A matter of taste

We all have different opinions about what tastes good. This variation is normal. But our dogs tend to have very different opinions from us. Have you ever wondered why that is?

For example, consider a time-honoured delicacy of the dog world.

Why do dogs often enjoy eating what their own bodies have determined to be waste? It appears to be more than just a matter of taste, since it is a common behaviour across the species.

A natural answer comes from thinking about the domesticated dog's ancient ancestors.

Like modern wolves, the ancestors of the dog gave birth in a den. Wolf puppies live in the den for 6-8 week before they are finally weaned and begin the process of joining a pack. That den is vulnerable to predators. So, in order to prevent predators from detecting the smell of accumulated feces, the mother eats the excrement of her young.

This behaviour may have been further enforced by the fact that dogs are often scavengers, capable of eating something of meager nutrition in order to survive.

In both cases, there is a simple explanation of why dogs appear to enjoy eating what they do. There was probably more variation of preference among some of the modern dog's ancient ancestors. But the ancient dogs that enjoyed eating their own stink were more likely to survive and pass on their genes. In contrast, over the generations, competing preferences simply died off.

Now, compare this to the human case. Suppose an ancient human child were to have developed a similar preference. What would happen to the child?

Since human excrement carries disease and does not contain enough nutrition to be of much use to a human, a human child with this preference would not have been likely to survive. If such humans ever existed, they died off.

It thus seems possible to explain what tastes good to us is evolutionary terms. But if evolution selects for taste, might it select for higher brain capacities as well? These are questions of Evolutionary Psychology.

Introducing Evolutionary Psychology

In the Conclusion of his 1859 Origin of the Species, Charles Darwin wrote,

"In the distant future I see open fields for far more important researches. Psychology will be based on a new foundation, that of the necessary acquirement of each mental power and capacity by gradation. Light will be thrown on the origin of man and his history."

That future turned out to be more distant than Darwin might have imagined. Psychology took a sharp turn away from biology in the 20th century following the work of Freud. Darwin's suggestion about the use of evolutionary theory to explain specific "mental powers" only flourished late in the 20th century.

Leda Cosmides and John Tooby at the University of California in Santa Barbara helped launch modern evolutionary psychology in a number of works published in 1992. They were concerned with the following kind of example.

Puzzle: A prisoner's dilemma

Suppose a town contains two kinds of people, gardeners and babysitters.

All of the people in this town agree to carry out their special job for someone else once per week. Once per week, each gardener will tend to some babysitter's garden. And once per week, each babysitter will tend to a gardener's child. Being provided with this service is a great benefit to each of them, and helping each other out is a comparatively small cost.

The problem is that, strictly speaking, the choice that provides the greatest benefit for an individual is to cheat, by accepting a service while renigging on one's obligation to help. If a benefit at a small cost is good, then a benefit at no cost is better! This is a version of the Prisoner's Dilemma: deciding renig strictly dominates the decision to cooperate, and so nobody can expect the benefit of cooperation.

Decision Matrix. No matter what a babysitter chooses to do, the gardener gets more benefit (or less harm) by renigging on her obligation.

Babysitter Gardener	Garden for Babysitter	Renig on Babysitter
Babysit for Gardener	Babysitter: Gardener:	Babysitter: Gardener:
Renig on Gardener	Babysitter: Gardener:	Babysitter: Gardener:

This is puzzling, for it looks as if no reasonable society would ever develop social contracts of this kind. But human society is built around social contracts. How is this possible?

Solution? A Cheater-Detection Module

Social contracts make more sense if there is a very bad punishment for cheating. Suppose society includes what Hobbes called a "Leviathan", an enormous power (like a central justice system) that imposes a severe penalty on broken contracts. This would seemingly encourage more cooperation among the gardeners and the babysitters above. The trouble is that people seem to cooperate even in the absence of a Leviathan, and did so historically as well.

Not only that, but non-human animals display social cooperation as well. For example, the group of Meerkats below display a "lookout" behaviour while the others forage. This behaviour isn't easily explained by self-interest alone. Might this kind of cooperation and the social contracts adopted by humans have a common explanation?

Evolutionary Psychologists argue that they have a common cause: the evolution of a mental capacity.

In particular, the story goes, some of our primitive ancestors developed the cognitive capacity to identify cheaters. These individuals would ostracise and punish cheaters once identified, which would discourage the cheating behaviour.

In particular, babysitters who renigged on gardeners would be identified and rejected by the community.

This cognitive capacity was moreover a fitness benefit: individuals that developed it would enjoy the benefit of cooperation, and thus be more likely to pass on their genes and the cheater-detection ability to their descendants. Those without the ability to detect cheaters, in contrast, would die off over the generations.

The capacity to detect and ostracise cheaters is an example of what Evolutionary Psychologists call a mental module: a roughly independent cognitive capacity that is devoted to some particular kind of task, which is postulated to have developed in a genetically-determined manner.

Evolutionary psychologists have even argued that there is empirical evidence for cheater-detection modules. They claim some studies to have shown that humans have enhanced cognitive abilities related to the presence of social contracts, such as the enhanced ability to correctly identify logical possibilities when social contracts are involved. Leda Cosmides in particular exhibited a striking experiment in which subjects were more likely to correctly perform a Wason selection task when social contracts are involved. (These famous experiments were introduced in Cosmides 1989.)

