Enlarge Shadow of Future
“Enlarge the shadow of future” is one of the recommendations given by Robert Axelrod (1984) to help promote cooperation in social dilemmas, based on evolutionary models.
Based on evolutionary models of cooperation and computer simulations, Axelrod (1984) proposed three mechanisms to promote cooperation between selfish individuals: (a) make future outcomes more important than the present ones, (b) change payoffs of possible outcomes in an interaction, and (c) teach people to care about others. This entry concerns the first of these mechanisms, which refers to the increase of the likelihood of future interaction with others and is called “the shadow of the future” by Axelrod and others (Axelrod 1984; Axelrod and Dion 1988; Bó 2005; Oates et al. 2010b; Sebastián-Enesco and Warneken 2015).
Evolutionary Models of Cooperation
Cooperation, or altruism, has long been considered an evolutionary puzzle (Nowak 2006). If one assumes that natural selection will tend to favor selfish strategies in a population, how could cooperative strategies have evolved? A solution to this problem was the kin selection theory (Hamilton 1964) which demonstrated that individuals will tend to cooperate with closer kin, because the cooperators’ genes will be transmitted indirectly through their relatives. However, this theory does not fully explain why some animals, and specially humans, cooperate with nonrelatives. Thus, a second solution to the puzzle of cooperation was the theory of reciprocal altruism (Trivers 1971). This theory proposes that cooperation can evolve if there is a high chance that the individual’s actions will be reciprocated in the future. The theories of indirect reciprocity (Alexander 1985) and network reciprocity (Nowak 2006) contribute to this framework by adding that this return can come from third parties and not necessarily from the individual who benefited from the altruistic act in the first place.
Axelrod (1984) modeled repeated pairwise reciprocal interactions to investigate whether and how cooperative strategies would evolve among a variety of other strategies. He and his collaborators (Axelrod 1984; Axelrod and Dion 1988) ran a series of computer tournaments in which several game theorists submitted their strategies to compete against each other in repeated rounds of an economic game called the prisoner’s dilemma. This game simulates a social dilemma in which an individual is always better off in the short term if he chooses to not cooperate with others. This dilemma is illustrated in the following way: two accomplices are arrested, hence becoming prisoners. They are put in separate rooms and cannot communicate with each other. They can either confess and receive a lighter sentence or remain silent and receive a harsher one. Regardless of which one decides, the confession always remains the best choice, if the individuals involved will never meet again. However, when there is a high chance of future interaction, remaining silent, and therefore cooperating with the other criminal, might be more important for the long term.
In Axelrod’s tournaments, the programmed strategies interacted repeatedly with each other, with some of them choosing always the same strategy (e.g., always defect or always cooperate) and some of them using some sort of decision criteria. The strategies could win points by either choosing to cooperate with or defect against their partners in each round, and the payoffs were dependent on both their decisions and those of their partners. The strategies that won more points in the long run would survive and reproduce more than (i.e., out-compete) the other strategies. The winning strategy in the first tournaments was the tit for tat, which had a simple heuristic: always cooperate in the first move and replicate the other player’s previous move – that is, cooperate when someone had cooperated previously and retaliate when someone had defected previously. Thus, in a mixed strategy population, tit for tat can survive and spread. But this will only happen if the shadow of the future is large enough, so that tit for tat can capitalize on the risk of cooperating with potential defectors. Other tournaments (Axelrod and Dion 1988; Wedekind and Milinski 1996) also showed that tit for tat loses to other more forgiving strategies, but only after winning over all the other previous ones. Thus, the better strategy for cooperation in repeated social dilemmas with the same partners involves a combination of being a bit generous, thus giving others the chance to change their strategy and engage in cooperation while still eventually punishing those who refuse to cooperate.
There are two main types of models that attempt to explain the evolution of cooperation: partner control models, in which cooperators can attempt to control their partners by cooperating with other cooperators and cheat or punish noncooperators (Axelrod 1984; Trivers 1971), and/or partner choice models in which cooperators can choose their partners, thus preferring to interact with other cooperators (Baumard et al. 2013; Noë and Hammerstein 1994). The shadow of the future is especially relevant to the former type of model, which accounts for situations in which individuals technically can’t choose with whom they interact. Thus, if the probability of encountering the same partners is relatively high, making future interactions more predictable, and in the long run the benefits of cooperating with others surpass those of cheating, individuals will tend to cooperate with each other. Moreover, in real-world situations similar to the prisoner’s dilemma, increasing the chances of meeting the same people in the future increases the probability of cooperating with them, more specifically when these repetitions appear infinite – that is, with no clear indication of when the last interaction between partners will occur (Bó 2005). Thus, enlarging the shadow of future is a recommendation applied to contexts in which cooperation is not always desirable, and individuals cannot choose with whom they interact with based on past interactions or any other judgment.
