An Integrative Theoretical Framework for Understanding Sexual Motivation, Arousal, and Behavior

Beginning of article

In trying to explain sexual motivation and behavior, a bewildering array of terms is employed for the excitatory concepts--for example, lust, arousal, desire, libido, drive, fantasy, attraction, and incentive. A slightly shorter list describes the concepts offering restraint--for example, superego, satiation, fatigue, .inhibition, and self-regulation. Therefore, by what means are sexual motivation and behavior organized on the basis of such opposing contributions? I suggest how the underlying processes can be characterized and combined. I propose a framework that can help to organize the terms and empirical findings. In the language of evolutionary psychology (Tooby & Cosmides, 1990), I suggest how the psychobiological processes underlying sexual motivation and its expression are designed.

To understand complex psychobiological systems involving motivation, sex researchers sometimes employ theoretical models as explanatory tools and means to organize scientific thinking systematically. These summarize the interactions between the parts of a system and thereby suggest how the properties of behavior arise. Some models are expressed simply in words (Beach, 1976; Hardy, 1964), whereas others take the form of diagrams showing boxes and flows of signals between them (Bancroft, 1999; Barlow, 1986; Bem, 1996) or even computer-based models (Freeman & McFarland, 1974).

The most-favored general model of motivation appears to be the incentive motivation model (Berridge, 2001; Bindra, 1978; Bolles, 1972; Depue & Collins, 1999; Depue & Morrone-Strupinsky, 2005; McClelland, 1987; Toates, 1986). This model has been applied to sexual motivation and behavior (Singer & Toates, 1987; Toates, 1980, 1986; Toates & O'Rourke, 1978) and, in so doing, has been a guide to subsequent theorizing (e.g., Basson, 2003; Both & Everaerd, 2002; Both, Everaerd, & Laan, 2003, 2007; Pfaus, Kippin, & Centeno, 2001).

The basic incentive motivation model of sex shown in Figure 1 is applicable widely across a range of mammalian species. Incentives and cues associated with them (conditional stimuli) impinge on the nervous system, which triggers sexual motivation. Motivation links to both autonomic effects and behavior. Sexual behavior first exerts positive feedback in terms of enhancing motivation but, subsequently, negative feedback reduces motivation as a function of orgasm and ejaculation. Satiety is assumed to strengthen the future power of the incentive in its capacity to trigger sexual motivation.

As useful as this model has been, it captures in a simplified form only some of the processes that underlie sexual motivation and behavior. Here, I develop the model and suggest that further insights require (a) looking more closely at the pathway of information between stimulus and response and (b) considering how the processes captured by the original model are embedded within other processes.

[FIGURE 1 OMITTED]

This development arose from several closely related considerations:

1. It is now some 30 years since the original application of an incentive-based model to sex, and there have been enormous experimental and theoretical developments since. I bring the model up-to-date and show how an extended model can establish links with more recent research findings and theoretical developments.

2. The model was developed primarily with reference to nonhumans. Although some of its basic features are applicable across vertebrate species, refinements are needed, particularly to apply it to humans.

3. Sex research suffers from fragmentation and lack of over-arching theoretical synthesis. For example, restraint processes are studied within cognitive (Graham, Sanders, Milhausen, & McBride, 2004), clinical (Barlow, 1986), evolutionary (Bjorklund & Kipp, 1996), biological (Bancroft, 1999), social (e.g., Baumeister & Vohs, 2003), and developmental psychology (e.g., Steinberg, 2004). Hence, there is a need for integration across traditional boundaries. The model was developed within a broad framework of motivation theory. By its extension, sex research might benefit from the study of general principles of motivation and the control of behavior.

4. Suggestions regarding evolutionary function have been a guide in the investigation of the motivational processes underlying aspects of sexual behavior and cognition--for example, sex differences (Symons, 1979), arousal (Spiering & Everaerd, 2007), fantasy (Ellis & Symons, 1990), inhibition (Bjorklund & Kipp, 1996), and rape (Thornhill & Palmer, 2000). However, there has been little corresponding input to evolutionary psychology arising from a causal (proximate) perspective on motivation. I aim to offer such insights.

Figure 1 involves a single route from the stimulus input, through motivational processes, and out to autonomic and behavioral reactions. However, accumulating evidence shows the situation to be more complex. For example, there is parallel processing of sexual information (Janssen, Everaerd, Spiering, & Janssen, 2000). In humans, some processing is done with full access to conscious awareness (controlled processing), whereas some is performed unconsciously (automatic processing; Janssen et al., 2000; Spiering & Everaerd, 2007). There can be considerable dissociation between genital reactions and subjective sexual arousal (Laan, Everaerd, van der Velde, & Geer, 1995).

