Abstract
The visual system can encode many stimuli across the visual field simultaneously, but the number of objects that can be fully identified in parallel is limited. At the extreme, some stimuli must be identified serially. One useful tool for distinguishing parallel from serial processing is the redundant target paradigm, which compares responses to displays containing one target to displays containing two targets. Parallel models generally predict a redundancy gain: faster responses to two targets. The standard serial model (improved to incorporate response accuracy as well as speed) predicts either no effect or slower responses to two-target than to one-target displays. We conducted several experiments that measured performance for three different judgments of written words: color detection (detecting colored letters), lexical decision (detecting real words among pseudowords), and semantic categorization (detecting nouns that refer to living things). In one set of experiments, the words were above and below fixation. In another set, the words were to the left and right of fixation to better match natural reading. The color detection tasks yielded a strong and positive redundancy gain, while the lexical and semantic tasks yielded zero or negative effects. These results are consistent with low-level features (color) for two words being processed in parallel, while the meanings of two words are processed serially. Altogether, this study informs models of reading and suggests opportunities for the redundant target paradigm to investigate other aspects of vision.