A probabilistic, distributed, recursive mechanism for decision-making in the brain.

Javier A Caballero,Mark D Humphries,Kevin N Gurney,Jean Daunizeau

doi:10.1371/journal.pcbi.1006033

Javier A Caballero, Mark D Humphries + Show 2 more

Open Access

https://doi.org/10.1371/journal.pcbi.1006033

Copy DOI

Abstract

Decision formation recruits many brain regions, but the procedure they jointly execute is unknown. Here we characterize its essential composition, using as a framework a novel recursive Bayesian algorithm that makes decisions based on spike-trains with the statistics of those in sensory cortex (MT). Using it to simulate the random-dot-motion task, we demonstrate it quantitatively replicates the choice behaviour of monkeys, whilst predicting losses of otherwise usable information from MT. Its architecture maps to the recurrent cortico-basal-ganglia-thalamo-cortical loops, whose components are all implicated in decision-making. We show that the dynamics of its mapped computations match those of neural activity in the sensorimotor cortex and striatum during decisions, and forecast those of basal ganglia output and thalamus. This also predicts which aspects of neural dynamics are and are not part of inference. Our single-equation algorithm is probabilistic, distributed, recursive, and parallel. Its success at capturing anatomy, behaviour, and electrophysiology suggests that the mechanism implemented by the brain has these same characteristics.

Highlights

Decisions rely on evidence that is collected for, accumulated about, and contrasted between available options
During the dot motion task, neurons in the middle-temporal visual area (MT) respond more vigorously to visual stimuli moving in their “preferred” direction than in the opposite “null” direction [24]
The multi-hypothesis sequential probability ratio test (MSPRT) is a feedforward algorithm. It cannot account for the ubiquitous presence of feedback in neural circuits and, as we show ahead, for slow dynamics in neural activity that result from this recurrence during decisions

Summary

Introduction

Decisions rely on evidence that is collected for, accumulated about, and contrasted between available options. Neural activity consistent with evidence accumulation over time has been reported in parietal and frontal sensorimotor cortex [1,2,3,4,5], and in the subcortical striatum [6, 7]. Multiple models of decision making match aspects of recorded choice behaviour, associated neural activity or both [8,9,10,11,12,13,14,15,16]. While successful, they lack insight into the underlying decision mechanism. Other studies have shown how exact inference algorithms may be plausibly implemented by a range of neural circuits [17,18,19,20,21]; none of these has reproduced experimental decision data

Methods

Results

Discussion

Conclusion