What is behavior?

The Question of Behavior

What exactly is behavior?  What is it that we are examining when we inquire about the behavior of animals?  This question is not as straightforward as one might think.  There is a significant amount of disagreement on the subject that benefits from consideration.  In a paper by Levitis et al. (2009), the question of ‘what is behavior?’ was asked of professors and professionals from three different scientific societies.  Of the eight operational definitions the authors proposed from various texts, the highest agreement on any single definition only included 44% of the respondents.  Even more problematic is that 52% of the respondents showed contradictory answers—implying that the respondents’ conceptions of behavior was not guided by an operational definition of behavior but of their personal biases (most likely due to their individual areas of interest).  For instance, even though 99.1% of the respondents agreed that “geese flying in a V formation” constitutes behavior, 31% of the respondents answered that only individuals, not groups, are capable of behavior.  In argument form, this line of deduction would look like this:

Premise 1:       Only individuals, not groups, are capable of behaving
Premise 2:       Geese flying in a V formation constitute a group
Therefore:       The group of geese is behaving

This is a flawed argument to say the least. The benefit of operational definitions is that they allow us to organize our observations through logical lines of reasoning, thus contradictions in logic are a significant red flag that bias has taken place of a scientific definition.  One potential explanation for this is that in Levitis et al.’s (2009) background research to gather published definitions for the term “behavior,” over one hundred publications that they felt should have had a definition for behavior had completely avoided defining the term, either by choice or due to the assumption that people intrinsically know what behavior is.  I hope it has been successfully illustrated that the latter conclusion is clearly not the case.

The Definition of Behavior

A quick thumb in the dictionary informs us about a common preconception: “behavior is the way in which an animal or person acts in response to a particular situation or stimulus.”  In fair agreement with this definition, Tinbergen defined behavior as “the total movements made by the intact animal” (Tinbergen, 1955).  However, what if we were to talk about the behavior of a plant?  Some species of orchids (e.g. Ophrys speculum) use chemical signals to mimic virgin females of their insect pollinators to deceive male insects into attempting reproduction with their flowers: thus causing the male insect to pollinate the orchid (Ayasse et al., 2003). Deception (the action of altering the perception of another to believe something that is not true in order to gain some personal advantage) is most certainly a phenomenon most anyone would associate with behavior.  However a plant is not an animal, so evidently the concept of behavior has to expand beyond just animals if it is going to include deception.


Ophrys speculum evolved features which mimic female reproductive organs in order to deceive male insects, thus increasing their ability to be pollinated.

Conceivably then we could expand the parameter of the definition to multi-cellular organisms.  Having said that, such a classification would then exclude bacteria.  Bacteria are single-cell organisms that meet all eight of the requirements that define life: (1) they have an internal organization; (2) they store and retrieve information through the organization of nucleotides; (3) they have a metabolism that converts energy from an unusable to a usable form; (4) they respond to the environment; (5) they grow and develop; (6) they regulate and maintain homeostasis; (7) they reproduce; and finally, (8) they evolve.  One could argue that if you momentarily exclude reproduction and evolution, all of the requirements of life facilitate just one organismic phenomenon: homeostasis.  So the first argument from the skeptic must be to show that a living entity that is responding to its environment is not capable of behavior.

It might seem unorthodox for many readers to discuss the behavior of bacteria, however even Skinner was not without his philosophical views of the behavior of single-cell organisms and discussed the stretching of an amoeba’s limbs as a an example of early behavioral evolution (Skinner, 1984).  In fact, amoebas are part of the group Protozoa which comes from the Greek for “first animal.”  Protozoa were named this because of their animal-like behavior, from their movement to their reliance on consuming other organisms for survival.  While I share no disagreement with Skinner regarding the behavior of single cell organisms, where I do disagree is in the view that the work of Jacques Loeb (1915) is sufficient in providing a sound argument that the behavior of “lower” organisms only needs explication in what the organism does as a whole—a recurrent theme which Skinner also applied to the behavior of larger organisms as well (Skinner, 1987).

This view is incredibly limiting.  In the hundreds years following Loeb, we have discovered more about the behavior of all organisms (prokaryotes [archaea & bacteria] and eukaryotes) by investigating internal cellular functions.  Indeed, researchers recently discovered that bacteria are able to activate a host body’s immune response to attack an invading virus that threatens the survival of their host they are living symbiotically with (Ichinohe et al., 2011).  Researchers discovered this behavioral response by analyzing signals that lead to the synthesis of mRNA essential to the immune response—a discovery that has launched an entirely new field of inquiry to other ways medicine might fight deadly diseases.  However, this extraordinary level of symbiotic behavior is still poorly understood, so for our purposes let us take a look at a simple environmental response in a single cell of Escherichia coli as an example of adaptive behavior in bacteria.

