The birth this year of the new Cultural Evolution Society is an exciting and historic development, and I am privileged to have served on the steering committee responsible for its initial conception and gestation. Cultural evolution research faces many challenges in the years to come. One of the most fundamental, perhaps, is to establish the extent to which cultural evolution is Darwinian.

This could be broken down into many sub-questions. For example, is cultural variability due to cultural evolution or some other process? If culture evolves then what are the units of selection? Does the evolutionary process involve random variation and selective retention as observed in natural selection? To what extent does it depend on deliberate design and innovation? To what extent is culture evoked by biologically evolved mechanisms or transmitted? While all of these questions and more can be addressed using theoretical models and running lab experiments, there is also an inescapable need for field research. Indeed, restricting the study of cultural evolution to university campuses would arguably be equivalent to trying to study biological evolution exclusively in a zoo or aquarium.

To illustrate the high importance of field research in the study of cultural evolution I propose to focus here on just one of the fractionated questions above: Is culture evoked or transmitted? Culture is evoked to the extent that some putatively innate behavioural tendency (let’s say incest avoidance) is triggered by the presence of some standard environmental cue (e.g. sharing the same mother). By contrast, culture is transmitted to the extent that some putatively learned behavioural tendency (let’s say incest avoidance again) is passed down through the generations as part of a set of traditional beliefs and practices (e.g. sexual mores and sacred taboos). Surely nobody doubts that there is evoked culture and transmitted culture, but Evolutionary Psychology (hereafter EP) strongly emphasizes the former over the latter whereas Evolutionary Theories of Culture (hereafter ETC) tend to place the emphasis the other way around. To adjudicate on this question we need to seek data beyond (as well as within) the lab. In explaining why I will draw heavily on my personal experience of field research as an anthropologist. But let’s begin by fleshing out some key features of the evoked-versus-transmitted problem.

Culture as Evoked or Transmitted

A concise account of the differences of emphasis between EP and ETC was conveyed by the debate held in San Diego earlier this year between Leda Cosmides and Joe Henrich at an SPSP Annual Convention symposium entitled “Big Questions in Evolutionary Science and What They Mean for Social-Personality Psychology” (see

Cosmides, a captivating exponent of the EP tradition, argued that much of the content of culture is evoked rather than learned. That is, many cultural representations are the way they are because they are anticipated by evolved psychological architecture and, as such, would be motivating or memorable and therefore ‘catchy’ in any normal human being placed in a suitable environment. Henrich, an equally captivating exponent of cultural evolution theory, argued that much of our cognitive architecture evolved to facilitate the acquisition of useful information that could not have been inherited genetically. That is, we have evolved to recognize and preferentially learn useful information wherever we may find it. Both Cosmides and Henrich clearly agree that many specialised cognitive adaptations have evolved through natural selection, and both agree that culture provides an important context for the activation of these cognitive systems. The points of disagreement between advocates of EP and ETC, however, are as subtle and multifaceted as they are theoretically portentous.

Firstly, Cosmides emphasizes the role of evolved psychological capacities that emerge similarly in development across all human populations, being somewhat ‘canalized’ or genetically pre-specified (Waddington, 1957). Henrich, by contrast, emphasizes the role of evolved psychological capacities for learning in flatter epigenetic landscapes (Whitehouse 2013). So while both acknowledge that human psychology is an outcome of biological evolution, for Cosmides the emphasis is on inherited cognitive specialisations (a ‘modular mind’) whereas for Henrich the emphasis is on learning capacities (if not a more general intelligence, then at least a mind specialised for learning new skills rather than simply pulling out preformed gadgets to suit the terrain).

Secondly, to the extent that having a certain socially learned cultural skill (e.g. cooking) can have significant consequences for anatomy, cognition, and behaviour (e.g. digestive system), Henrich argues that culture and genes can co-evolve. But whereas for Henrich this insight should have profound implications for our understanding of human psychology, Cosmides argues that most cultural innovations are too recent to have had much effect on cognitive evolution via natural selection.