Methodology: Identifying structure from function

The above explanation of social cooperation is a paradigm example of the Evolutionary Psychology research programme. The aim is to understand the structure of the mind. One proceeds by identifying some cognitive ability or behaviour. The function of that ability or behaviour is then explained by proposing a specialised psychological mechanism that arose through natural selection. One then ideally proposes some experimental evidence to support that explanation.

Brain regions thought to be associated with empathy

In short, one aims to understand the structure of the mind by studying the development of mental functions.

The philosophical critique of Evolutionary Psychology

Evolutionary Psychology has been the subject of critique by both philosophers of cognitive science and philosophers of biology. One notable pair of critics is James Woodward at the University of Pittsburgh and Fiona Cowie at the California Institute of Technology.

Woodward and Cowie criticise what they take to be the central premise of Evolutionary Psychology known as the Massive Modularity Hypothesis: that the mind consists of a system of modules or "mental organs" that are shaped by natural selection. The "massive" part refers to the fact that the entire mind is supposed to be made up of such modules.

The Woodward and Cowie critique of this view is a three-pronged attack, which may be summarised as follows.

They take these assumptions to be central to the methodology of evolutionary psychology. If they are right, then it would suggest major problems in the foundations of the field.

Let's consider each of the criticisms in turn.

Evolution does not favour mental modules

The first concern is about whether or not evolution really favours mental modules. There are both conceptual and empirical conerns.

Favouring mental modules: Conceptual concerns

1. Evidence about mental structures? Evolutionary psychologists argue that functional capacities like "cheater detection" provide a fitness advantage, and so are favoured by evolution.

However, their conclusion is that a corresponding mental structure that is favoured by evolution. You can see why this is needed for their story to work: in order for a trait to be favoured by evolution, it has to be passed on from parent to offspring. For evolutionary psychologists, a "mental structure" is just the physical trait that is passed on from parent to child and gives rise to the functional ability.

But does a cognitive function correspond to a unique mental structure? There are some examples that suggest that it is not.

For example, an evolutionary psychologist might identify memory as a cognitive capacity that is explained by evolution. But it is hard to identify a unique mental structure corresponding to memory. Memory does not appear to be localised in the brain, but is rather distributed across many disparate regions. It also appears to correspond to a very complex collection of different structures.

So, focusing on the cognitive ability of "memory" alone is in danger of conflating numerous distinct structures corresponding to short-term memory vs long-term memory, spatial information vs propositional information, and many other things.

2. Individuating modules? It is also difficult to see how one could realistically individuate one mental module from the next.

For example, take our ability to judge numerosity: whether there is a lot or a little of something. This is a highly complex ability, based on an enormous number of mental structures:

• Object detection and individuation
• Depth and edge perception
• Perception of luminance and color boundaries

It would be unjustified to stipulate that judging numerosity corresponds to one mental module that is passed on genetically. Rather, such a mental capacity corresponds to a complex network of further capacities that are inextricably interconnected.

Favouring mental moduels: Empirical concerns

Suppose our cognitive capacities correspond to mental modules that are favoured by evolution. Then what are we to make of learned cognitive capacities? These are capacities that are not passed on genetically, but rather picked up from our environment.

For example, the cognitive capacity to play the piano appears more likely to have been made enabled by other genetically inherited features but still ultimately learned.

Another example is that of re-purposing of mental structures.

The brain is highly flexible. For example, the brains of people who go blind or deaf often repurpose certain regions that were previously devoted to those senses, for use on other tasks.

It is hard to understand how a cognitive capacity corresponds to a fixed mental module when the brain is capable of changing which mental structure is responsible for that cognitive capacity.

The mind is not a system of modules

Evolutionary psychologists sometimes claim to be able to solve a problem known as the frame problem: the amount of actual information that is collected by human senses is exceedingly complex. How does the mind prevent it from being perceived as a disorganised jumble?

The answer, according to evolutionary psychologists, is modules. The mind consists of a system of well-defined 'slots' corresponding to different cognitive capacities like colour perception, judgement of human faces, and so on. These slots serve to "frame" the information that we collect, and allows us to organise it more easily.

However, this solution to the frame problem just hides the complexity somewhere else. The individuation of modules is claimed to be simple, but the interaction between the modules is then exceedingly complex. How does the brain prevent that from descending into a disorganised jumble? It appears that the complexity problem has just been pushed to a different location.

If evolutionary psychology is correct at all, it would seem possible that a single module might arise — call this 'slight modularity' as opposed to 'massive modularity'. For example, the ability to focus on faces appears to be inherited, while the ability to identify faces is learned. So, one might say that face recognition is slightly modular, rather than massively modular.

Thus, by demanding the brain as a whole consist in a system of modules, evolutionary psychology may go too far.

Modules aren't easily defined

Although cognitive capacity or "function" of the brain is clear enough, it is hard to see how a mental module can be defined. We have already seen one difficulty in the fact the brain is capable of repurposing one aspect of the brain for another use.

For example, one might try to explain it in terms of neural behaviour.

However, that doesn't seem to be enough, because neural behaviour is rampantly non-modular. For example, the posterior parietal cortex is involved in attention and spatial awareness, but has recently also been shown to be involved in the planning of goal-directed behaviour.

The role of evolution in psychology

These criticisms are not enough to destroy all hope in Darwin's suggestion. Cognitive capacities are the product of a brain, which is part of an organism that was shaped by evolution. In this sense, mental capacities are the product of evolution.

The difficulty that remains for the foundations of evolutionary psychology is to identify the processes that produce what we observe, and in particular what the units of selection are. This is an interesting foundational problem not just for philosophers, but for biologists, cognitive scientists, and psychologists alike.