It is commonly believed that planning for the future is a skill unique to humans. Could other animals, even those as evolutionarily distant as western scrub-jays, share this skill with us?
Can it ever be said that animals plan ahead? Animals do show behaviour that prepares them for the future, but in general that behaviour reflects unlearned or conditioned responses to predictive cues. For example, a swallow flying south or a marmot entering hibernation is reacting to a cue that has foretold the seasons for its ancestors. A hungry rat pressing a lever that provides food in ten seconds, rather than a lever providing food later, does so because rewards are more effective after short than after long delays. Two requirements1 for genuine future planning are that the behaviour involved should be a novel action, or combination of actions (thus ruling out migrating and hibernating), and that it should be appropriate to a motivational state other than the one the animal is in at that moment (thus ruling out the rat's lever pressing). In their report of two experiments with western scrub-jays (page 919 of this issue2), Raby et al. describe the first observations that unambiguously fulfil both requirements.
The scrub-jay (Fig. 1) naturally caches food. In Raby and colleagues' research, jays were first allowed to acquire information about where food would be available in the morning. Then, in a test in the evening, the authors found that the birds behaved as if they were planning for breakfast by caching food items in the place where the food was most likely to be needed. The birds lived in large cages with three compartments (rooms) (see Fig. 1 of the paper on page 919). In the first experiment, each evening they ate powdered pine nuts, food they were unable to cache, in the central room. Then the next morning each bird was confined to one of the end rooms for two hours. In the 'breakfast room', a bird was always fed, whereas in the 'no-breakfast room' no food was given.
The test of planning came after several cycles of this treatment. For the first time, whole pine nuts were provided in the central room in the evening, along with sand-filled trays for caching in the two end rooms. The authors found that the birds cached three times as many pine nuts in the no-breakfast room as in the breakfast room. Importantly, all the data came from this one test: learning how their choices determined the next day's breakfast could not have influenced the jays' behaviour.
In the second experiment, the birds learned to expect breakfast in both rooms, peanuts only in one and dog kibble only in the other. On their first opportunity to cache peanuts and dog kibble in the evening, they distributed their caches so as to provide each room with the kind of food it usually lacked.
The results of two recent studies have been proposed as evidence for planning in primates. In one3, monkeys chose between eating four dates and one date. Eating dates makes monkeys thirsty, and the animals received water after a shorter delay if they chose one date. They gradually reversed their natural preference for four dates as if taking account of future thirst. However, this study falls short of the demonstration with the western scrub-jays2 because the monkeys underwent repeated trials in which they learnt the consequences of their choices. In the other study4, bonobos and orangutans were taught to use a tool to obtain a treat and were then allowed to choose a tool to take out of the testing room for use when they returned later. Most animals did take the appropriate tool on some trials. But in addition to other problems5, individual animals' patterns of success were far from consistent with a true understanding of the task.
Not much more than 100 years ago, interpreting any of these observations as human-like planning would not have been problematic. Indeed, Darwin's6 programme for documenting evolutionary continuity between human minds and those of other species encouraged anthropomorphic interpretations of animal behaviour7. This attitude was largely replaced, in both experimental psychology7 and ethology8, by a bias against 'mentalistic' explanations. But recent years have seen a resurgence of attempts to document processes in animals that in humans are accompanied by distinctive conscious states. Besides thinking about the future, examples include awareness of other individuals' states of mind, understanding how tools work, intentional deception, and empathy. This trend has been impelled by several developments. These include Donald Griffin's9 exhortations to ethologists to tackle animal consciousness; legitimization of the study of consciousness in the cognitive sciences generally; and the prospect of understanding the neural and genetic bases of conscious processes using 'animal models'. This last interest partly explains the current resurgence of attempts to study future planning in animals.
In humans, the ability to imagine future events and consciously recollect past ones (episodic memory) is impaired in patients with damage to the region of the brain known as the hippocampus. Furthermore, new imaging studies show that some of the same brain areas are active during both planning and remembering in normal adults10. Future planning and episodic memory are thus increasingly seen as part of a single human faculty for mental time travel, a faculty that other species have been proposed to lack1. Of course, as Raby and colleagues2 acknowledge, we will never know if a non-verbal animal is actually 'mentally time travelling' anywhere — future or past. Still, it is interesting that birds of the same species that show evidence of future planning also show episodic-like memory11. In both cases, researchers have cleverly exploited scrub-jays' specialized food caching to tackle questions of general interest in the study of cognitive evolution. But people can plan for, or think back on, much more than breakfast. It remains to be seen whether the abilities that scrub-jays show when caching are similarly applicable in other contexts.
About this article
Scientific Reports (2016)