(Note: this is a “brief” summary of my MA dissertation. It’s still rather long, but hopefully of interest to those studying the cognitive science of religion)
In this piece I attempt to develop an account of the cognitive origins of religion; firstly, by arguing that the emergence of modern human consciousness during the “creative explosion” of 30,000 – 60,000 years ago caused the fusing of several human cognitive adaptations, an event which led to a sense of personal identity and a susceptibility to certain kinds of cultural ideas, and secondly, by invoking the concept of “memes” to explain why, following the initial emergence of religious ideas (brought on by “sparks” of creativity, i.e. novel multi-modular idea development), certain kinds of ideas become more successful at “exploiting” the susceptibility of the human mind, and hence remained numerous in the meme-pool. My theory is an attempt at a “third-way” – a theory based on the insights of spandrelist arguments, but one which invokes ideas which are generally seen as incompatible with spandrelist arguments (i.e. memes). I argue that such a synthesis is possible, and that the assumption of incompatibility stems from misunderstanding of certain terms.
Many aspects of human cognition provide obvious functional benefits for our species. For example, social cognition enables us to perceive and understand the intentions of others, thus enabling us to navigate the social world and create, maintain and develop social ties with others. The evolutionary functions of religion, however, are not so obvious. Religious cognition and behaviour is elaborate and costly. If one assumes that gods do not actually feature as part of the physical world, then the existence of such costly cognition and behaviour is rather puzzling. Religious people are known to take part in bizarre, painful rituals, invest in expensive architecture and objects, go celibate for their entire lives, wage war against other religions, and misperceive the world as filled with spiritual entities, including angels, demons, ghosts and gods. Furthermore, religions exist in some shape or form in every single known human society. What could possibly be the evolutionary benefit of such cognition/behaviour, and why is it so prevalent?
Currently, answers to this question tend to fall into one of two camps: adaptation theories and spandrel theories. Adaptation theories of religion argue that religion evolved for a specific purpose – an answer book to life’s riddles, an existential purpose generator, a solution to problems of social exchange, and many others. For adaptation theorists, religion is a coherently “designed” system which emerged to solve one (or several) ecological problems faced by early humans. Spandrel theories of religion argue that religion has no adaptational function per se but is rather a collection of side-effects of other adaptations; a “spandrel”. Religious beliefs may arise from innate cognitive heuristics and biases that in themselves are adaptations, but taken as a whole lead to a propensity for seemingly irrational belief systems. There are many fascinating theories within these two camps, and it would take too much time to properly discuss them here. For the purposes of this piece, I will briefly discuss a couple of theories which fall into the spandrel camp.
Hyperactive agent detection device
Justin Barrett argues that humans possess a hyperactive agent detection device (HADD) which animates the world with human-like agents, even where there are none. This is, again, the over-zealous application of a cognitive ability – in this case, the ability to recognise persons, which is a useful cognitive ability and an essential building block of theory of mind. Hyperactive agent detection occurs in many situations. For example, people see faces in the clouds, a man on the moon, and the face of Jesus in their toast. People anthropomorphise their pets, toys, cars, and computers. People see ghosts in darkened rooms, elves out the corner of their eyes, and aliens flying through the skies. Most of the time, however, we see actual people – our friends, co-workers, spouses etc. The agency detection device seems to work just fine in the vast majority of situations. We are perfectly able to tell the difference between human beings and other animals, plants, insects and inanimate objects. In fact, the over-zealous agency detection is relatively uncommon. Because it is uncommon, however, it stands out to us.
Assuming that selection has worked to improve our ability at recognising persons, then it seems strange that it would misfire in these ways. There are certain benefits, however, to having a slightly over-zealous system. Firstly, it is much more dangerous to detect an agent 9 times out of 10 than 11 times out of 10. That 1 time out of 10 where you don’t detect an agent could mean that you end up dead. Secondly, it incurs extra cognitive costs to hone a very accurate agent detection device. If the over-zealous application of agency detection is harmless, then there is no reason to incur the extra cognitive costs of an entirely accurate system. It is better to have a system which detects all possible agents and a little bit more than a system which either only detects 90% of agents or is entirely accurate, should the hyperactive agent detection prove to be harmless. It is a rule of thumb.
