Cumulative Culture Evolved to Rapidly Coordinate Novel Behaviours
In the deliberations over humanity and its perceived uniqueness, a link is frequently made between our ability to support a rich, diverse culture and the origin of complex human behaviour. Yet what is often overlooked in our view of these two, clearly connected phenomena is the thread that weaves them together: the ability to coordinate behaviour. We need only look at the products of our culture, from language to religion, to see that any variant we may deem successful is contingent on coordinating the behaviour of two or more individuals. Still, what is truly illuminating about this ability is that, far from being a uniquely human feature, the ability to coordinate behaviour is ubiquitous throughout the many organismal kingdoms.
So, what does it mean to coordinate behaviour?
If we take behaviour, in the context of my question, to simply refer to the actions or reactions of organism, then coordinated behaviour is when two or more organisms display similar behavioural patterns. Take, for instance, two squirrels that both conceal their cache of nuts in secret locations. Despite both squirrels choosing to hide their nuts in different locations, perhaps as a strategy to stop each other from hording the whole nut supply for themselves, they are displaying an instance of coordinated behaviour: the squirrels are performing practically identical behaviours to solve a problem. Importantly, coordinated behaviour should not be confused with cooperative behaviour, which is part of the reason why I choose to highlight Squirrel nut hiding: in this particular instance, coordinated behaviour led to non-cooperative actions.
Thus, we can initially view this problem from two timescales: across evolutionary time (phylogenetic) and the developmental lifecycle of an organism (ontogenetic). Furthermore, these timescales also have profound consequences from both the perspective of an individual and for an entire population. Broadly summarised, coordinating behaviour is when an organism is in behavioural sync with the larger population. That is, when an individual organism’s prior behavioural biases are constrained, through both natural selection and development, in a direction that coordinates its posterior behavioural biases with two or more organisms (the population). As we shall see, coordinating behaviour takes place over different timescales, and can be achieved through a plethora of methods, with the central thesis being: how our ability to coordinate behaviour was key in sustaining cumulative cultural evolution. First, however, I will outline why coordinating behaviour is a frequent feature of the natural world.
Coordinated Behaviour as an adaptive solution
Having established what coordinated behaviour is, we now move on to the question of why organisms should coordinate their behaviour. Quite simply, to coordinate behaviour is an old adaptive solution that’s a recurring theme throughout evolution. Take various dances employed by the Honeybee (Apini Apis) as an example: here, individual bees have hardwired into them, via genetic constraints, the ability to use a specific signal (the particular dance employed) to relay a specific spatial location external to the hive. As Chater and Christiansen (2009) highlight, this is a problem of coordination which has been primarily solved “over phylogenetic time, via natural selection, rather than solving it over ontogenetic time, via learning” (pg.8).
Through natural selection it is quite clear why coordinating behaviour is adaptive: if performing a specific behavioural action is advantageous in a certain environment, then, assuming the selection pressures are strong enough, this trait will sweep throughout the population. By fine-tuning an organism to a niche, natural selection allows for a rapid, innate behavioural response to the environment in which it is operating – a distinct advantage over those organisms that take longer to reach an equally useful solution. However, we can develop this argument further, showing how the ability to coordinate behaviour leads to greater degrees of interconnectedness, which in turn leads to group-level phenomena.
One such phenomenon is that of swarm intelligence. Here, we see instances of coordinated behaviour in organisms – from ant colonies and bird flocking to bacterial growth and fish schooling – resulting in collective, decentralised systems of organisation. Bird flocking is a particularly familiar example of swarm intelligence being adaptive: individually, birds outside the flock would be easy prey for predators, but by flying together as a flock, birds are able to operate as a single entity, ducking and diving and swerving to avoid being caught (as the video below shows). By adopting parallel patterns of behaviour, organisms operating as a swarm intelligence show how self-organising principles provide adaptive solutions.
Coordinating Behaviour over different timescales
There are a range of ways in which an organism may modify its behaviour. Learning strategies, such as emulation and imitation, are such instances. Furthermore, the degree to which a behaviour is influenced by the environment and/or genetics is yet another facet to consider. But for the purposes of simplicity, and to keep the argument relatively succinct, I am going to divide this section across three timescales, specifically: phylogenetic, ontogenetic and cultural historical.
Coordinating behaviour not only takes place across these aforementioned timescales, it also, in humans at the very least, is a complicated dynamic involving all three. We have already seen one example of phylogenetic effects of coordinating behaviour in the honeybee. Looking at the problem more abstractly, however, we can see that behaviour in a population is coordinated through selection for genes that constrain the collective biases in a certain direction. As such, a particular organism will be adapted, or pre-equipped, to the environment in which it operates, circumventing the need for a long and expensive developmental period.
