Thursday, February 15, 2007

Remembering the Past to Imagine the Future

By: Karl K. Szpunar and Kathleen B. McDermott Ph.D.

Remembering experiences in our past and imagining ourselves in some future event both involve a kind of mental "time travel" based in our brains.

Remember for a moment a recent gathering with friends. Who was there? When and where did the event take place? Now imagine a future gathering with friends. Who is there? When and where does the event take place? As with mentally reliving the past, mental representations of the future include images of personal episodes taking place in specific settings.

According to cognitive psychologist Endel Tulving, Ph.D., the ability to envision specific future scenarios (called episodic future thought) may be closely related to the ability to recollect specific episodes from our past (that is, episodic memory). Indeed, evidence from neuropsychology, clinical psychology, and developmental psychology indicates that people who cannot remember specific details from their past also appear to be impaired in their ability to mentally envision personal future experiences.

Efforts to understand the relationship between memory and future thought are in their infancy. The potential payoff, however, is considerable: it will likely inform our basic understanding of human memory, as well as of brain disorders in which that most essential human ability is impaired. 

Evidence from Damaged Brains

Consider the patient known in the scientific literature as KC. KC, who has been studied extensively by Tulving and his colleagues at the University of Toronto,1 has global amnesia caused by diffuse brain damage that he sustained in a motorcycle accident. Many of KC’s cognitive abilities are intact, but he can neither remember any single episode from his past nor project himself mentally into the future. When asked to do either, he states that his mind is “blank”; when asked to compare the kinds of blankness in the two situations, he says it is the “same kind of blankness.”

DB, another brain-injured patient who has been studied by Stanley Klein, Ph.D., and his colleagues at the University of California at Santa Barbara, exhibits a similar profile. Following DB's cardiac arrest, his brain sustained damage due to lack of oxygen. DB can no longer recollect his past, nor can he project himself into the future.2 Both KC and DB have self-concepts consistent with descriptions An inability to envision a “brighter future” may contribute to sustaining depression.  of their personalities given by others who know them well. Although they cannot remember specific events from their own pasts, their overall self-knowledge (for example, “I am generally comfortable in social settings”) is reliable and can even be changed by new experiences. Moreover, both patients understand the concept of time: they know that there is a future and a past, they can tell time with an analog clock, and they know about their past and their future in a vague sense. What they lack is the ability to perform mental time travel.

Eleanor Maguire, Ph.D., and her colleagues at University College, London, have recently replicated what was observed with KC and DB and extended the observations in a more systematic fashion.3 Five amnesic patients were tested for their ability to form mental images of novel experiences that might take place in the future in a familiar setting, such as a possible event in their lives over the next weekend. These patients were markedly impaired in their ability to do this—their mental images were vague and highly fragmented when compared to those reported by a control group of people of the same average age and level of education but without amnesia.

Another example can be found in people with severe depression. Researchers have known for some time that such individuals have difficulty in bringing to mind personal details from their past. Mark Williams, Ph.D., and his colleagues at the University of Wales showed in 1996 that people with clinical depression are also impaired in their ability to engage in episodic future thought, a discovery that may have important implications for understanding how prolonged depressive states are maintained. For example, an inability to envision a “brighter future” may contribute to sustaining depression.  

Developing Mental Time Travel

A growing line of research has demonstrated that episodic memory emerges in children sometime between the ages of four and five. Around this age children display evidence of vividly recollecting details related to their memories, as opposed to simply remembering something in a more general sense. For example, a three-year-old is likely to have greater difficulty than a five-year-old in remembering which of two uncles gave him a treat the day before, even though both children would be quite confident that the event in question had occurred.