Thirdly, lurking somewhat in the background of this particular debate is a question about whether or not culture itself evolves. For Henrich, and perhaps many other founding members of the new Cultural Evolution Society, it may seem obvious that culture evolves. Even Cosmides recognizes that of course cultural representations can accumulate in a population so as to form distinctive cultural traditions and that particular domains of culture, such as technology, can become progressively more effective and efficient via processes of winnowing and selection. But she doubts whether such processes constitute a separate system of inheritance, alongside genetic inheritance, such that the two might be said to co-evolve (see also Sperber, 1996). In her talk in San Diego, Cosmides says that her main reasons for doubting this is that the notion of a cultural inheritance system either requires or tends to lead to a ‘mind-less’ (psychologically implausible) view of cultural transmission (see also Powell and Clarke, 2012).

Of course, the discussions in San Diego were designed around an adversarial debating format, veiling much common agreement. And it would surely be a mistake to reduce the differences between entire subfields such as EP and ETC to the views expressed by only two individuals at a single event. Nevertheless, a puzzling conundrum surely lurks beneath the surface here. While some leading exponents of EP and ETC may indeed agree on many fundamental points of theory, by emphasizing different aspects they wind up concluding that what the other is studying is not what they think it is. I am reminded of the story of the three blind men who each feel a different part of the elephant (e.g. the tail, the trunk, and the ear) and, as a consequence, reach very different conclusions about the nature of the object before them (claiming respectively that the object is a rope, a branch, and a fan). In much the same way, proponents of EP and ETC arguably fail to identify the same objects or to agree on how they should be connected up.

Seeing the bigger picture is easier if instead of engaging in abstract theoretical debates one starts to grapple with the messiness of real-world observation. Maybe in the field we have a better chance of seeing the whole elephant…

The Experience of an Anthropologist in the Field
My life as an anthropologist began in the late 1980s, as a doctoral student at Cambridge University. My mission was to travel deep into Papua New Guinea’s rain forest and conduct participant observation among the Mali Baining, a group whose language had yet to be described and whose culture was unknown to anthropology (Whitehouse 1995). Houses in my village were built of bush materials and lacked electricity and running water. Because of limited access to medical care (the nearest aid post being too far to walk when seriously ill), many succumbed to malaria and other potentially treatable maladies. Materially the culture was simple and life was often brutish and short. But the rituals and beliefs of the community were contrastingly rich and vibrant.

In my village there were various temples: two large communal ones built close to the cemetery and numerous smaller ones tended by individual households. In each of these buildings, offerings to the ancestors of food and drink (and sometimes also money, if available) were laid out as part of an unremitting schedule of daily rituals associated with a secretive organization known as the ‘Kivung’. In tok pisin the word kivung means ‘a meeting’ and it is certainly true that my friends spent a lot of their time in meetings, discussing how best to prepare for the great day when their ancestors would return from the dead. It was often said that the returning ancestors would take the appearance of white men and women, wielding powerful magic and technology. They would summon a vast fleet of bulldozers to flatten the rainforest and construct, overnight, a vast urban sprawl with high-rise buildings and wealth beyond ordinary people’s imaginings.

At Kivung meetings the community would often dwell on the forces of darkness that prevented the ancestors from returning and endeavour to root out sinners and have them ritually absolved and cleansed. Sin was understood as any breach of the Kivung’s Ten Laws (loosely based on the Ten Commandments taught by the nearby Catholic Mission). Only when sinfulness had been completely eradicated among the living would they finally be reunited with the ancestors. A period of great plenty, known as the taim bilong kampani (period of the companies), would ensue during which Kivung members would be granted vast wealth. During this time, there would be a great temptation to indulge the sins of the flesh. At the Day of Judgement to follow, those who resisted temptation would enter an everlasting paradise on earth known as the taim bilong gavman (period of the government). The rest would be cast into to Hell to suffer eternal damnation. The leaving of offerings provided a measure of the community’s progress towards this goal. Consistent evidence that the ancestors were receiving the offerings indicated that they would soon return. Evidence that the offerings had been rejected showed that the community was still sullied by breaches of the Ten Laws, delaying the miracle.

