From edition

Metaphors, Models, and Modularity

Abstract

Conceptions of the human brain range from the blank slate to a strongly modular structure produced by adaptations to problems our ancestors encountered in African savannas. Metaphors and models can help us formulate hypotheses about the level of constraining structure built into the brain. Here I discuss David Sloan Wilson’s use of the immune system as a model for human cognitive flexibility. The immune system has evolved to cope with the challenges posed by disease-causing organisms that are highly diverse, short-lived, and rapidly evolving. I argue that this model is not a good guide to the kind of cognitive structure that has evolved in response to challenges from more stable features of the ancestral environment, for instance, coping with large predators and hostile conspecifics or finding and selecting food.

*********

Metaphors, Models, and Modularity

Few scientists doubt that evolved human traits are the result of the same processes that generated the great diversity of life on Earth. Most scientists agree that natural selection accounts for the characteristic features of human anatomy and physiology. Few doubt that the human brain and its interactions with the human body have evolved by the same processes. Nonetheless, as contradictory as this might sound, biologists and social scientists sometimes disagree strenuously about the degree to which evolutionary history constrains the structure and functioning of the brain. One might expect doubts about evolutionary constraints from the religious right and from post-modern philosophers and literary theorists. Similar doubts expressed by scientists are more puzzling and invite close attention. The debate over evolutionary constraints on the human brain now centers on one specific discipline: Evolutionary Psychology (EP). Within EP itself, as within any rapidly developing field, there is robust debate about specific features of mind and behavior. What defines EP as a discipline, though, is a set of shared assumptions: that the mind is the product of the brain and its interactions with the body; that the body-brain has evolved in an adaptive relation with its environment; and that “the adapted mind” produces human behavior. This set of assumptions has become the focus of intense debate. For example, in his lengthy book, On Human Natures (2000), Paul Ehrlich claims to debunk most of the claims of evolutionary psychologists. In their recent book, Making Sense of Evolution. The Conceptual Foundations of Evolutionary Biology (2006), Massimo Pigliucci and Jonathan Kaplan devote most of a chapter to a critique of EP.

EP is not based on an assumption that genes directly and rigidly control behavior. Evolutionary psychologists agree that an organism’s neural organization develops as the result of a complex interplay between genes and the environment at intracellular, body, social, and cultural levels. Evolutionary psychologists, in agreement with most other psychologists, view the brain as composed of programs that process information using complex arrangements of neural circuitry. Evolutionary psychologists expect some of these programs to develop reliably in most people if they experience an evolutionarily normal range of environmental conditions.

Evolutionary psychologists differ from most other psychologists, however, in that much, but not all, of their research is motivated by the view that the neural programs of the human mind were designed by natural selection to solve problems faced by our hunter-gatherer ancestors, for instance, finding a mate, cooperating with others, hunting, gathering food and other resources, protecting children and friends, and avoiding predators. They expect that natural selection has produced neural programs that respond to such problems reliably and quickly.

The scientific controversy that surrounds EP focuses on the concept of modularity, that is, functional specialization of the brain. Evolutionary psychologists postulate this functional specialization because the problems our ancestors had to solve were strikingly varied. Selection of a habitat in which to carry out specific activities, discriminating among types of food, dealing with environmental hazards, and choosing hunting partners, to name just a few, are such dramatically different problems that a general problem solver would be unlikely to be well designed to respond appropriately and effectively to more than one of them (Tooby & Cosmides 1992). Several of the metaphors used to describe the proposed modularity of the brain – Swiss army knife, jukebox, and computer software – imply the existence of cognitive modules designed to function appropriately to particular challenges in specific environments. A corollary is that, given that the environment in which people live today differs dramatically from the one in which our ancestors lived, some of our behavioral responses are not currently adaptive. That is, our minds are inhabited by “ghosts,” circuits established during our long residence on African savannas.