Recently, evidence has begun to accumulate suggesting that the ability to project oneself into the future emerges in concert with the ability to vividly recollect the past. Thomas Suddendorf, Ph.D., and his colleagues at the University of Queensland have shown that at about age five, children begin to be able to accurately report what they will or will not do in the future (for example, tomorrow), as well as what they have or have not done in the past (for example, yesterday).4 

Seeking How Past and Future Are Related

These examples all suggest that the personal past and future are related in some way, but what is the nature of the relationship? Understanding this relation should provide important insights into brain disorders in which both past and future thought are absent and also into the functioning of the healthy mind in general. One possibility is that envisioning the future involves sampling the contents of our memories. In fact, Arnaud D’Argembeau, Ph.D., and his colleagues at the University of Liege have shown that people tend to envision past and future events in similar contexts.5 For instance, if someone is asked to imagine a future event related to attending a lecture, he is likely to imagine himself sitting in a familiar classroom. What that classroom looked like, how big it was, and so on would play a large role in the imagined future scenario. One reason that amnesic patients, people with severe depression, and young children are unable to envision future events may be that they cannot rely upon memory for visual-spatial information when they attempt to construct images of the future.

In order to understand more about how this relationship between the past and the future operates, we recently conducted a functional neuroimaging study at Washington University in St. Louis.6 We were specifically interested in whether thinking about the future and remembering past episodes involved similar regions of the brain.

Functional neuroimaging techniques, such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI), allow neuroscientists to examine brain activity associated with mental activity. When participants in a research study engage in a given cognitive task, PET or fMRI can provide information about the level of cerebral blood flow (PET) or blood oxygenation level (fMRI) in the particular parts of the brain involved in performing the task. These techniques measure aspects of metabolism, rather than directly measuring the activity of neurons, but the information they provide allows researchers to infer which brain regions are involved in a given task.

In the typical design of a neuroimaging study, brain activity associated with two tasks is contrasted with the hope of isolating the brain regions that are important for the cognitive process of interest. In most cases, researchers attempt to contrast a pair of tasks that are similar to one another but vary in one key way. For instance, in order to identify the brain regions that are important for episodic memory, many researchers choose to contrast an episodic memory task with one that does not involve a specific personal episode. An episodic memory task might require a person to recount his experiences on the day of his college graduation. The comparison task might involve simply stating the name of the college he attended. Both tasks require a person to retrieve a personal memory, but naming the college does not involve recollecting experiences at a specific time and place in the past.

Our goal was to identify brain regions that might be important for representing oneself in time and then to examine those regions to see whether or not they are similarly engaged by past and future thought. To accomplish this goal, we asked study participants to perform a set of three tasks while lying in an fMRI scanner. In two of these tasks, participants viewed a series of event cues (for example, birthday party) and were asked to envision either a personal memory of that kind of event or one that might take place in the future. Brain activity common to both tasks (past and future) was contrasted with that observed during a third task. This task involved similar processes, such as mental construction of lifelike scenarios, but did not involve representing oneself in time. Our specific control task required participants to imagine former U.S. president Bill Clinton. Clinton was chosen because pretesting showed that he is easy to visualize in a variety of situations. 

The Brain During Past and Future Thought

As a result of our study, we learned that several regions in the brain’s posterior cortex were similarly engaged during personal past and future thought, but not during the control task. As shown in Figure 1, these regions were located in the occipital cortex (A), the posterior cingulate cortex (B), and the medial temporal cortex (C). We found this very interesting, because previous research had shown that these same regions are consistently engaged during such tasks as autobiographical memory and mental navigation of familiar routes. We hypothesize that asking study participants to envision a personal future scenario likely requires similar processes. That is, in order to effectively generate a plausible image of the future, participants reactivate images (such as a familiar lecture hall) stored in the posterior cortical regions. Questionnaires that study participants filled out after the experiment corroborated this hypothesis: participants tended to imagine future scenarios in the context of familiar settings and familiar people.

Hypotheses based on exploratory research, such as identifying the neural relation of past and future thought, are significantly strengthened if the results can be replicated. Indeed, Donna Rose Addis, Ph.D., and Daniel Schacter, Ph.D., at Harvard University recently presented a similar set of data.7 In their study, they demonstrated that mental construction of specific past and future episodes led to equivalent activity within brain regions similar to those that we identified.

