Key points are not available for this paper at this time.
Steep temporal discounting is a hallmark of a range of mental disorders, including substance use disorders and addiction. Episodic cueing during temporal discounting tasks reliably increases the selection of larger-but-later options, but the specific computational mechanisms through which episodic processes impact decision-making have remained elusive. Here, we addressed this issue using hierarchical Baysian computational modeling via temporal discounting drift diffusion models in a pre-registered multi-day study (https://osf.io/rhz2f/). Participants (n=20 males, n=21 females, n=1 non-binary) completed temporal discounting tasks with either future or past cues (compared to control trials), on two separate testing days. Model-agnostic analyses confirmed the prediced increase in larger-later choices following both future and past cueing. Computational modeling revealed that, in addition to reducing the discount rate, both future and past cues reliably increased decision thresholds in a temporal discounting drift diffusion model with non-linear drift rate scaling. Results highlight thatin addition to reducing the discount rate, episodic cueing elicits dynamic control adjustments that serve to counteract the impact of noise in the evidence accumulation process, thereby facilitating selection of options with better long-term value.
Brands et al. (Sat,) studied this question.