In the target article, Suddendorf & Corballis (S&C) formally propose a framework for linking cognitive/behavioral systems concerning memory (capacity for past experiences) with prospection (future-oriented behavior/cognition). Their primary argument is that the system, both for memory and prospection, is hierarchically organized according to levels of flexibility and singularity, and that the most flexible, “episodic” level heavily depends on the uniquely human ability of “mental time travel.” The argument is motivated largely by recent comparative studies of animal behavior and developmental studies in humans. Although these behavioral observations are amazing and extremely valuable for considering memory-prospection systems in humans and animals, a growing body of neurobiological studies also provides a seminal contribution to the theory. The target article refers to some of the most essential elements of such an approach, but seems to lack enough documentation on what the neurobiological data can suggest to the human ability of foresight. This approach has only recently begun to draw researchers’ attention, but a rapid expansion of both empirical data and theoretical concepts has taken place recently. In the following, I make some comments on the very recent advances in the neurobiology of human prospection.

S&C briefly introduce evolutionary change in the relative size of certain areas of the human brain, especially the prefrontal cortex, as a neurobiological index of differing capacity for mental time travel in humans and other animals. However, a more direct and informative way to understand how mental time travel functions in the brain is perhaps to seek distinguishable brain activity while participants are engaged in tasks requiring mental time travel. S&C mention potential involvement of cortico-basal ganglia loops in anticipation of future rewards evidenced by brain imaging studies employing tasks in which participants made a decision through learning of a stimulus-action-reward contingency. This series of processes involved with reward-based decision making is obviously one form of prospection, but, as was defined in the target article, it is best described as “procedural future cognition” that might be less flexible and stimulus-bound. That is, we cannot be sure whether the participants vividly imagine a particular future event of “gaining reward” in its unique detail upon deciding an action, when repeatedly presented with prototypical experimental stimuli such as geometrical shapes.

Probably a more direct approach to uncover the cerebral bases of mental time travel is brain imaging during stimulus-free, episodic recall. Okuda et al. (Reference Okuda, Fujii, Ohtake, Tsukiura, Tanji, Suzuki, Kawashima, Fukuda, Itoh and Yamadori2003) performed a simple but challenging experiment investigating brain activations while participants orally described particular events of certain time periods in the future and the past. They found that a basically similar brain network involving medial aspects of the prefrontal, temporal, and parietal cortices was activated during both thinking about the future and about past events, and that magnitude of the activity in each subregion was characteristically modulated by future/past orientation and temporal distance from the present.

Stimulated by this demonstration, several follow-up studies have recently been pursued. Addis et al. (Reference Addis, Wong and Schacter2007) controlled more strictly for phenomenological quality of remembered/imagined events so that they were truly “episodic” and that details, emotionality, personal significance, and field/observer perspective were constant across the future and past conditions. They utilized a technique of event-related functional magnetic resonance imaging (fMRI) to dissociate brain activity patterns during event construction (search and reconstruction of autobiographical event information) and elaboration (subsequent retrieval/imagination of supplementary details). Under such a well-controlled experimental setting, the common activation in the core network of the medial prefrontal-temporal-parietal cortices were observed particularly during the event elaboration, which might be considered a consequence of greater demand of the mental time travel during the event elaboration both for the future and the past. Szpunar et al. (Reference Szpunar, Watson and McDermott2007) also used event-related fMRI and confirmed common engagement of this core network in remembering/envisioning of self-relevant, autobiographical events in the past and future. Apart from the functional brain imaging, Hassabis et al. (Reference Hassabis, Kumaran, Vann and Maguire2007) have demonstrated striking evidence that amnesic patients with bilateral medial temporal damage cannot imagine new experiences in addition to their impairment in remembering past episodes. Importantly, the deficit was not restricted to personally relevant possible future events (e.g., a possible Christmas event) but was common to wide variety of commonplace scenarios (situation of standing in the main hall of a museum, etc.), with severe impairment in placing fragmented images into one coherent spatial context.

These sets of novel evidence and other related data have led Buckner and Carroll (2006) to propose that envisioning the future (prospection), remembering the past (episodic memory), conceiving the viewpoint of others (theory of mind), as well as shifting one's topographical perspective (some forms of spatial navigation) reflect the workings of the same core brain network, that is, the medial prefrontal-parietal-temporal network. Therefore, although the claim by S&C that evolutionary significance of episodic memory primarily lies in its adaptive advantage for future survival through flexible mental time travel can still be true, available neurobiological data appear to suggest that the mechanism should be extended to a broader context of “mental traveling,” not only in the time domain but also in the spatial context and others’ minds.

What, then, is the fundamental mechanism by which theses various domains of mental traveling work with the same core brain network? Buckner and Carroll (Reference Buckner and Carroll2007) assume “self-projection” as a basic facility. Burgess et al. (Reference Burgess, Gilbert, Dumontheil, Simons, Duncan, Phillips and MacLeod2005; Reference Burgess, Gilbert, Okuda, Simons, Prinz and Sebanz2006) propose a more explanatory model, although it only considers, at the least explicitly, a mechanism involving vicinity of the frontal part of the core regions. The model, based on careful reviews of both neuropsychological and functional imaging data, proposes that the anterior part of the prefrontal cortex acts as a gateway that modulates an attentional bias between stimulus-independent mental processes and stimulus-oriented immediate cognitions. It is plausible that cognitive operations, independent of environmental stimulus and motor responses to it, are the most fundamental requirement for mental travel to any other objectives that are not present in one's immediate perceptual-motor representations. This general idea is also confirmed by other brain imaging studies indicating roles of the anterior part of the prefrontal cortex in prospective memory or remembering mentally represented action plans during ongoing cognitive activities (e.g., Okuda et al. Reference Okuda, Fujii, Ohtake, Tsukiura, Yamadori, Frith and Burgess2007), as well as involvement of medial prefrontal and parietal cortices in stimulus-independent thought or mind-wondering (Mason et al. Reference Mason, Norton, Van Horn, Wegner, Grafton and Macrae2007a, but see also Gilbert et al. Reference Gilbert, Dumontheil, Simons, Frith and Burgess2007 and Mason et al. Reference Mason, Norton, Van Horn, Wegner, Grafton and Macrae2007b). Moreover, the argument of stimulus-independent and stimulus-oriented cognition is directly relevant to the distinction between “declarative” and “non-declarative” future prospection as discussed in the target article. Probably there is a corresponding neurobiological distinction between the two in the brain: The stimulus-oriented, non-declarative prospection may depend largely on cortico-basal ganglia loop, whereas the stimulus-independent, declarative prospection may recruit the medial prefrontal-parietal-temporal network. Then how can the two prospection systems coexist or switch one another? Again, the gateway model explains regulation between the two systems as a primary function of the anterior portion of the prefrontal cortex.

One final comment on the relation between the theater metaphor and the neurobiologically plausible mental traveling hypothesis I have raised earlier: How does the assumption of stimulus-independent mental traveling to other times, places, and individuals accommodate with the theater metaphor? It seems to me that the set, stage, and actor components roughly correspond to the time, place, and agent domains, respectively. Again, regulation between stimulus-oriented immediate cognition and stimulus-independent mental processes, possibly a role played by the executive producer, may be a critical process for well-organized mental traveling to function.