Men of the village process solemnly to one of the Kivung temples to lay out offerings to the ancestors (from Whitehouse 1995)

The procedure for laying out offerings in the larger temples begins with a task force of men processing solemnly through the front entrance of the building to arrange the food, water, money, and decorations on tables, rather as the servants of a grand house might have prepared for a banquet in Victorian England. Great care is paid to cleaning and straightening the containers of the offerings, and adjusting flowers and other adornments on the tables, all with a certain flourish and exaggerated attention to detail. Once the laying out of offerings has been completed to the satisfaction of all concerned, the temple is vacated except for one man (known as a ‘witness’) who remains behind, seated in solitude in a small cubicle in the heart of the building. It falls to the witness to note signs of ancestral presence such as the clinking of plates or cutlery, the sounds of eating, the creaking of benches, or even snippets of conversation among the visiting spirits. When his vigil is over the witness emerges blinking in the daylight before a gathering of the entire community. All are eager to hear whether the ancestors have accepted the offerings or, as disappointingly is so often the case, have refused them because of the sinfulness of one of more Kivung members (who must therefore be rooted out and encouraged to confess).

Temple rituals are only some of the many complex beliefs and practices associated with the Kivung. A similarly elaborated set of ideas and practices focus around communal gardens associated with ancestor heroes closely linked to the Genesis story as taught by the Catholic mission. Except that Eve’s misdemeanor was not to partake of a forbidden apple but to harvest the betel nuts of a taboo palm tree. As Eve slid down the trunk of the tree with her stash of nuts, she cut her groin on a sharpened stone that Adam had cunningly embedded in the trunk, causing her to produce a strong flow of menstrual blood from that time forward. She then passed on this curse to all her daughters and her daughter’s daughters, causing them to bleed every month. Kivung followers, unlike most neighboring groups in the region, repudiate betel chewing because they say the red substance it produces in the mouth is equivalent to menstrual blood.

Some years ago a group of us attempted to build a model of the Kivung meaning system in which some of the core ideas and practices were captured as nodes in an elaborated network (Whitehouse et al. 2012). The model incorporated four special nodes, depicted as black rectangles. We referred to them as ‘intuitive anchor points’ and what made these special was that they constituted universal implicit beliefs that are evoked by the environment (to use Cosmides’ terminology) rather than being sponged up via some sort of general learning capacity. The intuitive anchor points in question were selected for illustrative purposes and not because they were the only or even the most obvious intuitive beliefs that the Kivung evoked. Indeed, a more comprehensive model would have taken many other possible anchor points into account and the socially transmitted nodes in the meaning system would have been vastly more numerous and their crisscrossing interconnections unfathomably more complex. Nevertheless, our model helps convey the complex interplay between evoked and learned cultural representations in a given tradition.

In our model, one of the anchoring beliefs was mind-body dualism, the intuition that minds and bodies are distinct and can in principle be detached (Bloom 2004). This was clearly essential to the notion that ancestors could invisibly enter the temple and partake of the offerings without being physically present. Kivung members did not need to be taught that ancestors were bodiless, they inferred this from the fact that nobody ever entered or left the temple after the offerings had been laid out even though the witness might sometimes hear them talking or eating. The second was promiscuous teleology, the over-attribution of intelligent design to natural phenomena (Kelemen and DiYanni 2005). Consider, for instance, the Kivung creation myth about the causes of menstrual bleeding. The idea that this biological function was caused by the actions of primordial ancestors did not need to be explicitly taught but was simply inferred from the fact that Eve cut herself on a betel palm, women menstruate, and Kivung followers do not chew betel nut. The third anchoring belief in our model was the notion that offerings to the ancestors should be handled like potential contaminants, triggering hazard-precaution routines (Boyer and Lienard 2006). That the men entering the temples should walk slowly and deliberately, manipulate the offerings with great care (paying attention to separating, cleaning, and boundary maintenance), and communicate only in whispers were all intuitively obvious and did not need to be explicitly taught. All that was needed to generate these psychological responses and outward behaviours were cues that the food and the context for its preparation and laying out were somehow sacred and therefore potentially dangerous, serving as salient input to each participant’s hazard precaution system. And finally, the notion that God and the ancestors would punish sinners by delaying the miracle or casting them into Hell was implicitly informed by immanent justice intuitions (Callen, Ellard and Nicol 2006). Nobody needed to be explicitly taught that sinners, even the hapless ancestral Eve, deserved to be punished – this was immediately evident by virtue of their transgressions. The point, then, is that Kivung beliefs and practices were grounded in a set of deeper, maturationally natural intuitions (McCauley, 2011) delivered by people’s evolved psychological equipment rather than having to be explicitly taught. This much is entirely consistent with the EP view of culture.

