Evolutionary Psychology: Sources of variation
Author Paul Kenyon

Visit the SALMON Bookshop for recommended books on this topic

'Tinker, tailor, soldier, spy'

The psychologist John Watson knew he was being outrageous and intended to shock when, in 1924, he wrote :

"give me a dozen healthy infants, well -formed, and my own specified world to bring them up in and I'll guarantee to take anyone at random and train them to become any type of specialist I might select - doctor, lawyer, artist, merchant-chief, and, yes, even beggar man and thief , regardless of his talents, penchants, tendencies, abilities, vocations, and race of his ancestors" (see Boakes, 1984, pp226).

This lecture addresses some of the fundamental issues raised by this quote which is often used to illustrate the claims of behaviourism.

Genotype and phenotype

The genotype describes the genes inherited by an organism. Phenotype refers to an individual's anatomical structure,physiology and behaviour. The phenotype refers to everything that can be easily observed and measured about an plant, animal or human being.

Our phenotype is the product of

  • genes inherited from our biological parents,
  • our environment
  • and developmental noise

I deliberately used the word 'our' in the phrase "Our phenotype is the product of " because as psychologists we are interested in human behaviour. But debates about human phenotype are based on the science of genetics. A quick glance through a genetics textbook (e.g. Suzuki et al, 1981) reveals that genetics is founded on the study of plants and simpler animals such as the fruit fly Drosophila. The next section gives a flavour of these studies (see Lewontin (2000) Chapter 1 ).

It is relatively easy to study how the same genotype reacts to different environments in plants.

Some plants can be cut into several pieces that will grow if they are put into soil. In this experiment (Clausen et al. 1958 see Suzuki et al, 1981 p18) seven Achillea plants were each cut into three separate sections.

  • One section from each 'parent' plant was planted at low elevation (30 meters above sea level),
  • the second section was planted at medium elevation (1400 m)
  • and the third section from each donor plant was grown at high altitude (3000 m).

The three plants in each column have identical genotypes. Each parent plant (#4.....#16) has a unique genotype.

Consider the plants that resulted from parent plant #4. The clones grew well at low and high altitude, but at medium elevation this genotype failed to thrive.

Maybe the conditions are not suitable for growing Achillea at 1400 meters. But this explanation is not true because the cutting from parent #24 thrived at this level compared to the performance of this genotype at low and high altitude.

What does this experiment tell us about genotype. Quite clearly the phenotype of each cloned plant depends on the environment in which it grows.

Also it is impossible to make any firm conclusions about altitude as an environmental factor in this study. For example, some of the plants grew well at the intermediate site (#23,24) where others fared less well (#4,16).

The norm of reaction

This graph is based on the data presented in the picture above. Individual lines on the graph shows how the three cuttings taken from one plant grew at various altitudes. Each line is called the 'norm of reaction' .The term 'norm of reaction' refers to how a particular genotype (from a parent plant) develops into a phenotype (cloned offspring) as a function of the environment in which the clone develops.

It is immediately obvious that identical genetic material produces different phenotypes.

This result is not peculiar to plants. Similar results have been seen in the fruit fly (see Lewontin, 2000)

Suzuki et al (1981) draw the following conclusion from this type of experiment:

  • A single genotype can produce many different phenotypes, depending on the environment.
  • A single phenotype may be produced by various different genotypes, depending on the environment.

Can you identify examples on the graph that illustrate these point?

What is the consequence of 'norms of reaction' for human behaviour? Imagine that a political proposed a perfectly plausible proposal to modify the environment to improve childrens' intelligence. Would it work? Well it might, but again it might not if intelligence is the product of each child's genes and environment.

I have used the norms of reaction from the Achillea experiment in this diagram.

The child with a genotype represented by the red line would benefit from the 'enriched' and 'hyper-enriched' environments. But the other child would suffer a progressive loss in IQ score if raised outside the normal environment.

The point of this example is not to suggest that we should not seek to improve our childrens' environment, but to suggest that we need to be aware that outcomes may not always be as expected.

Incidentally the phenotypes in his hypothetical example have been given very high IQs in order to turn-up the 'ethical screws'.


Fluctuating asymmetry

If you look around your world at flowers, animals and fellow human beings you will see that most things appear to be symmetrical. If you draw a line down the middle of your body, the left hand side is a mirror image of the right hand side.

But if you look more carefully it quickly becomes clear that one side is not exactly like the other. The cheek on one side of the face may be slightly smaller than on the other; the lobe of one ear may be larger than the other, and so on. Relatively minor variation from perfect symmetry is called fluctuating asymmetry.

Fluctuating asymmetry is normally distributed with a mean of zero. For example, if you were to measure ear lobe size in a large number of people, the number of people with bigger left lobes would be the same as the number of people with larger right lobes. Furthermore the majority of the differences between left and right sides of the body are relatively small (about 1 or 2% see Kowner, 2001).

Facial asymmetry

In most people the two sides of the face are not perfectly symmetrical - facial asymmetry. These three pictures below show how the left and right sides of a face are not perfectly symmetrical. The picture in the middle is the person's real face. The pictures to the left and right have been artificially created on a computer. The image on the left was constructed by cutting the face down the middle and keeping the left hand side. Then I created a 'new' right hand side by flipping the left half of the face. Then the two sides: original left hand side and artificial right hand side were positioned side by side to create an artificial face that is symmetrical. The same procedure was used to create an artificial face based on the right hand side of the real face.

If the original face was symmetrical about its midline then there should be no difference between the three images. If you doubt this repeat the procedure with a perfect circle or square.

