How do lesion studies elucidate the role of the hippocampus in intertemporal choice?

In a thought-provoking article, Boyer (2008) hypothesized that the remarkable human feat of imagining the future may be a crucial driving force in guiding goal-directed behavior. According to his proposal, episodic future thinking serves as a footbrake to offset the natural human tendency to engage in nearsighted behaviors–those that have short but not necessarily long-term gains. In this regard, episodic future thinking may be an intrinsic “look before you leap” mechanism; you may be more likely to take the future into consideration if you first imagine what it would be like. An obvious implication of Boyer’s proposal is that episodic future thinking has adaptive value for humans. In support of this idea, two intertemporal choice studies have shown that individuals indeed discount future monetary rewards less (i.e., are more likely to choose a larger future reward over a smaller present reward) when they first imagine consuming these rewards in a future event (e.g., spending the money at a sporting event; Benoit et al., 2011; Peters and B€ uchel, 2010). These studies also provide evidence that this “cueing effect” is supported by brain structures previously implicated in episodic future thinking, particularly the hippocampus (reviewed in Schacter and Addis, 2009). Beyond the relevance to normal cognition, an additional implication of these findings is that patient populations with reduced episodic future thinking capacity may be more vulnerable to making maladaptive choices. Several recent studies have explored this topic in patients with amnesia due to medial temporal lobe damage and concomitant episodic future thinking deficits (Kwan et al., 2012, 2013, in press; Palombo et al., 2015). In this issue, Kwan et al. (in press) report a study in which amnesic patients and healthy controls were asked to make intertemporal choices (e.g., $50 now or $100 in 1 month) under two conditions: a standard condition and one that involved episodic future thinking cueing. Healthy controls demonstrated the expected attenuation in temporal discounting following episodic future thinking cueing, replicating prior work (Benoit et al., 2011; Peters and B€ uchel, 2010). Yet, the majority of their amnesic participants also demonstrated this attenuation, despite the fact that many of them performed poorly on an episodic future thinking task. In our own laboratory, we have also investigated the role of episodic future thinking cueing in intertemporal choice in amnesic patients with medial temporal lobe lesions (Palombo et al., 2015). As in Kwan et al., our healthy controls showed the expected attenuation in temporal discounting following episodic future thinking cueing. However, in contrast to their findings (but consistent with our prediction), our patients’ intertemporal choices were not altered by cueing. The discrepant findings between these studies are puzzling, given that the patients in the two studies have broadly similar demographic, neuropsychological, and neuroanatomical profiles. One possibility is that these differences may be due to the nature of the cues used to elicit episodic future thinking. Whereas we asked participants to imagine engaging in specific hypothetical events (e.g., imagine spending $54 at a bar in 1 month), Kwan et al. asked their participants to imagine real-life plans through the provision of day-planners and calendars (e.g., imagine your granddaughter’s birthday party in 1 month). Kwan et al. suggest that the highly personal nature of their cues may have enabled amnesic patients to leverage semantic future thinking instead of episodic future thinking. The notion that decision making can be supported by different types of future thinking, only some of which depend on the hippocampus, is an intriguing possibility that requires future empirical work. Critically, such work will require independent verification of the strategies participants bring to bear on envisioning the future during decision making. To this end, our study examined whether cueing future events in our participants indeed elicited episodic future thoughts (Palombo et al., 2015). We found that it did: Both patients and controls largely provided descriptions of specific events, albeit patients’ 1 Memory Disorders Research Center, VA Boston Healthcare System, Boston, Massachusetts; 2 Department of Psychiatry, Boston University School of Medicine, Boston, Massachusetts; 3 Department of Psychology, Wellesley College, Wellesley, Massachusetts *Correspondence to: Daniela J. Palombo, Memory Disorders Research Center, VA Boston Healthcare System, 150 South Huntington Ave, 151-A Boston, MA 02130, USA. E-mail: palombo@bu.edu Accepted for publication 16 February 2015. DOI 10.1002/hipo.22433 Published online 23 February 2015 in Wiley Online Library (wileyonlinelibrary.com).