Abstract
The neural control of pointing movements is strongly associated with the dorsal processing stream in primate visual cortex. Since the dorsal stream exhibits very little selectivity for color, it was our interest to test the precision of manual pointing movements towards isoluminant objects, whose locations are defined only by color differences. Small Gaussian blobs of different colors and varying saturation were presented in random positions on an isoluminant gray background on a computer monitor. Subjects had to point to these targets with their index finger as accurately and as fast as possible. A touchscreen was used to measure the accuracy of pointing. Additionally, we measured reaction time and movement time. In addition to pointing to the targets, subjects also had to identify their color. This way, we determined psychometric functions for color naming and for pointing accuracy. We found that pointing accuracy for the colored objects was equal to accuracy for objects defined by luminance contrast. The only difference between color and luminance was a 10 msec latency difference that was also seen when subjects performed a simple reaction time task where the motion was not directed to a visual target. Furthermore, the functions relating pointing accuracy and color naming to saturation were identical. Subjects could accurately point to the targets only in those trials where they also could correctly name the color of the object. We conclude that in humans color signals can be efficiently used by the motor system to guide pointing movements.