2.1. Foundations

Established knowledge of the ecology and evolution of parasitic disease (=infectious disease=pathogenic disease) provides a foundation for the parasite-stress theory of sociality. Infectious diseases were a major source of morbidity and mortality, and hence of natural selection, in human evolutionary history (Anderson & May Reference Anderson and May1991; Dobson & Carper Reference Dobson and Carper1996; Ewald Reference Ewald1994; McNeill Reference McNeill1998; Wolfe et al. Reference Wolfe, Dunavan and Diamond2007). Human adaptations that defend against parasites comprise the biochemical, cellular, and tissue-based classical immune system, as well as the behavioral immune system that includes (a) anti-parasite psychology and behavior (Schaller & Duncan Reference Schaller, Duncan, Forges, Haselton and Von Hippel2007), and (b) psychology and behavior that manages infectious diseases when they occur. The behavioral immune system is comprised of ancestrally adaptive feelings, attitudes, and values about and behaviors toward out-group and in-group members, caution about or unwillingness to interact with out-group people, and prejudice against people perceived as unhealthy, contaminated, or unclean (Curtis Reference Curtis2007; Curtis et al. Reference Curtis, Aunger and Rabie2004; Faulkner et al. Reference Faulkner, Schaller, Park and Duncan2004; Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Navarrete & Fessler Reference Navarrete and Fessler2006; Oaten et al. Reference Oaten, Stevenson and Case2009; Park et al. Reference Park, Faulkner and Schaller2003; Reference Park, Schaller and Crandall2007; Schaller & Duncan Reference Schaller, Duncan, Forges, Haselton and Von Hippel2007; Thornhill et al. Reference Thornhill, Fincher, Murray and Schaller2010). The behavioral immune system also includes the same types of bias against contact with nonhuman animals that pose human infectious disease threats (Prokop et al. Reference Prokop, Usak and Fancovicová2010a; Reference Prokop, Usak and Fancovicová2010b).

Hosts and their parasites coevolve in antagonistic and perpetual races with adaptation, counter-adaptation, and counter-counter-adaptation for both hosts and parasites (Ewald Reference Ewald1994; Haldane Reference Haldane1949; Ridley Reference Ridley1993; Thompson Reference Thompson2005; Tooby Reference Tooby1982; Van Valen Reference Van Valen1973). In the human case, this dynamic, enduring, antagonistic interaction is illustrated by the observation that, despite the huge somatic allocation made to the immune system, people still get sick and even small reductions in immunocompetence increase vulnerability to infectious disease.

Furthermore, host-parasite races are geographically localized across the range of a host species, creating a coevolutionary mosaic that involves genetic and phenotypic differences in host immune adaptation and corresponding parasite counter-adaptation across a host's range (Thompson Reference Thompson2005). An important outcome of this is that host defense works most effectively against the local parasite species, strains, or genotypes, and not against those evolving in nearby host groups. Hence, out-groups may often harbor novel parasites that cannot be defended against by an individual or his or her immunologically similar in-group members (Fincher & Thornhill Reference Fincher and Thornhill2008a; Reference Fincher and Thornhill2008b). Out-group individuals pose the additional infectious-disease threat of lacking knowledge of and therefore violating local customs or norms, many of which, like hygiene and methods of food preparation, may prevent infection from local parasites (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Schaller & Neuberg Reference Schaller, Neuberg, Crawford and Krebs2008). Norms of many types – culinary, linguistic, moral, sexual, nepotistic, religious, dress-related, and so on – are used by people both to portray in-group affiliation and associated values and to distinguish in-group from out-group members. Norm differences between groups are often the basis of intergroup prejudice and hostility (i.e., xenophobia). Likewise, norm similarity is the basis of positive valuation and altruism among people (Norenzayan & Shariff Reference Norenzayan and Shariff2008; Park & Schaller Reference Park and Schaller2005).

Evidence for geographically localized host–parasite coevolutionary races is convincing. On the parasite side of the race, parasite geographical mosaics were found, for example, in recent research on the important human parasite Leishmania braziliensis. Rougeron et al. (Reference Rougeron, De Meeus, Hide, Waleckx, Bermudez, Arevalo, Llanos-Cuentas, Dujardin, De Doncker, Le Ray, Ayala and Banuls2009) described the high genetic diversity and subdivided population structure of this parasite across both Peru and Bolivia. They found high levels of microgeographic variation identifiable by at least 124 highly localized, physiologically and genetically distinct strains. The strains showed strong evidence of high degrees of close inbreeding and thus resembled genetic clones. This extremely fine-grained geographic mosaic in L. braziliensis implies a similar microgeographic population and genetic mosaic in human hosts. This type of spatial variation in host adaptation against local parasites, or, said differently, in host immune maladaptation against out-group-typical parasites, is a general pattern in the animal and plant infectious disease literature (e.g., Corby-Harris & Promislow Reference Corby-Harris and Promislow2008; Dionne et al. Reference Dionne, Miller, Dodson, Caron and Bernatchez2007; Kaltz et al. Reference Kaltz, Gandon, Michalakis and Shykoff1999; Thompson Reference Thompson2005; Tinsley et al. Reference Tinsley, Blanford and Jiggins2006). Human cases showing this include the caste-specific infectious diseases and corresponding specific immunity among sympatric Indian castes (Pitchappan Reference Pitchappan2002). Indeed, McNeill (Reference McNeill1998) and Mattausch (in press) have suggested that the castes of India formed in part from differential cultural responses to parasite-stress. Another case is found in the village-specific immune defenses against leishmania parasites in adjacent Sudanese villages (Miller et al. Reference Miller, Fadl, Mohamed, Elzein, Jamieson, Cordell, Peacock, Fakiola, Madhuri, Eltahir, Elhassan, Musa, Muntaser and Blackwell2007). In particular cases the localization is so fine-grained that human hosts inbreed, risking the potential costs of inbreeding depression, in order to maintain coadapted gene complexes important for their offsprings' defense against local parasite infection, as Denic and colleagues have found for malaria (Denic & Nicholls Reference Denic and Nicholls2007; Denic et al. Reference Denic, Nagelkerke and Agarwal2008a; Reference Denic, Nagelkerke and Agarwal2008b; also Hoben et al. Reference Hoben, Buunk, Fincher, Thornhill and Schaller2010).