One salient difference between past and future thought is that recollecting a personal memory requires reconstructing, as a whole, an event that has already taken place, but imagining a future event requires actively and continuously constructing a new scenario. The event has not yet taken place, so it is up to the individual to decide on a moment-to-moment basis where the event is taking place, who is there, and what they are doing. This difference was also demonstrated in our research. 

fMRI memory scans 
Figure 1. Brain regions showing similarities (blue) and differences (red) during episodic future thought and remembering.  Regions showing similarities appear within the superior occipital cortex (A), posterior cingulate cortex (B), and medial temporal lobes (C). Regions showing differences appear within the lateral premotor cortex (D), medial posterior precuneus (E), and right posterior cerebellum (F).

Courtesy of Karl Szpunar and Kathleen McDermott. 


[Caption for illustration]

Figure 1. Brain regions showing similarities (blue) and differences (red) during episodic future thought and remembering.  Regions showing similarities appear within the superior occipital cortex (A), posterior cingulate cortex (B), and medial temporal lobes (C). Regions showing differences appear within the lateral premotor cortex (D), medial posterior precuneus (E), and right posterior cerebellum (F).

Courtesy of Karl Szpunar and Kathleen McDermott. 

As can be seen in Figure 1, we learned that several brain regions were more engaged when participants envisioned the future than when they recollected the past, although in both cases they were more involved than during the control task. These regions were found within the lateral premotor cortex (D), the medial posterior precuneus (E), and the posterior cerebellum (F). Previous research has suggested that each of these regions plays an important role in imagining a sequence of behavioral actions (for example, imagining tapping one’s fingers or retracing one’s jogging route). In short, we found that the act of mentally creating a future behavioral sequence—on the fly—appears to engage such brain regions to a greater extent than simply reflecting on a sequence of actions that have already taken place. Addis and Schacter also reported similar patterns of activity.

We were interested to find that not a single brain region was more engaged while recollecting the past than during thinking about the future. Addis and Schacter, whose research corroborated this, reported that the hippocampus (a structure important for pulling together the contents of memory to form a coherent mental image of the past) was more engaged during future thought than during recollection, and Maguire and her colleagues have also suggested that the hippocampus might be important in imagining future scenarios.  

Insights from Neuroimaging

Advanced neuroimaging techniques can provide new insights into long-standing questions about cognition. Researchers who study memory have long known that a variety of factors influence how well we are able to learn (that is, encode) new information. For instance, a person who is asked to tie a new piece of information to knowledge about herself will generally remember that information better than someone who does not engage in such self-referential processing. Participants in a memory experiment who rate a list of adjectives (for example, “old”) for self-descriptiveness will remember the list better than participants who think about the adjectives in a purely objective manner (for example, One of the most common complaints associated with aging is memory loss. But do the elderly experience similar difficulties in thinking about the future? deciding whether “old” means the same as “aged”). Until recently, this effect on encoding has only been inferred from memory tests showing that information processed during learning in a self-relevant manner is remembered better during a later test than information that did not have this added “boost.” Neuroimaging techniques now allow researchers to examine the influence of such variables during the actual learning phase. By observing brain activity associated with learning, we can find out whether there are any unique neural signals that may enhance encoding when people relate incoming information to themselves. This will be an exciting area for future research and has implications for education and learning of all kinds.

We have seen how brain regions known to be important for reinstating past experiences appear to play an important role in constructing future personal episodes. This discovery confirms what makes intuitive sense: thinking about the personal past and thinking about the personal future are closely related activities and involve similar processes. Of course, the story is never completely black and white, because we have also observed brain regions that show differences in neural activity as study participants think about the past and the future. Converging results from many studies will surely enhance understanding of how our brains discriminate memories of the past and imagination of the future. 

