The target article provides compelling evidence that culture is a real phenomenon that varies across groups and evolves over time. The authors go on to argue that cultural evolution, with selection occurring at the level of the group, is necessary for explaining human cooperation. In doing so, however, they overlook the importance of individual-level cultural evolution.
Cultural evolution need not be focused on group-level selection. As long as variation exists in ideas (or strategies, in a game-theoretic context), and ideas or strategies that are more successful are more likely to be adopted by others, then cultural evolution is occurring. From a mathematical perspective, such a process is formally equivalent to (individual-level) genetic evolution, and it can explain both interpersonal cooperation and the emergence of institutions that govern human behavior.
With respect to human cooperation, consider the theoretical perspective offered by the social heuristics hypothesis (SHH), which combines cultural evolution with dual-process models of cognition. The SHH posits that individuals internalize strategies that are typically advantageous in daily life (Rand et al. Reference Rand, Peysakhovich, Kraft-Todd, Newman, Wurzbacher, Nowak and Green2014b) and carry these automatic and intuitive responses with them into atypical social situations (such as laboratory experiments). Therefore, imitation and learning, as shaped by one's physical and cultural environment, determine an individual's internalized norms and preferences. Such a process can lead to substantial variation in norms across groups (as different groups inhabit different environments) and support the evolution of cooperation in one-shot anonymous settings (via spillovers from the cooperation that is typically advantageous due to reciprocity, reputation, and punishment).
Empirical support for the SHH comes from evidence that experimentally inducing greater intuitive thinking promotes cooperation in one-shot anonymous economic games relative to inducing deliberation: Applying time constraints (Cone & Rand Reference Cone and Rand2014; Rand et al. Reference Rand, Greene and Nowak2012; Rand & Kraft-Todd Reference Rand and Kraft-Todd2014; Rand et al. Reference Rand, Newman and Wurzbacher2014a; Reference Rand, Peysakhovich, Kraft-Todd, Newman, Wurzbacher, Nowak and Green2014b), conceptually priming intuition (Lotz Reference Lotz2014; Rand et al. Reference Rand, Greene and Nowak2012), inducing cognitive load (Cornelissen et al. Reference Cornelissen, Dewitte and Warlop2011; Roch et al. Reference Roch, Lane, Samuelson, Allison and Dent2000; Schulz et al. Reference Schulz, Fischbacher, Thöni and Utikal2014), or increasing the immediacy of payments (Dreber et al. Reference Dreber, Fudenberg, Levine and Rand2014; Kovarik Reference Kovarik2009) can increase participants' willingness to pay costs to provide benefits to strangers. Furthermore, cultural factors, such as trust of daily-life interaction partners, alter participants' intuitive responses as predicted by such a theory: Intuition only promotes cooperation among those who come from a culture of cooperation and trust, where cooperation is thus advantageous (Capraro & Cococcioni Reference Capraro and Cococcioni2015; Rand & Kraft-Todd Reference Rand and Kraft-Todd2014). Direct evidence of the spillover effects that are at the heart of the SHH comes from laboratory experiments where participants are randomly assigned to interact under rules that make cooperation or defection advantageous: Participants experiencing a social environment that favors defection are dramatically less prosocial in subsequent one-shot anonymous interactions, but only if they rely on intuitive, heuristic processing (Peysakhovich & Rand Reference Peysakhovich and Rand2015).
This theory accommodates the innate learning structures used in norm acquisition (Hamlin et al. Reference Hamlin, Wynn, Bloom and Mahajan2011), while also explaining cross-cultural differences in norms and preferences (Ellingsen et al. Reference Ellingsen, Herrmann, Nowak, Rand and Tarnita2012; Gächter et al. Reference Gächter, Herrmann and Thöni2010; Henrich et al. Reference Henrich, Boyd, Bowles, Camerer, Fehr, Gintis and McElreath2001; Reference Henrich, Heine and Norenzayan2010b; Herrmann et al. Reference Herrmann, Thöni and Gächter2008). Importantly, the SHH accomplishes this without an appeal to group selection or intergroup conflict. An individual benefits from internalizing successful strategies, and in typical settings where mechanisms for the evolution of cooperation are operating, what is good for the individual winds up also being beneficial for the group. Thus, improving the group's welfare occurs as a side effect of selection maximizing individual payoffs, rather than as the cause of that selection.