But at the same time, many of the details of the Kivung belief system certainly did have to be explicitly taught and learned. If one were to dig deep into each of the core concepts of the Kivung, such as the Ten Laws, or the movement’s eschatology, one would soon find many explicit beliefs that were relatively remote form the intuitive anchor points described above and some downright difficult to conceptualize and remember. For example, the law proscribing murder was interpreted to refer to a great many different kinds of sins involving a kind of metaphorical ‘killing’ including gossip, certain forms of which were understood as a form of ‘character assassination’. In Kivung meetings details of what the ancestors would regard as homicidal behavior were elucidated at great length and repeated with such frequency that the risks of garbling or forgetting were greatly reduced. In our agent-based model such repetition served to ‘refresh’ the nodes in our network, allowing them to persist in a stable form. But when the frequency of repetition in a simulation was reduced, the links between nodes furthest away from intuitive anchor points would be at risk of fading and disappearing, with the possibility that a node could become isolated and so be extinguished from the system altogether. Our model also took account of the motivational strength of particular nodes and other variables that were affected not only by intuitive foundations but also how recently they had been first encountered and other relevant factors. Nevertheless, my concern here is not with details of a computational model. What matters for the present argument is that some beliefs, qua the EP viewpoint championed by Cosmides, are anchored in evolved intuitions and so are largely evoked rather than learned. But, at the same time, cultural systems also incorporate beliefs that are more distant from intuitive anchor points and, qua the ETC viewpoint advocated by Henrich, have to be explicitly learned, practised, and rehearsed if they are to be culturally transmitted.

For any meaning system to achieve stability it must ensure not only that its less intuitive constructs have adequate mnemonic support but also, more challengingly still, that the complex webs of interconnections between component nodes are sustainable over time within bounded populations. If a regional tradition is to maintain homogeneity across the landscape it must overcome various kinds of threats to its integrity and coherence as a belief system. Among these threats are forgetting, innovation, and demoralization. Much of my own previous research has been devoted to showing how highly routinized religious traditions reduce the risk of forgetting simply by repeating the creed over and over. Moreover, frequent repetition also makes it easier to spot unauthorized deviations from the orthodox canon (Whitehouse 2004), and mechanisms leading to tighter norm-enforcement may also contribute to the stabilization of beliefs and practices (Gelfand et al., 2011).  But faithfully remembering and accurately reproducing a set of beliefs and practices also presents motivational challenges – not least that routinization can become mind-numbingly boring! The insistence that the faithful turn up day in, day out and week in, week out to hear the same speech acts over and again can lead to the so-called ‘tedium effect’ (Whitehouse 2000), lowering motivation and risking splintering events or other expressions of rebellion.

To the extent that the contents of beliefs systems stray from their intuitive anchorages, they may easily stretch credulity to the limit. Doubters pose a threat to retention of the group’s membership not only because they may vote with their feet but also because they may also cause others to lose their faith. The ETC perspective has generated a number of theories to help explain how cultural traditions can inoculate themselves against such problems. An instructive example is the theory of Credibility Enhancing Displays or ‘CREDs’ (Henrich, 2009).

