I decided to use vocalizations as an example of the ways in which potential sexual incentive stimuli need to be analyzed before any conclusion as to their incentive properties can be drawn. We should be able to learn a few lessons from this. Several of these lessons are applicable to any sexual incentive stimulus, while some are limited to those incentive stimuli that are emitted in response to the presence of a conspecific. First, however, we need to discuss the analysis of sexual incentive stimuli from an epistemological point of view. This discussion will appear in the following paragraph.
Odor, or body shape, are stimuli emitted whether there is any potential partner present or not. They are not, therefore, responses to the presence of a potential mate but context-independent characteristics of the subject. At difference, vocalizations are, according to abundant data mentioned above, emitted in response to the presence of an individual of the opposite sex. They are also emitted in other situations, like aggressive encounters, but there also they are a response to a conspecific. Vocalizations are, then, context-dependent at difference to odor and shape. The fact that they are emitted in quite specific situations easily leads to the notion that they communicate something in this situation. It may also provoke the idea that they are emitted with an intention, for example that the female rat vocalizes with the intention of attracting a male or to regulate some aspect of sexual interaction. In other words, she vocalizes with the purpose of influencing some aspect of the male's behavior. Here we are immediately back to a teleological explanation of vocalizations. Explanations in terms of intention are not immediately given to the emission of context-independent stimuli. We do not easily maintain that the female rat smells in a particular way because she has the intention of attracting a male. In case of a human female, however, we could say so, and sometimes with good reasons.
Although we do not immediately attribute an intention to the female rat's smell, we might almost spontaneously attribute it a purpose. The purpose of her particular smell is to attract a male, for example. As I did extensively in Chapter 1, I will continue to insist on the inadequacy of teleological explanations. Instead, we can maintain that the female vocalizes because some stimulus from the male, like his odor, excites a particular central motive state, and that particular central motive state somehow excites the motor neurons required for vocalization. This sequence of neural events can be conceived as automatic, as a part of the functional structure of the central nervous system, where intention or purpose neither is required nor useful. Likewise, there is no need to think that the male produces odors with the purpose of making the female vocalize. His smell just happens to have this effect.
We have now replaced teleological explanations for ultrasonic vocalizations with a mechanistic cause-effect explanation within incentive motivational theory. It is possible to go a step further in reductionism, as brilliantly shown some years ago (Blumberg, 1992). The following account is little more than a commented summary of the Blumberg (1992) paper. As explained early in this chapter, ultrasounds are produced when air is forced through the strongly constricted vocal cords. This means that they are emitted during expiration, the part of the respiratory cycle where the thoracic cage reduces its volume. It so happens that the respiratory cycle is coordinated with locomotion in quadruped species. This coordination is necessary because respiration and locomotion partially use the same parts of the body, for instance the thoracic cage and associated musculature. Expiration in a running quadruped starts at the moment the first forelimb hits the ground. Mechanical forces acting upon the foreleg lead to extra compression of the thorax, facilitating expiration. However, thoracic compression should not be too large or too fast. A way to avoid this is to constrict the larynx. The idea here is that constriction of the vocal cords in the larynx is a response to thoracic load caused by locomotion and that ultrasounds are a side effect of this physiological reaction. In support of this notion, it is reported that vocalizations in copulating animals are closely associated with intense locomotion, for example immediately before a mount or intromission.
In these moments the male normally runs behind the fast moving female, which suddenly stops just before a mount is made. Likewise, observation of females in slow motion shows that sounds are emitted shortly after the forepaws hit the ground, exactly as would be predicted from the locomotion-expiration hypothesis. The idea that the ultrasounds are side effects of a physiological process does not exclude that they may have effects on a perceiving organism. What it does exclude, though, is that the emitting subject has any intention.
This intriguing line of reasoning illustrates something extremely important and often overlooked: that animals may emit stimuli by accident or coincidence. I presented these arguments only to show that there are several ways to get rid of intentions or purposes as explanatory elements.
A lesson we can learn from our analysis of ultrasonic vocalizations has to do with their uncertain function in relation to sexual behavior. They may contribute to an animal's incentive value, but the data are scarce and mixed. Vocalizations do not appear to modulate copulatory behavior in any functionally relevant way. Despite this, almost every paper on the subject ends with a long list of the possible functions of ultrasonic vocalizations. This is natural, since most scientists certainly want their work to have some meaning. It might appear rather ridiculous to invest large efforts and much time in a long series of experiments just to conclude that the subject of study was, at the end, a non-subject. This applies, evidently, not only to studies of vocalizations and sex, but to all areas of scientific inquiry. It might be insinuated, though, that a more rational approach would have been to first establish the function of these vocalizations and then proceed with studies of their exact stimulus control, hormone-dependency, central nervous control, etc. The inverse procedure was used here as in so many other cases. The lesson is, then, that before performing detailed studies of a phenomenon we should assure that the phenomenon really is worth studying. Otherwise we make irresponsible use of our own and our students' time as well as of the tax-payers' money, provided we are supported by public funds. A suggestion as to how to proceed when judging the importance of a phenomenon will be outlined in the next lesson.