We do know, of course, that the minds and bodies of all animals are replete with ghosts: ghosts of past habitats, predators, parasites, competitors, mutualists, and conspecifics; ghosts too of past meteors, volcanic eruptions, hurricanes, and droughts. All adaptations reflect selection in past environments; natural selection cannot anticipate future environments. But do we have reason to expect that contemporary human behavior could still reflect adaptations of our ancestors to conditions on African savannas? Some scientists think that too much time has elapsed for behavioral adaptations molded in African savannas to reside in the human mind.

Traits may persist either because the duration of relaxed selection is too short for genetic mutations and genetic drift to erode them or because they are enmeshed in multiple regulatory pathways that channel development and buffer them from disruptive mutations. Empirical evidence suggests that these processes have resulted in long persistence times of behavioral traits. Adaptations have persisted in some species for thousands of generations after selection on those traits weakened or ceased (Coss & Goldthwaite 1995). Assuming an average generation time of 20 years, no more than 350,000 individuals separate us from “Lucy.” In the absence of strong selection against a trait, its persistence under today’s radically altered conditions is at least plausible. To demonstrate the existence of particular Pleistocene ghosts, however, we must develop specific hypotheses and test them with potentially falsifiable predictions. The existence or nonexistence of ghosts cannot be demonstrated by theoretical arguments.

The controversies surrounding the modularity of the brain and the existence of ghosts are inter-related. If the human brain is indeed a “blank slate,” it can house no ghosts. Any lingering ghosts can reside only in genetically programmed neural circuits. Clearly, the existence of cognitive models also can be settled only empirically. Nevertheless, concepts and metaphors have an important role to play by helping us conceive of possible modules, the problems they are supposed to solve, and the properties of the neural circuits that might serve that purpose. Such research is, of course, being conducted and much evidence of modular components of the brain is emerging (Gazzaniga 2008). My concern here, however, is not with the products of this research. My goal is to explore how we can employ metaphors and models to accelerate progress in understanding the structure and functioning of the human brain.

The dominant metaphor of the Standard Social Science Model (SSSM) is a blank slate on which almost anything can be written. According to this view, the human brain has few significant preprogrammed neural circuits other than some built-in psychological mechanisms that cause organisms to select among initially undirected responses. B. F. Skinner, a prominent early exponent of the blank slate view, encapsulated his position in his influential but controversial novel, Walden Two (1948), in which he describes a society in which human behavior does not reflect evolved predilections to seek status and compete for resources and mates–two universal human traits that evolutionary psychologists judge to be extremely difficult to over-ride. Skinner did not pay much attention to what structure underlying psychological mechanisms must have to take environmental inputs and translate them into the adaptive behavioral outputs that he thought emerged. Recently, David Sloan Wilson has suggested that the immune system might serve as a metaphor and model for the kinds of open-ended processes that Skinner thought could yield adaptive responses (Wilson 2010).

The immune system evolved to enable the body to respond to attacks by disease-causing organisms. It does this by enabling the body to distinguish self from non-self. Disease-causing organisms are remarkably diverse and geographically variable. Their very short life cycles enable them to evolve much more rapidly than the hosts they infect. The immune system functions by producing a vast array of antibodies that can detect and respond to a correspondingly vast array of potential pathogens, most of which are never encountered by a given individual. The central processes that generate the great diversity of the system are based on DNA rearrangements and clonal selection. The cells that form the major component of the immune system are manufactured by stem cells, mainly in the bone marrow. The 5000–10,000 white blood cells per mm3 of blood live only 5-9 days. About two and one-half million red blood cells are produced each second; each lives about 120 days.

Almost every gene that codes for major histocompatibility complex (MHC) proteins has multiple alleles, ranging from a handful to more than 100. Each person inherits and expresses a unique set of MHC alleles. There are millions of different combinations of MHC alleles; very few people are likely to have exactly the same set. The human immune system may be able to respond specifically to 10 million different antigens. The diversity of millions of different cells, each able to produce only one kind of antibody is not generated by millions of genes, each one coding for one antibody molecule. That would require a genome seven times the size of the entire human genome! Rather, the genome of a differentiating cell has a number of alleles for several regions of the protein. Combinations of these alleles generate the diversity.