Artificial symmetrical face constructed from left-hand side of image of real face Real asymmetrical face Artificial symmetrical face constructed from right-hand side of image real face

Developmental noise and fluctuating asymmetry

Some fluctuating asymmetry may be the result of variation in the timing of cell division. We know that cell division does not occur 'like clockwork'. For example, division of a group of otherwise identical cells under laboratory conditions is a continuous process, rather than a series of discrete steps at which all the cells divide simultaneously. These slight random imperfections are occurring at the molecular level within cells.

Because these are random events, their effect on the phenotype is independent of the organism's genotype or environment. Consequently the phenotype is the product of random events (developmental noise) as well as genetic and environmental influences.

It is important to realize that development happens from the moment of conception to death. Cell division is a continuous process throughout our lives.


  • Other terms for developmental noise are developmental instability and stress.
  • Developmental stability or developmental homeostasis refers to an organism's ability to withstand the events that produce fluctuating asymmetry.

Our reactions to facial asymmetry

Reis and Zaidel (2001) presented 24 (12 female, 12 male) undergraduates with composite (left-left and right-right) faces constructed from pictures of 21 women and 17 men. Participants were asked to judge which composite appeared healthier, or if there was no difference between the faces.

Experimental design

Face composite Left-Left face composite Right-Right face composite  
Stimuli for participants
Health judgement task Healthier? Healthier? Same
The results reveal that:
  • the health judgements made by male and female participants did not differ
  • composites of the right hand side of female faces faces were judged significantly healthier than composites created from the left side of faces
  • relatively few left and right composites were judged the same in terms of their healthiness.

The authors concluded that:

  • there is a sex-related left-right asymmetry in the appearance of health in the human face
  • the results are consistent with a notion from evolutionary psychology that the human face provides a clue to the health of the genotype

Evolutionary psychology

It has been said that "the eyes are a window into the soul". To what extent is the face a window into the genotype? Does the face contain reliable valid information about the ability of the genotype to withstand stress as claimed by the 'good-gene' hypothesis derived from evolutionary psychology?

According to evolutionary psychology facial symmetry may be a marker for health associated with 'good-genes'. During development the body is subject to stressors that can cause deviations from symmetry. Some individuals are able to withstand this stress and show less asymmetry. Humans have evolved

People with asymmetric features are judged as less healthy and less attractive as partners because (Buss, 1999, p 118):


A very important question is: Are the differences between individuals in fluctuating asymmetry due to the genotype?

According to opponents of evolutionary psychology there is very little evidence that the factors responsible for fluctuating asymmetry are inherited in humans. Consequently if fluctuating asymmetry is not passed on from generation to generation it could not have been the subject of sexual selection during evolution.

Members of a family often display similar characteristics because they share the same environment. You probably speak the same language as the rest of your family. But this is a familial trait caused by growing up in the same country as close relatives.

Heritability deals with:

Heritability is the proportion of the variance of a population's phenotype which is due to genetic variance within the population. Heritability values range between 0 and 1.

Heritability is not a measure of how much of an individual's phenotype is due to their genotype. For example, my height is a function of the interaction between my genes and my environment. The first five feet of my height are not due to my genes, and the remaining eight inches are not due to my environment!

Heritability is a statistical concept studied by geneticists but 'adopted' by psychologists interested in understanding the differences between people. A textbook written by geneticists for geneticists (Suzuki et al, 1981) makes sobering reading for any psychologist entering this minefield.

Suzuki et al (1981) make the following points about the heritability of traits:

To what extent is fluctuating asymmetry heritable?

In a review of fluctuating asymmetry (FA) from a psychological perspective, Kowner (2001) wrote:

"Most of the studies using FA to estimate the heritability of developmental stability from conventional parent-offspring or sibling analyses and from the results of selection have yielded very low and statistical non-significant results (Leamy,1997). Nonetheless, Moller and Thornhill (1997a), who conducted a meta-analysis of 34 studies in 17 species, calculated the overall mean effect size of heritabilities of individual FA as 0.19. Hence they argued, there is a small but statistically significant additive genetic component to developmental stability. It is undecided, however, whether this is due to genetic variation in the machinery of development per se, genetic variation in resistance to disease, or other unknown aspects of viability (for further critique, see Markow & Clarke, 1997; Palmer & Strobeck, 1997)."

Palmer and Strobeck's (1997) critique of the meta-analysis paper by Moller & Thornhill is available online

In a discussion of the usefulness of heritability Suzuki et al (1981, p 836) commented:

"Despite its limited meaning, H2 (heritability) has been used over and over again, especially by human geneticists and psychologists. They have done so in the erroneous belief that they can estimate H2 without strict adoption studies and that, if they could estimate H2, it would tell them something important about clinical or social policy. Students should not suppose that a practice is justified just because many professionals adhere to it. Sometimes scientists simply do not know what else to do, so they continue to persue useless and even incorrect lines. Sometimes they literally do not understand the basic structure of assumptions that underlay this practice. Sometimes they use incorrect methods and concepts in an attempt to justify an ideological or social prejudice (as is clearly the case for some of the studies of heritability of human temperamental and cognitive functions, such as IQ performance"

Class discussion questions

This sheep "Dolly" was the first fully grown mammal to be cloned. There is currently much debate about the ethics of cloning a human being. The chances are that before long someone will pay to have themselves cloned.
  • "To what extent will the phenotype of the clone resemble the genotype donor?" A cruder way of posing the question would be "Are they going to get their money's worth?".
  • Why are symmetrical faces attractive?

Online resources


Copyright Dr. C.A.P. Kenyon 1994-2006