Aligning behaviour through natural selection is a fantastic solution for very specific niches. Yet, when confronted with diverse and fluctuating environments, natural selection alone is not sufficient to tackle the problem (but see: balancing selection). The difficulty for natural selection in this case is two-fold: first, behaviour that’s highly constrained in a particular direction, or niche, makes it difficult for an organism to rapidly adapt to new environmental challenges; second, it takes more than a single generation for a population to embed themselves within a specific environmental niche.
This is why we see animals with more complex behaviours also undergo a greater degree of development. Over the course of an organism’s lifespan they are further constraining their pre-equipped biases into more narrowly defined behavioural patterns. As such, organisms with successively prolonged development are different in that they likely have behaviour that’s not as initially constrained in a certain direction, when compared with behaviourally simpler organisms, meaning they are capable of adapting to novel situations without being highly specified to certain niches. Instead, natural selection operates on more general cognitive capacities – e.g. neural plasticity, processing capabilities, encephalization etc – allowing for a protracted developmental cycle to fine-tune the behavioural biases. This is not to say natural selection cannot operate on specific behavioural biases within a population. If environmental conditions are stable then natural selection can, hypothetically speaking, constrain the prior biases at a genetic level.
Still, if we consider coordinated behaviour to be a common adaptive solution, then the logic can be extended to argue it is advantageous for organisms to coordinate their behaviour far more rapidly. Humans are perhaps the most poignant example, with Dr John Skoyles of UCL claiming that three fundamental features underlie our advanced cognition: “(1) Human evolution by expanding brain size would have provided extra cortical space for neural plasticity to underlie nonevolved cognitions. (2) Human evolution by changing the body’s input and output capacities (articulate hands, bipedalism, vocal tract and breathing modifications) would have provided new types of information and control for the development of such cognitions. (3)Cultural processes, increased in complexity by these new cognitions, would then through their material, content (symbols), and motivational effects have led to the development of further nonevolved cognitive capacities.” (2009, pg.1)
As previously noted, coordinated behaviour can lead to group-level phenomena. Culture is unique in that it is a group-level phenomenon but is also an evolutionary system in itself. Essentially, it is a way of interacting with the environment far quicker than through natural selection.
We can view culture as “information that is capable of affecting individuals’ behaviour, which they acquire from other individuals through teaching, imitation and other forms of social learning… ‘information’ includes knowledge, beliefs, values and skills” (Laland, Odling-Smee & Myles, 2010, pg. 138). Culture then, given the right evolutionary and developmental conditions, can be highly advantageous as it coordinates novel behaviours rapidly. This means a single generation can solve a problem (e.g. climatic change) far faster than natural selection (e.g. growing fur) by creating a cultural product (e.g. a coat). In truth, the situation is far more complicated than humans having simply bypassed natural selection; rather, the combination of the genes and culture often reveal “[...] patterns and rates of change that are uncharacteristic of more traditional population genetic theory. Gene-culture dynamics are typically faster, stronger and operate over a broader range of conditions than conventional evolutionary dynamics” (Laland, Odling-Smee & Myles, 2010, pg. 137).
As such, the survival of a cultural variant is contingent on it being successful at coordinating behaviour. The more coordinated a population, the easier it is for cultural variants created within that population to spread. Still, humans are not the only animals capable of hosting a culture, as indicated by studies into the behavioural variation of wild Chimpanzee populations (see Lycett, Collard & McGrew, 2009). However, it takes certain demographic conditions for novel cultural variants to arise and spread. Perhaps, what differentiates humans is our ability to coordinate behaviour via cumulative cultural evolution. It is with this in which we turn to our next section.
Coordinating Behaviour, Population Dynamics and Cumulative Cultural Evolution
When examining the complexities of human behaviour, and its range of variation, the role of cumulative culture cannot be underestimated. A succinct definition of cumulative cultural evolution is offered by Boyd & Richerson (1994), whereby a culture permits “learned improvements to accumulate from one generation to the next” (pg. 134). But how do humans maintain this unparalleled form of cumulative culture?
With culture in place to rapidly spread novel behaviours, and thus being an adaptive solution, ancestral human populations were able to develop highly coordinated behaviour within single generations. Through coordinating behaviour in this manner, human groups began to develop greater degrees of interconnectivity. This laid the foundations for population growth. As Lycett and Norton (2010) illustrate in their demographic model of Palaeolithic technological evolution, “Different conditions of density and effective population size will have profound consequences for the degree of social interconnectedness, the latter of which will essentially equate to the number of skilled practitioners of a given craft” (pg. 57).
Specifically, they argue demography “is a reflection of three inter-related factors: population size, density and interconnectedness… Social interconnectedness reflects the likelihood of encountering a given craft skill and the regularity of such encounters. Social interconnectedness is thus somewhat proportional to the parameters of effective population size (i.e. number of skilled craft practitioners) and population density (i.e. probability of encounter due to degree of aggregation)” (pg. 57, my emphasis). By varying these population conditions, Lycett and Norton use their model to show how different stages of lithic technologies may be sustained (see figure below). Crucially, demographic levels may decrease and lead to a situation where sustaining already-created technological innovations may not be tenable for the long-term.
