Authors
Jessica M. Moon, [1] Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA, [2] Lindenwood University, College of Science Technology and Health, St. Charles, MO, USA
John Pinette, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
Aubrey Fontenot, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
Violette Gibbs, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
Destiny Northcutt, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
Trevor J. Dufner, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
Adam J. Wells, Exercise Physiology, Intervention, and Collaboration Lab, School of Kinesiology and Rehabilitation Sciences, University of Central Florida, Orlando, FL, USA
International Journal of Exercise Science 19(4): 4004, 2026.
Abstract
Previous research has demonstrated that the Dynavision D2 Mode A reaction time (RT) assessment requires extensive familiarization to reach peak performance beyond initial training effects. This suggests that other commonly used D2 assessments, such as the Mode B test, may require familiarization periods to accurately assess changes in performance. This distinction is critical to ensure that observed changes reflect true adaptations rather than continued task learning, particularly for practitioners monitoring performance or recovery. This study examined the effects of repeated training and detraining on reliability and performance on Mode B tests. Twenty-six recreationally active adults completed 15 training sessions (two Mode B tests per session), followed by a detraining period of one or two weeks, before completing another 15 training sessions. Total hits (Hits), number of misses, average RT per hit (AvgRT), and the number of numerical T-scope challenges correct (TSC) were used to determine performance. Reliability was assessed using intraclass correlation coefficients (ICC), standard error of measurement (SEM), and minimal detectable change (MDC). Significant improvements in performance (p <.05) were observed across the first 24 tests, with no further changes thereafter. Hits ICC to .968 and standardize spacing. Corrected sentence: “Hits (ICC = .968), AvgRT (ICC = .958), and misses (ICC = .902) demonstrated excellent test-retest reliability, while TSC (ICC = .785) demonstrated good reliability. There were no performance differences based on detraining duration during the retraining period. These findings indicate that at least twenty-four tests are necessary to establish a reliable baseline and minimize training effects when Hits, AvgRT, Misses, and TSC are used as performance outcomes to avoid misinterpretation.