Individual differences in effort discounting and adjustments in volitional control

A central question of the control dilemma framework of the present CRC is which variables determine the settings of meta-control parameters in order to optimize performance. While, e.g., learning and environmental features are important factors, individual differences might also play a role in the modulation of cognitive control and the balance between complementary control modes. So far, major approaches have focused on inter- and intraindividual differences in the availability of an internal ressource, the depletion of which results in suboptimal performance. Recently, however, the idea that opimization of performance is largely driven by motivational parameters that result from cost-benefit calculations has received renewed attention and empirical support. In this context, differences between and within individuals in the degree to which control is exerted can be seen as differences in the representation of task difficulty and of the amount of rewards associated with competing goals. This may result in differences in the willingness to invest effort in the pursuit of a goal, given its costs and benefits. To generate insights into the neural and physiologic computations that underlie this so-called effort discounting phenomenon, approaches that explicitly model cost-benefit computations on the one hand and discounting on the other hand are required. The Expected Value of Control (EVC) model provides an ideal framework for this purpose, as it posits that the strength of a control signal that drives adjustments in cognitive control depends on cost-benefit calculations. Moreover, its formalization already includes weight terms for probability and delay discounting, but interestingly none for effort discounting. Based on recent studies on the relation between dispositional differences in motivational constructs and effort discounting and further evidence relating individual differences in these motivational constructs to performance differences in cognitive control tasks, we assume that individual differences in effort investment (assessed via self-report) and effort discounting (assessed via behavioral measures) may qualify as a supplemental weight term to the equations of the EVC model. In order to validate this assumption, a series of five experiments will be performed to address six research aims: (a) the nature of dispositional effort investment will be systematized via a large-scale online assessment of dispositional constructs related to cognitive motivation and self-regulation (Study 1); (b) it will be determined whether effort discounting can be considered as a disposition itself via an assessment of behavioral measures of effort discounting at two time points and latent-state-trait modeling (Study 2); (c) the relationship of (self-reported) effort investment and (behaviorally derived) effort discounting will be replicated in a more systematic way using more than one measure of effort investment, and discounting, respectively, and relating their trait-specific variance portions to each other (Study 2); (d) it will be examined whether individuals high in effort investment and effort discounting actually invest more effort in task processing, using effort-related measures derived from electroencephalography (EEG) and eye-tracking (Study 3); (e) the assumption that individual differences in effort investment and effort discounting moderate adjustments of cognitive control in a typical cognitive control task will be tested, again using EEG and eye-tracking (Study 4A) and functional magnetic resonance imaging (fMRI, Study 4B) to track the control demand imposed by varying task difficulty at different payoff levels, thereby manipulating both costs and benefits, which should be mirrored in physiological parameters assumed to index the need for control; finally (f) these control adjustments will be modeled based on individual differences in cost representations using the EVC framework (using data from Studies 4A and 4B), which in addition will be complemented by the so-called active inference framework (Study 5) to be developed within another CRC project. Overall, it is expected that the present project will provide a formalized account for the role of individual differences in effort discounting in the adjustment of cognitive control and thus will foster better predictions of individual control adjustments across a variety of tasks.

Project Members

Principal Investigators

Staff