The HADD can explain why people are liable to see ghosts, elves and other strange supernatural entities in the environment. It can also explain why people are liable to see things which empirically “confirm” their religious beliefs, i.e. Jesus on toast, or god’s face in the clouds. I have heard anecdotal reports of children who believe that distantly observed rays of sunlight shining through gaps in the cloud cover are shining because god has performed a miracle there. The clouds, in this case, take on the form of a powerful agent in the eyes of the child.
Minimally counter-intuitive structures
Pascal Boyer notes that a defining feature of any religious idea or symbol is its minimally counter-intuitive content. Religious concepts minimally violate our intuitions of how natural kinds are expected to be. “Minimal” in this sense means that only several of our relevant intuitions are violated while most of our intuitions regarding natural kinds are left intact. Many gods both ancient and modern are portrayed as being very human-like, but unlike humans they possess one or several unnatural abilities which break the laws of nature in startling ways. The Christian god is usually portrayed as a white, bearded, middle-aged male who also happens to be the all-powerful creator of the universe. The Hindu god Ganesha has fewer powers than the single Christian god (he is able to remove and place obstacles in the paths of those deserving of this) but he is usually depicted as having the head of an elephant. All religious symbols appear to be minimally counter-intuitive in this way – this is what makes them memorable. A religious idea which violates too many intuitive concepts, e.g. god as described by many professional theologians and philosophers, is unlikely to be easily remembered, even by those who have studied such work extensively.
It makes adaptive sense to be surprised by something counter-intuitive. Things that violate your expectations of the world could easily be dangerous, and if you are surprised by such things then they are more likely to be paid attention to. This then diverts cognitive resources to assessing the counter-intuitive object and deciding whether it is likely to be dangerous or beneficial. This can explain why religious ideas take on the forms they usually do. It may seem initially quite baffling that someone would believe in an elephant headed humanoid god, but it seems to make more sense in light of this. Of course, actually believing in something is different to just remembering the concept because it is minimally counter-intuitive. Boyer, however, does not claim that the theory of minimally counter-intuitive structures is the only explanation for religious cognition. It is just another aspect of human cognition which – along with other select cognitive adaptations – leads to the emergence of religion.
There are many other theories which could be discussed here (Deborah Kelemen’s theory of promiscuous teleology for example). These various cognitive “quirks” are adaptational in their own right but are hypothesised to work together to lead to the emergence of religious cognition and behaviour. Each one individually cannot explain the entire phenomenon.
The problem with spandrel theories of religious cognition
The various modules I have discussed so far all plausibly play a part in religious cognition. As I have described, various aspects of religion activate each module in some way. They each contribute to the “naturalness” of certain religious ideas, i.e. we find these ideas “make sense” because of the workings of these various modules. What, though, does this actually allow us to conclude about the origins of religion? Well, not actually very much. Firstly, no spandrel theorist (to my knowledge), has presented a case as to how spiritual cognition actually emerged as a side-effect. In fact, Boyer makes this explicit in Religion Explained, when he states that “…note how all this is not so much caused as made more likely by the cognitive processes I described” (p. 343). The usual strategy has been to unearth as many cognitive systems and sub-systems that could be linked to religious cognition as possible, and then argue that, following this, it is wholly explicable that religion would emerge as a side-effect. Religious ideas, on this view, just occur “naturally” as a side-effect of “normal” cognition. But there is just no obvious reason that they would. It seems that the only conclusion that can legitimately be drawn from this evidence is that if religious ideas exist, then human minds are particularly receptive to them. Spiritual ideas seem to resonate with human minds because of the aforementioned mental modules, i.e. humans have a predisposition to finding religious ideas “natural” or easy to believe. This does not, however, explain the evolutionary origin of such ideas.
