Monthly Archives: June 2013

Gene-culture Coevolution: Application for the evolution of religion

This is an image of the American anthropologis...

David Sloan Wilson has resurrected the notion of group selection, and believes religion may play a crucial role

Having introduced the key concepts of gene-culture coevolution last time, we now explore their utility for studying the evolution of religion.

Application for the evolution of religion

Like Boyd and Richerson, David Sloan Wilson (2003) puts credence in the notion of group selection (discussed last time).  He believes intrinsic motivation and enforcement via reward and punishment enables sufficient conformity for group selection to occur, and that religions play a key role in motivating this conformity.

Religions, according to Wilson, fulfill the function identified by Durkheim, i.e. uniting a people into a moral community conducive to cooperation.  As was revealed by sociobiology’s application of game theory, those who cooperate by reciprocal altruism fare better overall than those who do not.  Religion, therefore, may function to establish and normalize reciprocal altruism as an evolutionarily stable strategy in a group, which would increase the fitness of the group over those adopting lesser strategies.

This would suggest that religious change may be driven by the optimization of motivation toward reciprocal altruism, a process occurring in groups of people responding to changes in their environments and circumstances as well as to competition with other groups.  Group selection may thus prove a crucial concept for understanding religious change.

The key concepts of coevolution contribute much to a theory of change in religion.  Cultural selection, guided variation, and biased cultural transmission refine previously established concepts, and group selection introduces an entirely new dimension to the equation.  Further, rigorous mathematical models address the problem posed by memetics, that of failing to discern the relative influence of genes and memes.  The models pit the relative strengths of memes against their genetic counterparts, as precisely as possible.


Unfortunately, these models also make coevolution a difficult approach requiring considerable math, which may detract from its growth as a method.  Laland and Brown report that coevolution remains a small coterie of researchers, and no well-established procedure exists at the time of their writing (Laland & Brown, 2002).  Nevertheless, this method, more than any other, shows promise for addressing the problem of religious change.  It incorporates many of the useful concepts from the other approaches already considered, while enhancing the level of rigor and introducing new concepts.

This post continues a series investigating methods in the study of religion from an evolutionary perspective.  Various approaches will be considered in turn, and evaluated for their strengths and weaknesses.

Tagged , , , ,

Gene-culture Coevolution: Key concepts for the evolution of religions


This is a reproduction of the map made by Luig...

This is a reproduction of the map made by Luigi Luca Cavalli-Sforza in his book The history and geography of human genes. It displays the genetic relationship between human populations. (Photo credit: Wikipedia)


The principle thesis of gene-culture coevolution (a.k.a. dual-inheritance theory) is that genes and culture both contribute to evolution in the case of humans.  Luca Cavalli-Sforza and Marc Feldman at Stanford led the way in this field, and were later joined by Robert Boyd and Peter Richerson at UCLA.  To determine precisely the relative contributions of genes and culture, these researchers began to create highly technical, detailed mathematical models.


Cultural selection vs. natural selection


Coevolution begins with a distinction between cultural selection, which operates by the differential replication of the meme, and natural selection, which affects culture by the differential reproduction of a meme’s host as a consequence of having it (Cavalli-Sforza & Feldman, 1981).  These concepts clearly distinguish two different mechanisms by which cultural entities, including religions, may change.


Durham (1991) expands on this by identifying at least five ways in which these mechanisms interact.  The first two are genetic mediation, whereby cultural variation is driven by genetic differences, and cultural mediation, in which cultural differences drive genetic evolution.  The latter has been used to explain the spread of genes for lactose absorbance among Europeans due to the cultural habit of dairy farming (Durham, 1991).  The other three interactions, enhancement, neutrality, and opposition, are similar to Dennett’s three symbiotic relationships: mutualist, commensal, and parasitic, respectively.  In these five ways, genes and culture interact to produce human behavior.