Time consuming and materially costly rituals, such as the laying out of offerings to the ancestors in temples, serve to advertise the commitment of adherents to Kivung teachings. According to the CREDs hypothesis, beliefs are more likely to spread in a population if those espousing them act in a way that would be costly if they were only pretending to believe. Beliefs may be said to be costly if they reduce the holder’s fitness, if they make great behavioural demands, or if they stretch credulity to the limit. Such costs are commonly exacted by supernatural beliefs. For example, the Kivung belief that the ancestors will soon return from the dead and magically transform the rainforest into a modern city, incorporates many assumptions that fly in the face of everyday observation and experience. Declaring one’s commitment to such a belief system is relatively easy to do but also likely to prompt skepticism. When such declarations are combined with costly behaviour such as an unremitting regime of ritual observance, however, they appear more sincere and therefore imitable. By backing up one’s beliefs with actions, by ‘walking the walk’ as well as ‘talking the talk’, CREDs may facilitate the spread of otherwise incredible beliefs. Insofar as groups with different kinds of beliefs come into competition, those with beliefs that are better adapted to cooperation may prevail. So, for example, a group that can stabilize beliefs in an all-knowing punitive ancestors may be better able to deter cheaters and defectors allowing more effective cooperation in communal projects (Bulbulia et al. 2013).  The idea is that CREDs can play a crucial role in such a process, leading to the spread of cultural traditions like the Kivung, at the expense of others.

Rituals serve as admirable CREDs because they are actions rather than merely words. Whereas advocates of EP, like Cosmides, emphasize that what causes beliefs to spread and persist in a population is their ‘catchy’ cognitive content, exponents of ETC, like Henrich, can point to mechanisms like CREDs that allow cultural groups to override various kinds of content bias, making implausible or personally deleterious beliefs more transmissible. One of the great advantages of conducting qualitative fieldwork is that it allows us to see how both EP and ETC perspectives can be right, that evoked intuitive beliefs as well as elaborated systems of belief that need to be learned and supported in various ways can and are combined in the real world. From this field-based perspective, EP is just one element, extended-able by advocates of ETC, and culminating in something much larger and more impressive – if not an elephant exactly, then something roughly as majestic and impressive.

Building a Cross-Cultural Field-Based Approach: The AnthroLab Model
Participant observation affords us the opportunity to describe cultural systems holistically and that is perhaps its greatest strength. Nevertheless, observation and description are not the same as explanation. To understand the causes of the phenomena we observe in the wild requires carefully controlled experiments and systematic comparison across space and time. Field sites have a vital role to play at this level as well.  But when experiment and comparison are the goal, fieldwork starts to look very different. Research teams get larger, field sites need to start communicating with each other, and the whole enterprise of data gathering and analysis needs to be scaled up and, to some extent, centrally coordinated. In this section we briefly consider a couple of examples of collaborative research in the field based on a hub-and-spokes model, pioneered by AnthroLab in Oxford’s Centre for Anthropology and Mind.

AnthroLab is currently collecting data in 12 countries worldwide, in some of these countries at multiple field sites. We chose this particular spread of countries and regions for several reasons. Above all, we wanted to maximize cultural variability, something that is important whatever one’s theoretical orientations. The question whether culture really does evolve in a genuinely Darwinian fashion – that is via generated variation, inheritance, and selection – is clearly an empirical question that requires field-based research as well as theories, lab experiments, and secondary data analysis to resolve. As we have seen, however, another of the big questions facing the evolutionary human sciences is whether particular behaviours are evoked or transmitted. Adjudicating on this question requires empirical studies that put these possibilities head to head. It is all too easy to fall prey the widespread, if implicit, misconception that just because behaviour is universal it must be intuitive/ evoked whereas if it is variable is must be learned/ transmitted. For example, Henrich argues that because fairness judgments differ cross-culturally this must reflect differences in transmitted norms but as Cosmides observes evoked culture can also vary. Indeed, to the extent that behaviours vary as a consequence of environmental differences the fact that physical and social environments have diversified so dramatically over the past ten thousand years or so should be a powerful motivation for advocates of the EP approach to get out of the lab and enter the field.

Doctoral candidate Veronika Rybanska (background left) and a local research assistant (foreground right) help me run a priming study with and adult male participant (foreground left) while another research assistant, Joseph Watts (then a student of Atkinson’s) participates off-camera by managing props, keeping notes, prepping participants and carrying out other essential tasks.