Looking to the Future

Beginning with the pioneering work of Hermann Ebbinghaus,8 a late-nineteenth-century German psychologist, students of psychology and neuroscience have expended more than 100 years of thought and careful experimentation toward an understanding of human memory. However, they have made surprisingly little inquiry into our ability to mentally represent the future. The coming years will surely see an increase in the number of studies of how the brain enables us to mentally represent future thought and how future thought may be related to memory. Interesting questions and answers will likely arise from a wide range of research disciplines.

For example, one of the most common complaints associated with aging is memory loss. But do the elderly experience similar difficulties in thinking about the future? The answer to this question could have important implications for programs designed to improve older adults’ health-related behavior. For instance, older adults are better able to remember to take their medication if they have previously imagined performing the task. This implies that they can imagine the future, but how well? If older adults do have deficits in thinking about the future, would it be possible somehow to enhance their ability to do so?

As another example, think about your friends and colleagues. Is there someone who stands out as having a particularly good memory? What about someone with poor memory? People differ greatly in their capacity for remembering. What about future-directed thought? Are some people better able to imagine specific instances in their future and, consequently, better able to focus their behavior in order to accomplish their goals?  On the other hand, might deficits in the ability to vividly envision the future have a negative impact on motivation? 

We have had many enthusiastic discussions with our colleagues about directions for future research.  In one of these, Washington University researcher Jason Chan suggested that understanding the relation between memory and future thought may provide insight into the creative capacity of the human mind.  Authors of science fiction novels, for instance, are able to envision extremely detailed future scenarios.  Might such an ability to vividly imagine the future be related to a corresponding strength in memory?  Research on memory and on creativity are both part of studying human cognition, but until now little crosstalk between these two disciplines has taken place.

We believe that this is an exciting new approach to the study of the human mind and one that will greatly benefit from behavioral experimentation, clinical and neuropsychological observation, and recent advances in neuroimaging techniques.


1.         Tulving E. Memory and consciousness. Canadian Psychologist. 1985; 26: 1-12.

2.         Klein SB, Loftus J, Kihlstrom JF. Memory and temporal experience: The effects of episodic memory loss on an amnesic patient's ability to remember the past and imagine the future. Social Cognition. 2002; 20: 353-379.

3.         Hassabis D, Kumaran D, Vann SD, Maguire EA. Patients with hippocampal amnesia cannot imagine new experiences. Proceedings of the National Academy of Science USA. 2007; 104: 1726-1731.

4.         Busby J, Suddendorf T. Recalling yesterday and predicting tomorrow. Cognitive Development. 2005; 20: 362-372.

5.         D'Argembeau A, Van der Linden M. Phenomenal characteristics associated with projecting oneself back into the past and forward into the future: Influence of valence and temporal distance. Consciousness and Cognition. 2004; 13: 844-858.

6.         Szpunar KK, Watson JM, McDermott KB. Neural substrates of envisioning the future. Proceedings of the National Academy of Science USA. 2007; 104: 642-647.

7.         Addis DR, Wong AT, Schacter DL. Remembering the past and imagining the future: Common and distinct neural substrates during event construction and elaboration. Neuropsychologia. 2007; 45: 1363-1377.

8.         Ebbinghaus H. Memory: A contribution to experimental psychology. New York: Dover; 1885.

About Cerebrum

Bill Glovin, editor
Carolyn Asbury, Ph.D., consultant

Scientific Advisory Board
Joseph T. Coyle, M.D., Harvard Medical School
Kay Redfield Jamison, Ph.D., The Johns Hopkins University School of Medicine
Pierre J. Magistretti, M.D., Ph.D., University of Lausanne Medical School and Hospital
Robert Malenka, M.D., Ph.D., Stanford University School of Medicine
Bruce S. McEwen, Ph.D., The Rockefeller University
Donald Price, M.D., The Johns Hopkins University School of Medicine

Do you have a comment or question about something you've read in CerebrumContact Cerebrum Now.