What about institutions? By enforcing norms of cooperation, institutions play a critical role in stabilizing payoff structures that simultaneously optimize social and individual welfare. Can the emergence of such institutions be explained by individual-level forces? Indeed, it can. Consider the institution of democratic voting. Under a median voting rule, for instance, individuals engaged in a group cooperation endeavor each vote for a contribution amount, and then all participants are forced to contribute the amount specified by the median voter (Bernard et al. Reference Bernard, Dreber, Strimling and Eriksson2013; Deacon & Shapiro Reference Deacon and Shapiro1975; Hauser et al. Reference Hauser, Rand, Peysakhovich and Nowak2014; Walker et al. Reference Walker, Gardner, Herr and Ostrom2000). Under such an institution, even self-interested people would vote for contribution to a public good, because individuals earn higher payoffs in a group where everyone contributes. Since the institution forces all participants to behave in the same way, the free-rider problem is eliminated, and individual and collective interests are aligned. Furthermore, similar logic would cause individual-level forces to favor joining or founding such an institution, rather than preferring a scenario where free-riding was possible (and thus cooperation was doomed). Indeed, experiments show that people “vote with their feet” and learn to choose institutions with sanctions over a sanction-free alternative (Gürerk et al. Reference Gürerk, Irlenbusch and Rockenbach2006; Rockenbach & Milinski Reference Rockenbach and Milinski2006). These studies capture the essential components of institutionalized cooperation: When institutions can homogenize individual behavior through norm enforcement, social welfare and individual welfare are maximized by the same strategy.
Thus, cultural evolution and learning need not be linked to group-level selection, and genetic evolution to individual-level selection, as is done in the target article (and in much of the literature on cultural evolution more broadly). Both of these dimensions of the debate regarding human evolution can vary independently (Rand & Nowak Reference Rand and Nowak2013). A helpful analogy comes from American politics, where social and fiscal conservatism, two independent dimensions of political attitudes, tend to strongly co-vary because of the United States' two-party system. However, these two variables do not need to co-vary: An individual can be socially liberal and fiscally conservative (e.g., libertarian) or socially conservative and fiscally liberal. Similarly, scholars can advocate for the importance of cultural evolution without invoking group selection, an intellectual space which is often left unoccupied. It is largely due to historical accident that cultural evolution and group selection have come to be linked (the most successful proponents of the importance of culture are also advocates of group selection). Continuing to develop individual-level accounts of cultural evolution is essential for deepening our understanding of human evolution and prosociality.
The target article provides compelling evidence that culture is a real phenomenon that varies across groups and evolves over time. The authors go on to argue that cultural evolution, with selection occurring at the level of the group, is necessary for explaining human cooperation. In doing so, however, they overlook the importance of individual-level cultural evolution.
Cultural evolution need not be focused on group-level selection. As long as variation exists in ideas (or strategies, in a game-theoretic context), and ideas or strategies that are more successful are more likely to be adopted by others, then cultural evolution is occurring. From a mathematical perspective, such a process is formally equivalent to (individual-level) genetic evolution, and it can explain both interpersonal cooperation and the emergence of institutions that govern human behavior.
With respect to human cooperation, consider the theoretical perspective offered by the social heuristics hypothesis (SHH), which combines cultural evolution with dual-process models of cognition. The SHH posits that individuals internalize strategies that are typically advantageous in daily life (Rand et al. Reference Rand, Peysakhovich, Kraft-Todd, Newman, Wurzbacher, Nowak and Green2014b) and carry these automatic and intuitive responses with them into atypical social situations (such as laboratory experiments). Therefore, imitation and learning, as shaped by one's physical and cultural environment, determine an individual's internalized norms and preferences. Such a process can lead to substantial variation in norms across groups (as different groups inhabit different environments) and support the evolution of cooperation in one-shot anonymous settings (via spillovers from the cooperation that is typically advantageous due to reciprocity, reputation, and punishment).