As important as incentives are, humans are not simply passive until triggered by external incentives. Not captured in Figure 1 is that humans exhibit highly complex cognition involving conscious awareness, goals, and intentions (G. A. Miller, Galanter, & Pribram, 1960; Ramanathan & Menon, 2006). For example, sexual desire can, of course, be triggered within the imagination. Also, the model shows a single type of inhibition, whereas there is evidence that different forms of inhibition can act at more than one level (Bancroft, 1999; Bancroft & Janssen, 2001). When the same term is used for some very different forms of inhibition, there is the potential for great confusion (Geer, 2007); I hope that I can clarify the issue.

One way of assimilating these observations into a theory of sexual motivation, arousal, and behavior is to apply a hierarchical model in which stimuli and cognitions interact in determining behavior. Such hierarchical principles have yet to be applied to sexual behavior and an extension of the incentive motivation model is a means to do so.

Incentive motivation theory was originally developed on the basis that more than one motivation can relate to the attraction of a given incentive (Bindra, 1978). However, within the study of the incentive motivation of sex, this consideration has so far not been taken into account. For example, the links between attachment and sexual motivations have been recognized (Diamond, 2003), and animal models of this have been proposed (Young & Wang, 2004). Such interactions provide powerful inputs to sexual motivation, and suggestions are made as to how they can be fitted to incentive motivation theory.

First, I consider the bases of the original incentive motivation model and then suggest several additions that allow it to be applied to a wider range of phenomena.

The Basic Incentive-Motivation Model

In the case of sexual motivation, the incentive process interacts with a parameter termed arousal. In this section, I consider the nature of the interaction between these two processes.

The Principle of Incentive Motivation

Arising within biological psychology, the central idea of incentive motivation theory is as follows. The strength of motivations, such as those toward sex, food, or water, depend on the strength of stimuli (e.g., food or a mate) impinging on a nervous system that is sensitized by such physiological states as hormone levels and nutrient deficits (Berridge, 2001; Bindra, 1978; Bolles, 1972; Toates, 1986). This idea is compatible with a range of theoretical formulations and empirical data within biological, social, and developmental psychology: The physical presence of an incentive strongly increases the level of motivation directed to that incentive, an "immediacy effect" (Baumeister & Vohs, 2003; Hardy, 1964; Metcalfe & Mischel, 1999; Mischel, 1974; Steinberg, 2004). It fits a number of conceptualizations of sexual motivation (Byrne, 1983; Hardy, 1964). Hence, the model differs from that of Fisher (1998) in which distinct sex drive and attraction systems are involved. Similarly, in arguing against intrinsic drives, Both et al. (2007) suggested that "sexual motivation does not emerge through a deficit signalled by the hypothalamus but through the attractiveness of possible rewards in the environment" (p. 329).

The Link with Sexual Arousal

The term sexual arousal is used here to refer to the activity within a particular circuit of neurons, which is functionally tied to both sexual motivation and controlling the state of the genitals. This circuit links processes of sexual motivation within the central nervous system (CNS) to autonomic processes of blood vessel dilation and constriction at the genitals. The circuit also underlies orgasm. Although changes at the genitals depend on sexual arousal, there is not necessarily a simple one-to-one link. In principle, there could be central arousal but with a blockage of small arteries at the genitals or, conversely, there could be local vasodilation in the absence of arousal. It is assumed that sexual arousal can enter a positive feedback process, with changes at the genitals affecting the state of arousal.

The exact contribution, in terms of the relative weight of sympathetic and parasympathetic branches, remains controversial; but, by definition, the autonomic nervous system (ANS) is assumed to form the basis of genital arousal (Meston & Bradford, 2007). Although some general changes in sympathetic activity can affect sexual arousal, an assumption of a uniform change in sympathetic activity accompanying arousal would be simplistic. For example, it appears that decreased sympathetic tone at the penis is accompanied by increased sympathetic tone elsewhere, such as to shift blood to the penis (McKenna, 2007).

In the model of Figure 1, incentives and genital arousal act reciprocally in a dynamic relation (links 5 and 6): Arousal can be triggered by incentives and the motivational value of incentives can be increased by arousal. Such interdependence forms an integral part of a number of models of sexual motivation and behavior. For example, in response to Kaplan's triphasic model, Meston (2000) noted: "Yet in actual clinical practice, sexual desire, arousal and orgasm difficulties coexist more often than not. For example, hypoactive sexual desire disorder frequently occurs secondary to other sexual disorders such as arousal disorder, anorgasmia, or dyspareunia (sexual pain)" (p. 8). Meston posited a feedback effect such that incentive value and arousal depend on the individual's past history of sexual experiences.