Further evidence of localized immunity derives from events where humans from isolated groups interact with novel groups by conquest or trade and infectious disease transmission ensues, sometimes with drastic effects. This has occurred after intra- and inter-continental movement of individuals brought about inter-group contact (Diamond Reference Diamond1998; Dubos Reference Dubos and Dubos1980; Good Reference Good1972; Jenkins et al. Reference Jenkins, Dimitrakakis, Cook, Sanders and Stallman1989; McNeill Reference McNeill1998). Other human examples of localized immunity are reviewed in Fincher and Thornhill (Reference Fincher and Thornhill2008a) and Tibayrenc (Reference Tibayrenc and Tibayrenc2007). Final evidence for local host adaptation to parasites is found in the literature showing that the hybridization between adjacent, closely related conspecific populations results in hybrid offspring with reduced immunocompetence (e.g., house mice: Sage et al. Reference Sage, Heyneman, Lim and Wilson1986; cottonwood trees: Floate et al. Reference Floate, Kearsley and Whitham1993; also see Thompson Reference Thompson2005 for other examples).

Because of localized host immune adaptation in an ecological setting of high disease stress, xenophobia, ethnocentrism, and reduced dispersal are adaptive preferences/values and behaviors for avoidance of novel parasites contained in out-groups and the management of local infectious disease. Philopatry – the absence or low level of dispersal away from a natal area for reproduction – reduces contact with out-groups and their habitats that may contain new parasites. Likewise, xenophobia – the avoidance and dislike of out-group members – discourages contact with out-groups and their likely different parasites. Ethnocentrism – in-group favoritism and embeddedness entailing nepotism toward both nuclear and extended family, as well as altruism toward unrelated, yet immunologically similar, in-group members – focuses prosociality among in-group members, and fosters the supportive social networks for coping with present infections in members of the in-group. Thus, philopatry, xenophobia, and ethnocentrism – elements of in-group assortative sociality – are predicted to be strongly held values in areas of high parasite-stress (also see Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Thornhill et al. Reference Thornhill, Fincher and Aran2009).

Parasite-stress is not the same across the globe nor has it been the same across time. Humans have experienced parasite gradients throughout history and continue to do so today (Crawford Reference Crawford2007; Dobson & Carper Reference Dobson and Carper1996; Guernier et al. Reference Guernier, Hochberg and Guégan2004; Lopez et al. Reference Lopez, Mathers, Ezzati, Jamison and Murray2006; Low Reference Low1990; McNeill Reference McNeill, Stanley and Joske1980; Reference McNeill1998; Smith et al. Reference Smith, Sax, Gaines, Guernier and Guégan2007; Wolfe et al. Reference Wolfe, Dunavan and Diamond2007). Hence, we expect that the benefits and costs of in-group assortative sociality will shift along the parasite-stress gradient such that in some areas (of high parasite-stress) high levels of in-group assortative sociality will be more beneficial than in other areas (of low parasite-stress). As parasite-stress declines, the infectious-disease contagion risks to individuals from interaction with out-groups decrease. Consequently, for individuals in areas that are relatively low in parasite-stress, out-group contacts and alliances may provide greater benefits than costs. The benefits of out-group interactions include gains through intergroup trade, new and better ideas and technology, and diversified and larger social networks for marriage and other social alliances (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Thornhill et al. Reference Thornhill, Fincher and Aran2009).

McElreath et al. (Reference McElreath, Boyd and Richerson2003) and Nettle (Reference Nettle1999) argued that assortative sociality could cause cultural isolation and, hence, cultural divergence and emergent new cultures in the absence of geographic barriers (e.g., unfavorable habitats or unsurpassable mountain ranges) that fractionate a culture's distribution. Building on this, we argued that given the ecological localization of host defenses against parasites, the three components of assortative sociality – limited dispersal, ethnocentrism, and xenophobia – fractionate populations and thereby cause their cultural and evolutionary independence (Fincher & Thornhill Reference Fincher and Thornhill2008a; Reference Fincher and Thornhill2008b). Therefore, the parasite-stress model includes a theory about the genesis of cultural or ethnic diversity, and some of the predictions related to this aspect of the model have been empirically supported. We have shown that endemic religion diversity (both major religions and ethnoreligions), as well as indigenous language diversity, across contemporary countries worldwide are related strongly and positively to parasite-stress (Fincher & Thornhill Reference Fincher and Thornhill2008a; Reference Fincher and Thornhill2008b). Also consistent with the ethnogenesis aspect of the parasite-stress model was Cashdan's (Reference Cashdan2001a) finding for traditional peoples in the ethnographic record that high parasite-stress regions have more ethnic groups than low parasite-stress regions.