A technical note: modularity
Evolutionary psychologists claim that the mind is a collection of mental adaptations, or mental “modules”. For the purposes of the rest of this piece it will be useful to use the language of modularity. As I have just argued, spandrel theorists argue that there are many discrete innate cognitive systems at work when spiritual thinking occurs. For evolutionary psychologists, these innate cognitive systems are modules. Modules are computational systems that are hypothesised both to aid and constrain cognition. The paradigm example of a module is the language acquisition device which was postulated by Noam Chomsky to explain several striking features of language learning and use. Children learn language very quickly, with little or no explicit tuition and from relatively little data. Furthermore, once a child has learned a language and grown into an adult, the task of learning a second language becomes extremely difficult and the process becomes very slow, with lots of explicit tuition and a need to be exposed to very large amount of linguistic data. Chomsky argued that this could be explained by an innate mechanism that contains a large amount of grammatical parameters and directs the learning of a pre-linguistically competent child towards important features of the linguistic data available to them in the first few years of life. This is referred to as the “critical period” for language learning. Once it has passed, the language acquisition module becomes inactive, making it very difficult to learn a language later in life.
For evolutionary psychologists, the ability of humans to use language is a cognitive adaptation which evolved to meet the communicative needs of the human species. Cosmides and Tooby hypothesise that many other aspects of human cognition and behaviour are similar to language in this sense – i.e. they believe that underlying the overt behaviour and cognition of humans are a set of modules, the basic structures of which are innate, heavily constrain our learning, and are evolutionary adaptations. The innate cognitive systems that spandrel theorists argue give rise to spiritual thinking, then, are part of many cognitive modules that have evolved to meet specific ecological problems faced by humans in our evolutionary past.
My “third-way” theory
Now I have set the stage, it’s time to develop my own theory. This theory is a “third-way”, in the sense that I am not claiming that religion is an adaptation or a spandrel, but something else altogether. I will be using the insights into the cognitive aspects of religiosity provided by spandrel theorists but, however, that doesn’t mean that my theory is a spandrel theory.
Global workspace theory
Bernard Baars has proposed a theory of consciousness entitled “global workspace theory” which argues that consciousness might help to mobilize and integrate brain functions that are otherwise separate and independent. The subjective experience of consciousness, on this view, is essentially a spotlight of selective attention. This selective attention has a limited capacity, within which only one kind of consistent content can be visible at any given moment. Input into this spotlight of selective attention comes from one of the many specialised networks (modules) of which the brain is constituted. Once information enters the spotlight, it is globally broadcast across the brain, allowing communication across many different modules. The limited capacity spotlight, then, is a central information exchange, allowing certain regions (such as the sensory cortex) to broadcast their information to many different areas of the brain.
The spotlight is referred to as the global workspace because globally broadcast information can affect the actions of unconscious processes occurring in the specialised local networks. The result of this is that any information that enters the global workspace can affect behaviour, essentially allowing us to exert executive control and voluntary action.
It is only recently that this theory has garnered any actual scientific evidence, with the advent of modern brain imaging techniques. Even now, designing experiments to test consciousness is notoriously tricky. One such experiment has been conducted recently. In this experiment, neural activity was recorded directly from the brains of epileptic participants who had previously had neural implants inserted because of impending neural surgery. They were looking to test the global workspace theory, by seeing whether conscious processing of words ignited coherent neural activity all across the brain. They used a visual masking paradigm, which has been used in many behavioural and brain-imaging studies over the last 30 years. Words were flashed briefly into the participants’ vision before being “masked” by nonsense syllables. In this condition, the words were only visible for 29ms before being masked. In another condition, words were flashed for 29ms but then were not masked. In this “unmasked” condition, words were consciously reportable, and were categorised better than chance in a categorisation task on the emotional valence of words, in which participants were forced to choose between several categories. Masked words were actually processed, albeit non-consciously. This was observed in multiple cortical areas, mostly within an early time window, but without coherent long-distance neural activity. In contrast, unmasked words were consciously processed, and this conscious processing was characterised by the convergence of four distinct neurophysiological markers: sustained voltage changes, particularly in prefrontal cortex, large increases in spectral power in the gamma band, increases in long-distance phase synchrony in the beta range, and increases in long-range Granger causality.
