Behavior Genetics

I. What IS a Gene?

A gene is a coded section of the molecule DNA (deoxyribonucleic acid). DNA comes in long strands called chromosomes; each species has a species- typical number of these. A gene is a particular section of a chromosome that contains the instructions for producing a specific protein. So contrary to how people talk in the news, on TV, etc. every day, there is no "gene for schizophrenia," no "gene for depression," no "gene for asthma" (as reported only this past month!), no "gene for breast cancer." All genes do is code for protein.

I. Genes and Alleles

Genes can have different forms, or alleles. For example, the gene that codes for coat color in rabbits has several alleles. The particular coat color any one rabbit displays depends on which pigment alleles it has. Each allele of a gene codes for a different variety of protein that is produced at that locus. Although many alleles can exist with a population for a particular gene, an individual can have only two alleles for a gene at a specific locus. One comes from the mother, and one from the father. The two alleles could be the same, or they could be different. An animal has two copies of each gene (two alleles) because chromosomes exist in homologous pairs. Homologous chromosomes have the same genes along their length. One chromosome in each pair is inherited from each parent. If we look at a specific locus on each chromosome, we may find that you inherited the same allele from each parent, in which case we say that you are homozygous for that gene. If you inherited a different allele from each parent, we say that you are heterozygous for that gene. We should note that members of one sex (eg. Male mammals, female birds) usually have one pair of chromosomes that are not homologous, such as the X and Y chromosomes of humans. These two chromosomes are composed of different genes, and are therefore not homologous.

For many genes, one particular kind of allele is more often expressed in the phenotype, no matter what the other allele is. When this happens (when one allele in a heterozygous pair is expressed over the other), we call the allele that is being expressed the dominant gene. Dominant alleles are typically always expressed in the phenotype if they are present in the genotype. Six fingers and six toes, for example, are the dominant allele in humans. When they are present, they are invariably expressed. Why do most people have 5 fingers and 5 toes? Because the allele for 6 is not very common in the population of human genotypes. (Don't confuse genetic dominance with genetic frequency!) Alleles that are not expressed in the phenotype when they are present in a heterozygous pair are called recessive. In order for a recessive trait to be expressed, the animal must be homozygous for it; that is, both alleles must be the recessive allele.

Now, in order for natural selection to cause evolutionary change, there MUST BE GENETIC DIFFERENCES BETWEEN INDIVIDUALS. How does such genetic variation arise? The somatic (nongamete) cells of each animal are called diploid (because they contain 2n chromosomes), while the gametes (ovum and sperm) are haploid and contain only n chromosomes. The value of n varies for each species. For humans, n = 23, so that most of your cells have 2 x 23 = 46 chromosomes, while your gametes have 23. Specifically, gametes contain one chromosome from each homologous pair. At fertilization, the "paternal" chromosomes from a sperm cell join with the "maternal" chromosomes of an ovum to reestablish the diploid set of chromosomes that will be found in each of the new organism's somatic cells.

One way that variability in individuals can arise is through the process that results in haploid gametes. During the process of meiosis (the formation of gametes), the diploid gonadal cells that give rise to gametes divide so that only ONE homologue from each homologous pair of chromosomes ends up in each sperm or egg cell. During the division process, chromosomes come into contact with each other so that part of the maternal homologue may exchange with part of the paternal homologue. This process is called recombination, and it is something that can only happen in a sexually reproducing species. Recombination doesn't change which genes are on either homologue, but it MAY change the combinations of alleles at different loci on the two "new" homologues. Additional variability in genotype is introduced at each generation because the maternal and paternal homologues separate randomly and independently during the formation of gametes, producing new combinations of homologous chromosomes and therefore new combinations of alleles at the many loci an individual has. Keep this idea of recombination in mind when we talk later in the course about the evolutionary advantages and disadvantages of sexual reproduction.

As your text asks, how are the processes taking place at the level of the gene translated into something as complex and far-removed as behavior? Genetic analysis of behaviors is especially difficult for several reasons.