The general conclusions drawn from the model are that given certain demographic conditions – non-dispersed population and a large effective population size – humans show a greater degree of social interconnectedness. In turn, cumulative cultural learning is possible “because the effect of having a larger number of models from which to pick the most skilled, exceeds the losses resulting from imperfect copying of that skill.” (Lycett & Norton, 2010, pg. 57). By undergoing stronger incidences of social contact and cultural transmission, we can infer social interconnectedness directly relates to how well a group is behaviourally coordinated. Interestingly, the demographic processes described mirror those conditions under which human biological evolution may have accelerated. Meaning, human evolution has been advantageous for both the spread of novel cultural variants and adaptive genes. This may explain the reason why we are seeing growing evidence of gene-culture coevolution in modern human populations (see Laland, Odling-Smee & Myles, 2010).
Factoring in my previous point of successfully propagating cultural variants being those which are good at coordinating behaviour, and assuming the right population conditions are in place, then three conclusions can be proposed: 1) cultural variants are adapted to the general population; 2) cultural variants shape an individual’s biases; and, 3) an individual can build upon previous cultural variants to create new versions. This means the next generation are given time to refine the previous generation’s cultural variants, with innovations being more likely to spread through a population that’s highly interconnected. As such, certain groups may become tightly knit, and resistant to outsider cultural variants, by perpetually spreading their own cultural variants that subsequently serve to further coordinate behaviour.
In humans, for instance, we can argue belief begets behaviour and behaviour begets belief: that is, an individual’s prior bias is shaped by exposure to the religious environment, which in turn produces a particular behavioural output. This combination of exposure and behavioural output leads to a dynamic, reinforcing interplay between the religious environment and behavioural bias. Given that the religion is maintained by, and adapted to, an already established group of individuals, then an individual being exposed to this environment is also coordinating their behaviour with the group at large. This snowballing effect explains why religions are keen to ‘get them while they’re young’, because children are still undergoing development, and as such have comparatively unrestrained biases to that of adults, making them more susceptible to the environment.
However, if the goal of cultural variants is to coordinate behaviour, then particular forms of cultural variants can evolve and lead to maladaptive behaviours. Suicide contagions are such a phenomenon. Here, exposure to the idea to commit suicide (cultural variant) spreads from individual to individual (transmission vector), resulting in them committing suicide (the behavioural output). Interestingly, research by Dr Madelyn Gould, indicates: “While clusters have included friends or acquaintances in the same school or church, it is not necessary for the decedents to have direct contact with each other: sometimes knowledge of the first suicides were obtained through the news media. Other mechanisms, such as a shared environmental stressor, may also underlie suicide clusters.”
Of course, further study is needed to say whether the success of particular suicide contagions is contingent on having a shared socio-cultural environment. If it is, then the determining factor is based on the level of interconnectedness in how the coordinated behaviour is produced. For instance, teenagers, as a peer group, tend to have more in common with one another than they do with OAPs. This shared culture, which is reinforced through a vast number of cultural variants (music being a good example), can lead to numerous instances of coordinated behaviour. Therefore, a suicide contagion that’s adapted to spreading through this socio-cultural group will be successful in getting some members to actually carry out the act of suicide. But as we know, not all teenagers commit suicide when exposed to the variant, which is because their behaviour is coordinated with other elements of society that increase their immunity. Simply put: humanity is not a unified blob of conformity. On the contrary, we’re in fact the product of many competing biases, which are subjected to many influences (one recent example I read was concerning noisy genes during development).
Largely, this is mostly conjecture and what I’d like to see is something specifically testing a set of hypotheses relating to coordinated behaviour. The spread of suicide contagions would be a particularly useful case, and something I’d personally like to pursue. For now though, the next step in developing the argument is to show how coordinated behaviour can, given certain conditions, lead to cultural speciation in human populations. For this I will focus on two particular cases: Tasmanian Aborigines and the Pirahã.
Chater, N. & Christiansen, M.H. (2009). Language Acquisition Meets Language Evolution Cognitive Science : 10.1111/j.1551-6709.2009.01049.x
Laland KN, Odling-Smee J, & Myles S (2010). How culture shaped the human genome: bringing genetics and the human sciences together. Nature reviews. Genetics, 11 (2), 137-48 PMID: 20084086
Lycett, S., & Norton, C. (2010). A demographic model for Palaeolithic technological evolution: The case of East Asia and the Movius Line Quaternary International, 211 (1-2), 55-65 DOI: 10.1016/j.quaint.2008.12.001