An Accumulator Model Account of Semantic Interference in Memory Retrieval

To explain latency effects in picture-word interference tasks (Glaser & Düngelhoff, 1984, cf. Stroop, 1935), cognitive models need to account for both interference and SOA effects. As opposed to most ACT-R models, which model the time course in a ballistic manner, the RACE model (Retrieval by ACcumulating Evidence) presented here accounts for semantic interference during the retrieval interval. This way, RACE offers a better account of semantic memory retrieval latencies in different interference and SOA conditions. We describe the architectural assumptions of RACE and simulations of a Glaser and Düngelhoff experiment, illustrating the validity of the RACE model.