Comparing Discounting of Potentially Real Rewards and Losses by Means of Functional Magnetic Resonance Imaging

The sign effect in temporal discounting does not require the hippocampus

When considering future outcomes, humans tend to discount gains more than losses. This phenomenon, referred to as the temporal discounting sign effect, is thought to result from the greater anticipated emotional impact of waiting for a negative outcome (dread) compared to waiting for a positive outcome (mixture of savoring and impatience). The impact of such anticipatory emotions has been proposed to rely on episodic future thinking. We evaluated this proposal by examining the presence and magnitude of a sign effect in the intertemporal decisions of individuals with hippocampal amnesia, who are severely impaired in their ability to engage in episodic mental simulation, and by comparing their patterns of choices to those of healthy controls. We also measured loss aversion, the tendency to assign greater value to losses compared to equivalent gains, to verify that any reduction in the sign effect in the hippocampal lesion group could not be explained by a group difference in loss aversion. Results showed that participants with hippocampal amnesia exhibited a sign effect, with less discounting of monetary losses compared to gains, that was similar in magnitude to that of controls. Loss aversion, albeit greater in the hippocampal compared to the control group, did not account for the sign effect. These results indicate that the sign effect does not depend on the integrity of hippocampally mediated episodic processes. They suggest instead that the impact of anticipatory emotions can be factored into decisions via semantic future thinking, drawing on non-contextual knowledge about oneself.

Envisioning the future: An ALE meta-analysis on neural correlates of future thinking, prospective memory and delay discounting

Our representations of the future are processed in the service of several different cognitive functions, including episodic future thinking, prospective memory, and temporal discounting. The present meta-analysis used the Activation Likelihood Estimation method to understand whether there is a core network underlying future-oriented cognition and to identify the specific brain regions that support future-related processes in each function. Following the PRISMA guidelines, a total of 24, 19, and 27 neuroimaging studies were included for future thinking, prospective memory, and temporal discounting, respectively. Results showed that there is no specific region or network for the future. Instead, the 'future' seems to be represented on an anterior-posterior tangibility gradient, based on the level of abstractness/concreteness of the simulated scenario. Additionally, future-oriented cognition is mediated by two distinct networks: the Default Network and the Salience Network. The Default Network is mainly active in supporting future thinking, whereas the Salience Network is primarily involved in prospective memory and delay discounting.