This experiment appears to support the idea that consciousness involves the global broadcast of information, and that conscious processing involves distinctly different coherent neural activity to unconscious processing.
The creative explosion: the emergence of consciousness?
Stephen Mithen, in The Prehistory Of The Mind argues that the crucial thing which set modern humans apart from other humans is that cognitive domains, which had previously been separate, became fused together and allowed for “cognitive fluidity”. Previous human species had highly developed modules for many different adaptational problems, but these modules were independent of one another and so information could not flow readily between them. Mithen uses this hypothesis to explain several intriguing aspects of the archaeological record of early humans. Early Homo sapiens did not make tools out of bone or antler. Mithen argues that this is because whilst early humans had advanced domains of technical and natural history intelligence, there was no connection between the two. Therefore, pre-cognitive fluidity, it was not possible for early humans to think of using animal parts to make tools. Early humans also did not make multi-component tools, or specialised tools for taking down specific animals. Mithen argues that these oddities also make sense if one considers that the technical and natural history domains were not linked in any way.
The archaeological record shows that there was relatively little human cognitive or cultural development for an extremely long period of time (I wish to stress the word “relatively” here – it is only relative in relation to what was to come). When considering early humans in the Lower Palaeolithic (the period spanning from about 2.5 million years ago to about 100,000 years ago) the technological evidence in the archaeological record is one of prolonged monotony, patterns of technological traditions being maintained for tens of thousands of generation. However, the period about 60,000 to 30,000 years ago, referred to by John Pfeiffer as “the creative explosion”, was a period of enormous change for human beings. During this time, the archaeological record becomes suddenly saturated by an abundance of artistic and religious artefacts. This period is, arguably, when religion appeared in human societies. It is during this time when we first begin to see elaborate burials, the manufacture of bone artefacts, cave paintings, and artefacts designed for personal decoration, amongst many other things.
During the creative explosion, then, previously separate modules were fused and allowed for cognitive fluidity. Mithen argues that artefacts such as totems (human bodies with animal heads) demonstrate the integration of natural history and technical domains. Similarly religion, for Mithen, arises out of an integration of social and natural history domains. Whilst I think that Mithen is basically on the right lines, he does not explain how the integration occurs, nor how new ways of thinking emerge from domain integration. He does, however argue that “consciousness adopted the role of an integrating mechanism for knowledge that had previously been ‘trapped’ in separate specialized intelligences”. He seems, then, to be anticipating the role of the global workspace in facilitation of domain integration. Retaining Mithen’s main argument, we can see that the creative explosion is the period where the consciousness of modern humans arose – the separate specialised networks began to fuse and transfer information in a central information exchange. Domains could finally be integrated in a way necessary for the development of modern human artefacts. The question now is: how specifically does consciousness lead to the creation of novel ideas, such as religious ideas?
The stage is now set for the initial emergence of religion. As I mentioned previously, spandrel theories have so far focussed on explaining how religious ideas are likely to seem “natural” to humans. No specific theories have been put-forward to explain how religious ideas arose in the first place. Whilst normal cognition may make people receptive to religious ideas, it cannot account for their initial creation. Rather, what is needed is the initial spark of creativity.
