Abstract
Lavie's perceptual load theory (1995; Lavie & DeFockert, 2001; Lavie & Tsal, 1994) holds that observers can efficiently filter out task-irrelevant distractors when performing under high levels of perceptual load (e.g., large display size), but fail to do so under low levels of perceptual load (e.g., small display size). On this account, perceptual load modulates selectivity by way of limiting or freeing processing capacity. When load is high, capacity is exhausted and there are no available resources left to process irrelevant distractors; while the reverse holds true for low load. Central to load theory (Lavie, 1995) are critical hypotheses regarding the construct of capacity and its dissociation from sensory and perceptual processes (Lavie & DeFockert, 2001). Inferences on capacity changes under different levels of load have been made only at the level of mean RTs. The present work subjected this theory and its core assumptions to rigorous scrutiny in three ways: (a) By testing hypotheses regarding changes in processing capacity using the hazard function of the RT distributions, allowing a very fine-grained assessment of processing capacity (Townsend & Wenger, 2004; Townsend & Ashby, 1978; Wenger & Gibson, 2004). (b) By testing assumptions regarding sensory mechanisms using signal detection theory (Green & Swets, 1966). (c) By applying these analyses at the level of the individual observer approach in order to assess the extent to which individual differences in critical characteristics of processing may vary (e.g., Estes & Maddox, 2005). Our results indicate that (a) there are considerable individual differences in capacity as a function of varying perceptual load, and (b) there are significant violations of some of the foundational assumptions of perceptual load theory, such as the assumption that capacity limitations are independent of sensory limitations.