Psychophysiological considerations in theory, research, and treatment

Premature ejaculation: Psychophysiological considerations in theory, research, and treatment

David L Rowland

In this review we summarize recent progress in the understanding and treatment of premature ejaculation (PE). Long understudied because of the existence of presumably effective treatment procedures and an assumption of an underlying psychological etiology, premature ejaculation has increasingly become the subject of physiological and psychological investigations by sexologists. And rightfully so: PE is the most frequently occurring sexual dysfunction in men, affecting nearly 30% at some time in their lives (Laumann, Gagnon, Michael, & Michaels, 1994; Spector & Carey, 1990), and, based on our own study of over 150 men, one that severely impacts the quality of the sexual relationship (Rowland, Cooper, & Slob, 1996; Rowland, Cooper, Slob, & Houtsmuller, 1997). Coupled with a long-term prognosis generally considered less promising than that for erectile problems (Hawton, 1992), these conditions suggest that there is clear need both for reexamination of the theoretical and empirical underpinnings of this dysfunction and for application of novel perspectives and techniques to its study. Current thinking about PE has been summarized in several recent critical reviews (Grenier & Byers, 1995; Madakasira & St. Lawrence, 1997; Metz, Pryor, Nesvacil, Abuzzahab, & Koznar, 1997). Our goal, more limited in scope, is to examine PE specifically within the context of a psychophysiological model, one that emphasizes the interactive nature of the mind-body relationship in the etiology and treatment of sexual dysfunction.

Ejaculatory Mechanisms

Understanding premature ejaculation is facilitated by comprehension of the mechanisms underlying the ejaculatory process (see Figure 1). Ejaculation may be viewed as the efferent (motor) component of a reflex process beginning with sensory stimulation around the glans penis. Although the sensory component may not be critical to the process (e.g., spontaneous ejaculation), it clearly facilitates the reflex and is considered important to sexual response by most men

(Rowland & Slob, 1992). Whether penile receptors conveying erotic (i.e., pleasurable) sensory information are identical to those relaying simple surface somesthetic information is not known. Nor is it clear whether specific sensory receptors play distinctive roles in mediating erection, ejaculation, or both. Free nerve endings and corpuscular receptors, both present throughout the glans, are possible candidates for receptors mediating these reflexes (Halata & Munger, 1986).

Afferent signals from the penis traverse the sensory component of the pudendal nerve, entering the spinal cord at the sacral level. The efferent component of the reflex is multifaceted (deGroat & Booth, 1980; Kedia, 1983), involving both autonomic and somatic output. Autonomic sympathetic nerves (primarily a-adrenergic neurons arising from the thoracolumbar spinal regions) elicit smooth muscle contractions in the vas deferens, seminal vesicles, and prostate to effect seminal emission and bladder neck closure. This process triggers a secondary reflex, commencing with urethral sensation of seminal emission (which accounts for the subjective sense of “inevitable” ejaculation), then triggering the somatic component of the reflex (Biester & Howards, 1988). Rhythmic contractions of the striated (somatic) bulbocavernosus and ischiocavernosus muscles are mediated through motor neurons of the pudendal nerve (exiting the spinal cord at the sacral level), and this cholinergically mediated response is typically associated with the subjective experience of orgasm (Davidson, 1980).

Even at this simple “reflexive” level, many mysteries still require unraveling. For example, even though seminal emission (or the smooth muscle contractions underlying it) appears to trigger the striate contractions associated with seminal expulsion and orgasm, a small subset of men (some of whom are premature ejaculators) experience seminal emission without subsequent seminal expulsion. Furthermore, just how or why striate muscle contractions engender the orgasmic experience is not clear. Perhaps sexual stimulation stretches these muscles to the point of vigorous contraction, a process that might be facilitated by the posterior pituitary hormone, oxytocin (Carmichael et al., 1987). Finally, the ejaculatory contractions themselves must trigger yet another set of sensory neurons that mediate the pleasant experience of orgasm. Indeed, a rare and poorly understood dysfunction is that of anesthetic ejaculation, a condition in which men ejaculate with no sensation of orgasm (Williams, 1985). Complexity is added as one attempts to understand the modulating role of centrally mediated processes, either facilitatory or inhibitory, over the ejaculatory reflex. Most men indicate having at least moderate control over the ejaculatory process (Rowland, Cooper, Haensel, & Slob, 1997), although the relevant psychological mechanisms have yet to be explicated or even carefully examined. The person’s level of psychological sexual arousal is one factor that may affect the ejaculatory threshold, but this construct lacks strong explanatory power as it provides little or no insight into the multitude of factors that might underlie it. More pertinent to the question are the conscious cognitive (e.g., attentional focus) and affective (e.g., emotional excitement) strategies used to enhance or attenuate arousal and thereby influence the ejaculatory threshold. Although the physiological mechanisms underlying these central processes in the human male are difficult to investigate, studies in rats suggest that descending signals from the brain overcome the prevailing spinal inhibition over sexual reflexes (e.g., Marson & McKenna, 1992).

