Sexual approach conditioning: tests of unconditioned stimulus devaluation using hormone manipulations

The behavior system for sexual learning

In this paper we review and update evidence relevant to formulating a behavior system for sexual learning. We emphasize behavioral rather than neurobiological evidence and mechanisms. Our analysis focuses on three types of responses or response modes: general search, focal search, and consummatory or copulatory behavior. We consider how these response modes are influenced by three categories of stimuli: spatially distributed contextual cues, arbitrary localized stimuli, and species-typical cues provided by the sexual partner. We characterize behavior control by these types of stimuli before and after various Pavlovian conditioning procedures in which the unconditioned stimulus is provided by copulation with a sexual partner. The results document extensive Pavlovian modifications of sexual behavior. These conditioning effects reflect new conditioned responses that come to be elicited by various categories of stimuli. In addition, the conditioning of contextual cues and localized stimuli facilitate sexual responding to species-typical cues. Thus, learning experiences enhance how the species-typical cues of a sexual partner stimulate sexual behavior. These modulatory conditioning effects not only produce significant behavioral changes but also increase rates of fertilization of eggs and numbers of offspring produced. These latter findings suggest that sexual learning can lead to differential reproductive success, which in turn can contribute to evolutionary change.

The origins of individual differences in how learning is expressed in rats: A general-process perspective

Laboratory rats can exhibit marked, qualitative individual differences in the form of acquired behaviors. For example, when exposed to a signal-reinforcer relationship some rats show marked and consistent changes in sign-tracking (interacting with the signal; e.g., a lever) and others show marked and consistent changes in goal-tracking (interacting with the location of the predicted reinforcer; e.g., the food well). Here, stable individual differences in rats’ sign-tracking and goal-tracking emerged over the course of training, but these differences did not generalize across different signal-reinforcer relationships (Experiment 1). This selectivity suggests that individual differences in sign- and goal-tracking reflect differences in the value placed on individual reinforcers. Two findings provide direct support for this interpretation: the palatability of a reinforcer (as measured by an analysis of lick-cluster size) was positively correlated with goal-tracking (and negatively correlated with sign-tracking); and sating rats with a reinforcer affected goal-tracking but not sign-tracking (Experiment 2). These results indicate that the observed individual differences in sign- and goal-tracking behavior arise from the interaction between the palatability or value of the reinforcer and processes of association as opposed to dispositional differences (e.g., in sensory processes, “temperament,” or response repertoire).

Sign Tracking, but Not Goal Tracking, is Resistant to Outcome Devaluation

During Pavlovian conditioning, a conditioned stimulus (CS) may act as a predictor of a reward to be delivered in another location. Individuals vary widely in their propensity to engage with the CS (sign tracking) or with the site of eventual reward (goal tracking). It is often assumed that sign tracking involves the association of the CS with the motivational value of the reward, resulting in the CS acquiring incentive value independent of the outcome. However, experimental evidence for this assumption is lacking. In order to test the hypothesis that sign tracking behavior does not rely on a neural representation of the outcome, we employed a reward devaluation procedure. We trained rats on a classic Pavlovian paradigm in which a lever CS was paired with a sucrose reward, then devalued the reward by pairing sucrose with illness in the absence of the CS. We found that sign tracking behavior was enhanced, rather than diminished, following reward devaluation; thus, sign tracking is clearly independent of a representation of the outcome. In contrast, goal tracking behavior was decreased by reward devaluation. Furthermore, when we divided rats into those with high propensity to engage with the lever (sign trackers) and low propensity to engage with the lever (goal trackers), we found that nearly all of the effects of devaluation could be attributed to the goal trackers. These results show that sign tracking and goal tracking behavior may be the output of different associative structures in the brain, providing insight into the mechanisms by which reward-associated stimuli—such as drug cues—come to exert control over behavior in some individuals.

Differential c-fos expression in the brain of male Japanese quail following exposure to stimuli that predict or do not predict the arrival of a female

We investigated the effects of presenting a sexual conditioned stimulus on the expression of c-fos in male Japanese quail. Eight brain sites were selected for analysis based on previous reports of c-fos expression in these areas correlated with sexual behaviour or learning. Males received either paired or explicitly unpaired presentations of an arbitrary stimulus and visual access to a female. Nine conditioning trials were conducted, one per day, for each subject. On the day following the ninth trial, subjects were exposed to the conditional stimulus (CS) for 5 min. Conditioning was confirmed by analysis of rhythmic cloacal sphincter movements (RCSM), an appetitive sexual behaviour, made in response to the CS presentation. Subjects in the paired condition performed significantly more RCSM than subjects in the unpaired group. Brains were collected 90 min following the stimulus exposure and stained by immunohistochemistry for the FOS protein. Significant group differences in the number of FOS-immunoreactive (FOS-ir) cells were found in two brain regions, the nucleus taeniae of the amygdala (TnA) and the hippocampus (Hp). Subjects in the paired condition had fewer FOS-ir cells in both areas than subjects in the unpaired condition. These data provide additional support to the hypothesis that TnA is implicated in the expression of appetitive sexual behaviours in male quail and corroborate numerous previous reports of the involvement of the hippocampus in conditioning. Further, these data suggest that conditioned and unconditioned sexual stimuli activate different brain regions but have similar behavioural consequences.