First of all, unlike the characteristics of the pea that Mendel studied, such as "smooth" and "wrinkled," most behaviors are not "either/or" dichotomies. We are not smooth or wrinkled, psychologically.

Second of all, unlike most classic Mendelian genetics, most behaviors are influenced by many, many, MANY genes, each with small effects.

Finally, behavior is a phenotype, as we said above, that is influenced by many non-genetic factors!

Today, most genetic research employs the theory and methods of quantitative genetics. Quantitative genetics attempts to identify genetic influence, even when many genes and environmental influences are involved. It is an approach steeped, the, in the mechanistic, reductionistic tradition, and it assesses genetic influence in part by assuming that Mendel's law of discrete inheritance may also apply to such complex characteristics as behavior (IF we assume also that many genes, each with a small effect, contribute to produce observable differences among individuals in a population).

Quantitative genetics (QG) determines the sum of heritable genetic influence on a behavior, regardless of the COMPLEXITY of the genetic modes of action. It does not tell us WHAT genes are involved in that genetic influence. Let's talk for a minute about this: heritability, and what it means.

Heritability is a term from QG. It comes from an equation that attempts to answer what accounts for the differences between individuals (so critical for Darwin's theory of evolution by natural selection):

Behavior geneticists use this equation to assess how much variation is due to one, or the other of the values in the equation.

One way to do that is by producing (via identical twins, cloning, inbreeding) critters with identical genotypes. In that case, any observed difference in phenotype can be assumed to be a consequence of environmental factors. For example, if the genotypes of two groups of individuals were identical, that means that there would be NO variance in phenotype due to genetic factors. Thus, going back to our equation:

(0 times anything is 0, so V_I would = 0, since V_I = V_G * V_E).

Thus, any observed differences between groups would HAVE to be due to the influence of the environment alone. Studies of identical twins in humans, for example, are among those used to produce assessments of the heritability of things like performance on an IQ test, personality, hours of television viewing, and tendency to be homosexual.

Another approach used by behavior and quantitative geneticists is to try to hold the environment constant, but to vary the genotype, so that:

Still a third approach to determining how much genetic influence there is in behavior involves selective breeding. Just think about the various breeds of domestic animals we have produced for specific purposes (ex. sheep dogs, cutting horses, guard dogs, racing pigeons). The different traits associated with these different breeds provide plenty of evidence to suggest that behavior can be influenced to some degree by genotype.

OKAY--so it looks like there is SOME genetic influence, on at least SOME behaviors. How can the degree of that influence be measured?

V. Heritability

Heritability is the measure used in studies that attempt to answer the question about the relative contribution of genes to behavior. Heritability (h²_b) is an estimate of the degree of genetic determination of any particular trait. Since it is dependent on variance, it cannot be used to describe an individual (in whom genes don't vary!) but rather, can only be used to compare different populations of individuals (who could vary from one another genetically).

Specifically, h²_b = V_G/ V_G + V_E.

The closer V_E is to 0, the bigger h²_b is. (That is, the closer h²_b is to 1.0, the more the variability in the characteristic of interest is genetically influenced). The bigger V_E is, the smaller h²_b is. Often, it is multiplied by 100 and reported as a figure representing "percent of variance explained by genetic influences."

Let's go back to a popular topic in the media these days about the heritability of intelligence, and whether or not there is a "gene" for IQ. We can see, I hope, why we cannot say that there is a "gene" for behavioral traits, but what about the degree of influence that genes have, or how heritable a given characteristic is? Now THAT we CAN ask.

The most common technique sued to answer these questions (for obvious reasons) in people are twin studies. The essence of the twin-study method is that there are, biologically speaking, two different kinds of twins. Some twin pairs are produced by separate sperm fertilizing two different eggs, which happen to be in the fallopian tubes at the same time. These types of twins are called fraternal, or dizygotic ("Di"-two; "Zygote"-egg) twins, and they have no more genetic material in common than any other ordinary brothers and sisters--an average of 50%.