This massive, expensive, and continuing lifetime production of antibodies has been favored both because disease-causing organisms have always been major sources of morbidity and mortality and because which pathogens are likely to be encountered and their genetic composition cannot be predicted. An expensive, open-ended system may be the only effective response to such threats. Moreover, until the twentieth century humans were unaware of their existence, much less that they caused diseases.

How suitable is the immune system as a model for thinking about flexibility and modularity of the brain? To what other problems and challenges might a similar open-ended system be adaptive? As a metaphor the immune system is useful because it shows that such open-ended systems are possible. They cannot be dismissed on theoretical grounds. Nonetheless, the system as a model has very limited applicability because the problem to which it is a response differs dramatically from the other challenges that evolutionary psychologists study. Its only function is to distinguish self from non-self.

In addition, as Wilson points out, sometimes the process breaks down and the immune system attacks self-cells, causing autoimmune diseases like multiple sclerosis, lupus, and some forms of arthritis and diabetes. The immune system also may respond inappropriately to innocuous substances. The immune system possesses highly constrained flexibility and regularly misfires.

The immune system is probably not broadly applicable as a model for brain evolution because none of the other challenges are posed by massively diverse, mostly undetectable, and rapidly evolving entities. For example, only a few species of macro-predators attack humans. These animals were familiar to our ancestors. They all share readily identifiable features, such as eyes and possession of dangerous, pointed weapons. Attention to and detection of these features does not require mechanism having the open-ended complexity of the immune system. What have evolved are strong responses, both positive and negative, to sharp, pointed forms. Facemasks that enlarge canines evoke strong fearful emotions in most people. Aichmophobia, dread of sharp objects, such as scissors, knives, and needles, is a common human phobia. Snake phobias and rapid, selective responses to snake forms are common human traits, even among people who have never seen a snake.

Selection of food is a more complex challenge because the range of foods available to a dietary generalist varies considerably among environments. Ingesting new, unfamiliar foods is risky, yet a dietary generalist needs to sample and determine the value of potential new food sources. Failure to learn about new foods reduces food intake. A varied diet is likely to be more nutritionally balanced and to have lower levels of particular toxins than a diet composed of only a few species. Dietary generalists evolve to be both curious about, but suspicious of, new foods. “It’s part of the biology of being an omnivore to have to learn almost everything about what is edible (Rozin 1997, p. 27).” Yet, the underlying response mechanisms are relatively simple. Children learn what is edible mostly by observing adults and being instructed by them. We can immediately tell whether most types of objects are dangerously hot or cold, sharp or irritating, soft or hard, but if we ingest toxic food, we do not get sick until hours later. As do other omnivores, we automatically and unconsciously associate sickness with food we ingested hours earlier. We develop aversions to that food, and typically avoid it in the future. Even odor of a food that made us sick may evoke nausea. Disgust, a powerful human emotion, is directed almost exclusively to animal foods even though animal tissues are more nutritious than plant tissues. Several feeding experiences are generally required to develop flavor preferences, but aversions are learned more rapidly (Rozin 1990). In short, a small number of neural circuits and a propensity to imitate suffice to enable us to deal with the challenging problem of knowing what to eat.

Many of the challenges for which evolutionary psychologists have expected and looked for neural modules concern our relationships with other people. During recent human history we have not shared the planet with another species of Homo; it is easy for us to recognize other members of our species. What is important is determining other people’s intentions. We have evolved sophisticated mechanisms for doing so. All body parts may provide cues to likely future behavior of an individual, but the human face is especially information-rich. Ability to detect subtle facial clues offers advantages, as does an ability to regulate facial expressions to give false information about one’s motivational state and, hence, one’s likely future behavior. A simple but powerful neural circuit is based on the fact that the angles in the eyebrows, cheeks, chin, and jaw in an angry human face resemble a downward pointing V, whereas the curves in the cheeks, eyes, and mouth of a happy face are round. A very simple shape, a downward pointing V, conveys the essence of threat and triggers greater activation of the amygdala and other brain regions than do presentations of an identical V-shape pointing upward (Larson et al. 2009). People detect downward pointing V-shapes embedded in a field of other shapes more readily than they detect identical shapes pointing upward (Larson, Aronoff & Stearns 2007).

