A neural mechanism mediating the impact of episodic prospection on farsighted decisions

Abstract

Humans can vividly imagine possible future events. This faculty, episodic prospection, allows the simulation of distant outcomes and desires. Here, we provide evidence for the adaptive function of this capacity and elucidate its neuronal basis. Participants either imagined specific events of spending money (e.g., £ 35 in 180 days at a pub), or merely estimated what the money could purchase in the scenario. Imagining the future biased subsequent monetary decisions toward choices associated with a higher long-term pay-off. It thus effectively attenuated temporal discounting, i.e., the propensity to devalue rewards with a delay until delivery. Using functional magnetic resonance imaging, we implicate the medial rostral prefrontal cortex (mrPFC) in this effect. Blood oxygen level-dependent signal in this region predicted future-oriented choices on a trial-by-trial basis. Activation reflected the reward magnitude of imagined episodes, and greater reward sensitivity was related to less discounting. This effect was also associated with increased mrPFC-hippocampal coupling. The data suggest that mrPFC uses information conveyed by the hippocampus to represent the undiscounted utility of envisaged events. The immediate experience of the delayed reward value might then bias toward farsighted decisions.

Figure 1.

Trial structure. Each trial started with a cue indicating the task (i.e., Imagine, Estimate) and the scenario (e.g., pub). Next, the delayed reward option was presented (e.g., £35 in 180 days). For the following 14 s participants either vividly imagined a specific episode that involved spending the money or estimated what the money would purchase in the scenario. Afterward, they indicated their preference for either the delayed reward option or a smaller fixed amount (£25) that they would receive immediately after the session. Participants then rated (1) their experienced emotional intensity during the task interval and (2) their feeling of vividly experiencing a coherent episode. The trial concluded with the presentation of a fixation cross that remained onscreen until 28.5 s after trial onset.

Figure 2.

Behavioral results. A, Phenomenological ratings for (1) the feeling of experiencing an episode and (2) the emotional intensity while imagining (white) or estimating (gray) (p values from two-tailed paired-samples t tests). B, Discounting behavior as quantified by the ratios of (1) the frequency of delayed options chosen versus all options chosen (i.e., delayed plus immediate) (choice index), and (2) the accumulated reward over trials versus the maximal possible reward (reward index) (p values from one-tailed t tests). C, Correlation between the Consideration of Future Consequences scale (CFC) and the ratios of the reward indices for the Imagine versus the Estimate task (p value from two-tailed test). Error bars indicate the SEM.

Figure 3.

Neuroimaging results. A–C, BOLD signal increases for the Imagine versus the Estimate task (A), the parametric modulation by reward magnitude in the Imagine versus Estimate task (B), and the conjunction of both contrasts (C). (For display purposes, the SPM is thresholded at p < 0.001, uncorrected, at least 10 contiguous voxels.) D, E, Parameter estimates, averaged from a 5 mm sphere centered on the peak of the conjunction contrast, for the task effect (D) and the parametric modulation of the Imagine task by reward magnitude (E). Connected blue diamonds indicate the mean-centered natural logarithm of the reward magnitudes. Error bars reflect the SEM.

Figure 4.

Brain–behavior relationships. A, B, Correlations between (1) BOLD signal sensitivity to reward magnitude and (2) the ratios of reward indices in the Imagine versus the Estimate task. C, A PPI analysis revealed stronger coupling between mrPFC and right hippocampus during the Imagine task for those participants who also exhibited a greater effect of episodic prospection on discounting. (For display purposes, the PPI image is thresholded at p < 0.005, uncorrected, at least 5 contiguous voxels.) dTC, Dorsal temporal cortex.

Figure 5.

Subsequent choice analysis. A, B, The only region showing greater BOLD signal during trials of the Imagine task that were followed by choices of the delayed versus immediate option. (For display purposes, the SPM is thresholded at p < 0.001, uncorrected, at least 10 contiguous voxels.) C, Parameter estimates, averaged from a 5 mm sphere centered on the peak of the subsequent choice effect. Error bars indicate the SEM.