Exploring the relationship between spatial and semantic attention: Two limited capacity systems or one?

Abstract

Attention is associated with benefit and cost due to competition for limited capacity resources involved in access for awareness. Spatial cues guide spatial attention producing benefit for attended stimuli and cost for unattended stimuli. Semantic relatedness shows a similar pattern: primes produce benefit for related and cost for unrelated stimuli, suggesting they work through semantic attention. An unanswered question is how semantic and spatial attention interact: Does the benefit and cost for related stimuli occur because they attract spatial attention? Three accounts of the relationship between semantic and spatial attention were examined: Semantic attention a) works through spatial attention, b) delays disengagement of spatial attention, and c) is independent of spatial attention. In Experiments 1 to 3, participants performed a visual search task for a prime target in a word display and/or a probe discrimination task on a probe target. Experiments 1a and 1b showed that the word and probe tasks were suitable for examining semantic and spatial attention, respectively. Experiments 2a and 2b showed that in the absence of explicit influences on spatial attention, related distractors led to higher word search task accuracy on target present and lower accuracy on target absent trials. Importantly, related distractors did not attract spatial attention as measured by probe task performance, although they did slow responses on target absent trials. Experiment 3 showed that related distractors also do not attract spatial attention when an abrupt onset spatial cue is presented. Instead, spatial attention was slower to disengage from related versus unrelated distractors on target absent trials. Experiments 4 and 5 examined spatial and semantic attention using the temporal order judgment paradigm in which participants judged which of two stimuli occurred first (or second). Related words and abrupt onset spatial cues had different effects on performance: Relatedness did not influence