Why Do Humans Form Long-Term Mateships? An Evolutionary Game-Theoretic Model

Daniel Conroy-Beam*; Cari D. Goetz†; David M. Buss*,1    * Department of Psychology, University of Texas at Austin, Austin, Texas, USA
† Department of Psychology, California State University San Bernardino, San Bernardino, California, USA
1 Corresponding author: email address: dbuss@austin.utexas.edu

1 Introduction

Pair-bonding species such as humans represent just 3% of all mammals (Kleiman, 1977). We are the only great apes that engage in long-term pair-bonding. Chimpanzees and bonobos, our closest relatives, mate promiscuously and do not form long-term bonds (Dixson, 1998). The fitness costs of long-term mating account for its rarity. Committing to one or a few mates risks the large opportunity cost of forgoing other beneficial mating opportunities (Hurtado & Hill, 1992). Long-term mating males face paternity uncertainty because of internal female fertilization and gestation, which creates the adaptive problem of investing resources in the children of same-sex rivals (Buss, 2000). Women who commit to one man often fail to secure the best possible genes for their children, in part because men with good-genes indicators are often reluctant to commit to one woman and because women's own mate value limits the quality of the long-term mate they are able to attract. Males and females both risk significant costs at the hands of jealous or controlling long-term partners in the form of physical violence, emotional abuse, or manipulation, adding yet another cost to long-term committed mating (Buss & Duntley, 2011). Despite its costs and infrequency in nature, long-term mating is a major mode of mating in all human cultures (Jankowiak & Fischer, 1992). Human long-term mating psychology is therefore an important part of human life as well as our species’ unique evolutionary trajectory.

The evolution of our long-term mating psychology must have been driven by selection pressures that were at least somewhat specific to humans. Once a long-term mating psychology began to evolve, it would have generated an additional suite of novel adaptive problems, resulting in further selection for new psychological solutions. A complete understanding of human long-term mating psychology ideally includes an understanding of (1) the selection pressures that favored long-term mating psychology in humans initially, (2) the additional selection pressures that long-term mating, once formed, exposed our ancestors to, and (3) the psychological machinery these selection pressures produced. We propose a conceptual framework that integrates these essential components by modeling mating relationships as public goods games. This framework provides precise predictions about the long-term mating adaptations humans evolved to solve adaptive problems. Public goods modeling also connects research on long-term mating psychology to the origins of long-term mating itself. This novel connection provides a more complete understanding of the problems encountered within long-term relationships and the possible psychological solutions. By identifying and uniting the adaptive problems long-term mating solved and posed, public goods analysis yields a thorough and productive picture of human long-term mating psychology.

2 Part I: Adaptive Problems and Evolutionary Game-Theoretic Models

Adaptive problems are recurrent challenges from the physical, biotic, social, or internal environment—such as extremes of temperature, parasites, hostile conspecifics, or caloric needs—whose solution increases reproduction (Cosmides & Tooby, 1995). Knowledge of the adaptive problems a species has faced is essential because natural selection, by definition, favors traits that solve adaptive problems. Researchers can identify previously unrealized adaptive problems a species faces in social relationships with evolutionary game-theoretic modeling. Here, we review the relation between adaptive problems and their evolved solutions, as well as the ways in which evolutionary game-theoretic models identify adaptive problems solved and faced in long-term mating.

2.1 Adaptive problems and adaptations

Natural selection is the only known causal process capable of creating psychological and physiological systems that are complex, efficient, and reliable in solving adaptive problems (Tooby & Cosmides, 1990; Williams, 1966). Traits that solve adaptive problems increase the reproduction of their own genetic bases and thus actively contribute to their representation in future generations. This positive feedback process is more likely to favor traits that efficiently solve adaptive problems than are chance or random processes. The set of possible human traits is infinitely large—much larger than the delimited set of traits that could, in principle, solve adaptive problems. Unguided chance evolution, due to processes such as random mutations and genetic drift, picks randomly from this array and consequently almost never produces complexly functional traits; it tends instead to produce nonfunctional or even fitness-detrimental alternatives. Complex design for solving adaptive problems is therefore the hallmark of evolution by selection (Williams, 1966).

Researchers can exploit the fact that selection favors adaptive problem solving traits to generate predictions about the nature of human psychology. If a psychological mechanism is an adaptation, it must have demonstrable features that would have made it an improbably good solution to an adaptive challenge humans recurrently faced throughout their evolutionary history (Pinker, 2003). The more design features a mechanism has, the more likely that mechanism was the product of natural selection rather than serendipity. Hypothesizing that a psychological mechanism solves some candidate adaptive problem therefore provides predictions about features that psychological mechanism must have: those that coordinate improbably well with the design specs of the problem like a key in a lock. Confirming these predictions also allows conclusions about a psychological mechanism's ultimate origins.

Consider the adaptive problem of thermoregulation. A psychological or physiological mechanism could have any of a vast array of design features: a bright color pattern; a computational system for tracking social exchanges; an aerofoil shape for producing lift. From the array of possible features, only a tiny subset is capable in principle of solving adaptive problems in thermoregulation: for instance, producing and exposing watery secretions to body surfaces in order to shed heat through evaporative cooling. If a researcher hypothesizes the existence of thermoregulatory adaptations (e.g., sweat glands in humans; panting in dogs), that researcher immediately knows to look for mechanisms that embody this small subset of efficient thermoregulation design features. Thus, correctly identifying an adaptive problem dramatically reduces hypotheses about candidate evolved solutions, physiological or psychological.

Jealousy in long-term relationships provides a useful psychological example. Buss, Larsen, Westen, and Semmelroth (1992) recognized that the sexes confronted distinct adaptive problems in the face of infidelity. When it comes to offspring, women are certain of their maternity but men face the threat of genetic cuckoldry and investing substantial resources in the offspring of rivals. Women, on the other hand, incur large reproductive costs relative to men in the form of internal fertilization, gestation, and breast feeding. For women more than men, a partner's infidelity thus risks the diversion of essential investment away from the woman and toward another woman and her children. Buss et al. (1992) thus proposed that men and women would have...