Dichotomies of desire and arousal that describe which factor comes first might be misleading (cf. Pfaus, 2007b). It appears that sometimes desire can be high, associated with urges and fantasies, and yet arousal, as measured by penile tumescence can be low (Barlow & Dur.and, 1995). Laan and Everaerd (1995) wrote: "We conclude that functional women do not tend to use feedback from genital arousal in order to assess their subjective feelings of sexual arousal. Rather, external information is used to label an internal event" (p. 65). This analysis suggests the possibility that internal arousal acts as a modulator of the efficacy with which external information and cognitive interpretations are linked to subjective sexual arousal (i.e., a multiplicative relation exists).

Arousal that is specifically sexual interacts with arousal of a more general kind. Depending on how the arousal is interpreted centrally, even that arising from, say, anxiety can sometimes contribute to sexual arousal (Bancroft, 1999). When it is made artificially high, false feedback on heart rate can increase the attraction rating of an incentive (Valins, 1970), a situation indicating the role of cognitive interpretation.

It is logical to assume that the link between incentive value and arousal has a developmental and learning history, to which I now turn.

Developmental Factors

It is commonly assumed that "incentive value" is acquired, at least in part, by experience: Arousal plays a role in labeling incentive value (Bem, 1996; Money, 1986; Pfaus, 2007a; Storms, 1981). Exposure to particular potential sexual incentives in early life accompanied by general arousal (e.g., from nonsexual sources) is thought to attribute actual incentive value to them. Evidence on boys aged 11 to 12 years points to arousal from nonsexual sources triggering erection (Ramsey, 1943), a response that could set the conditions for a link with incentives. It has been suggested that genes have the influence of setting certain tendencies on which arousal further acts to consolidate sexual orientation. A similar idea is that exposure to a potential incentive or a media representation is followed by rehearsal of the imagery in fantasy accompanied by arousal, which reinforces its actual incentive value (Hardy, 1964; Storms, 1981; Wincze, 2000). Any actual sexual experience with the incentive might further consolidate the strength (Hardy, 1964).

For some individuals, such labelling might never occur, in which case the person would not come to experience sexual desire. Under other conditions, labelling might occur in conjunction with deviant stimuli and lead to a paraphilia (Geer, Lapour, & Jackson, 1993).

Laws and Marshall (1990) suggested that humans are prepared to form particular associations between stimuli and sexual arousal, most obviously corresponding to a heterosexual partner. Other associations, such as pedophiliac arousal, represent a less probable location on a continuum of possibilities. Laws and Marshall described the process of forming sexual associations as one that is "primitive" (not cognitive) and one that cannot be modified in the light of new information. In hierarchical terms (described shortly), this would correspond to a low level of organization. Baumeister (2000) argued that, in such an imprinting-like process, males are relatively flexible regarding the preferred sexual incentive when young but inflexible when adult, whereas females retain greater flexibility throughout life.

Once the initial pattern is set, a rich variety of conditional stimuli arising both outside the individual and in the imagination can then come to form triggers to arousal. Masturbation can powerfully act to consolidate particular courses of desire (Barlow & Durand, 1995; Laws & Marshall, 1990).

Strengthening and Maintaining Incentive Value

Incentive value appears to be consolidated and maintained in part via the consequences of interaction with the incentive, particularly if these involve orgasm (Depue & Morrone-Strupinsky, 2005; Laws & Marshall, 1990; Pfaus et al., 2001). "Interaction" is meant both in reality and in the imagination. For people in a negative mood, an even temporary lifting of mood in association with sexual interaction is likely to increase the value of the incentive. Strengthening of the power of the incentive arises not only from orgasm per se but also from such associated factors as lowering the level of anxiety or depression (Bancroft, 2007; Janssen & Bancroft, 2007). It is possible that negative mood could act as a contextual factor to increase the salience of sexual incentives (Bancroft, 2000).

If punishment (e.g., a gesture of disapproval) follows arousal to a particular sexual stimulus, its incentive value can sometimes decline (Laws & Marshall, 1990). Laws and Marshall saw this occurrence as one possible basis of sexual deviancy: "Becoming sexually aroused to unpunished deviant acts...while being punished for appropriate behaviors will lead to an increased tendency to act deviantly and a reduced tendency to act appropriately" (p. 217).

So much for the principles of incentive motivation and arousal seen somewhat in isolation; it is now necessary to view these within the context of the hierarchical organization of the controls of behavior.