Variation in Ejaculatory Latency and Its Sources

Variation in ejaculatory latency can arise from any of the known or unknown afferent, central, or efferent processes described above. A heuristic, though somewhat contrived, dichotomy has characterized the study of ejaculatory latency such that causes have frequently been relegated to either the biological or the psychological domain. Although operationalization of these domains has not been offered, one might assume that the biological refers to those “hard-wired” neural pathways involved in the ejaculatory reflex, whereas the psychological pertains to those less precisely defined constructs usually associated with central processing, involving motivation, arousal, affect, and cognition. Part of the challenge facing sexologists is that of identifying probable factors, either biological or psychological, that contribute to variation in ejaculatory latency. Yet, it is important to recognize that these domains are not independent, and that at the level of the central nervous system, the distinction between psychological and biological factors is often blurred. A simple illustration of this tangled connection can be drawn from animal models investigating ejaculatory response. DPAT, a serotonergic agonist, when acting upon the diencephalic region of the male rat brain, greatly decreases ejaculation latency during copulation with a female rat in heat (Ahlenius, Larsson, Svensson, Hjorth, & Carlsson, 1981; FernandezGuasti, Escalante, Ahlenius, Hillegaart, & Larsson, 1992; Haensel, Mos, Olivier, & Slob, 1991). This effect, however, depends partly on the rat’s sexual experience (Mos, Olivier, Bloetjes, & Poth, 1990), and indeed sexual experience by itself can achieve an effect similar to low doses of DPAT (Rowland & Houtsmuller, in press). Yet sexual experience represents a psychological construct that proxies for a series of chemical processes in the brain, as yet only partly understood. Despite this parallel effect between DPAT and experience, few researchers would, given the interdependence of these two domains, advance a dualistic interpretation, attributing one effect to physiological factors and the other to psychological factors.

Nevertheless, in men, where the complexity of central biopsychological relationships is undoubtedly greater than that of the male rat, a dualistic approach may be useful and sometimes even necessary for addressing basic questions about the ejaculatory process. There are, for example, rare occasions when biological factors are primarily implicated, and psychological factors presumably have little or no role-certain types of urological pathology (prostatitis) or spinal cord injury are known to affect ejaculatory control (Kuhr, Heiman, Cardenas, Bradley, & Berger, 1995; Williams, 1984). Investigators may also find it expedient to separate the psychological from the biological domain when dealing with such factors as arousal, attention, and self-efficacy, constructs that resist simple reductionistic biochemical characterizations. Such psychological constructs, redefined in more operational ways, will undoubtedly continue to have value in the investigation of sexual dysfunctions. Nevertheless, despite the convenience of isolating the biological and psychological domains, the view of an increasing number of sexologists (McCarthy, 1994; Rowland & Slob, 1995; Strassberg, 1994) is that in the absence of obvious pathophysiology, psychological and biological processes interact to generate a dysfunctional response cycle (Bancroft, 1989; Buvat, BuvatHerbaut, Lemaire, Marcolin, & Quittelier, 1990). Such a perspective cautions against interpreting findings in one domain without reference to the other.

Subtypes of Premature Ejaculation

Recent investigators of PE increasingly support the notion that this dysfunction may have various etiologies, and, indeed, most behavioral scientists have long since abandoned the idea that behavioral endpoints of this sort can be attributed to a single cause. Thus, using different etiologies constitutes one strategy for defining subtypes of PE. But unlike erectile dysfunction, the mainly psychogenic and mainly organic classifications are hypothetical for PE. Therefore, given this lack of clear differentiating etiologies, an alternative method of identifying PE subtypes is to base distinctions upon the differential response characteristics of PE men. Traditionally, several broad classifications based upon response characteristics have been made. Most researchers differentiate between primary (lifelong) and secondary (acquired) PE, and between global and situational PE. Many also specify whether there is coexisting erectile dysfunction (ED). Recently, for example, Cooper, Cernovksy, and Colussi (1993) described important differences between men with primary and secondary PE: Those with primary PE showed above average anxiety, whereas those with secondary PE exhibited diminished sex drive and arousal, and higher probability of concomitant ED. Even these rudimentary distinctions, however, have not been exploited to their maximum potential in clinical and experimental research. For example, although it is known that secondary PE frequently coexists with ED, little research has focused on the distinction between secondary PE in men with and without ED. Nor have researchers attempted to distinguish between PE that is secondary to ED, versus PE that is primary with ED secondary. Symptoms having different temporal sequences and comorbidities may well have different etiologies (Buvat et al., 1990; Gospodinoff, 1989; Williams, 1984).