Guided variation


As for the means by which memes change, analogous to mutation among genes, Boyd and Richerson (1985) propose guided variation, or the process by which cultural information about a behavior is received and then modified by the individual on the basis of personal experience.  This concept seems to fall short of the mark, however, as it begs the question of what causes the individual to make these modifications.  A more thorough meme’s-eye view interpretation might posit that modification decisions result from competition and interaction among memes within the individual’s mind, as part of the normal processes that constitute human consciousness.  Indeed, Dennett has proposed that consciousness may be none other than the result of memes modifying their environment, i.e. the human mind, to better suit their needs (Dennett, 1993).


Biased cultural transmission


In any case, variation exists among memes, and certain forces influence choice between alternatives.  This is called biased cultural transmission (Boyd & Richerson, 1985), which comes in two types.  First, direct bias predisposes the individual to prefer certain types of information, “based on their judgments about the properties of the variants themselves” (Boyd & Richerson, 1985).  This may well result from genetic adaptation resulting in innate psychological predispositions, as theorized by evolutionary psychology.  It may also result from other means, such as prior integration of certain memes modifying reception of other memes.  For example, someone who has accepted the Creationism meme will thereafter prefer memes consistent with that worldview.  The other type of biased transmission is indirect bias, a.k.a. frequency-dependent bias, whereby the probability of accepting a meme is influenced by its prevalence in the population.  Humans have a tendency toward conformity, and this significantly affects the spread of memes in a group (Boyd & Richerson, 1985).


Group selection


This leads to yet another important concept in coevolution, group selection.  This form of selection has been disfavored by biology for some time, but recent studies suggest it may play a greater role than previously thought.  The main problem is that group selection requires that traits of individuals within a group remain relatively consistent over time, but within-group variation destabilizes these traits, such that group selection seems possible only under exceptionally rare conditions.


Boyd and Richerson (1985) propose that conformity provides the means by which within-group variation is overcome in the case of cultural evolution.


The next post will apply these concepts to the evolution of religion.


This post continues a series investigating methods in the study of religion from an evolutionary perspective.  Various approaches will be considered in turn, and evaluated for their strengths and weaknesses.


Tagged , , , ,

This week’s studies: Reading comprehension and statistics

 book open pages library books knowledge reading, by ksheltonTo keep myself accountable, I’m posting my weekly readings.  The deal is: at least seven articles or chapters a week, no excuses.This will also double as a nice record of book and article references, and a trail for those interested in the topics of Evolutionary Religious Studies, Cognitive Science of Religion, and Culture and Cognition.This week’s studies


Morrow, D.  “Situation Models and Point of View in Narrative Understanding.”  In: Van Peer, W. and Chatman, S. B., eds.  (2001).  New Perspectives on Narrative Perspective.  Albany, NY: State University of New York Press.

– explores role of protagonist point of view (deictic perspective, or “here/now” point in space and time), and summarizes many predictions on how switching between points of view should affect reading times, accessibility of information, causal inferences, and emotional reactions to events (p. 228).


Therriault, D. J. and Rinck, M.  (2007).  “Multidimensional Situation Models.”  In: Schmalhofer, F. and Perfetti, C., eds.  Higher level language processes in the brain: inference and comprehension processes.  Mawah, NJ: Lawrence Erlbaum Associates, Inc.

– Summarizes much research on the situation model of discourse processes theory, and evaluates the various dimensions of it.


van den Broek, P., Risden, K., and Husebye-Hartmann, E.  (1995).  “The Role of Readers’ Standards for Coherence in the Generation of Inferences.”  In: Lorch, R. F., and O’Brien, E. J., eds.  Scources of Coherence in Reading.  Hillsdale, NJ: Erlbaum.


Stone, J.  (2008).  Religious Naturalism Today: Rebirth of a Forgotten Alternative.  Albany, NY: State University of New York.  (introduction)


Utts, J. M., and Heckard, R. F.  (2012).  “Chapter 1: Success Stories and Cautionary Tales.”  Mind on Statistics, 4th Edition.  Boston, MA: Brooks/Cole.