Hierarchical Organization of Behavior

Basic Principles

Across species, evidence suggests that the controls of emotion (Bonanno, 2001; Rolls, 2004, 2005), motivation (Berridge, 2001), and action (Dehaene & Naccache, 2001; Gray & McNaughton, 2000; Hughlings Jackson, as cited in Taylor, 1958; G. A. Miller et al., 1960; Toates, 1995, 1998, 2006) are organized hierarchically. Evolutionarily newer ("high-level") structures coexist and share control with evolutionarily older ("low-level") structures. Corresponding to this structure, emotion (Bonanno, 2001; Ochsner & Barrett, 2001), motivation (Berridge, 2001), and behavior are controlled by a combination of (external) stimuli and (internal) cognitive events (Carver, 2005; Metcalfe & Mischel, 1999; Strack & Deutsch, 2004). These triggers sometimes act in the same direction ("cooperatively") but at other times pull in opposite directions (Berridge, 2001; Toates, 1995, 1998, 2005, 2006). A model of how this dual control can be realized within a hierarchical structure has been proposed (Toates, 1995, 1998, 2006).

The hierarchical model is based on the assumption that behavior arises from a complex interaction between direct (as in Figure 1) and indirect controls. The direct control is excited by the basic perceptual analysis of the physically present stimuli. In humans, the indirect control is based on refined cognitions ("representations") held in memory (e.g., the meaning attached to the potential interaction) and is associated with common-sense notions of conscious intentions, for example, in the sexual domain of the kind "I intend to pursue this sexual goal because of the consequences that I expect to follow from it." Motivation depends on a combination of such stimulus-based ("online") and cognition-based ("offline") information.

Looking broadly at brain and behavior, evidence suggests that the relative weighting given to online and offline controls varies with a number of factors:

1. Learning: With repetition, behavior tends to become more online ("habit-like"; G. A. Miller et al., 1960; Zuckerman & Kuhlman, 2000).

2. Age: In general, development consists of the emergence of a greater role for cognition (Metcalfe & Mischel, 1999), sometimes expressed as increasing voluntary control relative to "reactive" determination of behavior (Derryberry & Rothbart, 1997).

3. Phylogeny: Relative to other species, humans have a particularly large capacity to use offline controls (McClure, Laibson, Loewenstein, & Cohen, 2004).

4. Brain damage: Damage to a specific brain region can be associated with disruption to either direct or indirect controls (McDonald & White, 1993).

For example, damage to the prefrontal cortex has a particularly disruptive effect on offline controls, whereas damage to the dorsal striatum disrupts online controls.

5. Chemicals: Various chemical manipulations change the weighting. For example, alcohol tends to induce alcohol myopia, the result of which is that immediately present stimuli assume more weight than representations of remote factors (Baumeister & Vohs, 2003; Steele & Josephs, 1990). In some circumstances, testosterone can lower the weight of offline controls relative to online (van Honk & Schutter, 2007).

6. Emotion and stress: These tend to decrease the weight of offline controls relative to online (Arnsten, 1998; Mischel, Ayduk, & Mendoza-Denton, 2003).

7. Gender: In the domain of sexual motivation (reviewed later), evidence suggests a greater weight of offline controls in women as compared to men.

Consideration of such a hierarchical structure provides insight into some of the more peculiarly (offline) human features of motivation and how they interact with the more general lower (online) layers of control. As such, the model permits cross-fertilization with some powerful theoretical models advanced in recent years in the psychology literature (Baumeister & Vohs, 2003; Carver & Scheier, 1990; Metcalfe & Mischel, 1999; Ramanathan & Menon, 2006).

It is possible to fit such ideas to current theories on the nature of conscious and unconscious processing of information (Baars, 1988; Dehaene & Naccache, 2001; Toates, 2006). Intentional action is said to be associated with conscious processing as organized by a workspace within the brain. Conscious states are based on coherent activity between various brain regions. Linking this to ideas of incentive motivation, one might suggest that as the strength of the incentive increases, so its ability to lock into such coherent circuits of brain activity and form the goal of intentional action also increases.

According to a variety of experimental evidence, we do not necessarily have conscious access to all the determinants of our behavior, even at the high level of intentions (Berridge, 2001; Fitzsimons & Bargh, 2003; Nisbett & Wilson, 1977; Robinson & Berridge, 1993). For example, a person might find himself pursuing a particular route home that is associated with a drug or sexual incentive, but without being consciously aware of having made such a choice of route.

Application to Human Sexuality

Several features of human sexuality are entirely congruent with the idea of layered and hierarchical organization, as described next. Behavior is determined by a combination of excitatory and inhibitory processes (Bancroft, 1999; Bancroft & Janssen, 2001); it is argued that these act at various levels. Cognitive processes, involving representations of (a) incentive stimuli and (b) consequences of interaction with them, can offer either excitation or inhibition of behavior depending on the quality of the representation and the circumstances in which the person finds him- or herself. The top level of the …