When E. coli are in a high solute, hypertonic environment, their life is instantly endangered by diffusion’s closest relative: osmosis.  Osmosis is the movement of water from an area of higher free energy to an area of lower free energy (Saupe, 2013).  Solutes (dissolved molecules) affect this process by affecting the free energy of the water the same way that rush hour traffic decreases the kinetic energy of a freeway—the more traffic in the solution (presence of solutes), the less available free energy.  If the solutes from the E. coli’s environment diffuse across its membrane through its pores, osmosis might destroy and kill the cell because freely permeable water will pass across the membrane until the free energy inside the cell is equal to that outside.  If the presence of solutes is dense enough, the E. coli will explode and die.

E. coli would not have survived for the eons they have been around for if a handful of solutes made them explode, thus their behavioral response is an extremely effective defensive mechanism rooted in their DNA.  When E. coli senses the presence of solutes, they transmit a signal to begin transcribing and translating sections of DNA into proteins that will block the pores and prevent solutes from diffusing across the membrane—thus maintaining homeostasis.  There is no question that these actions of E. coli are essential for their survival and fit perfectly well along side the adaptive behaviors of plants and animals.

Permitting then that we include single-cell organisms in our definition of behavior, what then do we do about viruses?  Viruses have no cells.  They are purely microscopic packets of genetic code wrapped in a protein shell.  Consequently, viruses lack the essential organelles required for replication, thus in order for a virus to reproduce it literally breaks open the membrane of a host cell by using spikes on its protein shell that act like a medieval battering ram to break down the gates and invade.  This allows the virus to use the mechanisms within the fully functioning host cell to replicate its viral genetic code like nefarious molecular zombies (see Grove & Marsh, 2011).  Interestingly, scientists analyze viruses in terms of their (1) physical structure, (2) proximate causation, (3) temporal effect on the fitness of a host which impacts the virus’s capacity for survival, and (4) evolutionary development—Tinbergen’s four questions of behavior (Erkoreka, 2010; also see Tinbergen, 1963).  If viruses are incapable of behavior, are Tinbergen’s four questions irrelevant?  Such a claim would create an upheaval in the behavioral literature.


Here is another way we might ask ourselves about the behavior of viruses.  Let us for the sake of argument claim that viruses are not capable of behavior.  If this is true and we are to exclude these molecular zombies from the definition of behavior, why then do we commonly say that viruses attack cells?  Are we just attributing the behavioral characteristics of animals to non-animal entities or is it really valid to say that these molecular zombies are truly attacking and threatening the homeostasis of an organism?  Even the dictionary cites a definition for “attack” as the aggressive action of a disease.  Viruses obviously meet this criterion so then are we to exclude the behavior of attacking from the definition of behavior just so we can deprive these mysterious molecules of behavior?  While colloquialisms are generally poor grounds for scientific exploration, in this case I would argue that they bear important consideration to the question at hand.  After all, it would be ignorant to deny that common expressions reflect our experiential perceptions of the world (known in science as empirical observation).

The fact that these molecules are nothing but packets of genetic material and literally straddle the definitional requirements of life yet have actions attributable to behavior quickly leads us into questions about non-living entities—for example, do rocks behave?  If an individual was to pick up a rock and throw it, is the rock behaving?  A definition that relied purely on the reactions of matter to its environment would require us to include every rock on Earth as a behaving entity since rocks are quite literally orbiting the Sun as the Earth moves through our solar system.  Using Newtonian physics, we would stipulate that the rock is exerting a force on the Earth equal to that of the force of the Earth on the rock.  Instead of thinking about a small rock, what if we were to look at hydrogen dioxide (water).  The behavior of water makes it one of the most unique molecules in existence.  Three of waters most important behavioral characteristics (i.e. properties) are that it is a powerful solvent, it sticks to itself because it is polar and thus can generate incredible surface tensions, and it is denser as a liquid than as a solid.  If any one of these three properties were different, life would not exist as we know it today.  Molecules have observable, quantifiable properties which respond to other molecules in the environment, therefore why can’t molecules behave?


One might argue for a definition that stipulated some kind of internal propulsion or energy source which might allow us to exclude a rock.  But would this criterion be valid?  Even though the energy used by animals is converted from an unusable to a useable form through internal metabolic processes, we have already established that the source of all energy for every organism comes from the Sun.  The survival of any animal would be immediately threatened without the energy mitochondria (prokaryotic organisms) produce or the symbiotic benefits offered by gut flora (prokaryotic organisms like E. coli), thus the behavior of animals cannot be observed without the impact of the organisms acting on them.  In a similar fashion, no organism lives in a void.  We shouldn’t forget, as silly as the concept of a rock behaving may seem, all animal behavior is subject to the physical laws of nature.  A bird cannot fly without overcoming the necessity to generate lift, neither could energy be converted to facilitate homeostasis or motion without adhering to the intrinsic properties of atomic theory.  If an owl is diving on a field mouse, prey capture could not occur without the effects of gravity—thus the very action pattern depends on the same phenomenon that determines the travel path of a rock that is thrown.  It should be mentioned that the skill of a bird negotiating with the physical properties of lift is equal to any human pilot, even though the mechanisms are obviously different, and while the Border Collie is unlikely to recite Newton’s laws of motion, they have a skill in predicting the motion of a Frisbee equal to any Ultimate Frisbee competitor.

