Cautions about Evolutionary Thinking

With all the solid research supporting evolutionary influences on behavior, it is perhaps no wonder that evolutionary perspectives are achieving widespread acceptance within the discipline of psychology. However, students of evolutionary psychology should keep in mind several cautionary points:

What are a few complications in relating genes to behavior?

—Almost all behavior is the result of multiple genetic influences (i.e. most behavior is polygenic).

—A single genetic change almost always results in multiple behavioral changes. This is called pleiotropy. It is important because all the changes produced by a genetic alteration (not just the most obvious) can affect evolution.

—A behavior that evolved in the past to serve one function may, in later times, serve entirely different functions.

The last point is sometimes called evolutionary opportunism. Evolution always works with what is already there. A behavior or body part that is already existing may have previously unexploited uses or benefits. If so, it can fall under a new set of selective pressures.

For example, ground-dwelling birds may use a pre-flight movement in their mating displays, even though members of the species can no longer fly. Some time in their ancestral past, the pre-existing behavioral pattern (pre-flight movement of the wings) was assimilated into the mating display. Why? Because it was available in the male bird's behavioral repertoire, and some females liked it. Now it serves a new function: attracting a mate. Consequently, the genes that control muscle development and expression of that behavior are maintained by new selective pressures. In other words, the behavior serves a new purpose (helping to attract a mate) and that is what keeps a flight movement programmed into the species DNA of a flightless bird. Such a behavior is said to be emancipated from its original function.

How might meaningful dreaming in humans be the result of opportunism?

The twin concepts of opportunism and emancipation greatly complicate the interpretation of evolved behavior. One cannot assume the present function of a behavior (or body alteration) is the same as its ancestral function.

For example, dreaming may have evolved to serve some basic biological purpose unrelated to cognition, but in humans the vivid nighttime experiences may have influenced daytime behavior. If the influences were positive (aiding differential reproduction) then dreaming may have taken on a new function for humans, such as helping with problem solving. In that case, the occurrence of meaningful dreaming could be an example of opportunism: emergence of a new function for an existing behavior.

Dreaming is probably not emancipated from its original function, because it probably still serves its original purpose as well as occasionally offering guidance for decision-making. But the newer function of solving problems (and giving warnings, etc.) may have emerged well after the original function of dreaming in pre-human mammals. Warnings and inspirations from dreams can have considerable impact on reproductive success, so dreaming may now be adaptive for humans in several different ways: aiding memory, guding decisions, keeping people out of trouble during night hours, and many other possible functions.

The above factors are all complications of evolutionary thinking. Keeping them in mind can help a student avoid simplistic thinking. In addition, students should avoid several illogical ways of discussing evolution.

Why is it illogical to speak of evolution making something happen?

—Evolution should not be reified or treated as a thing or a force of nature. Evolution is just a word used to label all the various factors that influence differential reproduction and propagation of DNA. Evolution is not a thing or an agent with a separate existence. Therefore it is erroneous to speak of evolution making something happen. That type of language is a sloppy shorthand for saying that a genetic change leads some creatures to prosper and reproduce while others fail to pass on their genes.

In what sense is evolution unpredictable, but not random?

—Evolutionary change may be "blind" (unpredictable) but it is not random in a statistical sense. Consider how babies resemble their biological parents. The outcome is not predictable, but neither is it random.

In statistics, a random process in one in which all outcomes are equally likely. However, when a biological system is altered (even by a random process like a cosmic ray) the results are not random in a statistical sense. The evolutionary past of any system makes some outcomes much more likely than others. The range of possible outcomes (when genetic expression is altered) always depends on which components and arrangements already exist in the system. Those pre-existing structures have been cumulating and modifying for millions of years. (In the same sense, a human never has a truly "random" thought…all thoughts are influenced by a person's history of learning, etc.)

What does it mean to say two similar evolutionary outcomes are "analogous" vs. "homologous"?

—At the same time as the details of an evolutionary process are unpredictable, the general shape or outcome of an evolutionary process can be predicted, and when a given mutation or behavior is sufficiently adaptive, it may emerge independently several different times. Therefore one cannot assume that similar structures or behaviors are produced by the same ancestral process. Sometimes identical behaviors in different population grooups are analogous (resulting from similar evolutionary processes) rather than homologous (resulting from a common ancestors). One cannot distinguish analogous vs. homologous processes on the basis of observation alone. DNA analyses are required to resolve the issue of common ancestry.

What is convergent evolution, and why does it occur?

Convergent evolution is the term used to label strongly constrained, repeated evolutionary outcomes. Some adaptations appear more than once, as the result of independent processes that differ in their details. An adaptation will appear very reliably if there is a sufficiently compelling network of constraints or selective pressures.

When humans starting domesticating cattle, for example, it became highly adaptive for adult humans to tolerate lactose, so adults could benefit from drinking milk from their cattle. This was a life or death matter during famines and droughts. Lactose tolerance in adulthood required only a simple genetic modification to preserve an enzyme already produced in babyhood. Consequently, as shown by DNA evidence, lactose tolerance evolved five different times in the past 2-3,000 years, independently, in human populations of northern Africa (Wade, 2006). That is an example of convergent evolution.

Ironically, the parallel between learning and evolution has been discovered repeatedly by different scientists. Pringle, Ashby, Donald T. Campbell, Jean Piaget, and B.F. Skinner all noticed it. Both learning and evolution proceed by variation and selective retention of adaptive variants, the evolutionary pattern. It is certainly fitting that this insight occurred independently to so many scientists, because it emphasizes the parallel between learning and evolution itself. Convergent evolution can occur in scientific discoveries (a form of learning) just as it can in biological evolution.

Write to Dr. Dewey at

Don't see what you need? Psych Web has over 1,000 pages, so it may be elsewhere on the site. Do a site-specific Google search using the box below.

Custom Search

Copyright © 2007 Russ Dewey