For the shadow of future to be effective, individuals must be able to respond to environmental or social cues that hint on the likelihood of meeting their partners again. This can happen through simple mechanisms, such as territoriality (individuals interact with others in fixed territories, thus ensuring that the same individuals will meet periodically, Axelrod and Hamilton 1981) or cooperation with partners that are similar to themselves in some manner (Axelrod and Hamilton 1981; Riolo et al. 2001). However, more sophisticated skills are required for more complex interactions, such as the ability of recognizing others and remembering past interactions (Stevens et al. 2005). In all of the cases outlined above, the shadow of future can maintain cooperative relationships because these mechanisms ensure that noncooperators can be retaliated by desertion or punishment. The following sections present examples of the influence of the shadow of the future on nonhuman animal and human social interactions and lay out the limitations of the shadow of the future as a mechanism to promote cooperative behavior.
The Shadow of the Future in Nonhuman Animals
There are several examples of reciprocal interactions among nonhuman animals which are affected by the shadow of the future (Axelrod and Hamilton 1981; Buss 2008; Oates et al. 2010b; Trivers 1971). The vampire bat is a classic example of reciprocal altruism (Carter and Wilkinson 2013; Trivers 1971; Wilkinson 1984). Vampire bats live in small communities of females and their offspring and feed themselves in the night, by sucking the blood of other animals. Feeding is very costly to them, due to higher rates of failure. To overcome this, vampire bats regurgitate blood and share it with those that had given blood to themselves previously, hence favoring reciprocal interactions. In addition, bats only share blood with individuals with whom they had frequent interactions previously and share more often with those who helped them recently. Thus, it pays off for them share blood with those in need, because the chances that the favor will be reciprocated in the future are high.
The mutualistic relationship between the cleaner fish and its clients is another good example of reciprocal altruism, which occurs between species, and it is kept by means of territoriality cues (Axelrod and Hamilton 1981) and indirect reciprocity (Bshary and Grutter 2006). Cleaner fish feed themselves off parasites they remove from other fish’s and reptiles’ bodies; large fish are potential predators for the cleaners, which are very small, but the clients avoid eating them because of the benefits of the removal of parasites. A species of cleaner fish that remains in small areas (called cleaning stations) tend to cheat their clients less than another species that rovers over a larger area, which makes it possible for them to desert clients more often (Oates et al. 2010a,b). Despite that, the latter also has preferred areas for interacting with specific clients, more specifically where there is a higher probability of future repeated interactions, which indicates that they also establish stable reciprocal relationships and are affected by the shadow of the future (Oates et al. 2010b).
The Shadow of the Future in Humans
The social contract theory (Cosmides and Tooby 1992) proposes basic adaptations for reciprocal cooperative interactions in humans. Based on findings from experiments with a logic task applied to social contexts, Cosmides and Tooby (1992) proposed that humans have psychological adaptations to detect cheaters in social exchanges. They identified the risk of being exploited by cheaters as the main source of evolved adaptations for cooperation; in response to that, humans have developed: the ability to recognize different individuals, to remember past interactions, to communicate their values, to model others’ values, and to represent costs and benefits across different kinds of items. In other words, human have evolved a set of skills necessary for engaging in more sophisticated cooperative reciprocal interactions than other animals.
Humans engage in reciprocal interactions very early, more specifically by turn-taking interactions initiated by parents (Trevarthen 2011). There is evidence that these interactions increase prosocial behavior in young children (Cortes Barragan and Dweck 2014; Rabinowitch and Meltzoff 2017), which can be an indication of an early mechanism that responds to cues of future interactions. In other words, repeated interactions might hint at future probability of interactions, thus functioning as a cue for prosocial behavior. However, reciprocal prosocial behavior, such as taking turns in sharing goods with a partner, only seems to fully emerge during middle childhood. Five-year-old American children are able to delay gratification and share more goods with a partner in anticipation of future reciprocal interactions (House et al. 2013; Sebastián-Enesco and Warneken 2015). Eleven-year-old American children increased their donations in repeated prisoner’s dilemma games compared to one-off games (Blake et al. 2015). However, when in repeated interactions within a group, children’s donations tend to decrease with time when they are anonymous but not when they are public, especially among children older than 8 across different cultures (Dutra et al. 2018; Zarbatany et al. 1985) and similarly to adults (Bó 2005; Wang et al. 2017). Thus, the shadow of the future increases human prosocial behavior, but particularly when individuals make public decisions.