Utts, J. M., and Heckard, R. F.  (2012).  “Chapter 2: Turning Data into Information.”  Mind on Statistics, 4th Edition.  Boston, MA: Brooks/Cole.


Utts, J. M., and Heckard, R. F.  (2012).  “Chapter 5: Sampling: Surveys and How to Ask Questions.”  Mind on Statistics, 4th Edition.  Boston, MA: Brooks/Cole.

Tagged ,

Memetics: Applications for the evolution of religion

English: Daniel Dennett at the 17. Göttinger L...

Many memeticists have posed religions as viruses of the mind, but Daniel Dennett reminds us they can just as well be beneficial symbionts.


Having explored the key concepts of memetics last time, we now explore its applications for the evolution of religion, as well as criticisms of the field.


Parasitic religions


From the meme’s eye view, it appears plausible that religions could be parasitic entities.  Religious behaviors may serve none of the functions proposed by the social sciences, but may instead serve the function of replicating memes.


Of course, it could just as well be the case that memes cooperate with their hosts toward mutual benefit, serving functions for both meme and gene.  Daniel Dennett reminds us that there are three types of symbiotic relationships: mutualist, commensal, and parasitic.  In a mutualist relationship, both entities benefit, as in the case of the bacteria living in the human intestine, without which we could not digest food.  In a commensal relationship, the one is at least not harmful to the other, as with barnacles growing on the sides of whales.  It is only in a parasitic relationship, such as that of flu viruses infecting humans, that harmful exploitation emerges (Dennett, 2007).  If religious memes exist in symbiosis with humans, it could take the form of a mutualist or commensal relationship, not just a parasitic one akin to a virus.




Yet the virus is exactly the model many memeticists have employed (e.g. Lynch, 1998; Brodie, 2009).  It is perhaps this paranoid-sounding analogy that has most prevented memetics from being taken seriously.  While it has attracted a wide audience among popular readers, it has made little impact on academics.  A number of other criticisms have been leveled, but Laland and Brown suggest that most of them are also true of evolutionary theory in general.  For example, memetics has been faulted for the meme’s apparent lack of clear boundaries and propensity to merge one into the other, but in all fairness it is also difficult to tell where one gene ends and another begins (Laland & Brown, 2002).


A more serious charge is that memetics does not yet account for the possibility that our minds may more readily accept some memes than others.  Something like the innate modules of evolutionary psychology may predispose us toward certain memes.


Furthermore, memetics does not yet offer a way to clearly determine the relative strengths of genetic and memetic interests.  Without the clear incorporation of genetic interests, we are left with the same problem as the standard social sciences model: the mind becomes a blank slate equally receptive to all content, with culture explaining everything and biology nothing.  A method to weigh the relative power of genes and memes in a given situation is needed.


Despite these criticisms, and poor esteem by academia, memetics shows particular promise for understanding religious change.  From the point of view of religions, which may themselves be understood as complexes of memes or memeplexes (Blackmore, 2000), replication may be maximized by employing various strategies to attract followers, who may in turn pass on the religion to offspring and others to whom they proselytize.  The functions proposed by Durkheim, Malinowski, and Geertz may be how religions attract followers at the individual or societal level.  Religions offer the fulfillment of functions in exchange for propagation, in a largely if not entirely unconscious manner akin to how flowers offer honey to a bee in exchange for the spreading of its pollen.  At the same time, religions compete against other religions to fulfill these functions.  Religions may change in order to deal with changes in functional needs of followers as well as to contend with competing religions.  In many cases, the exchange between religions and followers may be more or less equitable, i.e. a mutualist relationship.  In other cases, however, conditions may make it more profitable from the religion’s perspective to exploit its followers, resulting in a parasitic relationship.  The situation would thus be a dynamic relationship between followers and their needs on the one hand, and religions considered as memes on the other.