The parasite-stress theory of sociality posits an adaptive (ancestrally) condition-dependent adoption of in-group and out-group social tactics by individuals dependent on local parasite-stress. This condition-dependent adaptation requires local variation in morbidity and mortality from parasite severity, as the selection that acted historically in favoring contingent assortative-sociality behavioral and psychological adaptations. The evolution of conditional response as an important feature of assortative sociality's design, rather than exclusive fixity of localized genetically distinct adaptations, is consistent with knowledge about infectious diseases. The dynamics of an infectious disease can generate high variation in incidence, prevalence, transmissibility, and pathogenicity of the disease agent across the range of its host species, as well as on a very fine-grained, local scale. Factors affecting this variability at a single locale are temporal changes in host group size, climate and weather, disease-vector abundance and behavior, and the number and dynamics of the different infectious diseases infecting a host (Anderson & May Reference Anderson and May1991; Corby-Harris & Promislow Reference Corby-Harris and Promislow2008; Ewald Reference Ewald1994; Guernier et al. Reference Guernier, Hochberg and Guégan2004; Prugnolle et al. Reference Prugnolle, Manica, Charpentier, Guégan, Guernier and Balloux2005). The dynamic nature of host-parasite coevolution itself creates localized variation across generations in parasite-stress (Hamilton & Zuk Reference Hamilton and Zuk1982).

In-group assortative sociality is an example, we argue, of adaptive phenotypic plasticity, that is, of a conditional strategy with multiple contingent tactics (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Schaller & Murray Reference Schaller and Murray2008; Thornhill et al. Reference Thornhill, Fincher and Aran2009). Such plasticity in traits is favored when phenotypic change allows the individual to modify phenotypic expression in directions that give higher inclusive fitness than that achieved by a single phenotype. Conditional strategies in behavior, psychology, development, and physiology are very common across animal taxa (West-Eberhard Reference West-Eberhard2003). Socially learned or cultural behavior in humans is a category of behavioral and psychological plasticity that evolved, at least in part, as a solution to the evolutionary historical fitness problem of local social complexity and change (Alexander Reference Alexander1979; Flinn Reference Flinn1997; Flinn & Coe Reference Flinn, Coe, Gangestad and Simpson2007). We have argued that a significant part of this complexity and change likely arose from local people's adjustments in in-group and out-group oriented behavior to deal adaptively with temporally varying parasite problems (Thornhill et al. Reference Thornhill, Fincher and Aran2009).

The characterization of human in-group assortative sociality as a contingent, “plastic” phenotype includes cultural transmission via social learning that is conditional on local optima in values. Accordingly, culture is not evoked and transmitted passively, but instead is actively evoked and transmitted by psychological adaptations for enculturation that discriminate cultural elements incorporating the elements that have local utility and discarding elements that do not. We hypothesize that the ontogeny of people includes design by Darwinian selection to choose ideas, ways of thinking, and attitudes – that is, choose values – that correspond to routes of ancestrally adaptive social navigation in their community/local culture. The recognition that individuals choose their values is quite old in the sociological and psychological literature (see Jost et al. Reference Jost, Federico and Napier2009). We add to this traditional sociological view with our proposal that the choices are by evolved design and, specifically in regard to values of assortative sociality, they are guided by psychological adaptations dedicated to the function of value acquisition to meet ecological adversity and demands pertaining to infectious-disease stress (for views compatible with this hypothesis for cultural acquisition, see Alexander Reference Alexander1979; Billing & Sherman Reference Billing and Sherman1998; Boyd & Richerson Reference Boyd and Richerson1985; Gangestad et al. Reference Gangestad, Haselton and Buss2006b; Henrich & Henrich Reference Henrich and Henrich2010; Schaller Reference Schaller2006).

However, we do not restrict application of the parasite-stress theory of sociality only to humans and other cultural animals. Therefore, we include in our article aspects of the behavioral immune system that can be considered even in acultural species.

Furthermore, our emphasis on adaptive contingency in the expression and adoption of human assortative sociality does not imply that we expect no region-specific variation across human societies in genetic adaptation for assortative sociality. Culture-gene coevolution may produce genetically differentiated cross-cultural variation in the values and behaviors of assortative sociality. For example, in areas of high parasite prevalence, cultural practices of philopatry, ethnocentrism, and xenophobia may effectively select for alleles affecting psychological features that promote the learning and use of these practices (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008). Our argument is that infectious disease problems are locally variable, and hence, significant conditionality will be favored and maintained by selection even in the presence of region-specific genetic adaptation functioning in local adoption and use of values and behaviors. Boyd and Richerson (Reference Boyd and Richerson1985), Cavalli-Sforza and Feldman (Reference Cavalli-Sforza and Feldman1981), and Lumsden and Wilson (Reference Lumsden and Wilson1981) treat culture-gene coevolution in detail. There is evidence that it may play a role in cross-national variation in the value dimension collectivism-individualism (Chiao & Blizinsky Reference Chiao and Blizinsky2010), a dimension related importantly to topics in this article. That genetically distinct adaptations for coping with an ecological problem and condition-dependent adaptation for the same problem domain can co-occur, is well established in the literature of alternative reproductive tactics (see recent review in Oliveira et al. Reference Oliveira, Taborsky and Brockmann2008).

