Working memory, inhibitory control, and cognitive flexibility are the core components of executive functions, which play a crucial role in sports and physical exercise by influencing problem-solving, logical reasoning, and decision-making. Physical activity also triggers neurobiological changes in the brain, such as increased oxygenation, neurotransmitter release, and elevated body temperature. Therefore, developing tests that assess specific executive function components during exercise is essential for understanding brain behavior in this context. This study aimed to evaluate the reliability and sensitivity of the computerized Digit Span task in assessing working memory during physical exercise. Eleven healthy men participated on three days with a one-week interval between visits. Participants completed questionnaires, anthropometric measurements, and a maximal cardiopulmonary exercise test during the first visit. On the second and third visits, they familiarized themselves with the computerized digit span before engaging in a 30 minute moderate exercise protocol. The task was then performed during exercise for an average of 10 minutes. The results demonstrated good reliability for minimum heart rate (HR) and excellent reliability for maximum and average HR. However, the reliability of the computerized digit span measures was moderate. Sensitivity analysis showed that typical error of measurement values were lower than the smallest worthwhile change thresholds of 0.2, 0.6, and 1.2 for the variables correct trials × span, correct trials, and Span, respectively. This study focused on adapting the traditional digit span for use during physical exercise, evaluating its reliability and sensitivity. HR measures remained consistent, but task reliability was moderate, particularly in the backward condition. These findings highlight the need for methodological improvements to enhance the task's reliability in sports contexts.