Empirical support for the SHH comes from evidence that experimentally inducing greater intuitive thinking promotes cooperation in one-shot anonymous economic games relative to inducing deliberation: Applying time constraints (Cone & Rand Reference Cone and Rand2014; Rand et al. Reference Rand, Greene and Nowak2012; Rand & Kraft-Todd Reference Rand and Kraft-Todd2014; Rand et al. Reference Rand, Newman and Wurzbacher2014a; Reference Rand, Peysakhovich, Kraft-Todd, Newman, Wurzbacher, Nowak and Green2014b), conceptually priming intuition (Lotz Reference Lotz2014; Rand et al. Reference Rand, Greene and Nowak2012), inducing cognitive load (Cornelissen et al. Reference Cornelissen, Dewitte and Warlop2011; Roch et al. Reference Roch, Lane, Samuelson, Allison and Dent2000; Schulz et al. Reference Schulz, Fischbacher, Thöni and Utikal2014), or increasing the immediacy of payments (Dreber et al. Reference Dreber, Fudenberg, Levine and Rand2014; Kovarik Reference Kovarik2009) can increase participants' willingness to pay costs to provide benefits to strangers. Furthermore, cultural factors, such as trust of daily-life interaction partners, alter participants' intuitive responses as predicted by such a theory: Intuition only promotes cooperation among those who come from a culture of cooperation and trust, where cooperation is thus advantageous (Capraro & Cococcioni Reference Capraro and Cococcioni2015; Rand & Kraft-Todd Reference Rand and Kraft-Todd2014). Direct evidence of the spillover effects that are at the heart of the SHH comes from laboratory experiments where participants are randomly assigned to interact under rules that make cooperation or defection advantageous: Participants experiencing a social environment that favors defection are dramatically less prosocial in subsequent one-shot anonymous interactions, but only if they rely on intuitive, heuristic processing (Peysakhovich & Rand Reference Peysakhovich and Rand2015).
This theory accommodates the innate learning structures used in norm acquisition (Hamlin et al. Reference Hamlin, Wynn, Bloom and Mahajan2011), while also explaining cross-cultural differences in norms and preferences (Ellingsen et al. Reference Ellingsen, Herrmann, Nowak, Rand and Tarnita2012; Gächter et al. Reference Gächter, Herrmann and Thöni2010; Henrich et al. Reference Henrich, Boyd, Bowles, Camerer, Fehr, Gintis and McElreath2001; Reference Henrich, Heine and Norenzayan2010b; Herrmann et al. Reference Herrmann, Thöni and Gächter2008). Importantly, the SHH accomplishes this without an appeal to group selection or intergroup conflict. An individual benefits from internalizing successful strategies, and in typical settings where mechanisms for the evolution of cooperation are operating, what is good for the individual winds up also being beneficial for the group. Thus, improving the group's welfare occurs as a side effect of selection maximizing individual payoffs, rather than as the cause of that selection.
What about institutions? By enforcing norms of cooperation, institutions play a critical role in stabilizing payoff structures that simultaneously optimize social and individual welfare. Can the emergence of such institutions be explained by individual-level forces? Indeed, it can. Consider the institution of democratic voting. Under a median voting rule, for instance, individuals engaged in a group cooperation endeavor each vote for a contribution amount, and then all participants are forced to contribute the amount specified by the median voter (Bernard et al. Reference Bernard, Dreber, Strimling and Eriksson2013; Deacon & Shapiro Reference Deacon and Shapiro1975; Hauser et al. Reference Hauser, Rand, Peysakhovich and Nowak2014; Walker et al. Reference Walker, Gardner, Herr and Ostrom2000). Under such an institution, even self-interested people would vote for contribution to a public good, because individuals earn higher payoffs in a group where everyone contributes. Since the institution forces all participants to behave in the same way, the free-rider problem is eliminated, and individual and collective interests are aligned. Furthermore, similar logic would cause individual-level forces to favor joining or founding such an institution, rather than preferring a scenario where free-riding was possible (and thus cooperation was doomed). Indeed, experiments show that people “vote with their feet” and learn to choose institutions with sanctions over a sanction-free alternative (Gürerk et al. Reference Gürerk, Irlenbusch and Rockenbach2006; Rockenbach & Milinski Reference Rockenbach and Milinski2006). These studies capture the essential components of institutionalized cooperation: When institutions can homogenize individual behavior through norm enforcement, social welfare and individual welfare are maximized by the same strategy.
Thus, cultural evolution and learning need not be linked to group-level selection, and genetic evolution to individual-level selection, as is done in the target article (and in much of the literature on cultural evolution more broadly). Both of these dimensions of the debate regarding human evolution can vary independently (Rand & Nowak Reference Rand and Nowak2013). A helpful analogy comes from American politics, where social and fiscal conservatism, two independent dimensions of political attitudes, tend to strongly co-vary because of the United States' two-party system. However, these two variables do not need to co-vary: An individual can be socially liberal and fiscally conservative (e.g., libertarian) or socially conservative and fiscally liberal. Similarly, scholars can advocate for the importance of cultural evolution without invoking group selection, an intellectual space which is often left unoccupied. It is largely due to historical accident that cultural evolution and group selection have come to be linked (the most successful proponents of the importance of culture are also advocates of group selection). Continuing to develop individual-level accounts of cultural evolution is essential for deepening our understanding of human evolution and prosociality.