How many channels does it take to integrate features?

Watson and Robson (1981) showed that observers can discriminate large frequency differences at detection threshold, but require more contrast to discriminate fine differences. They suggested that the extra contrast allows the observer to use two channels. However, critical band masking of identification of letters, objects, faces, and gratings always reveals a single channel (Majaj et al. ARVO '98). Although appropriate stimuli reveal any channel at any location, it seems that each particular stimulus is identified by only one channel at each location. We ask observers to identify objects composed of two Gabor patches that differ in position and frequency. Correct identification requires integration of information from both components. We find that fine discrimination of orientation and frequency is impaired if, and only if, the components have similar position and different spatial frequency.