such as consciousness or intentions as causes of behavior. However, it has gradually become clear that intrinsic causes can be studied scientifically and that any explanation of behavior that only takes the effects of external stimuli into account will be incomplete.
One of the earliest attempts to incorporate intrinsic causes into the scientific study of behavior was made by Skinner (1938). His concept of the operant is a motor unit of behavior that occurs originally due to unspecified, intrinsic causes. It is only as a result of conditioning that the operant comes to be controlled by specific stimuli. The motivational model of Lorenz was another attempt (Figure 3.2). Lorenz postulated that motivational energy builds up as a function of time. His model predicts that the probability that a particular behavior pattern will occur increases with the time since its last occurrence. One might imagine that as the pressure in the reservoir increases, it becomes more and more difficult to prevent the energy from escaping through the valve. In fact, behavior does sometimes occur in the absence of any apparent external stimulus. Such behavior has been called a vacuum activity . Lorenz described the behavior of a captive starling that performed vacuum insect hunting. This bird would repeatedly watch, catch, kill, and swallow an imaginary insect.
Lorenz’s model implies a continuously active nervous system kept in check by various kinds of inhibition. A particularly striking example concerns the copulatory behavior of the male praying mantis (Mantis religiosa). Mantids are solitary insects that sit motionless most of the time waiting in ambush for passing insects. Movement of an object at the correct distance and up to the mantis’s own size releases a rapid strike. Any insect caught will be eaten, even if it is a member of the same species. This cannibalistic behavior might be expected to interfere with successful sex, because the male mantis must necessarily approach the female if copulation is to occur. Sometimes a female apparently fails to detect an approaching male and he is able to mount and copulate without mishap, but very often the male is caught and the female then begins to eat him. Now an amazing thing happens. While the female is devouring the male’s head, the rest of his body manages to move round and mount the female, and successful copulation occurs.
In a series of behavioral and neurophysiological experiments, Roeder (1967) showed that surgical decapitation of a male, even before sexual maturity, releases intense sexual behavior patterns. He was then able to demonstrate that a particular part of the mantis’s brain, the subesophageal ganglion, normally sends inhibitory impulses to the neurons responsible for sexual behavior. By surgically isolating these neurons from all neural input, he showed that the neural activity responsible for sexual activity is truly endogenous.
A more recent example of endogenous control is dustbathing behavior in fowl. Most animals possess behavior patterns that can be used for cleaning themselves or for keeping their muscles, skin, or feathers in good condition. These patterns range from simply stretching or rubbing up against some object to complex integrated sequences of behavior used for grooming in many species, such as dustbathing in fowl. This behavior comprises a sequence of coordinated movements of the wings, feet, head, and body of the bird that serve to spread dust through the feathers. One might suppose that this behavior is primarily a reaction to dirt or parasites in the feathers. However, a series of experiments, testing young chicks after periods of dust deprivation, has provided strong evidence that dustbathing is primarily endogenously controlled. In one experiment it was possible to test genetically featherless chicks (Figure 3.5); these chicks also showed a strong correlation between length of dust deprivation and amount of dustbathing (Vestergaard et al. 1999).
Figure 3.5 Genetically featherless chicks dustbathing in sand. Courtesy of Klaus Vestergaard.
The intrinsic factors just discussed are all related to the motivation of specific behavior patterns. One additional intrinsic factor is the pacemaker or oscillator cells that are thought to be responsible for biological clocks (see Chapter 4). These clocks do not control any specific behavior pattern, but rather modulate the behavior mechanisms that control many different types of behavior. Dustbathing in chickens provides an example. A bout of dustbathing can last for half an hour and usually occurs in the middle of the day (Vestergaard 1982). Experiments have confirmed that an internal clock is an important causal factor in the timing of dustbathing (Hogan & Van Boxel 1993). The timing of human sleep is also an important example of oscillator control of behavior (Borbély et al. 2001).
There are many other examples of oscillator control of behavior, but most of the experimental work has investigated the oscillators responsible for daily (circadian) rhythms, often at a neurophysiological or genetic level. There has also been considerable work on the oscillators controlling interval and hourglass timers (Buhusi & Meck 2005). Timing mechanisms and biological rhythms are discussed further in Chapter 4.
Interactions among Behavior Systems
Causal factors for many behavior systems are present at the same time, yet an animal can generally only do one thing at a time. This is a situation of motivational conflict. In this section I consider the kinds of behavior that occur in conflict situations, as well as some mechanisms that have been proposed for switching from one behavior to another. There have been two major ways of studying conflict behavior. One way to classify conflicts is in terms of the direction an organism takes from a goal object: either toward or away. Many psychologists have distinguished three basic kinds of motivational conflict, each designated according to the direction associated with the specific tendencies aroused: approach–approach, avoidance–avoidance, and approach–avoidance (e.g., Miller 1959). Other psychologists have used approach/withdrawal concepts as the basis for a theory of behavioral development (Schneirla 1965). An alternative way to classify conflict situations, used by ethologists, is to look at the specific behavior systems that are activated and analyze the behavior that is actually seen. Four major types of outcome have been studied: inhibition, ambivalence, redirection, and displacement. I will briefly discuss each.
Inhibition and Intention Movements
The most common outcome in a conflict situation is that the behavior system with the highest level of causal factors will be expressed and all the other systems will be suppressed. A male stickleback that is foraging in its territory will stop foraging when a female enters and will begin courting. The male’s hunger has not changed, nor has the availability of food. It follows that the activation of the systems responsible for courtship must have inhibited the feeding system. In general, behavior system inhibition can be said to occur when causal factors are present that are normally sufficient to elicit a certain kind of behavior, but the behavior does not appear as a result of the presence of causal factors for another kind of behavior.
Sometimes, inhibition of a behavior system is not complete, and incipient movements belonging to the suppressed behavior systems are seen. These provide an indication of the relative strength of the causal factors for other behaviors that are activated in the situation. They have been called intention movements because they suggest to an observer, human or conspecific, what behavior might occur next. Intention movements have played an important role in theories of the evolution of motor mechanisms (Tinbergen 1952).
Ambivalence
When a female stickleback enters the territory of a male, she is both an intruder and a potential sex partner. The appropriate response to an intruding conspecific is to attack it; the appropriate response to a sex partner is to lead it to the nest. The male essentially does both; he performs a zigzag dance (see Figure 3.4). He makes a sideways leap followed by a jump in the direction of the female, and this sequence may be repeated many times.