A considerable body of research supports the hypothesis of an evolved contingent assortative sociality in people that functions against infectious disease. For example, Faulkner et al. (Reference Faulkner, Schaller, Park and Duncan2004) and Navarrete and Fessler (Reference Navarrete and Fessler2006) provide evidence, based on numerous and diverse Western samples, that scores among individuals on scales that measure the degree of xenophobia and ethnocentrism correspond to chronic individual differences in perceived vulnerability to infectious disease; those who perceive high disease risk are more xenophobic and ethnocentric than those who perceive low disease risk. This research also shows that xenophobia and ethnocentrism within individuals increases under experimental primes of greater disease salience in the current environment. A related recent study (Schaller et al. Reference Schaller, Miller, Gervais, Yager and Chen2010) reports that research subjects who observed slides of people with disease symptoms (e.g., pox, skin lesions, sneezing) immediately mounted a classical immune response. Their white blood cells produced elevated amounts of inflammatory cytokine-interleukin-6 when exposed to bacterial antigens. This immune response was not seen in research subjects who viewed control slides, including slides depicting a person pointing a gun directly at the viewer. Hence, the immune response was not a general reaction to danger or threat, but was specific to cues of other people with symptoms of parasitic infection (Schaller et al. Reference Schaller, Miller, Gervais, Yager and Chen2010).

Moreover, a recent study by Mortensen et al. (Reference Mortensen, Becker, Ackerman, Neuberg and Kenrick2010) reports that subjects viewing slides with disease-salient cues immediately exhibited greater feelings promoting between-person avoidance (reduced extraversion and openness to experiences) in comparison to these subjects' feelings upon viewing control slides. These researchers also found that subjects with high scores on perceived vulnerability to disease showed greater feelings of interpersonal avoidance than did subjects with low scores on the same scale. Finally, this same paper reports that viewing parasite-salient slides resulted in increased avoidant arm movements when subjects viewed facial photos of strangers, especially for subjects high in perceived vulnerability to disease. In sum, the studies by Schaller et al. (Reference Schaller, Miller, Gervais, Yager and Chen2010) and Mortensen et al. (Reference Mortensen, Becker, Ackerman, Neuberg and Kenrick2010) reveal that visually perceiving cues pertinent to risk of parasitic infection generates an immediate immune response and a change in perceptions of one's own personality and behavioral actions that defend against or avoid infectious people. Hence, such cues markedly activate the classical immune system as well as the behavioral immune system.

In sum, there is considerable evidence of both inter-individual chronic differences as well as within-individual conditionality in xenophobic and ethnocentric values and related personality features and behaviors, and that both the inter-individual consistency and within-individual contingency are caused by infectious-disease problems in the local environment.

Proximate mechanisms by which individuals assess local parasite-stress – and thereby ontogenetically and contingently express the locally adaptive degree of defensive assortative sociality – may include immune system activation (such as, the frequency of infection; Stevenson et al. Reference Stevenson, Case and Oaten2009) and social learning of local disease risks, as well as direct observation of parasite threat (as evidenced in the studies mentioned just above). All of these mechanisms may act in combination and account for both the inter-individual and within-individual variation in values affecting in- and out-group behavioral preferences comprising assortative sociality. The recent research by Stevenson et al. (Reference Stevenson, Case and Oaten2009) reports that people with high contamination sensitivity and disgust sensitivity, which are thought to be emotional defenses against parasitic infection, had fewer recent infectious diseases than people with low sensitivities, providing evidence of a protective function of these emotions against these diseases. These researchers also found that high contamination sensitivity, in particular, was associated positively with a person's history of contracting infectious diseases (but not with recency of infections), implying that an ontogeny and conditionality of repeated activation of the classical immune system may underlie the adoption of the values associated with assortative sociality.

2.2. Cross-cultural evidence for the parasite-stress theory of sociality

We mentioned in section 2.1 that the parasite-stress theory of sociality led to the discovery of global patterns in the diversity of religions and languages. The theory has been tested and supported by additional, recent cross-cultural studies, which we describe next.

The parasite-stress model predicts that philopatry should be positively associated with infectious disease stress. We tested this prediction by examining human societal range size (a measure of dispersal) in relation to parasite-stress for a large sample of traditional societies in the anthropological record (339 societies), and found that range size was smaller where pathogen stress was higher, indicative of adaptive philopatry as defense against parasites (Fincher & Thornhill Reference Fincher and Thornhill2008b). It is possible that the reduced range size in high pathogen-stress regions is due to the malaise and inactivity of parasitized individuals. However, this apparently is not the case. According to our analysis in Fincher and Thornhill (Reference Fincher and Thornhill2008b), people in traditional societies move more often but over shorter distances in high parasite-stress areas than do people in low parasite-stress areas.

High xenophobia and ethnocentrism correspond to the Western value system referred to as conservatism, whereas low xenophobia (i.e., high xenophilia toward out-groups) and restricted ethnocentrism (i.e., focus on nuclear family) correspond to more liberal values (see Navarrete & Fessler Reference Navarrete and Fessler2006; Thornhill et al. Reference Thornhill, Fincher and Aran2009). Furthermore, conservatism–liberalism overlaps considerably with the well-studied cross-cultural value system referred to as “collectivism–individualism” by cross-cultural sociologists and psychologists. Conservatism and collectivism are similar in their heightened xenophobia and ethnocentrism inclusive of the extended family and other in-group members with similar conformist and traditional values, whereas liberalism and individualism are similar in nuclear-family-focused nepotism and relatively high xenophilia (Barnea & Schwartz Reference Barnea and Schwartz1998; Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Gelfand et al. Reference Gelfand, Bhawuk, Nishii, Bechtold, House, Hanges, Javidan, Dorfman and Gupta2004; Georgas et al. Reference Georgas, Mylonas, Bafiti, Poortinga, Christakopoulou, Kagitcibasi, Kwak, Ataca, Berry, Orung, Sunar, Charalambous, Goodwin, Wang, Angleitner, Stepanikova, Pick, Givaudan, Zhuravliova-Gionis, Konantambigi, Gelfand, Marinova, Mcbride-Chang and Kodic2001; Oishi et al. Reference Oishi, Schimmack, Diener and Suh1998; Sagiv & Schwartz Reference Sagiv and Schwartz1995; Schwartz Reference Schwartz, Vinkin, Soeters and Ester2004; Triandis Reference Triandis1995). In contemporary societies, collectivists and individualists differ significantly in their view of the social structure of the society in which they reside. Collectivists emphasize the boundary between in-group and out-group and are distrusting of and unwilling to contact out-group members; individualists make less distinction between in- and out-groups, and are more trusting of and show more willingness to contact out-groups (Gelfand et al. Reference Gelfand, Bhawuk, Nishii, Bechtold, House, Hanges, Javidan, Dorfman and Gupta2004; Oishi et al. Reference Oishi, Schimmack, Diener and Suh1998; Sagiv & Schwartz Reference Sagiv and Schwartz1995).

