Remembering the past and envisioning the future rely on episodic memory which enables mental time travel. Studies in young adults indicate that past and future thinking share common cognitive and neural underpinnings. No imaging data is yet available in healthy aged subjects. Using fMRI, we scanned older subjects while they remembered personal events (PP: last 12 months) or envisioned future plans (FP: next 12 months). Behaviorally, both time-periods were comparable in terms of visual search strategy, emotion, frequency of rehearsal and recency of the last evocation. However, PP were more episodic, engaged a higher state of autonoetic consciousness and mental visual images were clearer and more numerous than FP. Neuroimaging results revealed a common network of activation (posterior cingulate cortex, precuneus, prefrontal cortex, hippocampus) reflecting the use of similar cognitive processes. Furthermore, the episodic nature of PP depended on hippocampal and visuo-spatial activations (occipital and angular gyri), while, for FP, it depended on the inferior frontal and lateral temporal gyri, involved in semantic memory retrieval. The common neural network and behavior suggests that healthy aged subjects thought about their future prospects in the past. The contribution of retrospective thinking into the future that engages the same network as the one recruited when remembering the past is discussed. Within this network, differential recruitment of specific areas highlights the episodic distinction between past and future mental time travel.
Episodic memory is the only memory system that allows individuals to mentally travel in subjective time, into either the past or the future (Tulving, 2002 and Tulving, 2005). This ability depends on autonoetic consciousness which mediates an individual’s awareness of his or her existence and identity in subjective time. Converging lines of evidence from different fields of research indicate that remembering the past or envisioning the future share common cognitive and neural underpinnings. First, developmental studies suggest that the level of awareness for episodic remembering and the ability to identify with future interests develops around ages three to four (Atance and O’Neill, 2001, Levine, 2004 and Wheeler et al., 1997). Second, age-related changes seem to affect similarly the quality of past and future mental evocations, with older adults generating fewer details for past and future events compared to younger adults (Addis, Wong, & Schacter, 2008). Third, neuropsychological case studies have shown that patients with hippocampal lesions have difficulties in remembering their personal past, but also in foreseeing their personal future (patient KC, Tulving, 1985; patient DB, Hassabis et al., 2007 and Klein et al., 2002), their productions lacking in episodic details compared to age-matched controls (Addis et al., 2009 and Gamboz et al., 2010). Fourth, certain phenomenological characteristics similarly affect past and future mental thinking, such as positive emotional valence and temporally close events which are associated with a stronger feeling of re-experiencing or pre-experiencing (Addis et al., 2008, D’Argembeau and Van der Linden, 2004, D’Argembeau and Van der Linden, 2006 and Gamboz et al., in press). Most recently, a growing number of neuroimaging studies detect a common neural network when thinking about the past or the future (Buckner and Carroll, 2007, Hassabis and Maguire, 2007, Hassabis and Maguire, 2009 and Schacter and Addis, 2007).
This study showed that thinking about the past and the future recruits the same set of brains regions in healthy aged subjects reflecting the use of similar cognitive processes (visual imagery, semantic retrieval, episodic binding). When planning a future event beforehand, thinking retrospectively about it engaged the same regions as those recruited when thinking about a past event. However, past and future events differed on certain phenomenological aspects: past events were more episodic (i.e., contained more details), engaged a higher state of autonoetic consciousness and mental visual images were clearer and more numerous compared to future events. Regression analyses showed that the episodic nature of past events was particularly dependent on activation in the right hippocampus (critical in episodic recollection) and visuo-spatial areas (implicated in the retrieval of the spatial context of events), while the episodic nature of future events was dependent on the inferior frontal and lateral temporal gyri (involved in semantic retrieval). Furthermore, future events elicited greater activation in parietal regions compared to past events, reflecting the use of higher attentional resources to envision future scenarios. Although sharing common neural processes, past and future thinking differed most notably at the episodic level, likely due to the fact that past events have actually happened, while future events are mental constructions.
