد بود.

a pilot study on the evaluation of brain signal features to provide an index related to visual attention in semi-professional table tennis players

background and purposeexcellence in sports performance relies not only on physical and motor abilities but also on sensory and cognitive skills. given the high cognitive demands in sports, numerous studies in recent years have highlighted the critical role of these skills in athletic success (shoxrux et al, 2023). reaction time, in particular, is a key factor across all sports, often determining the outcome of competitions where athletes possess comparable levels of physical training (reigal et al, 2019). the emerging field of sports neuroscience aims to deepen our understanding of the brain-behavior relationship, with the ultimate goal of informing training practices and enhancing athletic performance. one approach to exploring performance enhancement involves studying brain activity through electroencephalography (eeg). by extracting relevant features from eeg signals linked to cognitive performance, these insights can be applied to cognitive training systems, such as neurofeedback. this study aims to identify eeg features associated with reaction time in amateur table tennis players.materials and methodsthe study recruited eight healthy male amateur table tennis players aged 15 to 25 to investigate the relationship between cognitive performance and eeg-derived features. participants’ cognitive abilities were assessed using the test of variables of attention (tova), which measures response time, response time variability, errors of commission, errors of omission, and anticipatory errors. brain activity was recorded from the cz electrode, following the international 10-20 system, using the flexcomp infinity device. recordings were performed under resting and decision-making states to capture diverse brain activity patterns. the protocol began with a three-minute baseline recording under resting conditions with eyes closed, conducted both before and after the cognitive test. during the tova assessment, additional eeg data was collected. to familiarize participants with the procedure and reduce variability, a two-minute practice session preceded the actual recordings. after data acquisition, eeg signals underwent artifact removal and preprocessing. feature extraction encompassed four domains: time domain, including mean and variance; frequency domain, encompassing metrics such as maximum, minimum, mean, absolute power, and various power ratios derived from fourier transforms; nonlinear domain, which analyzed fractal dimension, entropy, maximum lyapunov exponent, and correlation dimension; and time-frequency domain, assessing wavelet coefficients for eeg bands through average, standard deviation, and ratio calculations. to pinpoint the most relevant eeg features, correlation coefficients were computed between tova test scores and the extracted features. features demonstrating the strongest correlations were identified as optimal for further analysis. this methodological approach enables an in-depth understanding of the neural signatures associated with cognitive performance, particularly in tasks requiring sustained attention and rapid decision-making. these findings could contribute to developing tailored training protocols to enhance cognitive and neural performance in athletes.findingsthe test results of one participant in this study significantly differed from those of individuals in the general population. to ensure consistency in analyzing the target population's processes, this participant’s data was excluded. feature selection was performed to identify the most relevant variables. using matlab, the correlation values of all features were ranked from highest to lowest. features with correlation values below 0.5 were excluded from further analysis, narrowing the focus to the most impactful variables. frequency and time-frequency features demonstrated the strongest correlation with the test indicators. however, the relationships between test indicators and features varied across the states considered for the target population. this lack of consistency suggests that it is challenging to derive a unified understanding of feature presentation across different states. in simpler terms, each subject's state does not provide a coherent representation of the features. in the first half of the test, which assessed reaction speed to stimulus presentation (the second mode), frequency and time-frequency features consistently exhibited the highest correlation with reaction speed indicators. based on this finding, instead of conducting comprehensive tests for each subject, it is possible to estimate a subject's reaction speed by recording their brain signals during a single session. if the frequency and time-frequency features display a high correlation, they can reliably predict the individual's reaction speed to the stimulus. the second half of the test was designed to assess each subject's tension levels. by leveraging the strong correlation of frequency and time-frequency features with the reaction speed and tension indicators, the study suggests a streamlined approach for measuring these parameters. this methodology reduces the need for repeated testing while still providing accurate insights into the subject’s reaction dynamics and emotional states. these findings highlight the utility of frequency and time-frequency features in understanding brain signal responses, paving the way for more efficient and targeted data collection methods in neurophysiological research.conclusionexamining the correlation between the features extracted from the brain signal and reaction time shows that the frequency and time-frequency features have a higher correlation with the reaction time and the correlation value is positive and on average equal to 0.84, while the correlation of the nonlinear features in the test is not logical and it has not followed a certain relationship. other studies also confirmed the significant changes in frequency components during the activities of table tennis athletes (pineda et al, 2022; tsai et al, 2022). among the limitations of previous research, is the evaluation of cognitive functions only using non-objective tests (christie et al, 2024; ceylan et al, 2020) or based on protocol design in the simulation space. (babiloni, et al, 2010; del percio et al, 2010) that this issue lowers the reliability of the approaches used. identifying indicators related to cognitive functions based on brain signal enables the evaluation of these functions objectively.