Research Article: Concreteness and emotional valence of episodic future thinking (EFT) independently affect the dynamics of intertemporal decisions

Date Published: May 28, 2019

Publisher: Public Library of Science

Author(s): Cinzia Calluso, Annalisa Tosoni, Loreta Cannito, Giorgia Committeri, Alessio Avenanti.


During intertemporal decisions, the value of future rewards decreases as a function of the delay of its receipt (temporal discounting, TD). Since high discount rates have been associated with a series of problematic behaviours and clinical conditions, current research has focused on possible modulators of TD. Specifically, a reduction of individual discount rates has been shown during episodic future thinking (EFT), wherein time intervals are anchored to personal future events. However, it is not entirely clear whether this effect is mediated by a change in the representation of future events (i.e., from abstract to concrete) or by a positive-emotion modulation. Here, we investigated this issue by manipulating the valence of the EFT (i.e., using negative, neutral and positive episodic tags), and by collecting explicit and implicit measures of behaviour. The results showed a significant reduction of TD in all the three emotional conditions compared to the baseline, with differences among them, thus suggesting the existence of a cumulative effect of the concreteness and affective components of the EFT. The analyses of implicit measures additionally revealed that this effect was mediated by a simultaneous increase/decrease of attraction toward the delayed/immediate alternative. Finally, these effects appeared to be modulated by participants’ baseline discounting preferences. These findings provide important insights on clinical applications in reward-related disorders.

Partial Text

The preference for smaller sooner rewards over larger delayed ones, known as temporal discounting, is not only a commonly observed behaviour during many daily-life situations but also the predominant choice pattern of a series of sub-optimal behaviours and clinical conditions [1–4]. Prior research indeed has demonstrated that steep delay discounting predicts initiation of smoking in adolescence [1,5,6], and that individual differences in intertemporal choice tasks are associated with dysfunctional behaviours (including alcoholism [7,8], drug abuse [9], pathological gambling [10,11], credit card borrowing [12]) and life-style habits [13].

A conspicuous amount of studies in the last few years have shown the attenuating effect of episodic future thinking on temporal discounting [25,31,32,35–38,68–73]. Despite the enthusiasm driven by the possibility to change individual preferences in discounting behaviour, however, the modulation elicited by episodic future thinking has also been discussed with some degree of scepticism. Indeed, it is not entirely clear how this effect works. Specifically, on the one hand, it has been proposed that this effect is a consequence of a change in the time construal associated with future time delays (i.e., from high level, abstract representations to low-level, concrete representations). On the other hand, since most of the previous works have used only positive tags, it has also been suggested that this modulation is guided by the positive emotions elicited by the episodic tags. Notably, moreover, even though some studies have tried to investigate this issue by introducing an emotional valence manipulation, the results are not yet conclusive, maybe due to the high inter-subject variability of the choice behaviour, which particularly affects between-subjects designs [35–38]. Furthermore, it should be noted that the procedure for EFT manipulation has not been always consistent across different studies. For example, while in some previous studies on episodic tagging participants were explicitly asked to imagine the future reward consumption in the circumstances signalled by the tag [31,71] or to imagine the future self immediately before the task [74], our procedure for the episodic tagging, which was based on the study by Peters & Büchel (2010) [25], did not involve an active/explicit mental reactivation of the tags. Additionally, with respect to the previous studies, here we manipulated the emotional valence of episodic future thinking using a within-subjects’ design in order to disentangle the contribution of concreteness and emotional valence to the observed effect of enhanced farsightedness induced by the EFT manipulation. To this aim we recorded both explicit (choice outcome) and implicit (mouse kinematics) measures. The results showed that: i) discount rates were significantly reduced in all the three emotional valence conditions with respect to baseline, with differences among them; ii) the effect of the EFT manipulation on subjective values was observed at both short and long time delays; iii) the effects were mediated by an increased attractiveness toward the delayed choice option and a decreased attractiveness toward the immediate one; iv) the effect of the EFT manipulation was differential in discounters vs. farsighted subjects (i.e., it was modulated by baseline discounting preferences) both at the level of the explicit and the implicit kinematic measures.

In line with relatively recent evolutionary considerations [77], the present results provide an additional piece of evidence in favour of the hypothesis that mental time travel and imaginative foresight may be crucial cognitive mechanisms for human decision making. In particular, according to this hypothesis, the ability to imagine and anticipate future events is thought to oppose the natural human tendency toward discounting, impulsive and opportunistic behaviour, nudging individuals towards choices which become advantageous in the long term [77]. Furthermore, the current implicit and explicit results also underlie the role of emotional valence in shifting individual discounting preferences, thus bringing support to integrated cognitive-affective models of intertemporal decision [14,15] and offering important insights into the clinical domain of reward-related disorders.