The authors of the target article have done an admirable job of describing the current understanding of the concept of mental time travel (MTT). Especially important is the creation of a taxonomy of future-oriented cognition because it will serve to further guide research in this area. However, the behavioral correlates of MTT (and specifically MTT into the future) require elaboration. Although the authors are clear about the various components of MTT, they are somewhat less clear about the various future-oriented behaviors that MTT undergirds. According to Suddendorf & Corballis (S&C), the sine qua non of MTT (into the future) is the ability to act in the present in anticipation of a need or state that is not currently experienced (e.g., imagining being thirsty when one is quenched). We are sympathetic to this view and agree that it is a powerful test of MTT. In fact, some of our recent research has explored this ability from a developmental perspective (Atance & Meltzoff Reference Atance and Meltzoff2005; Reference Atance and Meltzoff2006). But, there are at least three reasons why neither we nor S&C should fixate on this one example alone.

Other markers of future-oriented behavior

First, as research with both adults (e.g., Gilbert et al. Reference Gilbert, Gill and Wilson2002) and children (e.g., Atance & Meltzoff Reference Atance and Meltzoff2006; Suddendorf & Busby Reference Suddendorf and Busby2005) demonstrates, this particular variety of MTT is extremely difficult – especially when one must anticipate a future state that directly conflicts with a current one as was the case in Atance and Meltzoff (Reference Atance and Meltzoff2006). This might best be viewed as a high-end marker of MTT. But other markers also exist. For instance, it would also seem that the myriad of planning behaviors in which humans engage, along with the capacity to delay gratification and to remember to perform a future intention (i.e., prospective memory), all reflect MTT. Also important is whether there exist future-oriented skills crucial to human functioning that do not involve MTT. For example, the planning required to succeed on the Tower of Hanoi (ToH) task is quite sophisticated, but it is unclear whether it requires the individual to actually mentally project the self into the future. On the other hand, planning what one will have for breakfast the next day seems much less taxing than planning moves on the ToH but may draw heavily on MTT. Developmental studies would provide useful evidence regarding which of these different tasks “cluster together.” For instance, is the child who successfully reports what he plans to do tomorrow the same child who delays gratification now in order to obtain a larger reward in the future? Or, is the child who is able to report what she'll do tomorrow the same child who performs well on the ToH? Answers to such questions may aid in determining which behaviors reflect MTT, at which point in development they emerge, and provide hints about the neural systems required.

The importance of current desires for future success

Second, S&C's criterion that a person acts in the present in anticipation of the future makes sense as applied to such visceral states as hunger and thirst, but it may not fully elucidate the cognitive complexities that underlie our strivings for more elusive states such as happiness, or our achievements of long-term goals such as job stability and the like. One of the most adaptive features of future thinking in humans is that we can hold in mind a desire or goal state and work tirelessly to achieve it. Part of the reason that this is possible is precisely because we experience the motivating desire or goal in the present. For example, some instances of prospective memory rely on the ability to translate a current desire into a plan for the future. That is, my (current) desire to take my medicine at 7 p.m. (future) ensures that at 7 a.m., I place my medicine beside my drinking glass. Similarly, when a child or adult is able to delay gratification, the desire for the larger (or end) reward is present. Arguably, it is when we lose sight, or stop experiencing, these drives and emotions that we cease to work towards them in a future-directed manner. Thus, it is important to consider not only how the anticipation of a future state can drive behavior in the present, but also how a current desire can ensure that an action, which brings the organism closer to a future goal, is carried out.

Partial knowledge and development

Third, there is the issue of partial knowledge. Even by age 5, children have not developed all the components that S&C argue to be necessary for MTT. Are we to infer that prior to this age children are incapable of MTT? Although young children may not be able to consistently act in a manner that takes into account their future states, it seems overly conservative to argue that they are not capable of any form of MTT. True, some instances of MTT may require more advanced imaginative, executive, or theory of mind skills, but not all are required for a legitimate instance of MTT. For example, it seems plausible that a young child's inability to tell you what he's going to do tomorrow is not due to underdeveloped executive function skills but perhaps because of the division of labor in the family – mother does this planning. Careful developmental research would illuminate this and related issues. One could measure individual differences in executive function abilities and determine whether these predict individual children's performance on various MTT tasks.

Developmental science is crucial for constructing a more comprehensive theory of MTT. It will be important to develop tasks that not only test the child's ability to anticipate future states but also measure a number of other future-directed behaviors. In the early days of theory of mind research, a narrow fixation on children's understanding of false belief obscured the importance of many behaviors that form the foundation of a theory of mind, including joint attention, social referencing, and reasoning about desires and intentions. We should not make the same error again in the domain of future thinking.