Psychophysiological markers of attention concentration as predictors of performance in student hockey forwards: ROC analysis

Abstract

Introduction. Attention is a key cognitive process in team sports. In hockey, competitive performance is critically dependent on rapid decision-making and efficient information processing under time pressure. However, the current literature lacks established threshold values for attention metrics that could be used to differentiate players according to their performance level. Objective. To evaluate the prognostic value of attention concentration and sustained attention metrics as predictors of athletic performance among student hockey forwards. Materials and methods. The cross-sectional study enrolled 50 forwards from the student hockey team of the Urals State University of Physical Culture, with a mean age of 21.9 ± 1.9 years. Sustained attention and concentration were assessed using psychophysiological testing with the NS-PsychoTest hardware-software complex; the results were expressed as quantitative and scoring metrics. Performance was evaluated based on an integral index of technical and tactical actions (TTA), calculated by the coaching staff from the data of official matches. Based on the TTA distribution, players were stratified into three groups: Group 1 (below the 25th percentile; low performance; n  = 16), Group 2 (between the 25th and 75th percentiles; moderate performance; n  = 18), and Group 3 (above the 75th percentile; high performance; n  = 16). Group comparisons were performed using nonparametric Kruskal–Wallis and Mann–Whitney U tests. The effect size was estimated using Cohen’s  d . The predictive ability of attention concentration metrics was assessed via Receiver Operating Characteristic (ROC) analysis, including calculation of the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Results. Statistically significant differences in attention concentration metrics were found between the low- and high-performance groups. For the quantitative attention concentration measure, p  = 0.003 and Cohen’s  d  = 0.65 (medium effect); for the scoring measure, p  < 0.001 and Cohen’s  d  = 0.82 (large effect). ROC analysis confirmed the high predictive potential of both metrics. For the quantitative measure, AUC was 0.805 (95% CI: 0.627–0.946), with an optimal threshold of ≤ 0.950, sensitivity of 62.5%, specificity of 93.8%, PPV of 90.9%, and NPV of 71.4%. For the scoring metric, AUC was 0.812 (95% CI: 0.674–0.938), with an optimal threshold of ≥ 2 points, sensitivity of 87.5%, specificity of 75.0%, PPV of 77.8%, and NPV of 85.7%. The scoring assessment of attention concentration (on a 3-point scale) demonstrated diagnostic accuracy comparable to that of the quantitative metric (AUC: 0.812 vs. 0.805, respectively). Age-related effects were controlled using rank-based ANCOVA, which confirmed the independence of the observed differences ( p  = 0.007, η 2  = 0.26). Conclusions. Attention concentration metrics demonstrated high predictive value and may serve as psychophysiological markers for differentiating forwards by performance level. Given its comparable diagnostic accuracy and ease of interpretation, the scoring-based assessment of attention concentration is recommended for practical application in the system of sports talent selection. The established diagnostic thresholds require validation in larger samples; however, they may already be applied in sports selection practice as additional objective criteria.