To illustrate the role that field research can play in these areas, consider the work we have been doing on ritual learning in early childhood. For several years we have been conducting experiments with 4-6 year olds in the USA showing that young children are very sensitive to cues that modelled behaviour is ritualistic rather than instrumental and tend to imitate the former more slavishly than the latter (Legare et al., 2015). In particular one of my doctoral students, Rachel Watson-Jones, spearheaded a series of studies suggesting that the imitation of rituals is motivated by a desire to affiliate with a group (Watson-Jones et al. 2014; Watson-Jones et al. 2015). Since this would appear to be an important mechanism of social learning, a crucial question from an ETC perspective is whether more ritualistic cultural environments (e.g. Melanesia) foster greater sensitivity to cues for conventional rather than instrumental learning. In 2012, Cristine Legare, Quentin Atkinson and I went to Vanuatu, a Melanesian archipelago, to lay the foundations for future fieldwork projects aimed at addressing this important question. In 2013, Atkinson and I returned with a larger entourage of research assistants to begin data collection. Although we each had our own teams dedicated to distinct projects, we were able to join forces to collect basic demographic data, covering more ground much more rapidly than would have been achievable using more conventional ethnographic methods. Moreover, teamwork was essential to carrying out field-based experiments. For example, we wanted to run a study of the effects of ritual versus instrumental primes on cooperation, trust, and preparedness to delay gratification. The procedure involved participation in an artificial ritual involving four experimenters and only one participant per session, a design that would have obviously been impossible to implement by a researcher working alone.

Conducting surveys and experiments as teams in countries like Vanuatu has enabled us to carry out rigorous comparisons between field sites. For example, we have been able to show that although Vanuatu is a more tradition-bound, ritualistic environment than much of Eurasia, children of the same age in both regions are more or less equally sensitive to the effects of ritualistic versus instrumental cues on executive control and delayed gratification tasks (Rybanska et al., in press). Currently we are studying the effects of ritual participation across multiple field sites simultaneously and planning even more ambitious research involving thousands of participants from scores of countries (more on this below).

Barriers to progress

Team-based field research in remote rural settings isn’t easy – or cheap. One of the most obvious barriers to progress is funding. While AnthroLab has been fortunate to secure continuous funding from large grants from the EU, Research Councils, and various charities since its establishment more than a decade ago, fundraising is always a perilous business. To be sustainable in the long-run, field-based approaches in the human sciences need to be embedded in universities serving the regions in which data collection takes place, via collaborations with permanent faculty and their renewable teams of student RAs. We cannot afford to rely only on postdocs supported by ‘soft’ money and fixed term contracts.

There are also undoubtedly intellectual barriers to overcome. We all know about the ‘two cultures’ problem that for many decades has made collaboration between scientists and humanist scholars so difficult or even impossible. But this, at least, is an obvious and longstanding problem being addressed from many angles. A potentially more pernicious barrier to progress – at least if it is allowed to persist unabated – is the unwarranted dislocation of EP and ETC perspectives. The need to combine these productively may seem less pressing from the comfort of one’s armchair or even the university lab, but for those of us committed to rolling up our sleeves and going into the field it is a high priority.

One solution is to get more EP and ETC people into the field working together. AntroLab is committed to training and hiring researchers to engage in scientific research all around the world based on collaborative fieldwork. To this end, we are continually hiring researchers to work on our many projects overseas and anticipate advertising several more full-time postdocs positions before the end of the current year as well as many opportunities for research assistants. Those interested in applying should look for updates here:

Our field-based approach not only combines the perspectives from EP and ETC

but more generally adopts a methodologically eclectic approach to problem solving. By doing so, our research will help to illuminate the essentials of being human – our language, rituals, religion, and morality. The best way to help research on cultural evolution inch forward is to wrestle with empirical problems, forcing the theorists catch up. That, in my experience, is what going into the field is all about.


I am grateful to Michael Buhrmester, Tom Currie, Oliver Scott Curry, Pieter Francois, Jonathan Jong, Chris Kavanagh, and David Sloan Wilson for commenting on a draft of this essay.  This work was supported by a Large Grant from the UK’s Economic and Social Research Council (REF RES-060-25-0085) entitled “Ritual, Community, and Conflict”.


