Abstract
The interaction of colour and motion cues for global motion integration across space has only recently been studied. To establish the tolerance limits of global motion we first determined global motion detection thresholds (81%) as a function of the chromatic contrast in the isoluminant cone-opponent colour-space (S-(M-L) space). We used random dot kinematograms with 300 coloured gaussian blobs (0.22°, 1°/s, 5.1° × 4°, 200cm viewing distance). Observers had to distinguish between an interval with random motion and an interval with 40% of the blobs moving either left or right (2IFC). Experiment 1: all blobs had the same colour and were presented on a grey background (luminance= 50cd/m2). Eight out of ten observers were not able to perform the task for S-cone isolating colours. In all other colour directions we found that the thresholds were determined by the stimulation of the red-green (R/G) mechanism. Experiment 2: we further tested this by adding chromatic noise (with a constant projection onto the R/G mechanism) to the red and green stimuli. We found that the width of the chromatic noise distribution did not affect the results and the thresholds were predicted by the projection onto the R/G mechanism. Experiment 3: To evaluate possible luminance artefacts we established observers' individual isoluminance by heterochromatic flicker and rerun Experiment 1. We obtained the same results. Experiment 4: To determine whether this is a single R/G mechanism, half the dots in the display were green and the other half red. We found that threshold coherence levels were not greater than 40%, indicating the presence of a single R/G mechanism. Conclusion: the S cone input in our global motion task is negligible for the majority of observers. Global motion extraction in the observer's isoluminant plane appears to be mediated exclusively by a single R/G mechanism andthe observed sensitivity to global motion is predicted by the projection onto this mechanism.
This research is supported by the WellcomeTrust.