Optimal multisensory decision-making in a reaction-time task

Abstract

Humans and animals can integrate sensory evidence from various sources to make decisions in a statistically near-optimal manner, provided that the stimulus presentation time is fixed across trials. Little is known about whether optimality is preserved when subjects can choose when to make a decision (reaction-time task), nor when sensory inputs have time-varying reliability. Using a reaction-time version of a visual/vestibular heading discrimination task, we show that behavior is clearly sub-optimal when quantified with traditional optimality metrics that ignore reaction times. We created a computational model that accumulates evidence optimally across both cues and time, and trades off accuracy with decision speed. This model quantitatively explains subjects's choices and reaction times, supporting the hypothesis that subjects do, in fact, accumulate evidence optimally over time and across sensory modalities, even when the reaction time is under the subject's control.DOI: http://dx.doi.org/10.7554/eLife.03005.001

Learning optimal decisions with confidence

“…Bayes-optimal decision-making with diffusion models One standard way (Drugowitsch et al 2012;Drugowitsch, Deangelis, et al 2014;Moreno-Bote 2010) to interpret diffusion models as mechanistic implementations of Bayes-optimal decision-making is to assume that, in each trial, a latent state (or drift rate in the terminology of diffusion models) is drawn from a prior distribution, ∼ N(0, -. ), with zero mean and variance -.…”

“…Therefore, efficient decisions require not only a mechanism to accumulate evidence, but also one to trigger a choice once enough evidence has been collected. An exceedingly popular model family known as diffusion models (or diffusion decision models; DDMs) (Ratcliff 1978) provides both mechanisms, and does not only yield surprisingly good fits to human and animal behavior (Bogacz et al 2006;Ratcliff and McKoon 2008;Ratcliff and Smith 2004), but is also known to implement the Bayes-optimal decision strategy under a wide range of circumstances (Bogacz et al 2006;Drugowitsch et al 2012;Drugowitsch, Deangelis, et al 2014;Frazier and Yu 2008;Gold and Shadlen 2002;Tajima, Drugowitsch, and Pouget 2016). Diffusion models assume a particle that drifts and diffuses until it reaches one of two boundaries, each triggering a different choice (Fig.…”

Distinct encoding of decision confidence in human medial prefrontal cortex

“…http://dx.doi.org/10.1101/251330 doi: bioRxiv preprint first posted online Feb. 14, 2018; visual-haptic percept, regardless of which response may eventually be required (47). Choiceindependent estimates of certainty can also be useful when the reliability of different cues varies with time and must be updated independently (48,49). However, after committing to a choice, it may be more efficient to combine these representations into a statistic that summarises the probability that a choice is correct (1).…”

How Optic Flow and Inertial Cues Improve Motion Perception

“…Psychophysical studies of heading discrimination using two-alternative forcedchoice tasks have reported vestibular heading discrimination thresholds in darkness that are as small as a few degrees Fetsch et al 2009;Butler et al 2010Butler et al , 2015de Winkel et al 2010;Drugowitsch et al 2014). Such threshold values are comparable (although larger) with those described in visual heading discrimi-nation tasks (Warren and Hannon 1990;Royden et al 1992;van den Berg and Brenner 1994;Stone and Perrone 1997).…”