Research report: Basketball expertise is associated with distinct relationship s among dynamic visual tracking, visuospatial working memory, and tactical decision-making

Skilled basketball players do not all lean on the same mental tools when reading a game, according to a new cross-sectional study of 90 male players published in Frontiers in Psychology. The research, led by Jinyi Yu of Pai Chai University with colleagues at Jiuquan Vocational and Technical University, Kyungil University and Henan University, found that the cognitive abilities tied to tactical decision-making differ sharply between expert and intermediate players. The authors are careful to frame the findings as exploratory: the cross-sectional design cannot establish causal mechanisms or developmental transitions.

The study recruited 90 men aged 18 to 25 and divided them into expert, intermediate and novice groups of 30 each, using competitive level and training history. Experts held national first-level athlete certification or belonged to a university high-performance team with extensive systematic training. Intermediate players came from collegiate intramural teams or local amateur leagues and generally reported three to seven years of structured training. Novices had no more than roughly three years of basketball experience and did not meet the competitive criteria for the higher groups. All participants had normal or corrected-to-normal vision, were right-handed, and reported no neurological or psychiatric disorders or concussion within the previous six months. Recruitment ran from July 1 to December 13, 2025, and the study was approved by the Institutional Review Board of Jiuquan Vocational and Technical University.

Each participant completed three laboratory tasks in a single session lasting about 70 to 90 minutes, with five-minute rests between tasks. The first was a multiple-object tracking task, a standard measure of dynamic attentional allocation. Participants tracked four target circles among four identical distractors moving within a circular boundary, with circles colliding elastically and bouncing off the walls. A staircase procedure adjusted the speed of motion to estimate each person’s tracking threshold at roughly 79.4 percent accuracy; higher thresholds indicate better dynamic tracking ability. The second task was a computerized Corsi block-tapping test of visuospatial working memory, in which participants reproduced sequences of illuminated squares, with span estimated adaptively between three and seven items.

The third task was the most basketball-specific: a video-based tactical decision-making simulation. Sixty clips of four to six seconds, drawn from professional-league games and selected in consultation with experienced coaches and professional players, ended with a freeze-frame at a critical decision point for the ball carrier. Participants chose among Pass, Shoot, Drive and a context-specific fourth option. Three nationally certified coaches independently evaluated each clip, and only clips with unanimous agreement on the optimal action were retained. A graded rubric assigned three points to the optimal response, two to a viable alternative, one to a plausible but lower-value choice, and zero to an inappropriate decision, yielding a total score transformed to a 0–100 scale with high internal consistency (Cronbach’s alpha of 0.87). Reaction time was recorded but not used in the primary analyses, which targeted decision quality rather than speed–accuracy trade-offs.

On group comparisons, experts clearly led. Their multiple-object tracking thresholds (5.38 ± 0.67 degrees per second) exceeded those of intermediate players (4.82 ± 0.60) and novices (4.45 ± 0.65), with a large effect size (partial eta squared of 0.269). Their tactical decision-making scores (77.60 ± 10.40) also topped intermediate players (68.90 ± 9.80) and novices (60.20 ± 11.10), and intermediate players in turn outscored novices. Working memory told a more graded story: experts had higher Corsi spans than novices, but the expert–intermediate comparison did not reach the adjusted significance threshold. All pairwise comparisons were adjusted using the Benjamini–Hochberg false discovery rate across nine comparisons.

The correlation patterns were where the study became genuinely distinctive. Within the expert group, tracking threshold was positively associated with tactical decision-making (r = 0.535, 95% CI [0.216, 0.750], FDR-adjusted p = 0.010), while the working-memory association was weak and nonsignificant. In intermediate players the pattern reversed: Corsi span was strongly associated with tactical scores (r = 0.726, 95% CI [0.495, 0.861], FDR-adjusted p < 0.001), and tracking showed no reliable link. Among novices, no within-group association reached significance, though the authors caution that the confidence intervals were wide, so these null results should not be read as evidence that the underlying associations are absent.

Because simple correlations can be confounded by age or the other cognitive predictor, the researchers ran a moderated regression as their primary inferential analysis, regressing tactical scores on standardized tracking and Corsi measures, expertise group, age, and the interaction terms. The four-term interaction block was significant, F(4, 80) = 3.331, p = 0.014, and the model accounted for 52.2 percent of the variance in tactical performance. Direct pairwise slope contrasts, Holm-adjusted within each predictor, showed that the tracking–decision slope was steeper in experts than in intermediate players (ΔB = 7.210, adjusted p = 0.043), and the working-memory–decision slope was steeper in intermediate than in expert players (ΔB = 6.682, adjusted p = 0.042). The other pairwise slope contrasts did not survive multiplicity correction, meaning the clearest regression evidence concerned expert–intermediate differences rather than a unique separation of any single group from both others.

Sensitivity analyses bolstered the main contrasts. Spearman rank correlations reproduced the two principal associations, heteroskedasticity-robust HC3 standard errors retained the expert–intermediate slope differences, and influence diagnostics including Cook’s distance, leave-one-out correlations and DFBETAs showed the interaction finding was not driven by a handful of influential observations. The authors also report a power caveat: with 30 participants per group, the study had roughly 80 percent power only for correlations of about 0.48 or larger, so smaller effects could plausibly have been missed.

The theoretical backdrop matters for interpreting these patterns. Accounts of chunking, long-term working memory, perceptual attunement and automaticity all suggest that extensive domain-specific practice changes how relevant information is encoded, organized and linked to action selection. The authors explicitly note that they did not formally test a sequential pathway in which tracking feeds working memory and working memory feeds decisions: no mediation or temporal-order model was fitted, and tracking and Corsi span were not significantly correlated within any group. They also stress that expertise-related differences in these associations need not imply a developmental reorganization of cognition; they could reflect differences in task demands, accumulated domain knowledge, perceptual familiarity or selection effects.

The limitations are laid out candidly. The cross-sectional design precludes conclusions about developmental change or causal direction — tactical expertise could shape cognitive-task performance, cognitive capacities could contribute to tactical performance, or both could reflect other training- and selection-related factors. The tracking and Corsi tasks are domain-general laboratory measures that may not capture basketball-specific tracking of players, spacing and tactical affordances. The TDM rubric’s intermediate categories were not independently validated against an external criterion, the tactical task was always administered last so order or fatigue effects cannot be excluded, and all participants were young, male and right-handed, limiting generalizability. The authors treat the patterns as hypothesis-generating signals for assessment design rather than prescriptive training recommendations, and they emphasize that the data do not establish that training tracking or working memory would improve tactical decision-making. Larger longitudinal studies using validated basketball-specific cognitive tasks, they conclude, are needed to determine whether these association patterns replicate and change with expertise development.

Subject of Research: The relationship between multiple-object tracking, visuospatial working memory and tactical decision-making across basketball expertise levels.

Article Title: Basketball expertise is associated with distinct relationship s among dynamic visual tracking, visuospatial working memory, and tactical decision-making

Article References: Yu, J., Xin, C., & Diao, Y. (2026). Basketball expertise is associated with distinct relationship s among dynamic visual tracking, visuospatial working memory, and tactical decision-making. Frontiers in Psychology, 17, Article 1988470. https://doi.org/10.3389/fpsyg.2026.1988470

Image Credits: AI Generated

DOI: 10.3389/fpsyg.2026.1988470

Keywords: basketball, multiple-object tracking, visuospatial working memory, tactical decision-making, sport psychology, perceptual-cognitive expertise, Corsi block-tapping, expertise, visual attention, cross-sectional study, athletes, cognitive performance