Androgen mediation of conditioned rhythmic cloacal sphincter movements in Japanese Quail (Coturnix japonica)

Demonstrations of increased reproductive success due to sexual conditioning in Japanese quail (Coturnix japonica) have been reported, although the mechanisms that underlie these effects have remained elusive. One possible mechanism is conditioned rhythmic cloacal sphincter movements (RCSM). Two experiments were conducted with male quail to determine whether associations between a conditioned stimulus (CS) and a hen would result in the ability of the CS to elicit RCSM, and to explore the androgen mediation of conditioned RCSM. The results suggest that a focal CS paired with visual access to a female will elicit RCSM via a representation of the hen activated by the CS. Further, the available evidence indicates that conditioned RCSM is androgen mediated and that this learning may transfer across breeding seasons.

A new model for sexual conditioning: the ring dove (Streptopelia risoria)

Sexual conditioning investigations have primarily used rats and domestic quail as subjects. Although much has been learned from such experiments, the relative simplicity of rat and quail reproductive behavior prohibits investigation of certain experimental questions about sexual conditioning. In contrast, the reproductive behavior of male and female ring doves (Streptopelia risoria) is complex, involving courtship, bonding, and parental care. In the present experiment, male ring doves that were presented with a visual conditioned stimulus paired with access to their pair-bonded mate (the unconditioned stimulus) showed sexual conditioned responding in the presence of the conditioned stimulus. These results represent the first evidence of sexual conditioning in ring doves and illustrate the potential of ring doves as a useful model for future sexual conditioning investigations.

Learning with arbitrary versus ecological conditioned stimuli: Evidence from sexual conditioning

Laboratory investigations of Pavlovian conditioning typically involve the association of an arbitrary conditioned stimulus (CS) with an unconditioned stimulus (US) that has no inherent relation to the CS. However, arbitrary CSs are unlikely to become conditioned outside the laboratory, because they do not occur often enough with the US to result in an association. Learning under natural circumstances is likely only if the CS has a preexisting relation to the US. Recent studies of sexual conditioning have shown that in contrast to an arbitrary CS, an ecologically relevant CS is resistant to blocking, extinction, and increases in the CS-US interval and results in sensitized responding and stronger second-order conditioning. Although the mechanisms of these effects are not fully understood, these findings have shown that signature learning phenomena are significantly altered when the kinds of stimuli that are likely to become conditioned under natural circumstances are used. The implications of these findings for an ecological approach to the study of learning are discussed.

Relevance of non-human animal studies to the understanding of human sexuality

There are several reasons for studying sexual behavior in animals other than the human. Prominent among these is probably the use of sexual behavior as a model system for analyzing hormone actions on the brain. In this kind of study, lordosis behavior in the female rat has frequently been used as the behavioral end-point. Recently, the mouse has become a popular subject because of the availability of strains lacking estrogen or progesterone receptors. Animal sexual behavior is also studied for its own sake by biologists, by agricultural scientists and by those interested in the management of wildlife resources. Finally, there are some researchers interested in human sexual behavior who prefer to approach their problems through studies in non-human subjects, as they thereby gain access to techniques of experimentation impossible in human subjects. This is particularly important when the neurobiological basis of sexual behavior is the subject of study. Two striking similarities between humans and other mammals are the hormone dependency of sexual motivation and the fundamentally bisexual organization of the nervous system. In addition, animal studies have revealed how learning with sexual reinforcement can be an important determinant of sexual behavior. The role of learning has been largely ignored in human studies, and this has led to many misconceptions. Paramount among these is the idea that sexual behavior is associated with reproduction.

Unconditioned sexual incentive motivation in the male Norway rat (Rattus norvegicus)

Sexual incentive motivation was evaluated in a procedure consisting of a large open field where incentive animals were confined behind wire mesh openings. When sexually inexperienced male rats (Rattus norvegicus) were exposed to the receptive female-male incentives, they spent more time close to the female. If the incentives were receptive female-nonreceptive female, the receptive female was preferred. However, when the alternatives were nonreceptive female-male, no preference was obtained. Castration abolished preference for the receptive female, and treatment with testosterone propionate restored it. Estradiol plus oil is as efficient as estradiol plus progesterone for giving the ovariectomized female incentive properties. The living female can be replaced with female odor. Sexual experience did not have any long-term effects on the female's incentive value, but immediately preceding limited sexual activity enhanced it

The behavior of the Japanese or domestic quail Coturnix japonica

This paper reviews the literature pertaining to the behavior of the Japanese or domestic quail Coturnix japonica. Details are given of the classification, characteristics, domestication and the economic and research potential of the species. Further sections deal with sensation and perception (including taste and smell, vision and hearing), maintenance behavior (including feeding and drinking, dust bathing and thermoregulation), development and aging (including vocalization, filial imprinting, sexual imprinting, fear and avoidance responses, sexual maturation and aging), adult learning (including habituation, instrumental conditioning, Pavlovian conditioning and observational learning), photoperiodism, reproductive behavior (including courtship and mating, hormonal control and ontogeny of sexual differentiation, and male and female sexual behavior), parental behavior (including nest-site selection and nest building, incubation behavior and its hormonal control, and hen-chick relationships), and aggressive behavior and dominance (including agonistic behavior and the hormonal control of aggressive behavior).