Something is still left out of the equation here, though.  Just because memes have their own interests does not mean they are able to overpower genes.  As suggested by the criticisms above, memetics must account for genetic influence.  The power of genes to enforce their interests may be greater than, less than, or equal to that of memes.  Further, relative power may well be different in different environments and situations.  Finally, the way that genes express their influence, whether through psychological modules that predispose humans toward certain memes, or by some other mechanism, remains to be explained.  Thus, memetics leaves our theory of religious change incomplete until genes are factored back into the picture.  An approach that responds to this problem is gene-culture coevolution.


This post continues a series investigating methods in the study of religion from an evolutionary perspective.  Various approaches will be considered in turn, and evaluated for their strengths and weaknesses.



Tagged , , , ,

This week’s studies: Domains, counterintuitiveness, and reading comprehension

 book open pages library books knowledge reading, by ksheltonTo keep myself accountable, I’m posting my weekly readings.  The deal is: at least seven articles or chapters a week, no excuses.This will also double as a nice record of book and article references, and a trail for those interested in the topics of Evolutionary Religious Studies, Cognitive Science of Religion, and Culture and Cognition.This week’s studies


Journal of Cognition and Culture – sifted through ten years of this journal, reading abstracts and downloading all articles relevant to my research interests, since I currently have access as a university student


Kinzler, K. D., and Spelke, E. S.  (2007).  “Core systems in human cognition.”  Progress in Brain Research, 164.

– This article gives a good summary of the “signature limits” of 4 core systems (inanimate objects, animate objects, numbers, and geometry) plus a possible fifth (social partners).  It does not discuss why other systems were excluded, however, such as biology or language.


Johnson, C. V. M., Kelly, S. W., and Bishop, P.  (2010).  “Measuring the Mnemonic Advantage of Counter-intuitive and Counter-schematic Concepts.”  Journal of Cognition and Culture, 10(109-121).

– Study confirms Boyer’s MCI hypothesis, but only for delayed recall.  It also uses items removed from a narrative context (addressing objections by Upal and Gonce), but also greater than 2-word lists (addressing objections by Norenzayan and Atran).  The 2-word lists are considered to be too ambiguous in meaning (weak “public representations”) to be memorable.  Here, lists of 23-word items are used.  The study also attempts to implement a more precise method of counting counterintuitiveness developed by Barrett (2008), but IMO fails miserably: 4 out of 10 of their items seem to me like they should be counted differently.  This underscores this line of research’s failure to adequately address the problem of subjectivity in parsing counterintuitiveness.


Keil, F. C.  “The Birth and Nurturance of Concepts by Domains: The Origins of Concepts of Living Things.”  In: Hirschfeld, L. A. and Gelman, S. A., eds.  Mapping the Mind: Domain Specificity in Cognition and CultureCambridge: Cambridge University Press.

– Keil presents evidence for a third core domain beyond physics and psychology: biology.  Actually, what he argues for are three cognitive stances: the mechanical (physics), intentional (psychology), and teleological (biology).  One thing I don’t understand about Keil’s work as well as that of others’ is the use of data from children of schooling age (usually elementary ages 4-10).  It seems to me the cognitive changes detected could be the result of schooling rather than the developmental process of evolved cognitive mechanisms.  Neither Keil nor other authors ever seem to address this question.


Magliano, J. P., Taylor, H. A., and Kim, H. J.  (2005).  “When goals collide: Monitoring the goals of multiple characters.”  Memory & Cognition, 33(8), pp. 1357-1367.

– Article assesses the tracking of the goals of multiple characters in two movies, Moonraker and Wrath of Khan, and finds that primary antagonist and protagonist goals are tracked equally, while secondary antagonist goals are tracked to a lesser extent.


Horton, W. S. and Rapp, D. N.  (2003).  “Out of sight, out of mind: Occlusion and the accessibility of information in narrative comprehension.”  Psychonomic Bulletin & Review, 10, pp. 104-110.