The second lesson is that behavioral research needs to be guided by coherent theoretical frameworks. If someone more than myself would take the trouble to go through the rather extensive literature on the subject of sexual behavior, this unlikely person would discover that most studies are descriptive and that at a rather low level of abstraction. This is legitimate, and even admirable. As we have seen, modern science is focused on careful description of the phenomena of nature. The problem in some of the literature on sex behavior is that this changes when we come to the conclusions or implications of the descriptive work. We are frequently exposed to far-fetched speculations in all directions, particularly concerning the function or adaptive value of the behaviors or behavioral mechanisms that were described. An explicit theoretical framework rejecting the use of teleo-logical explanations at the same time as it provides criteria for the evaluation of any phenomenon with regard to its possible behavioral importance would be extremely helpful. It could assist us in the process of understanding the observations we have made and the behavior described through these observations, and it could provide guidelines for the generation of testable hypotheses about cause-effect relationships. I will try to illustrate the usefulness of an explicit theoretical framework by applying the incentive motivational model of sexual behavior to an analysis of the many descriptions and tests of hypotheses concerning ultrasonic vocalizations presented in many preceding paragraphs.
The first question must be whether ultrasonic vocalizations enter into the framework of incentive motivation or if they are more appropriately analyzed within the theoretical context of ethological/Hullian motivation theory. The question could be restated by asking whether they should be considered a part of cop-ulatory behavior or of behaviors associated with approach and establishment of initial contact. We already know that copulatory behavior consists of a series of stereotyped, tactile reflexes in both males and females. Once initiated, copulatory behavior is not easy to modify (see Figure 3 in (Agmo, 1999) for an illustration of this). I repeat once more that the ease of activation of these copulatory reflexes, be it mount with pelvic thrusting or lordosis, is certainly determined by motivation. I also repeat that the incentive motivational framework is not particularly useful here. Ethological/Hullian models are more appropriate. Sounds emitted by the partner could, in principle, be powerful activators of copulatory reflexes in both males and females. The male's calls could be the stimulus triggering lordosis, or ear wiggling. The female's calls could be the immediate cause of the male's mounts. Alternatively, it could be assumed that ultrasonic calls do not trigger any copulatory reflex but instead have a modulating action. We now know that all this is false, but we know that because these hypotheses have been empirically tested. Even though ultrasonic vocalizations do not affect the stereotyped copulatory reflexes, it is still possible that they have some role in the processes of finding a mate and establishing initial contact.
The events determining approach to a potential mate are preferentially analyzed in terms of incentive motivation. The clear-cut question here would be if vocalizations function as sexual incentives or not. Again, an additional/alternative question could be whether they modify the impact of other incentive stimuli or not. Considering that ultrasonic vocalizations may be perceived by individuals at some distance, they seem particularly suitable for having sexual incentive properties. This hypothesis could easily be tested, and it was tested many years ago in one mouse experiment and in a few rat experiments. While the single mouse study found that the male's vocalizations indeed have incentive properties for the female, two of the rat studies failed to detect an effect (Thomas et al., 1982b; White and Barfield, 1990). The third maintained that male vocalizations had incentive properties (Barfield et al., 1979). The latter study was not reported in extenso but briefly mentioned in a review. Thus, the rat results are, as we already know, mixed. Moreover, the few existing rat studies have evaluated the role of male vocalizations but ignored those of females. That female vocalizations were ignored in mice is quite understandable, since female mice do not vocalize. Nevertheless, despite the obvious importance of the question of vocalizations as sexual incentives, the number of studies evaluating their incentive properties is only a small fraction of the number of studies evaluating their effects on copulatory behavior. The reason for the almost obsessive concentration of research on copulatory behavior and the surprising neglect of incentive motivational aspects has been and is a lack of a coherent theoretical framework applicable to sexual behavior. Without such a framework there is nothing on which to focus our attention and hang up our concepts.
