I'm happy to announce the publication of "Learning optimal decisions with confidence" with André Mendonça, @zmainen, @pouget_alex, about learning to make better decisions that require the temporal accumulation of evidence. Get it at pnas.org/content/early/… 1/ One normative approach to modeling such decisions is diffusion models, which commonly have one- to two-dimensional inputs. We increased the number of inputs, and asked how decision feedback can be used to learn how to weigh the various inputs. 2/

We're happy to announce that our work (with @satohirotajima, @Nisheet9, and @pouget_alex) on the optimal policy for multi-alternative decisions is finally out: nature.com/articles/s4159… (ReadCube link: rdcu.be/bM2Ej). 1/7 We asked how one would optimally trade off the speed and accuracy of a decision when having to decide between multiple alternatives. 2/7

We are out with a new, slightly more technical than usual, preprint on how to find closed-form solutions to Fokker-Planck equations of two-dimensional correlated race-to-boundary models arxiv.org/abs/1907.03341. Congratulations to Haozhe Shan and @MorenoBote! [1/4] These kind of models can be used to model decision-making, or neural activity, with two correlated sources of evidence/inputs. In the limit of perfectly anti-correlated sources, these models become equivalent to drift diffusion models. [2/4]