Trading mental effort for confidence in the metacognitive control of value-based decision-making

Abstract

Why do we sometimes opt for actions or items that we do not value the most? Under current neurocomputational theories, such preference reversals are typically interpreted in terms of errors that arise from the unreliable signaling of value to brain decision systems. But, an alternative explanation is that people may change their mind because they are reassessing the value of alternative options while pondering the decision. So, why do we carefully ponder some decisions, but not others? In this work, we derive a computational model of the metacognitive control of decisions or MCD. In brief, we assume that fast and automatic processes first provide initial (and largely uncertain) representations of options' values, yielding prior estimates of decision difficulty. These uncertain value representations are then refined by deploying cognitive (e.g., attentional, mnesic) resources, the allocation of which is controlled by an effort-confidence tradeoff. Importantly, the anticipated benefit of allocating resources varies in a decision-by-decision manner according to the prior estimate of decision difficulty. The ensuing MCD model predicts response time, subjective feeling of effort, choice confidence, changes of mind, as well as choice-induced preference change and certainty gain. We test these predictions in a systematic manner, using a dedicated behavioral paradigm. Our results provide a quantitative link between mental effort, choice confidence, and preference reversals, which could inform interpretations of related neuroimaging findings.

Evidence Accumulates for Individual Attributes during Value-Based Decisions

“…All model comparison results reported above were qualitatively identical (and quantitatively similar) when using final rather than initial ratings. The only difference was that M1 gained a small amount (6%) of support when using post-choice ratings (consistent with the idea that the accuracy of overall value ratings was refined during choice deliberation; (Lee & Daunizeau, 2021)).…”

“…See Figure 2 for an illustration of the different tasks. In addition to the aforementioned datasets, we also examined data from an unpublished pilot auxiliary task that were originally collected along with the primary data reported in (Lee & Daunizeau, 2021). In the main study, participants provided overall value ratings for 148 food options, then made 74 choices between pairs of options.…”

Locus coeruleus neurons encode the subjective difficulty of triggering and executing actions

“…In line with previous studies, we assumed that RTs could be affected both by sensory effects (how difficult it was to identify the target stimulus), motor effects (how difficult it was to execute the action), and cognitive effects, namely the amount of cognitive control required to trigger the action in the current condition (here, as a function of the natural tendency to disengage from the task in the current condition) [ 33 ]. Critically, the increase in RT in conditions associated with greater cognitive control could reflect both the difficulty itself and/or the mobilization of resources (i.e., time) in order to overcome that difficulty [ 7 , 34 , 35 ]. To capture the potential influence of sensory-motor and cognitive effects on RT, we compared behavior across 3 tasks manipulating both sensory-motor and reward parameters across conditions ( Fig 1 ).…”

Filling the gaps: Cognitive control as a critical lens for understanding mechanisms of value-based decision-making

“…People thus choose choice strategies that balance the value of the outcome itself against the (cognitive) cost of obtaining it (D. G. Lee & Daunizeau, 2021;Falk Lieder & Griffiths, 2017;Shenhav et al, 2013). This could for instance drive people to satisfice rather than put in the effort to maximize rewards, as they for instance increasingly do as their option set grows (Thomas et al, 2021).…”

“…Confidence in value-based choices integrates multiple cues related to choice difficulty (De Martino et al, 2013), changes of mind (Folke et al, 2016), goal congruency of options (Sepulveda et al, 2020), and value certainty (Boldt, Blundell, et al, 2019;D. Lee & Coricelli, 2020;D. G. Lee & Daunizeau, 2021).…”