Evaluation of the Psychometric Properties of the Gap-Overlap Task in 10-Month-Old Infants

Understanding the development of attention is key to understanding cognitive maturation. The gap-overlap task can be administered at all ages and is widely used to study the development of overt visual attention. However, studies using the gap-overlap task report different measures and little is known about the tasks’ psychometric properties, especially in infants. We tested the 1-week test–retest reliability of two frequently used gap-overlap measures of attentional disengagement in 10-month-old infants; the gap effect as measured by the difference between the gap and overlap condition and the gap effect as measured by the difference between the gap and baseline condition. Sixty-seven infants performed the gap-overlap task twice, of which 45 infants had sufficient data quality for further analyses. Test–retest reliability of the overlap-gap gap effect was higher (r = .50) than the baseline-gap gap effect (r = .29). Moreover, the shared variance between overlap and baseline saccadic reaction times was moderate to high across sessions. In light of these results and the methodological challenges and limitations of infant research, we consider the overlap-gap gap effect to be a good measure to study the development of attentional disengagement in infants and suggest the exclusion of the baseline condition in future studies.