Bloom, Paul (2004). Descartes’ Baby: How the science of child development explains what makes us human. New York: Basic Books.

Boyer, P. and Lienard, P. (2006). Why ritualized behavior? Precaution systems and action parsing in developmental, pathological and cultural rituals.  Behavioural and Brain Sciences, 29 (6), 595-612.

Bulbulia, J., Geertz, A. W., Atkinson, Q. D., Cohen, E. A., Evans, N., François, P., Gintis, H., Gray, R., Henrich, J., Jordon, F., Norenzayan, A., Richerson, P. J., Slingerland, E., Turchin, P., Whitehouse, H., Widlok, T. & Wilson, D. S. (2013). The cultural evolution of religion. In P. J. Richerson & M. H. Christiansen (Eds.) Cultural Evolution: Society, Technology, Language, and Religion (Strungmann Forum Reports). Cambridge, MA: MIT Press.

Callan M.J., Ellard J.H., and Nicol J.E. (2006). The belief in a just world and immanent justice reasoning in adults. Pers Soc Psychol Bull. 32(12):1646-58.

Gelfand, M., Raver, J., Nishii, L., Leslie, L., & Lun, J., and colleagues (2011). Differences between tight and loose societies: A 33-nation study. Science, 33, 1100-1104.

Henrich, J. (2009). The evolution of costly displays, cooperation, and religion: Credibility enhancing displays and their implications for cultural evolution. Evolution and Human Behaviour, 30, 244-260.

Legare, Cristine H., Nicole J. Wen, Patricia A. Herrmann, and Harvey Whitehouse (2015). Imitative flexibility and the development of cultural learning. Cognition. Vol 142: pp. 351-361. DOI: 10.1016/j.cognition.2015.05.020

Kelemen, D., and DiYanni, C. (2005). Intuitions about origins: Purpose and intelligent design in children’s reasoning about nature. Journal of Cognition and Development, 6, 3-31.

McCauley, Robert N. (2011). Why Religion is Natural and Science is Not. New York: Oxford University Press.

Powell, Russell and Clarke, Steve (2012). Religion as an Evolutionary Byproduct: A Critique of the Standard Model. In Brit. J. Phil. Sci., 63: pp 457-486.

Rybanska, Veronika, Ryan McKay, Jonathan Jong, and Harvey Whitehouse (In Press). Rituals improve children’s ability to delay gratification. Child Development.

Sperber, Dan (1996). Explaining Culture: A naturalistic approach. Oxford: Blackwell.

Waddington, C.H. 1957. The Strategy of the Genes. London: Allen & Unwin.

Watson-Jones, Rachel, Cristine H. Legare, Harvey Whitehouse and Jennifer Clegg (2014). Task-specific effects of ostracism on imitation of social convention in early childhood. Evolution and Human Behavior, Vol. 35, No. 3: pp 204 – 210.

Watson-Jones, Rachel E., Harvey Whitehouse, and Cristine H. Legare (2015). In-group ostracism increases high fidelity imitation in early childhood. Psychological Science. DOI: 10.1177/0956797615607205

Whitehouse, Harvey (1995). Inside the Cult: religious innovation and transmission in Papua New Guinea, Oxford: Oxford University Press.

Whitehouse, Harvey (2000). Arguments and Icons: divergent modes of religiosity, Oxford: Oxford University Press.

Whitehouse, Harvey (2004). Modes of Religiosity: a cognitive theory of religious transmission, Walnut Creek, CA: AltaMira Press.

Whitehouse, Harvey, Ken Kahn, Michael E.  Hochberg, and Joanna J. Bryson. (2012). The role for simulations in theory construction for the social sciences:  Case studies concerning Divergent Modes of Religiosity. Religion, Brain, and Behavior, Vol. 2, No. 3: pp 182-201.

Whitehouse, Harvey (2013). Rethinking Proximate Causation and Development in Religious Evolution. In P. J. Richerson and M. H. Christiansen. (eds.) Cultural Evolution: Society, Technology, Language, and Religion (Strungmann Forum Reports). Cambridge, MA: MIT Press.