Lingo
← Back to Neuroscience

Race Models

Overview

Race Models represent a broad class of decision-making frameworks where different response options are modeled as entirely independent accumulators. Rather than evidence for one choice subtracting from another (as in the Drift Diffusion Model), each alternative "races" toward its own threshold. The first accumulator to cross its threshold determines both the choice made and the response time.

Independent Accumulation

The defining feature of a pure race model is independence . Evidence supporting Option A does not interact with evidence supporting Option B. This architecture easily scales to multiple-choice decisions (e.g., 3, 4, or 10 options) simply by adding more accumulators.

Classic examples include the Vickers Accumulator Model and various Poisson counter models. Unlike the LBA, standard race models often incorporate within-trial noise (diffusion), and unlike the LCA, they lack lateral inhibition.

Statistical Facilitation

Race models naturally predict a phenomenon known as statistical facilitation (or the redundant targets effect). If a task presents two redundant targets that can both trigger a response, the overall response time is faster than the response time to either target alone. This occurs because you are taking the minimum completion time of two independent racing processes.

References

  1. Vickers (1970) — Ergonomics— Evidence for an accumulator model of psychophysical discrimination.
  2. Ratcliff (1978) — Psychological Review— Statistical facilitation and the classic race model formulation.

Examples