Abstract
The allocation of visual processing capacity is a key topic in studies and theories of visual attention. The Load Theory of Lavie (1995) has proposed that allocation happens in two stages where processing resources are first allocated to task-relevant stimuli and then remaining capacity 'spills over' to task-irrelevant distractors. Kyllingsbæk, Sy, and Giesbrecht (2011) previously showed that the two-stage allocation scheme is not valid and instead showed that processing capacity is allocated in a single step. Here we test another critical assumption made by Load Theory: task-relevant and task-irrelevant stimuli are sharply distinguished, usually based on spatial location, and this information is accurate and computed rapidly before the two-stage capacity allocation scheme is engaged. To test this assumption, six participants performed a flanker search task that varied in load (e.g., Lavie & Cox, 1997). We then constructed two models based on the Neural Theory of Visual Attention (Bundesen, Habekost, & Kyllingsbæk, 2005). The first model embodied the Load Theory assumption, such that location information was available immediately. In the second model, location information was processed in parallel with the processing of stimulus identity. Contrary to Load Theory, we find that the second model in which location information distinguishing task-relevant and task-irrelevant stimuli from each other is processed slowly provided a more accurate fit of the data. Our alternative model provides a detailed computational account of how bottom-up and top-down information is integrated to provide efficient attentional selection and allocation of perceptual processing resources to task-relevant and task-irrelevant information.
Meeting abstract presented at VSS 2014