Description
Responses to bimodal signals are typically faster than those to unimodal signals, a phenomenon known as the redundant signals effect (RSE). To explain the RSE, Raab (1962) proposed the race model, suggesting that responses to two redundant signals are determined by the faster of two parallel decision processes. Despite the simplicity, race models exhibit remarkable explanatory power, successfully predicting response time variations based on stimulus onset and signal strength manipulations. However, Miller (1982) demonstrated that the observed RSE often exceeds race model predictions, leading to small but systematic violations of the race model inequality (RMI). These violations have long been interpreted as evidence against race models, challenging their adequacy in fully capturing multisensory integration. Here, we introduce the relay model, a novel extension of Raab’s framework that reconciles these milestone findings. In this model, two race units operate sequentially, with the first unit providing a start signal for the second, which can be considered a mechanism analogous to two-factor authentication in the digital world. This simple modification preserves the explanatory power of traditional race models while naturally accounting for violations of Miller’s bound. Constrained only by unimodal response time distributions, we demonstrate that the relay model captures all critical aspects of the RSE, with RMI violations emerging as an intrinsic property of relay-based processing. This framework offers a new perspective on multisensory decision-making and refines our understanding of the computational principles underlying the RSE.| Period | 18 Jul 2025 |
|---|---|
| Event title | 23rd International Multisensory Research Forum |
| Event type | Conference |
| Location | Durham, United KingdomShow on map |
| Degree of Recognition | International |