A short note on the reliability of perceptual timing tasks as commonly used in research on developmental disorders

Perceptual timing tasks are frequently applied in research on developmental disorders, but information on their reliability is lacking in pediatric studies. We therefore aimed to assess the reliability of the four paradigms most frequently used, i.e., time discrimination, time estimation, time production, and time reproduction.

Based on the data from our recent longitudinal study by Marx et al. (Front Hum Neurosci 11:122, 2017), we estimated the internal consistency and test-retest reliability of these tasks in children with ADHD and typically developing children. Individual thresholds were used as dependent measures for the time discrimination task, whereas absolute error and accuracy coefficient scores were used for the other three tasks.

Although less commonly used, the time estimation paradigm was the most robust measure of perceptual timing in terms of internal consistency and test-retest reliability in both ADHD and typically developing children, whereas the most frequently used paradigms showed poor internal consistency (time reproduction) and poor test-retest reliability (time discrimination). Compared to the absolute errors, accuracy coefficients showed almost exclusively higher internal consistency and test-retest reliability.

Our findings call for more frequent use of the time estimation paradigm in studies of perceptual timing in ADHD. The time reproduction paradigm should be re-considered, avoiding pooling of a wide range of time intervals (2-48 s). The accuracy coefficient score is the more reliable and the more intuitive dependent variable and should be preferred in future timing research. To increase the reliability of the timing measurement, each experimental session should be performed twice, if possible.