I hope that the attentive reader has noticed that it was possible to give a meaningful summary of a sizeable part of the long discussion of ultrasonic vocalizations in just a couple of lines. Instead of long-winded descriptions of many experiments, sometimes rather disparate ones, we could put existing knowledge together in an easily understandable system. It was not by accident that I dedicated so many pages to a presentation of parts of the research performed on ultrasonic vocalizations. I gave the facts, to the best of my knowledge. The end result, at least for myself, was quite confusing. A suffocating amount of detailed information prevented me from getting a productive idea about the role of these vocalizations in sexual behavior. The confusion was replaced by insight when the facts were put into a theoretical model. I do believe that one of the basic impediments for a faster scientific progress, particularly in the area of sexual behavior, is the overabundance of data and the paucity of sound theoretical analysis. An additional impediment comes from the tremendous amount of quasiscientific notions of sexual behavior circulating in popular magazines and invading some scientific texts, perverting the minds of the feeble.
For the duration of the third lesson, it is convenient to assume that ultrasonic vocalizations indeed have strong positive incentive properties in the way that they activate approach behaviors on the part of a listener. An individual may approach another individual for many reasons. In the case of a rat, approaching and establishing close physical contact with another rat could be a thermoregulatory response activated by low ambient temperature. Perhaps it could also be a response to a predator. A group of rats might be less vulnerable to attack than a single individual. Approach could also be activated by social incentive motivation. The simple fact of being close to another rat seems to be rewarding. There are much data showing that rats actively seek social contact (Latane and Glass, 1968; Eckman et al., 1969; Latane, 1969; Latane et al., 1972, 1973; Sloan and Latane, 1974) and there is no doubt that a rat is a positive social incentive for other rats. However, and what might be of particular importance in the present context, ultrasounds emitted by a rat do not seem to contribute to its social incentive properties (Borden et al., 1977). Deafened rats approach and interact with other rats exactly as hearing rats do. We conclude that vocalizations do not contribute to a rat's social incentive properties.
Although nothing suggests that vocalizations are socially attractive among rats, we can for a while imagine that a rat approaches a vocalizing conspecific more than a non-vocalizing. This imaginary approach could be caused by any of the following central motive states: thermoregulatory, protection-seeking, social contact, sexual contact, and perhaps others. Under standard laboratory conditions there is no reason to believe that adult rats need to activate thermoregulatory responses. Likewise, there is no need for a rat to seek protection. There can be no perceived danger in an environment to which the rat has been thoroughly familiarized, as is the case with standard testing environments. Thus, there are two likely explanations for the rat's approach. One is a search for social contact and the other is search for sexual contact. In order to show that the approach is determined by sexual motivation rather than purely social motivation, we need to perform a couple of experiments. The most straightforward is probably to determine the effects of removal of the gonads from the subject. Neither males nor females show any sexual motivation in the absence of gonadal hormones. At the same time, it is known that the presence or absence of gonadal hormones do not affect social motivation, at least in some procedures (discussed in Âgmo et al., 2004). If a subject without adequate amounts of gonadal hormones, for example a castrated male or an ovariectomized female, approach the vocalizing subject as much as an animal with gonadal hormones, for example an intact male or an estrus female, then we can conclude that the intensity of approach is determined by social motivation. If the approach is more intense in the presence of gonadal hormones than in their absence, it can be suggested that sexual motivation is important for approach behavior. Observing animals that have engaged in extensive sexual activity immediately before testing could further support this notion. There are many reasons to believe that such activity temporarily reduces sexual motivation (Âgmo et al., 2004) while it does not affect social motivation (Sloan and Latané, 1974). If an effect of this manipulation is observed, then the idea of sexual rather than social motivation is reinforced. Although there certainly are several more ways to substantiate further sexual versus social motivation, the point should be clear: it is not self-evident that approaching another animal, even if it is of the opposite sex, is an indication of sexual motivation. Alternative explanations need to be ruled out by experimental tests. The message of the fourth lesson is, then, that the simple establishment of incentive properties of a stimulus is not enough. It must also be determined for what kind of motivation the stimulus is an incentive. In other words, which is the central motive state activated by the incentive.
The fifth, and fortunately last, lesson is very simple. We should always be humble in our conclusions and joyfully accept that we cannot answer all questions. In particular, we should avoid far-fetched speculations and never stretch our data beyond what they really show. We should also be aware of what the data do not show. Applying this to the ultrasonic vocalizations, a sufficiently humble conclusion would be that we do not know whether they have any function or not. In case they do have a function, we do not know which. We do not even know whether they are worth studying at all. Would the sound produced by a sneeze be worth studying? Another aspect of humility obliges me to accept that we cannot answer this question without having adequate data.
Was this article helpful?