Any general definition of behavior must allow for all of the nuances mentioned so far if it is to permit our inferences to take into account the immense diversity of behavior in the world.  Thus, returning to Tinbergen, I will modify and broaden his insight and offer that behavior is the total measurement of movements (both internal and external—regardless of physical or biological limitations) of an entity through its environment.  More simply: behavior is the response of a system to a stimulus.  Therefore, since every particle in the universe is in motion and endlessly responding to stimuli, by this definition we can conclude that everything—even a subatomic particle—is in a constant state of behavior.

Such an incredibly broad definition may at first seem either ridiculous or of little value, yet I would contend that there is nothing more valuable than to question our preconceptions with a logical and unbiased deductive thought process.  There is no reason to exclude rocks as behaving entities in the world, and a rejection of this has more to do with the novelty of it (“well I just don’t like it..”-type of response) than of the reasoning to the question.  Furthermore, I would argue that the benefit of such a definition provides a common concept to be discussed whether the system involves entire populations or just a single atom.

Previously I cited the opinion from Levitis et al’s (2009) study that 31% of participants believed that only individuals, not groups, are capable of behavior.  Problematically, the behavior of the human body is the result of the sum of movement of molecules from within—therefore how can we say that the movement of trillions of molecules inside a single organism is “behavior” yet preclude the sum of a group of geese performing an action (such as flying in a V formation) as a behaving entity?  I have no doubt that a five-minute discussion with a police officer that has had to respond to a riot can attest to the behavioral differences of an angry mob versus that of an angry individual.  The tendency for human nature to warp their perceptions to fit their preconceptions makes the ability to integrate new information very difficult; especially if that information negates beliefs we have held for a long time.  As Tinbergen states:

“… if we overdo this in itself justifiable tendency of making description subject to our analytical aims, we may fall into the trap some branches of Psychology have fallen into, and fail to describe any behaviour that seems ‘trivial’ to us; we might forget that naïve, unsophisticated, or intuitively guided observation may open our eyes to new problems.  Contempt for simple observation is a lethal trait in any science, and certainly in a science as young as ours.”  (Tinbergen, 1963)

A definition for behavior that includes all matter in the universe, while useful for expanding our perceptive lens, may not initially seem to provide a productive answer for our question of what is animal behavior.  If narrow perspectives risk losing the forest through the tree, this might be like losing the forest in the galaxy.  What should be emphasized is that the question of what is behavior is only productive if the nature of the entity exhibiting the behavior is definitively described.  The behavior of organisms is thus our aim, and this behavior is intrinsically tied to the fact that organisms—regardless of their complexity—are bound to their biological nature.


Ayasse, M., Schiestl, F. P., Paulus, H. F., Ibarra, F., & Francke, W. (2003). Pollinator attraction in a sexually deceptive orchid by means of unconventional chemicals. Proceedings of the Royal Society B: Biological Sciences, 270(1514), 517–522. doi:10.1098/rspb.2002.2271

Erkoreka, A. (2010). The Spanish influenza pandemic in occidental Europe (1918-1920) and victim age. Influenza and Other Respiratory Viruses, 4(2), 81–89. doi:10.1111/j.1750-2659.2009.00125.x

Freeman, S., Quiliin, K., & Allison, L. (2013). Pearson – Biological Science (5th edition.). San Fransisco, CA: Benjamin Cummings.

Grove, J., & Marsh, M. (2011). Host-pathogen interactions: The cell biology of receptor-mediated virus entry. The Journal of Cell Biology, 195(7), 1071–1082. doi:10.1083/jcb.201108131

Ichinohe, T., Pang, I. K., Kumamoto, Y., Peaper, D. R., Ho, J. H., Murray, T. S., & Iwasaki, A. (2011). Microbiota regulates immune defense against respiratory tract influenza A virus infection. Proceedings of the National Academy of Sciences, 108(13), 5354–5359.

Levitis, D. A., Lidicker, W. Z., & Freund, G. (2009). Behavioural biologists do not agree on what constitutes behaviour. Animal Behaviour, 78(1), 103–110. doi:10.1016/j.anbehav.2009.03.018

Loeb, J. (1915). On the role of electrolytes in the diffusion of acid into the egg of Fundulus. Journal of Biological Chemistry, 23(1), 139–144.

Saupe, Stephen G. (2013). Letter to the Editor. American Biology Teacher, 75(1), 4-5.  DOI: 10.1525/abt.2013.75.1.2

Skinner, B. F. (1984). The evolution of behavior. Journal of the Experimental Analysis of Behavior, 41(2), 217–221.

Skinner, B. F. (1987). Whatever happened to psychology as the science of behavior? American Psychologist, 42(8), 780–786. doi:10.1037/0003-066X.42.8.780

Tinbergen, N. (1955). The Study of Instinct. Oxford: Clarendon.

Tinbergen, N. (1963). On aims and methods of Ethology. Zeitschrift für Tierpsychologie, 20(4), 410–433. doi:10.1111/j.1439-0310.1963.tb01161.x