– Horton and Rapp (2003) find that response times are longer when asked about information no longer visible (“occluded”) from the protagonist’s perspective, as opposed to those still visible from that same perspective.  Related to this topic are O’Brien and Albrecht (1992) and Albrecht, et al. (1995), which both find that readers notice when a text contains details perceptually inconsistent with the perspective they are using to understand the story, but only when explicitly instructed to adopt the perspective of the protagonist.  The latter condition is puzzling in relation to Horton and Rapp’s study, which did not include explicit instructions to adopt the character’s perspective yet attained such results nonetheless.


– Read and summarized dozens of article abstracts on reading comprehension as part of a research assistant project on how readers track multiple characters’ perspectives.

Tagged , ,

Memetics: Key concepts for the evolution of religion

English: This is a pertinent image of Dawkins ...

Coined by Richard Dawkins in his book The Selfish Gene, the concept of memes has become wildly popular


The next approach to the evolution of religion in our investigation is memetics, a controversial field that has drawn harsh criticism.




When Richard Dawkins wrote his influential work on sociobiology, The Selfish Gene, he included a chapter proposing that genes may not be the only objects of evolution by selection.  Theoretically, any replicator in an environment of differential success may evolve by the same process.  Units of cultural information, or memes as he called them, are such replicators (Dawkins, 1976).  Dawkins was not the first to propose this, as Darwin himself supposed that language evolved (Darwin, 1871).  Yet it was Dawkins’ term that caught on, and now there is a fledgling science of memetics.


A meme, according to Dawkins, is an idea or parcel of information that replicates itself by leaping from mind to mind, evolving under the selection pressures of limited mental capacity.  Examples are “tunes, ideas, catch-phrases, clothes fashions, ways of making pots or of building arches” (Dawkins, 1976).  Like genes, memes have only one interest: their own replication.


The meme’s-eye view


Since Dawkins was a sociobiologist, it should come as no surprise that the procedure of memetics closely resembles that of sociobiology.  It is the role of the memeticist to take a meme’s-eye view, looking at how behaviors serve the interests of the unit of selection, which in this case is not the gene but the meme.  No longer is genetic reproduction the only factor to which behaviors must ultimately be traced, as it was with sociobiology.  Now they may be traced to memetic replication.  The implication is that some behaviors may serve the interests of memes only, and may in fact be detrimental to the interests of genes, let alone to the conscious desires of the individual people that are their hosts.


The next post will apply these concepts to the evolution of religion, and explore criticisms of memetics.



This post continues a series investigating methods in the study of religion from an evolutionary perspective.  Various approaches will be considered in turn, and evaluated for their strengths and weaknesses.


Tagged , , , , ,

This week’s studies: Minimally Counterintuitive Concepts and Modes of Religiosity

 book open pages library books knowledge reading, by ksheltonTo keep myself accountable, I’m posting my weekly readings.  The deal is: at least one study or chapter a day, no excuses.

This will also double as a nice record of book and article references, and a trail for those interested in the topics of Evolutionary Religious Studies, Cognitive Science of Religion, and Culture and Cognition.

This week’s studies


M. A. Upal (2007) What is More Memorable Counterintuitive Concepts Interpreted Metaphorically or Literally?, in Proceedings of the 29th Annual Meeting of the Cognitive Science Society, pages 1581-1586, Mahwah, NJ: Lawrence Erlbaum.

– Unfortunately, this is a terrible study terribly written.  It gives only the barest description of the experiments, with almost no methodological precautions at all, and mixes up the labels “literal” and “metaphorical” at least once (on Study 1) or possibly throughout the whole article (if Study 1 is actually correct and Study 2 is mistakenly labeled).  It also seems to confuse domain-level counterintuitiveness based on evolved cognition with cultural counterintuitiveness based on learning.  On all accounts, this article gets an F.


Whitehouse, H.  (2009).  “Graeco-Roman Religions and the Cognitive Science of Religion.”  In: Martin, H. L. and Panayotis, P., eds., Imagistic Traditions in the Graeco-Roman World: A Cognitive Modeling of History of Religious Research: Acts of the Panel held during the XIX Congress of the International Association of History of Religions, Tokyo, Japan, March 2005, Thessaloniki: Vanias Editions.