Creativity – that is, the ability of humans to invent, entertain, and transmit multi-domain ideas – appeared during the creative explosion because of the fusing of modules. Crucially, “true” creativity involves the creation of novel multi-domain ideas, i.e. ideas which have not been entertained by others previously (this is, to a degree, what is valued in artists). This kind of creativity is also essential for progress in other domains, for example the initial development of scientific theories, or the initial spark of an idea for a new engineering design. Why and how this “spark” occurs is a particularly puzzling question, but there is a possibility that it may occur during the mental representation of ideas in the global workspace. For example, say you entertain the idea of an elephant in your mind. An image of an elephant pops up into your consciousness, and thus enters the global workspace. In the global workspace, this image is broadcast across the brain, and multiple modules take the image as input. Now, the longer you consciously consider this image, then the more the likelihood that various mental modules unrelated to the normal processing of the concept of “elephant” will process the information, and churn out something odd as output, which also enters the global workspace. You thus have a brief flash of an elephant-but-not-elephant in your consciousness, perhaps, for example, a human with an elephant head. If this particular idea grabs your attention (which it would do, if it is truly novel to you, because it is a minimally counter-intuitive idea), then you are likely to further consider it. It, too, will eventually be subject to sparks of creativity – multi-modular workings taking the same input and churning out strange modified images as the output.
Personal identity and existential angst
Whilst globally broadcasting information would allow for novel ways of thinking, it would not guarantee the spontaneous emergence of specifically religious ideas. There would have to be a particular reason for the various modules involved in spiritual cognition to work together.
The continual stream of consciousness equates with our feeling that there is an individual persisting over time. The “spotlight” of consciousness, then, is effectively the same as our sense of personal identity, i.e. the feeling that there is an “I” rather than just a series of causally related cognitive events. This definition of personal identity is the one espoused by John Locke, who argues that a person is a being who can “consider itself as itself, the same thinking thing in different times and places; which it does only by that consciousness, which is inseparable from thinking, and as it seems to me essential to it: it being impossible for anyone to perceive, without perceiving, that he does perceive” (Essay II.27.9). He further argues: “Nothing but consciousness can unite remote existences into the same person; the identity of substance will not do it, for whatever substance there is, however framed, without consciousness, there is no person” (Essay II.27.23).
The fusing of mental modules was important for many reasons, but one of its most profound effects was to develop in humans a sense of personal identity and self-awareness. We are aware that we exist. We can conceive of ourselves in the future, present and past. We can do the same for others around us. An unfortunate consequence of this, however, is that we can conceive of ourselves dying, and can do the same for those around us. We can also conceive of ourselves as a materially existing being, occupying a place in the world. This, however, means we can also wonder why we occupy our place in the world. We can wonder why we do the things we do. Scott Atran refers to this as the “tragedy of cognition”, and that name seems rather apt. Humans, in gaining personal identity and self-awareness, gained something unfortunate as a side-effect: existential angst.
We can see now why religions prove so appealing. One of the adaptational candidate functions of religion is that it may function as an answer book to the complex existential questions of life. Religions around the world do indeed provide metaphysical explanations for the existence of life, the universe and everything. They explain that when you and your friends die, you don’t truly die, because your spirit or soul remains alive. They explain that you are in the world for a specific purpose (although this varies from religion to religion). They explain that all the apparent disorder and randomness in the world is caused by an agent or agents, and the world itself exists because it was created by an agent or agents. The existential questions which arise after the emergence of consciousness and personal identity are answered strikingly well by religious ideologies, and are answered in a way which resonates with many of our innate cognitive mechanisms – mechanisms whose workings are unified in the central information exchange, allowing us to consciously entertain religious ideas in a way which would have been impossible had the central information exchange not existed. It was existential anxiety, then, which drove the initial “need” for metaphysical answers to these questions, and the particular form which these answers take (i.e. spiritual thinking) is a result of these forms of answers seeming particularly “natural” to human beings.
Cultural transmission: memetics
Religion, then, is neither a coherent solution to an ecological problem nor is it a useless side-effect. Rather, natural selection left us with a kind of cognitive need for particular kinds of ideas – ideas which could eventually be supplied by the kind of creativity allowed by the global workspace.