Fincher et al. (Reference Fincher, Thornhill, Murray and Schaller2008) showed that the unidimension of collectivism–individualism across many countries of the world is predicted strongly by infectious disease prevalence. High parasite-stress is associated with high collectivism (low individualism), and low infectious disease risk with low collectivism (high individualism). Murray et al. (Reference Murray, Trudeau and Schaller2011) showed that cultural emphasis on conformity was positively related to pathogen prevalence across many countries. Other recent cross-national studies showed that collectivism, autocracy, traditional gender roles (women's subordination relative to men's higher status), and women's traditional sexual restrictiveness and continence are values that positively covary with one another, and occur in nations with high prevalence of infectious disease. The assortative sociality adaptations of xenophobia and ethnocentrism link these values to avoidance and management of parasites. Also, the antipoles of each of the values – individualism (hence, liberalism), democracy, and women's political rights, freedom, and increased participation in casual sex – are a positively covarying set of values and are found in countries with relatively low parasite-stress (Gangestad et al. Reference Gangestad, Haselton and Buss2006a; Murray & Schaller Reference Murray and Schaller2010; Schaller & Murray Reference Schaller and Murray2008; Thornhill et al. Reference Thornhill, Fincher and Aran2009).

Moreover, Schaller and Murray (Reference Schaller and Murray2008) found that important components of personality seem to be part of assortative sociality and associated behavioral immunity. They reported that extraversion versus introversion and openness versus closedness to new experiences and ideas correlated with variation in parasite prevalence across many countries of the world. People in high-parasite-stress nations showed cautious personalities conducive to avoiding exposure to contagion from conspecifics – high introversion scores and low scores on interest in new ideas and experiences – whereas people in low-parasite-stress nations showed high extraversion and openness to novelty.

In addition, Thornhill et al. (Reference Thornhill, Fincher, Murray and Schaller2010) have shown that, across countries, the relationships between parasite-stress and democratization, gender relations, sexual restrictiveness, and collectivism-individualism are much more strongly correlated with human infectious diseases that are transmissible between humans (called nonzoonotics) than with those that are not transmissible between humans (zoonotics). This is a strong test of the parasite-stress theory of sociality because only nonzoonotic human diseases can be contracted from conspecifics; and hence, assortative sociality and related values should be designed to respond primarily to these diseases.

The cross-national patterns we mentioned that support the parasite-stress theory of sociality, depending on the particular analysis, statistically controlled for potential confounders, including Murdock's (Reference Murdock1949) six world regions (see sect. 4.6), a particular region's history of colonization and conquest, and the respective countries' latitude, Gross Domestic Product (GDP) per capita, wealth disparity, population size, land area, and extent of democratization. The cross-cultural study of philopatry in traditional societies controlled for population size, big-game hunting, world region, and other variables. Relatedly, we reported that across a large sample of contemporary countries, collectivism is positively correlated with philopatry measured as adults remaining throughout life in their natal region (Fincher & Thornhill Reference Fincher and Thornhill2008b).

In sum, we and others have argued that parasite-stress generated past selection that crafted the assortative-sociality psychological adaptation of humans. Accordingly, this adaptation has a condition-dependent functional design, resulting in the contingent expression by the individual of ancestrally adaptive degrees of in-group assortative sociality – arising from its three basic components, philopatry, xenophobia, and ethnocentrism – along a gradient of experienced parasite-stress. Individuals who experience relatively high levels of parasite-stress show greater in-group assortative sociality than those who experience relatively lower levels of parasite-stress (Fincher & Thornhill Reference Fincher and Thornhill2008a; Reference Fincher and Thornhill2008b; Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Murray & Schaller Reference Murray and Schaller2010; Schaller & Duncan Reference Schaller, Duncan, Forges, Haselton and Von Hippel2007; Schaller & Murray Reference Schaller and Murray2008; Thornhill et al. Reference Thornhill, Fincher and Aran2009; see also Freeland Reference Freeland1976; Reference Freeland1979; Loehle Reference Loehle1995 for earlier discussion of related ideas).

3.1. Strength of family ties

No one is without parents; but individuals do differ in the amount that they rely on or invest in their parents and other members of their family. Some people invest only in themselves or maybe at most in their nuclear family (spouse and children), while others consider themselves completely interdependent on a much larger extended family that includes not only spouses and children but also their parents, siblings, grandparents, cousins, uncles, aunts, nieces and nephews, and so on. Why this variation exists has been frequently discussed by scholars and it is generally attributed to differences in relative economic prosperity. For example, Inglehart and Baker (Reference Inglehart and Baker2000) provided evidence suggesting that countries that have relatively low wealth are also characterized by people who generally rely on their extended families; whereas people who live in countries with high levels of wealth are more able to strike out on their own because of the greater opportunities associated with greater wealth (see also Gelfand et al. Reference Gelfand, Bhawuk, Nishii, Bechtold, House, Hanges, Javidan, Dorfman and Gupta2004).