Over most of human history both males and females formed hierarchies, but the hierarchies of the larger, more powerful males dominated those of females. In experiments, men and women were aversively conditioned to equally intense angry faces expressed by a male or female. As predicted, males showed stronger skin conductance responses to angry males than to angry females, whereas females responded equally to both conditions (Mazurski et al. 1996). A complex open-ended system that continually updates its responses to threats does not seem necessary to deal with challenges of mate selection and assessing intentions of other people.

Human brains, and presumably the brains of all other species that have them, are Darwinian machines that evolved to enable their possessors to respond to a rich array of contingencies with rapid, efficient, and appropriate responses. We live in a world awash with patterns that signal regularities in the world; predicting what will happen next can be a life or death matter. Most of the modules that evolutionary psychologists expect to exist and for which they develop hypotheses and devise tests, are circuits designed to detect and respond preferentially to patterns to which we should pay particular attention. If the patterns in nature share common features, simple circuits should suffice to yield appropriate responses. If the patterns are complex, difficult to detect, and rapidly changing, either no adaptive response may be possible or the response system may be open-ended, massive, continually updated. The immune system is a response to such difficult and rapidly changing conditions. None of the other major challenges that faced our ancestors appear to have these characteristics.

D. S. Wilson has performed a valuable service by stressing the importance of thinking clearly about the neurological structures likely to have evolved in response to life’s challenges. We can make further progress by carefully describing the features of various challenges and the possible structure of adaptive neural responses to them. With this process we should develop better hypotheses about whether neural modules are to be expected and, if so, what features they should have.

References

Coss, R. G. & R. O. Goldthwaite. 1995. The persistence of old designs for perception. In: Thompson, N. S. (ed.) Perspectives in Ethology, Volume 11: Behavioral Design, pp. 83-148. New York, Plenum Press.

Ehrlich P. R. 2000. Human Natures. Covelo, CA, Island Press.

Gazzaniga, M. S. 2008. Human. The Science Behind What Makes Us Unique. HarperCollins.

Larson, C. L., J. Aronoff, and J. J. Stearns. 2007. The shape of threat: Simple geometric forms evoke rapid and sustained capture of attention. Emotion 7:526-534.

Larson, C. L., J. Aronoff, I. C. Sarinopoulos, D. C. Zhu. 2009. Recognizing threat: A simple geometric shape activates neural circuitry for threat detection. Journal of Cognitive Neuroscience 21:1523-1535.

Mazurski, E. J., N. W. Bond, D. A. T. Siddle & P. F. Lovibond. 1996. Conditioning with facial expressions of emotions. Effects of CS sex and age. Psychophysiology 33:416-425.

Pigliucci, Massimo and Jonathan Kaplan. 2006. Making Sense of Evolution. The Conceptual Foundations of Evolutionary Biology. University of Chicago Press.

Rozin, P. 1990. Getting to like the burn of chili pepper: biological, psychological and cultural perspectives. In: Green, B. G., J. R. Mason, & M. R. Kare (Eds.) Chemical Senses. Irritation, Vol. 2, pp. 231-269. New York, Marcel Dekker.

Rozin, P. 1997. Why we eat what we eat, and why we worry about it. Bulletin of the American Academy of Arts and Sciences L (March 1997): 26-48.

Skinner, B. F. 1948. Walden Two. Indianapolis, Hackett.

Tooby J. Cosmides L. 1992.The psychological foundations of culture. Pages 19-136 in The Adapted Mind: Evolutionary Psychology and the Generation of Culture, edited by J. Barkow, L. Cosmides, and J. Tooby. Oxford, Oxford University Press.

Wilson, D. S. 2010. Learning from the immune system about evolutionary psychology.  The Evolutionary Review: Art, Science, Culture 1:13-17.

This entry was posted in Articles and edition . Bookmark the permalink. Both comments and trackbacks are currently closed.

Comments are closed.

  • Pages

  • Categories

  • Issues