Limitations to the Shadow of the Future
Enlarging the shadow of future refers to the increase of frequency of interactions and commitment between partners. However, this recommendation is based on models of reciprocal altruism which often assume that players have similar levels of power, cannot communicate with each other, and cannot opt completely out of the game (Buss 2000). Whenever one of these conditions is not met, enlarging the shadow of the future might not be as effective or even necessary. For instance, Ostrom (2000) found that communication can increase cooperation in collective social dilemmas similar to the public goods games, up until the last interaction, being more effective than external incentives or norms in similar settings.
Empirical evidence supports classical theoretical predictions that enlarging the shadow of the future – that is, increasing the probability of interacting with others in the future – is one significant factor in the increase of cooperation among partners in the long run. However, the influence of future interactions is constrained by certain conditions, such as individuals’ abilities to detect and interact with cooperative partners, environmental conditions that limit social interactions to certain areas and increase contact among individuals. Many human interactions are either one-off, or future interactions are uncertain, particularly in contemporary societies; other mechanisms such reputation and social control from institutions are in place to enforce cooperation among humans. Nevertheless, whenever the shadow of the future is large enough and individuals are able to retaliate by not cooperating with noncooperators, cooperation thrives.
- Alexander, R. D. (1985). A biological interpretation of moral systems. Zygon, 20(1), 3–20. https://doi.org/10.1111/j.1467-9744.1985.tb00574.x.CrossRefGoogle Scholar
- Axelrod, R. (1984). The evolution of cooperation (rev. ed). New York: Basic Books.Google Scholar
- Blake, P. R., Rand, D.G., Tingley, D., & Warneken, F. (2015). The shadow of the future promotes cooperation in a repeated prisoner’s dilemma for children. Scientific Reports, 5(1). https://doi.org/10.1038/srep14559
- Buss, D. (2008). Evolutionary psychology: The new science of the mind (3rd ed.). Boston: Pearson/Allyn and Bacon.Google Scholar
- Cortes Barragan, R., & Dweck, C. S. (2014). Rethinking natural altruism: Simple reciprocal interactions trigger children’s benevolence. Proceedings of the National Academy of Sciences of the United States of America, 111(48), 17071–17074. https://doi.org/10.1073/pnas.1419408111.CrossRefPubMedPubMedCentralGoogle Scholar
- Cosmides, L., & Tooby, J. (1992). Cognitive adaptations for social exchange. In J. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: Evolutionary psychology and the generation of culture (pp. 556–579). New York: Oxford University Press.Google Scholar
- Dutra, N. B., Boccardi, N. C., Silva, P. R. R., Siqueira, J. d. O., Hattori, W. T., Yamamoto, M. E., & Alencar, A. I. d. (2018). Adult criticism and vigilance diminish free riding by children in a social dilemma. Journal of Experimental Child Psychology, 167, 1–9. https://doi.org/10.1016/j.jecp.2017.10.007.CrossRefPubMedGoogle Scholar
- House, B., Henrich, J., Sarnecka, B., & Silk, J. B. (2013). The development of contingent reciprocity in children. Evolution and Human Behavior, 34(2), 86–93. https://doi.org/10.1016/j.evolhumbehav.2012.10.001.CrossRefGoogle Scholar
- Rabinowitch, T.-C., & Meltzoff, A. N. (2017). Joint rhythmic movement increases 4-year-old Children’s prosocial sharing and fairness toward peers. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.01050.
- Stevens, J. R., Cushman, F. A., & Hauser, M. D. (2005). Evolving the psychological mechanisms for cooperation. Annual Review of Ecology, Evolution, and Systematics, 36(1), 499–518. https://doi.org/10.1146/annurev.ecolsys.36.113004.083814.CrossRefGoogle Scholar