Intense loyalty and interdependence on the family (i.e., strong family ties) are generally considered important components of the cross-cultural dimension of collectivism-individualism, with collectivism positively associated with strong family ties in multiple cross-cultural studies (e.g., Alesina & Giuliano Reference Alesina and Giuliano2007; Gelfand et al. Reference Gelfand, Bhawuk, Nishii, Bechtold, House, Hanges, Javidan, Dorfman and Gupta2004). Often, collectivism is indexed by measuring the relative importance of the family. For example, Vandello and Cohen (Reference Vandello and Cohen1999) developed a measure of collectivism for each of the states in the USA that includes scores for family loyalty and interdependence. The distribution of collectivism-individualism across the globe has been explained by the distribution of wealth across the world in the form of GDP with the highest levels of collectivism associated most strongly with lowest levels of societal wealth (Hofstede Reference Hofstede2001; Kashima & Kashima Reference Kashima and Kashima2003; Triandis Reference Triandis1995). This is consistent with the argument that prosperity can explain patterns of family interdependence.

There is an alternative view, however (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008). Along with an emphasis on interdependence, collectivistic attitudes are generally associated with an unwillingness to contact or otherwise interact with out-group members (Gelfand et al. Reference Gelfand, Bhawuk, Nishii, Bechtold, House, Hanges, Javidan, Dorfman and Gupta2004; Oishi et al. Reference Oishi, Schimmack, Diener and Suh1998; Sagiv & Schwartz Reference Sagiv and Schwartz1995). As described in section 2.2, Fincher et al. (Reference Fincher, Thornhill, Murray and Schaller2008) argue that this reflects the importance of avoiding out-group members, who may carry infectious diseases that an individual is not able to cope with. Relatedly, collectivism is adaptive under high levels of parasite-stress by providing the benefits of disease avoidance and management, while widespread individualism under low levels of parasite-stress provides benefits in terms of increased out-group interaction (e.g., increased sharing of goods and innovations). Our studies of multiple, separate but conceptually related measures of collectivism-individualism support this view (Fincher et al. Reference Fincher, Thornhill, Murray and Schaller2008; Thornhill et al. Reference Thornhill, Fincher, Murray and Schaller2010). We make a similar argument for the importance of strong family ties – in areas with high levels of parasite-stress people will value strong family ties more than in regions with low parasite-stress. This reflects the importance of xenophobia for the avoidance of out-group members and of ethnocentrism for the development and maintenance of supportive in-group networks in the face of parasite-stress.

Current evidence that supports this comes from studies of traditional societies. To paraphrase Navarrete and Fessler (Reference Navarrete and Fessler2006), in human evolutionary history, when under parasite attack, in-group members were the only health insurance one had, and it was adaptive to have always paid your premiums – in terms of social investment and loyalty toward in-group allies that buffer an individual and his or her family against the morbidity and mortality of infectious disease. The support and loyalty toward in-group members was an individual's defense against the morbidity and mortality effects of parasites (Navarrete & Fessler Reference Navarrete and Fessler2006; Sugiyama Reference Sugiyama2004; Sugiyama & Sugiyama Reference Sugiyama and Sugiyama2003). Sugiyama (Reference Sugiyama2004) reported that among the Shiwiar, an Amazonian society without ready access to modern medicine, health care in the forms of food and other assistance from in-group members to persons suffering from infectious diseases was a major factor in lowering the mortality rate. This pattern, in general, seems to characterize numerous traditional human societies in the ethnographic record (Gurven et al. Reference Gurven, Allen-Arave, Hill and Hurtado2000; Hill & Hurtado Reference Hill and Hurtado2009; Sugiyama Reference Sugiyama2004; Sugiyama & Sugiyama Reference Sugiyama and Sugiyama2003).

Thornhill et al. (Reference Thornhill, Fincher, Murray and Schaller2010) show as well that parasite-stress was positively associated with a measure of family ties across modern countries. In this target article we explore a new measure of the strength of family ties at the cross-national level, using updated World Values Survey files from a recently produced, public dataset which was unavailable at the time of Thornhill et al. (Reference Thornhill, Fincher, Murray and Schaller2010). We also provide a novel empirical examination of parasite-stress in relation to a measure of collectivism across the states of the United States of America (Vandello & Cohen Reference Vandello and Cohen1999) and to a component of this measure that taps family-ties specifically.

3.2. Religiosity

Participation in a religion has certain costs for the participant, which include the time and effort involved in learning a religion and practicing it, the loss of opportunity to engage in other beneficial activities (opportunity costs), and risks such as the avoidance of modern medical care or extended fasting (Sosis et al. Reference Sosis, Kress and Boster2007). To learn the emotionality and associated language of a religion requires a long developmental (ontogenetic) exposure to the belief system. Opportunity costs include the inability to associate with other groups because one's specific beliefs may be considered irrational or contra-evidentiary to out-group members. (On irrationality as a functional component of religiosity, see Irons Reference Irons, Bulbulia, Sosis, Genet, Genet and Wyman2008).