The other kind of twin is produced when a single sperm fertilizes a single egg, and the fertilized ovum splits into two separate embryos. These twins are called identical, or monozygotic ("One egg") twins. And they share 100% of their genes in common. They are genetically identical.

If sexual preference, or intelligence, or even hours of TV viewing WAS inherited, then genetically identical twins should grow up to show the same sexual preference, or IQ test performance, or interest in TV-- no matter what environments they experienced. But fraternal twins should differ just as any two siblings would, even if they did experience the same environments. To control both genetic influences and environment requires finding some very special kinds of twins. Consider the diagram below:

Remember the equation for h²_b: h²_b = V_G/ V_G + V_E .

If all of the twins are reared in the same environments, then, the geneticists argue, V_E is 0. There is no variability due to differences in environments between each member of a twin set. Thus, any differences between the two should be due to genetic influence alone.

However, we really need to compare our identical and varying genotypes in different environments, in order to assess how much each (genes or environment) contributes to variability. Thus we have to find some twins who have been adopted away from one another at birth, and thus exposed to different environments.

These studies, of both identical and fraternal twins reared apart, are still conducted today. Granted, it is not EASY to FIND identical or fraternal twins reared apart. Thus, it should not come as a surprise that there is one center in the United States that has devoted itself to the tracking down and study of these rare families: the Minnesota Center for Twin and Adoption Research (MICTAR), in (where else?) the "twin-city" area of Minneapolis/St. Paul. This Center began data collection in 1979, and to date has found and studied more than 100 sets of twins who were reared apart. Just a few years ago, the MICTAR published an article in the prestigious journal Science (I can show you the article if any of you are interested) that summarized their most recent work. Among other data, they reported in this study that about 70% of the differences in IQ between identical twins reared apart was due to genetic variation. In other words, the heritability of IQ score in these twins is about 0.70.

What exactly does this mean? Using the data from another text book, let's look at this another way:

These data show that the correlation between IQ scores (meaning the score of one twin compared to the score of the other twin) for identical twins is higher than it is for fraternal twins. The assumption in these studies is that the environmental experiences of twins of either kind reared together is the same. Thus, the only thing that differs between the two groups, and which might explain these correlations is genetics.

What do you think about this? If intelligence was 100% heritable, what should the correlation be?

MICTAR concludes that :

1. General intelligence as measured by intelligence tests is strongly affected by genetic factors.

2. Identical twins are so similar in psychological traits because their identical genomes make it probable that their effective environments will be the same.

VI. A Final Reminder: What DO Genes Do, Anyway?

Genes code for proteins and enzymes, using the amino acids that your diet provides. There is no such thing as a "gene for IQ," or a "gene for schizophrenia." There MAY, however, be genes that code for amino acids that result in the construction of an organism with particular abilities. (like the difference between a Corvette and a Chevette--a lot of it has to do with materials, as well as design.) However, at every step in the line, from gene to amino acid to protein to cell membrane to organs to organism, there are environmental influences at work. You can have all the genes in the world coding for an enzyme that requires hydrogen, but if there is no hydrogen around to be used, that enzyme will not be built. Thus, a protein that will go into a cell membrane might be a little different in the animal with this environmental deficit. And that dell may then function differently, thus the organ that it is part of my function differently, etc.

I don't want you to think that we cannot talk about the degree of influence that genes can have on behavior. We certainly can--in fact, that is why folks calculate heritability coefficients!

The danger sometimes comes in thinking then that we can find a specific gene for a given behavior, and thus "cure" people of that behavior--for ex. sexual preference. Or in thinking that if IQ is somehow inherited, that we should abandon all affirmative action plans, as a waste of money. For most behaviors, it is a long way from the genotype to the behavioral phenotype, and many things can add their influence to that phenotype along the way. Finally, even genotypes may help to determine the very experiences that on organism has. Ex. an organism which, as a consequence of the effects of structural and regulator genes, has a "touchy" nervous system may be what we end up calling a cranky baby, and we treat cranky babies differently than we treat calm babies; thus the interaction between genes and environment begins from the start!