Ambivalent behavior and redirected behavior are appropriate responses to causal factors that are obviously present in the situation in which the animal finds itself. Sometimes, however, an animal shows behavior that is not expected, in that appropriate causal factors are not apparent. A male stickleback meets its neighbor at the territory boundary and shows intention movements of attack and escape; then it suddenly swims to the bottom and takes a mouthful of sand (which is a component of nest-building behavior). A young chick encounters a wriggling mealworm and shows intention movements of approach to peck and eat the mealworm and of retreating from the novel object; then, while watching the mealworm the chick falls asleep. A pigeon, actively engaged in courtship, suddenly stops and preens itself. A student studying hard for an exam, puts down her book, walks to the kitchen, and makes herself a sandwich. These behaviors are all examples of displacement activities that are controlled by a behavior system different from the behavior systems one might expect to be activated in a particular situation.
In the case of the stickleback, it is reasonable to show components of attack and escape behavior at the boundary of its territory because the neighboring fish is an intruder when it crosses into our subject’s territory, and our subject loses the security of home when it ventures into its neighbor’s territory. But why should it engage in nest-building behavior? The stickleback has probably already built its nest elsewhere and, in any case, would not normally build it at the edge of its territory. What are the causal factors for nest building in this situation? Similar considerations apply to the other examples as well. In all cases, causal factors for the displacement activity appear to be missing. It is this apparent inexplicableness of displacement activities that has caused so much attention to be focused on them. Why does this unexpected behavior occur?
There have been two main theories put forward to account for displacement activities: the overflow theory and the disinhibition theory. The original theory was proposed independently by Kortlandt (1940) and by Tinbergen (1940) and is usually called the overflow theory. They proposed that when causal factors for a particular behavior system (e.g., aggression) were strong, but appropriate behavior was prevented from occurring, the energy from the activated system would “spark” or flow over to a behavior system that was not blocked (e.g., nest building) and a displacement activity would be seen. The appropriate behavior might be prevented from occurring because of interference from an antagonistic behavior system (e.g., fear or escape) or the absence of a suitable object or thwarting of any sort.
This theory was formulated in the framework of Lorenz’s model of motivation, which accounts for the graphic metaphor of energy sparking over or overflowing. In more prosaic terms, this is actually a theory in which causal factors have general as well as specific effects. Many examples of displacement activities are described as being incomplete or hurried—the stickleback does not calmly proceed to build a nest during a boundary conflict—and such observations give support to a theory that posits general effects of causal factors. It can be noted that Freud’s (1940/1949) theory of displacement and sublimation of sexual energy (libido) is basically the same as the overflow theory: sexual energy is expressed in nonsexual activities such as creating works of art.
The alternative theory is called the disinhibition theory. In essence, it states that a strongly activated behavior system normally inhibits weakly activated systems. If, however, two behavior systems are strongly activated (e.g., sex and aggression), the inhibition they exert on each other will result in a release of inhibition on other behavior systems (e.g., parental) and a displacement activity will occur. The general idea was proposed by several scientists, but the most detailed exploration of the theory was made by Sevenster (1961). He studied displacement fanning in the male stickleback, which often occurs during courtship before there are any eggs in the nest. The sex and aggression behavior systems are known to be strongly activated during courtship. By careful measurements, it was possible to show that fanning occurred at a particular level of sex and aggression when their mutual inhibition was the strongest. Of special importance for the disinhibition theory, the amount of displacement fanning that occurred depended on the strength of causal factors for the parental behavior system. When extra CO2 was introduced into the water, there was an increase in fanning.
The primary difference between the two theories is that according to the disinhibition theory the displacement activity is motivated by it own normal causal factors and the conflict between systems merely serves a permissive role; whereas according to the overflow theory the displacement activity is motivated by causal factors for one or both of the conflicting systems. In the disinhibition theory, causal factors always have specific effects; in the overflow theory, they have general effects. Which theory is correct? As is so often the case, neither theory, by itself, is able to account for all the phenomena associated with displacement activities. The disinhibition theory is in many ways more satisfying because it only requires that causal factors have their normal and expected effects on behavior. Nonetheless, more general effects of causal factors must be invoked to account for the frantic or excited aspects of displacement activities seen in many situations.
It is frequently true that the causation of a behavior pattern is even more complicated. For example, ground pecking occurs as a displacement activity during aggressive encounters between two male junglefowl. Arguments for considering this activity as a displaced feeding movement include the fact that it is often directed to food pieces on the ground and the fact that it occurs more frequently when the animals are hungry. This same activity can also be considered redirected aggression, and experimental evidence also supports this interpretation. Thus, one behavior pattern can be both a displacement activity and a redirected activity at the same time (Feekes 1972). Each contribution to the causation of a behavior pattern can be analyzed separately, but the list of causal factors affecting the behavior pattern can be very long. Indeed, multiple causation of behavior is the rule rather than the exception. The causation of behavior is a very complex question, and it is unreasonable to expect a simple answer.
Mechanisms of Behavioral Change
What determines when a particular behavior will occur, how long it will continue, and what behavior will follow it? One can imagine that all an animal’s behavior systems are competing with each other for expression, perhaps in a kind of free-for-all. For example, if the level of causal factors for eating is very high, the hunger system will inhibit other systems and the animal will eat. As it eats, the causal factors for eating will decline while the causal factors for other behaviors, say drinking, will become higher than those for eating and the animal will change its behavior. If a predator approaches, the escape system will be strongly activated, which will inhibit eating and drinking, and the animal will run away. And so on.
Unfortunately, as attractive as this account appears, it is clearly an oversimplification of reality. Perhaps its most serious shortcoming is that if there were a real free-for-all and only the most dominant behavior system could be expressed, many essential but generally low-priority activities might never occur. If a hungry animal never stopped to look around for danger before the predator was upon it, it would not long survive. Since most animals do survive, this must imply that the rules for behavioral change are more complex than the “winner take all” model. Lorenz (1966) has compared the interactions among behavior systems to the working of a parliament that, though generally democratic, has evolved special rules and procedures to produce at least tolerable and practicable compromises between different interests. The special rules that apply to interactions among behavior systems have only begun to be studied, but a few principles are beginning to emerge.
One important mechanism for behavioral change arises from the fact that most behavior systems are organized in such a way that “pauses” occur after the animal has engaged in a particular activity for a certain time. The level of causal factors for the activity may remain very high, but during the pause other activities can occur. For example, in many species, feeding occurs in discrete bouts; between bouts there is an opportunity for the animal to groom, look around, drink, and so on. It appears that the dominant behavior system (in this case, the hunger system) releases its inhibition on other systems for a certain length of time. During the period of disinhibition, other behavior systems may compete for dominance according to their level of causal factors