This premise – that religious participation has costs – is a basis for studying religiosity from both the economic and the evolutionary science perspectives. From the economics viewpoint, Azzi and Ehrenberg (Reference Azzi and Ehrenberg1975) presented an analysis of individual religiosity as a function of the maximization of household allocation of time and found that people were attempting to rationally engage in religious behaviors such that they maximized the return on their time investments. Iannaccone has done much to formalize this economic investigation of religiosity (e.g., Iannaccone Reference Iannaccone1990; Reference Iannaccone1994, among others). Iannaccone (Reference Iannaccone1990) used economic theory to show that people employ sophisticated cost-benefit analyses often maximizing their investments when engaging in religious behaviors. Iannaccone (Reference Iannaccone1994) used rational choice theory to examine the relationship between a church's religious strictness and its strength or permanence, and concluded that, “Strictness reduces free riding. It screens out members who lack commitment and stimulates participation among those who remain” (p. 1204). In other words, paying-in indicates commitment but it also precludes desertion to other churches, because it is too costly to desert and then develop the same level of embeddedness in a new church. Therefore, individuals in strict churches exhibit higher rates of participation because they are assured, in comparison to individuals in less strict churches, a higher level of return on their investment through the reduction of free-riders and a higher level of investment by other individuals in the church. Iannaccone (Reference Iannaccone1994) observed that there is variation among churches in strictness and hence in the average religiosity among its members. That same research established that the strictest churches, those that require the highest costs for continued membership, have the tightest and most permanent collectives.

Many researchers have applied evolutionary costly-signaling theory to the understanding of religion and religious behavior (Bulbulia Reference Bulbulia2004a; Reference Bulbulia2004b; Cronk Reference Cronk1994; Henrich Reference Henrich2009; Irons Reference Irons1996; Reference Irons and Nesse2001; Johnson Reference Johnson, Bulbulia, Sosis, Genet, Genet, Harris and Wyman2008; Sosis Reference Sosis2000; Reference Sosis2003; Reference Sosis2005; Sosis & Alcorta Reference Sosis and Alcorta2003; Sosis & Bressler Reference Sosis and Bressler2003; Sosis & Ruffle Reference Sosis and Ruffle2003; Steadman & Palmer Reference Steadman and Palmer2008; Wilson Reference Wilson2002). This approach builds on the same foundation as the economic study of religious behavior – that religious participation has costs. Evolutionary science deepens our knowledge by providing methods that can pinpoint not only how benefits are maximized currently (also called the proximate-level of understanding), but also how benefits were maximized historically (also called ultimate-level of understanding). Through adaptationist study, evolutionary science has the ability to discover the historical setting in which a phenotypic feature that is an evolved adaptation, yielded net reproductive benefits (inclusive fitness) to its bearers (Andrews et al. Reference Andrews, Gangestad and Matthews2002; Thornhill Reference Thornhill, Bekoff and Jamieson1990; Williams Reference Williams1966). The researchers using costly-signaling theory propose that membership in a religious group is necessary for individuals to accrue certain social benefits not accessible independently, and that engaging in religious behavior is a signal of in-group allegiance to other individuals (both in-group and out-group individuals). The greater the costs of religious participation, the more honestly the participation signals allegiance to the religious in-group. The high costs of religiosity mentioned above are ideal for honestly signaling embeddedness in and commitment to an in-group with a particular spiritual belief system. Religion is often defined as a value system that is based on supernatural phenomena (Boyer Reference Boyer2001). This defining feature of religiosity has the high cost of displaying belief in the power of supernatural phenomena – phenomena that are generally considered antithetical to the empirical data humans gather from sensory experience (Irons Reference Irons, Bulbulia, Sosis, Genet, Genet and Wyman2008). Religious groups adopt their own distinct costly versions of supernatural beliefs in order to heighten costs of participation and distance themselves from out-groups.

Sosis (Reference Sosis2000) and Sosis and Bressler (Reference Sosis and Bressler2003) provided supportive evidence for the costly-signaling theory of religion by studying the longevity of 19th century United States communes. Sosis (Reference Sosis2000) found that religious communes in comparison to secular ones had longer life spans, and Sosis and Bressler (Reference Sosis and Bressler2003) found that longevity for religious communes was positively related to the magnitude of the costly acts required for membership within a commune (e.g., restriction from alcohol and sex). A logical prediction from the costly-signaling perspective, put in evolutionary theoretical terms, is that the adaptive value of religious signaling to signalers, and hence the magnitude and associated costs of the signal, will vary from place to place based on the underlying ecological necessity of in-group assortativeness for inclusive fitness maximization (also see, Sosis et al. Reference Sosis, Kress and Boster2007). According to the parasite-stress model, this ecological necessity is determined by parasite-stress variation across regions.

We argue that the maintenance of in-group assortativeness by practiced and signaled religious allegiance provides two benefits: (a) the protective barrier provided by separation from out-group individuals who may harbor novel infectious diseases and/or perform non-normative behavior; and (b) in-group embeddedness that reduces the morbidity and mortality caused when infectious disease invades the in-group. Hence, measures of the importance of religion for people in an area (religiosity) should be predictable based on the area's position along the parasite gradient, reflecting the average infectious disease stress experienced by people in the region. Therefore, we hypothesized that religious participation and commitment, indicating the importance of in-group assortative sociality, would be positively related to parasite-stress across regions.

