With respect to the described paradigm, we can think of three outcomes. 1. In addition to surface features, motion streaks help to establish object correspondence by identifying the targets’ spatiotemporal position. Motion streaks improve the performance at any stage, the maximal performance is only achievable when a motion streak is present and thus both, surface and spatiotemporal information is processed. In that case, we predict that the proportion of correct identified targets will increase and secondary saccade latency will get shorter. 2. Especially at earlier mask onset times – where surface information is temporally limited, and thus less informative – the motion streak information has a more important influence as an additional information to establishing object correspondence. The proportion of correct identified targets will be better and the secondary saccade latencies will get shorter when a streak is present in earlier onset times. When the onset times of the mask appearing gets longer, thus we have more time to process surface information, the spatiotemporal information of the streak is getting less important for identifying the right target. At one point we reach an asymptote where all secondary saccades are directed to the original target and no faster latencies are possible. The less information we have about the surface, the more important is the streak. 3. Motion streaks do not have any function in establishing object correspondence; our visual apparatus does not process the motion information, thus the perceived streak has no influence on the proportion of correct identified targets or secondary saccades latency. Only surface information is relevant for establishing object correspondence. Proportion of correctly identified targets will improve and secondary saccade latency will get faster if more surface information can be processed due to longer masking onset times.