Keynotes | K&Ts | GACs | Talks | Posters | Search

Poster B in Poster Session B: Tuesday, August 4, 2:00 – 3:45 pm, Kimmel Center, Shorin & Rosenthal Rooms

Speed vs. Accuracy in aDDM Inference: Where Shortcuts Fail!

Andrew Zhang¹, Alexander Fengler¹, Sicheng Liu¹, Matthew Harrison¹, Michael Frank¹; ¹Brown University

Presenter: Andrew Zhang

Attentional drift diffusion models (aDDMs) provide a mechanistic account of how visual attention dynamically influences value-based decision-making. However, full Bayesian inference has been computationally intractable due to within-trial drift rate dynamics precluding analytic likelihood estimates. This intractability lead researchers to rely on time-averaged drift approximations (TADA), where drift rate dynamics were reduced to a single "equivalent" drift. In previous work, Liu et al., (2026) demonstrated that TADA methods are statistically inconsistent and showed with targeted counterexamples how they can bias conclusions about attentional effects when the true data-generating process is the aDDM. They further derived an algorithm that speeds up proper aDDM likelihood computation by three orders of magnitude. Here we present an application and translation of this research. First, we provide a software integration bridging the gap toward practical application: our methods are now available to be used naturally via the probabilistic programming library PyMC. We extended our algorithm to an autodifferentiable JAX implementation, enabling gradient-based MCMC methods like NUTS. Second, we present a comprehensive, systematic Bayesian parameter recovery study comparing TADA against proper aDDM inference. This provides clear guidance for experimentalists on the consequences of using TADA and pinpoints specific risks to scientific conclusions when it is applied for computational convenience.

Topic Area: Methods, Tools, Theory & Neural Coding