Memetics is the study of memes (a concept first postulated by Richard Dawkins). A meme is a hypothetical element of cultural information, transmitted from one individual to another through various imitable methods (e.g. rituals, gestures, speech). In the same way that a gene is the basic unit of biological heredity, a meme is hypothesised to be the basic unit of cultural heredity. A gene works with many other genes to develop specific traits in a body (or, as Dawkins refers to it, a “survival machine” for that gene), be it physiological or psychological. If a particular gene happens to contribute to the development of a trait which causes the survival machine to have a better chance at surviving, then that survival machine is more likely to pass on its genes to its offspring thus increasing the chance that the gene will survive in another individual. In this fashion, certain genes happen to increase in frequency in a population. Memes are hypothesised to function in a similar fashion – they are also replicators which form the building blocks of bodies. Bodies, in the case of memes, are theories, songs, pictures, plays, etc., and the term for a memetic body is memeplex. A particular meme may help to make its memeplex attractive to a human being in some way, and – if successful – the human thus takes on the cultural information and memorises it. If the memes constructing the information have done their job particularly well, then that person is likely to transmit this aspect of culture to their friends. The memes which compose this aspect of culture may then go on to form the buildings blocks of new pieces of cultural information, and may “mutate” either a) through human creativity, or b) through errors in transmission of cultural information. Memes, then, are more likely to survive if they help construct cultural “bodies” which have elements that prove attractive to human minds. What a meme does, essentially, is exploit certain aspects of human minds. If it can do so, it is more likely to be internalised and transmitted, and thus increase in frequency in the meme pool.
What is the appeal of meme-based arguments? Well, they seem to fit very well with everything I have been describing about religious ideas. For example, one of the main aims of memetics is to ask why certain cultural ideas are so “sticky”, i.e. why certain ideas become so popular and predominant within a culture. Religious memes may be sticky because they resonate with the various mental modules I have been discussing. Religious ideas “tap into” specific kinds of cognition which arise from the workings of these modules, and are thus memorable. For example, humans have a tendency towards over-zealous intention-based reasoning and remembering minimally counter-intuitive structures, so a cultural idea about everything in the world being created by a creator god (who is like a human but much more powerful) is more likely to “stick” and be passed on to others. Thus the memes which compose this particular idea are more likely to survive and form the basis of subsequent cultural ideas which will most likely be very similar to this original cultural idea, but (thanks to mistakes and invention) may differ in some way. Hence the reason that religious ideas are so widespread, but so different in many ways.
The main appeal of memetics, for me, is that regardless of the usage of the term “meme”, cultural ideas are either successful or unsuccessful at catching on. Some ideas die out, some flourish. The ones that flourish must do so because they are particularly appealing to humans in some way. Certain ideas become linked to other ideas and either flourish more because of it or become less popular because of it. This is all relatively uncontroversial. Using the term “meme” simply makes it explicit that ideas are subject to selection according to how well they adapt to their environment (or how well they adapt their “bodies”, or memeplexes to the environment). Regardless of whether you want to use the term meme, ideas still die out or flourish depending on whether or not people find them interesting in some way. It just seems to me that using the term meme makes a lot more sense.
Why have memes fallen out of favour? A re-assessment of the criticism
Meme-based arguments are generally seen as being incompatible with religion-as-spandrel arguments (both Boyer and Atran make this claim). Atran refers to memetics as a “mind-blind” theory, and this appears to be his main reason for dismissing the role of memetics in religion. By “mind-blind”, he means that memetics is apparently unable to take into account cognitive constraints on religious belief and practice. As I have been discussing, there are many reasons to believe that certain cognitive mechanisms make human minds more likely to be receptive to religious ideas. If a theory cannot this aspect of the mind into account, then I agree with Atran that the particular theory is likely to be false. Memetics, however, can take this into account, and has done all along. One important aspect of memetics is that certain memes are attractive to human minds because of the design of human minds. Memes can only be successful if they can “exploit” certain features of human cognition so that they prove attractive. Religious memes, then, prove particularly attractive because they activate many specific aspects of the human mind – specifically, the cognitive mechanisms I have been describing in this essay. What Atran does not realise is that memes are entirely dependent on human minds for their success, and so memes are not “mind-blind” at all. In fact, memes which “understand” the human mind better than other memes are the ones that are likely to be successful and remain in the meme pool.
