Multiphasic value biases in fast-paced decisions

Abstract

Perceptual decisions are biased toward higher-value options when overall gains can be improved. When stimuli demand immediate reactions, the neurophysiological decision process dynamically evolves through distinct phases of growing anticipation, detection and discrimination, but how value biases are exerted through these phases remains unknown. Here, by parsing motor preparation dynamics in human electrophysiology, we uncovered a multiphasic pattern of countervailing biases operating in speeded decisions. Anticipatory preparation of higher-value actions began earlier, conferring a "starting point"- advantage at stimulus onset, but the delayed preparation of lower-value actions was steeper, conferring a value-opposed buildup rate bias. This, in turn, was countered by a transient deflection toward the higher value action evoked by stimulus detection. A neurally-constrained process model featuring anticipatory urgency, biased detection, and accumulation of growing stimulus-discriminating evidence, successfully captured both behavior and motor preparation dynamics. Thus, an intricate interplay of distinct biasing mechanisms serves to prioritise time-constrained perceptual decisions.

Balancing true and false detection of intermittent sensory targets by adjusting the inputs to the evidence accumulation process

“…3A, upper ). This is consistent with previous observations of an evidence-independent buildup component in signals related to motor preparation attributed to gradually increasing urgency ( Cisek et al, 2009 ; Hanks et al, 2014 ; Murphy et al, 2016 ; Standage et al, 2011 ; Thura et al, 2012 ; Thura and Cisek, 2014, 2016 ), seen sometimes in anticipation of evidence ( Corbett et al, 2021 ; Kelly et al, 2021 ). This growing urgency component, which in evidence-accumulation models combines with cumulative evidence in a decision variable building towards a constant action-triggering threshold, is mathematically equivalent to, and a well established neural implementation of, a collapsing decision bound ( Churchland et al, 2008 ; O’Connell et al, 2018 ).…”

“…Similar to Corbett et al ( 2021 ), the models were fitted in several stages: In the first stage, a minimal model was fitted that included all relevant free parameters for that model (see above) but only one free drift rate parameter for the Weak Coherence, which was then linearly scaled for the Strong conditions according to relative coherence (70/25). In this initial broad search, about 4,500 trials were simulated per model evaluation in separate SIMPLEX fits for 1,000 different initial “guess” parameter vectors.…”