A Comparison of Methods for Measuring Interocular Delays

Observers often exhibit eye dominance, in which information from one eye is prioritized over the other, a phenomenon that can be observed across several different visual processes. The literature has largely emphasized spatial asymmetries between the eyes, but previous work also points to temporal asymmetries (i.e., interocular delays), particularly in clinical populations. Several psychophysical methods have been developed to measure interocular delays, yet the agreement between them remains unknown. We systematically compared five interocular delay measures in normally sighted observers: two depth-based judgement tasks (based on the Pulfrich effect), binocular flicker integration, interocular temporal order judgements, reaction time, and a sensory eye dominance measure. In Experiment 1, several methods showed good internal consistency, yet no significant agreement across tasks emerged. This suggests that naturally occurring delays in normally sighted individuals may be too small to yield reliable correlations. To increase the range of delays, Experiment 2 introduced a neutral density filter to induce a delay in the right eye. Filter-induced delays showed significant repeated-measures correlations across most methods. The structure-from-motion Pulfrich tasks showed the largest filter-induced delays, the lowest between-subject variability, and the highest split-half reliability. In contrast, temporal order judgements showed lower reliability and weaker agreement with other tasks, likely reflecting the limited temporal resolution inherent in these judgements. These findings indicate that while multiple behavioural methods can detect interocular delays, agreement across tasks depends on the magnitude of the delay. Task selection is therefore crucial, particularly in clinical populations where delays may be larger and more variable.