Atran’s other main argument against memetics is a standard one, which is that memetic transmission introduces errors at an extremely fast rate – too fast, apparently, for selection to take place. I have seen this argument repeated often in various places and I am still confused as to what the problem is. Yes, memetic mutation occurs at a much faster rate than genetic variation. But isn’t it also possible that memetic selection occurs at a much faster rate than genetic variation? If a memeplex is transmitted and several memes miss out because of errors in transmission, then selection is at work. Every single time a memeplex is transmitted, there is mutation and error. Memes are added, memes miss out. Surely this is selection in action? The memes within the memeplex which don’t miss out during transmission are being selected for each time information is transmitted. This is, indeed, rather fast. It is significantly faster than genetic transmission. But no one argued that mimetic transmission had to occur at the same speed as genetic transmission. They are entirely different units of selection.
This does not mean, however, that memetic selection occurs entirely independently of genetic selection. Brains are constructed by genes, even though the cultural information within them is made up of memes. As I argued previously, memes are entirely dependent on the genetic construction of the human mind because is it this that they need to exploit. The speed of selection at each level though is very different, and so it is unlikely that meme selection can have an effect on gene selection. The kinds of memes that are selected for, however, will be the ones that have appropriately adapted to the effects of genetic selection on the minds of humans. Memetic selection is, foundationally, dependent on genetic selection.
The emergence of the first meme
Dawkins argues that the first gene arose as a basic biological replicator, a simple entity that was only remarkable in the sense that it was able to make copies of itself. It was a very improbable event that such a thing should arise by accident, but when considering timescales of millions of years, the occurrence of such improbable events does not seem so strange, especially considering that it only had to happen once. Once it came into existence, it began to make copies of itself. Over time, copying errors gradually crept in. Soon, there were various different kinds of replicators, each with an affinity for its own kind. He completes this story by explaining how such replicators would inadvertently end up “competing”, simply because the ones that didn’t compete would disappear. The ones that were left, then, became a gradually more and more competitive bunch.
A similar story can be told for the origin of the first meme. I have argued that the development of the global workspace during the creative explosion led to multi-modular mental representation and thus sparks of creativity. These sparks of creativity, then, created the first memes. This only had to happen once. These memes were then passed on to others, who mutated them and/or created their own memes. Memes inadvertently ended up competing with one another, because the ones that didn’t disappeared from the meme pool. Thus, by natural selection, memetic competition for the best way to exploit the human mind was born.
In sum: my “third-way” theory
Global workspace theory argues that consciousness involves the collective operation of widely distributed neural networks all fusing into a single stream. During the creative explosion, previously separate and evolutionarily old mental modules became linked through a central information exchange. This linkage of modules led to the fusing of various kinds of information, a novel development which allowed for disparate and unrelated modules to trade information. “Sparks” of creativity, caused by multi-modular idea representation, cause the initial development of novel ideas. The “stream” of consciousness links disparate cognitive events into a single unifying entity: personal identity. The existential questions which arise after the emergence of consciousness and personal identity are answered strikingly well by religious ideologies, and are answered in a way which resonates with many of our innate cognitive mechanisms – mechanisms whose workings are unified in the central information exchange, allowing us to consciously entertain religious ideas in a way which would have been impossible had the central information exchange not existed. Religious ideas are selected for because they exploit several aspects of the human mind, and hence are memes. Religious memes, then, prove particularly attractive because they activate many specific aspects of the human mind – specifically, the cognitive mechanisms I have been describing in this piece.