Expectations of reward and efficacy guide cognitive control allocation

Abstract

The amount of mental effort we invest in a task is influenced by the reward we can expect if we perform that task well. However, some of the rewards that have the greatest potential for driving these efforts are partly determined by factors beyond one’s control. In such cases, effort has more limited efficacy for obtaining rewards. According to the Expected Value of Control theory, people integrate information about the expected reward and efficacy of task performance to determine the expected value of control, and then adjust their control allocation (i.e., mental effort) accordingly. Here we test this theory’s key behavioral and neural predictions. We show that participants invest more cognitive control when this control is more rewarding and more efficacious, and that these incentive components separately modulate EEG signatures of incentive evaluation and proactive control allocation. Our findings support the prediction that people combine expectations of reward and efficacy to determine how much effort to invest.

Is the juice worth the squeeze? Learning the marginal value of mental effort over time

“…To this point, we have also found suggestive evidence that the relationship between task-evoked pupillary responses and task switch costs is stronger in low-NFC individuals (da Silva Castanheira et al, 2020). Interestingly, our results also dovetail with the recent observation that the efficacy of cognitive effort-the environmental contingency between performance and reward receipt, holding demand constant-is a critical determinant of individuals reward-induced effort investment in a Stroop-like task (Frömer et al, 2021). When efficacy, which was explicitly signaled to participants before each trial, was high, participants were more inclined to invest effort in the task in accordance with varying incentive level (as evidenced by faster and more accurate responses) compared to when efficacy was low.…”

“…The resource-limited nature of cognitive processing prescribes that people should exert cognitive effort only when it is worthwhile-that is, when the costs incurred by expending additional effort are justified by the benefits it may confer (Hull, 1943;Kool & Botvinick, 2018;Shenhav et al, 2017). For example, reward incentives mobilize cognitive processing resources across diverse cognitive control tasks (Bijleveld et al, 2010;Frömer et al, 2021;Hübner & Schlösser, 2010;Otto & Vassena, 2020;Padmala & Pessoa, 2011), which is thought to result from a cost-benefit tradeoff calculation (Shenhav et al, 2017): increasing the benefits tied to effort exertion offsets the cognitive costs of exerting effort. People also tend to avoid exertion of cognitive effort when given the choice, in line with the idea that mentally demanding behavior is experienced as costly (Kool et al, 2010;Vogel et al, 2020;Westbrook et al, 2013), and further buttressing the notion that effort allocation arises from a cost-benefit calculus.…”

Apparent preferences for cognitive effort fade when multiple forms of effort and delay are interleaved in a foraging environment

“…Even though participants decreased their acceptance rates for effort in both experiments, we would have expected participants in Experiment 2 to increase engagement with Wait trials, substituting Wait trials for effort trials in order to preserve their earnings. Our observations are unlikely to be due to the risk of failing to successfully complete effortful trials, as participants remained accurate throughout both experiments, and efficacy was consistent across cost conditions and experiments (Frömer et al, 2021). Finally, we cannot conclude that cognitive effort gained value from its rewarding outcomes, as has been proposed in the past (Eisenberger & Cameron, 1996).…”

Bored by bothering: A cost-value approach to pandemic boredom

“…First, when considering the impact of value and effort on recommended behaviors (first part of the proposed mediation model), the results revealed a more potent role of value for behavior (see for example Frömer et al 2021). Individuals are more likely to restrict their behavior if they value this choice.…”