One of the assumptions of our hypothesis is that there is a positive relationship between religiosity and out-group dislike or in-group preference. Evidence supporting this is found in a few studies. For example, Jackson and Hunsberger (Reference Jackson and Hunsberger1999) conducted a study of the relationships between individuals' religiosity and their prejudicial attitudes toward religious and non-religious others. They found that the religious participants showed significant positive attitudes towards in-group-religious others but negative attitudes towards non-religious others. As well, the magnitude of the prejudice was correspondent to the individual's own level of religiosity. That is, a participant who scored highly on religious fundamentalism also scored highly on out-group prejudice. In a separate study, Bulbulia and Mahoney (Reference Bulbulia and Mahoney2008) demonstrated that New Zealand Christians were more altruistic toward Canadian Christians than were New Zealand citizens to other New Zealand citizens. Similarly, Widman et al. (Reference Widman, Corcoran and Nagy2009) showed that individuals with strong Christian beliefs were more likely to rate others displaying a symbol of Christianity (a cross) as more kind and moral than others not displaying such a symbol. These studies support our assumption. They also suggest, on the one hand, the importance of religiosity as a marker of in-group membership, and, on the other hand, an underlying mental mechanism within individuals to measure religious similarity. Such a mechanism was indicated by Park and Schaller (Reference Park and Schaller2005), who found that when people experienced attitudinal similarity with others, they considered them more like family than when attitudes were dissimilar. Furthermore, there is convincing evidence that religious prosociality is primarily in-group altruism (Norenzayan & Shariff Reference Norenzayan and Shariff2008).

In a study supportive of the proposal we present (albeit this study was not designed to test the parasite-stress model), Saroglou et al. (Reference Saroglou, Delpierre and Dernelle2004) conducted a meta-analysis of the relationships between Schwartz's model of 10 cross-culturally stable, core values (Schwartz Reference Schwartz1992) and religiosity. Saroglou et al.'s (2004) meta-analysis focused on 21 samples from 15 countries (total n=8551 people). They discovered that religious people favored values that promoted social order (mainly the values Tradition and Conformity) but disliked values that promoted openness to change and autonomy (Stimulation and Self-Direction). This was true across a variety of religions (i.e., Christians, Jews, and Muslims) and countries from Europe, North America, and the Middle East. Interestingly, the positive correlation between religiosity and “conservation” (=conservative) values (Conformity, Tradition, and Security) and the negative relationship between religiosity and openness to change and autonomy (Stimulation and Self-Direction) showed greater effect sizes in a sample of Mediterranean countries in contrast to a sample of Western European countries. Mediterranean countries have greater levels of parasite-stress than Western European countries (Fincher & Thornhill Reference Fincher and Thornhill2008b; Guernier et al. Reference Guernier, Hochberg and Guégan2004).

The question may be raised: If a signal such as religious identity is strong enough to function as a social boundary, then why are multiple signaling modalities (e.g., language and religion) used to indicate in-group allegiance? Signaling systems across species typically show redundancy across components or modalities (Searcy & Nowicki Reference Searcy and Nowicki2005). This is thought to enhance communication, given that each signal is imperfect in information content but, combined, they provide greater accuracy. Redundancy is seen in human signaling of in-group affiliation and boundary. A combination of signals involving religiosity, language or dialect, word use, dress, music, smell, and so on comprise a redundant suite of honest signals about one's group membership and embeddedness.

4.1. Strength of family ties

We propose that individuals who value strong family ties will be found predominantly in areas with greater parasite-stress because of the benefits of in-group assortativeness promoted by family embeddedness. We predict, then, a positive relationship between the strength of family ties and parasite-stress across nations and across the states of the United States of America.

4.2. Religiosity

We predicted a positive association between religiosity and parasite-stress cross-nationally and across the states of the United States of America. To test this we indexed religiosity with two measures: (a) religious affiliation, and (b) religious participation and value. In the next sections, we describe how these variables were constructed for the cross-national and interstate analyses.

4.3. In-group assortative sociality

There is considerable conceptual overlap between religiosity and family ties that, we argue, reflects the importance of in-group assortative sociality within societies (e.g., the cross-national Religious Participation and Value was positively correlated with Strength of Family Ties, r=.79, n=72 countries, p < .0001). Because of this conceptual overlap, we generated synthetic in-group assortative sociality variables, one cross-national, which we called In-Group Assortativeness, and one for the states of the USA, which we called In-Group Assortativeness USA, to capture the common variation among our multiple dependent variables tapping in-group assortative sociality (ES 1.F). The cross-national data are in ES 2 and the interstate data are in ES 3.

4.4. Parasite-stress

4.5. Potentially confounding influences

As described in the introduction, there are features other than parasite-stress and assortative sociality (e.g., economic development) that have been proposed as explanations of the strength of family ties and religiosity. We next outline our approach used to explore alternative causal conjectures.

4.6. The problem of nonindependence

Geographically adjacent countries or U.S. states may be similar to each other due to common influences such as experiencing similar levels of infectious disease. Because of this, statistical independence among analytical units in the cross-national and the USA analyses may be questioned. To account for this potential problem, we used the following approach: We divided the countries into six world regions according to the method devised by Murdock (Reference Murdock1949), which is based on geographical proximity and cultural historical contact. Murdock's division of world cultures reduces the interdependence between societies among the six regions. The country assignments to the world regions are indicated in ES 2. Then, we conducted correlations using the mean values for each of the variables for each world region. This approach allowed us to characterize a region composed of multiple countries (or states) into a single value (thus deflating sample size). The small sample size makes the p-values suspect, but it does allow us to examine whether the correlations remain in the direction predicted by the parasite-stress theory after reducing the sample size. We also conducted a nested-effect linear regression that accounts for the nested design of our analysis. In the cross-national case, Combined Parasite-Stress was nested within each world region as the independent variable and used to predict the different dependent variables. Similarly, for the USA, we divided the states into the nine geographic regions used by the Census Bureau and used both approaches as we did for the cross-national analysis.