Abstract
Transparent motion occurs when multiple objects move through the same region of space without total occlusion. Previous work demonstrates that when direction differences are the only cue to transparency, observers can detect no more than two transparent motion signals simultaneously. This limit appears to occur because transparent motion detection requires coherence levels much higher than those required for uni-directional motion. If this is the case, it should be possible to increase the number of signals that can be detected by increasing their signal intensities. We increased signal intensity by distributing processing of the signals between two speed-tuned global motion systems. Because these systems perform signal-to-noise operations independently, dividing the signals between them increased the effective intensities by decreasing the amount of signals processed by each system. Observers were required to indicate which of two temporal intervals contained the greatest number of transparent motion signals. Simultaneous processing was ensured through brief durations (200ms) and comparisons between n and n+1 signals, eg. 3 vs. 4. With no speed differences, observers could detect no more than two transparent signals. When some signals moved at a low speed and others at high speeds, up to three could be detected. Thus, by raising the effective signal intensity, the number of transparent motion signals that could be processed simultaneously was increased. This is consistent with the signal-to-noise processing basis of the transparency limit. However, were signal intensity the sole constraint, the addition of speed differences should have allowed the representation of up to four signals. The fact that the limit could only be extended to three indicates further limitations on transparent motion processing.