Neither the number of feature dimensions nor their task relevance necessarily affect interocular grouping in binocular rivalry.

When one eye views half of stimulus A and half of stimulus B, and the other eye views the respective other halves, perception fluctuates. Among percepts reflecting monocular inputs (indicating typical binocular rivalry), percepts of undivided stimuli A and B occur. They indicate the grouping of parts of inputs from both eyes together, a process called "interocular grouping". Our study investigated if, similar to binocular rivalry, this process is affected by attentional modulation stemming from manipulating the task relevance of the stimulus features driving it. We used stimuli in which form (object identity), motion (motion direction), or both could drive interocular grouping. Contrary to our hypothesis, the grouping was not affected by designating either of these dimensions as task relevant. Moreover, unlike in previous studies, we did not observe stronger grouping when two features, instead of one, drove it. However, we found that our manipulation of task relevance influenced observers' eye movements, which, in turn, was related to the number of perceptual changes the observers reported. We interpret this pattern of results as suggesting that the dynamics of perceptual fluctuations differed depending on the task. Overall, we show that the factors influencing interocular grouping strength are more complex than previously thought and not necessarily aligned with the factors influencing binocular rivalry. The data from this study can be accessed via the following link: https://doi.org/10.5281/zenodo.18924437 .