– Whitehouse introduces the volume with a discussion emphasizing two main points: 1) the variation in religion can only be explained against a backdrop of universality in relation, to which end he connect his modes of religiosity theory with other cognitive theories such as that of counterintuitive concepts; and 2) Greco-Religion seems an anomaly insofar as it never developed a widely-shared doctrine.  The latter anomaly is not very clearly articulated or resolved.  Personally, I remain quite skeptical of Whitehouse’s imagistic/doctrinal ritual concepts, as a binary scheme seems far too course-grained and there seems too much room for bias in attributing a ritual as either one or the other.  Furthermore, Whitehouse is not at all reticent to classify a whole religion as either “doctrinal” or “imagistic”, even though these concepts supposedly apply to rituals, and nearly all religions employ some measure of both.


Gragg, D. L.  (2004).  “Old and New Roman Religion: A Cognitive Account.”  In: Whitehouse, H. and Martin, L. H., eds., Theorizing Religions Past: Archaeology, History, and Cognition, Walnut Creek, CA: Altamira Press.

– Finds Roman religion doesn’t fit into either the doctrinal or imagistic modes.  Emphasizes the durability and stability of traditional Roman religion, simultaneous with the attraction to imagistic mystery cults that attracted some but not all Romans.


Atkinson, Q. D., and Whitehouse, H.  (2011).  “The Cultural Morphospace of Ritual Form: Examining Modes of Religiosity Cross-culturally.”  Evolution and Human Behavior, 32, pp. 50–62.

– Uses a huge database to sample 654 rituals from diverse cultures across the world, in order to test the modes of religiosity theory.  The results come out in the predicted direction, with low-frequency rituals correlating with high arousal levels and high-frequency with low-arousal.  But comparison of euphoric and dysphoric arousal levels throws a monkey wrench in the works.  Only dysphoric arousal correlates (negatively) with frequency.  Euphoric arousal levels remain fairly similar across ritual functions, including doctrinal rituals.  Strangely, Whitehouse doesn’t see this as a problem, but just assumes dysphoric rituals are the thing to focus on.  On the contrary, I would argue the results could be explained much more simply by appealing to the nature of dysphoric rituals: you simply can’t have high-frequency dysphoric rituals – the body and mind can’t bear it.  It seems to me that the whole theory needs to be rethought now that euphoric arousal level has been shown irrelevant to imagistic/doctrinal modes.


Upal, M. A.  (2005).  “Role of Context in Memorability of Intuitive and Counterintuitive Concepts.”  In: Proceedings of the 27th Annual Meeting of the Cognitive Science Society (2005), pp. 2224-2229
– A frustrating study.  An intriguing idea, but I can’t see how the numbers come out the way Upal says they do.  He says results are significant and robust, but it looks to me like most results are either p = greater than .05 or else they are just barely significant.  Further, there is no indication here that Upal appreciates the fact that the counterintuitiveness theorized by Boyer and Barrett refers to domain-level counterintuitiveness based on evolved cognitive architecture.  Instead, Upal seems to interpret it as counterintuitiveness based on prior cultural expectations.


Mithen, S.  (2010)  “The domestication of water: water management in the ancient world and its prehistoric origins in the Jordan Valley.”  Philosophical Transactions of the Royal Society A, 368 (1931). pp. 5249-5274.
– argues that it was domestication of water, not agriculture, that led to urban civilization, because earlier agricultural efforts without water domestication proved unsustainable.


Hirschfeld, L. A.  “The acquisition of social categories based on domain-specific competence or on knowledge transfer.”  In: Hirschfeld, L. A. and Gelman, S. A., eds.  Mapping the Mind: Domain Specificity in Cognition and CultureCambridge: Cambridge University Press.

– While this article looked promising, effectively asking whether perceptions of race are intuitive or learned, there was very little in it I proved able to understand.  I’ll have to come back to it again at some later date when my background knowledge is more filled in.


Tagged , ,