To complicate matters, a number of other strategies for PE classification based upon response characteristics have recently been proposed within experimental investigations. For example, Colpi, Fanciullacci, Beretta, Negri, and Zanollo (1986) studied a group of “true” PE men, ones requiring 15 or fewer pelvic thrusts to ejaculation, independent of the sexual partner. And, in a study designed to treat PE with alpha-adrenergic blockers, Cavallini (1995) defined a group of men with “primitive” PE, men whose major “complaint was psychogenic PE but who proved unresponsive to psychotherapy” (p. 127). Whereas these groups were defined a priori by the researchers, Kim, Choi, and Lho (1992) delineated a posthoc system of classification whereby PE men with fluctuating penograms during sexual stimulation (a measure of dynamic blood flow changes in the penis) could be differentiated from PE men with stable penograms. Because most men with fluctuating penograms also had erectile problems, Kim et al.’s (1992) groupings may show significant overlap with more conventional classifications, such as primary PE and PE with coexisting ED. Although on one level this burgeoning of classification schema confuses the issue, it represents experimenters’ preliminary attempts to identify and to study homogenous subgroups drawn from a somewhat diverse PE population. At a broader level, such unconventional distinctions typify an area of study that, due to its infancy, must yet grapple with many unresolved issues regarding the nature of the disorder.

As seen above, classifications of PE types based upon genital response may be useful in that they suggest a particular etiology and/or avenue of investigation. Yet they do little to address more subtle variations in response that might help define subtypes of PE. For example, we (Rowland et al., 1996) recently found that the psychoaf fective response to penile stimulation in men with PE is often quite distinguishable from that of men with coexisting PE and ED on at least two dimensions. The latter group not only typically ejaculates with a flaccid or half-erect penis, but also exhibits an affective response to penile stimulation that differs from PE-only men. This latter characteristic (discussed later) may be important in revealing the causal factors responsible for these men’s rapid ejaculatory response.

Ejaculation Latency Versus Ejaculatory Control

Despite the interest in understanding factors that affect ejaculatory threshold, neither variation in ejaculation latency nor its source per se (whether biological or psychological) may be the critical factor in understanding and defining PE. Perhaps more important is the man’s sense of control over the timing of ejaculation-it is a lack of such control and its consequent duress to both the individual and the sexual dyad that renders the condition “dysfunctional” (see American Psychiatric Association, 1994; Grenier & Byers, 1995; Williams, 1984). In this respect, researchers who exclusively pursue afferent or efferent sources of variation in ejaculation latency may be addressing the issue of PE only indirectly, and thus may never succeed in completely accounting for rapid ejaculation as a dysfunction. That is, despite the difficulties in operationalizing “control,” as well as the reluctance of some researchers to accept the utility of such an idea (at least in its presently defined status: Ruff & St. Lawrence, 1985), the key issue may lie in the ability (or lack thereof) to exert control over the timing of ejaculation (McCarthy, 1988; Vandereycken, 1986). Accordingly, research efforts in the future might be directed toward delineating cognitive/affective strategies that enable control over any supraspinally influenced reflexes, not just those related to erection and ejaculation. An equally important approach lies in understanding the relationship between ejaculatory latency and ejaculatory control, an avenue that has only recently received attention in a study by Grenier and Byers (1997). Although these researchers found only weak relationships between perceived latency and control, their sample of college students was young, relatively inexperienced sexually, and did not include men diagnosed with PE. It may be, for example, that men with PE have less control over the timing of ejaculation compared with sexually functional men, or with men having coexisting PE and ED who presumably ejaculate out of fear of losing their erection rather than from loss of control. Actual versus perceived control are, of course, very different measures, the latter representing a highly subjective assessment that has no reference outside one’s own experiences; as such, identifying differences in perceived control between PE and functional men may not provide highly meaningful information. Relying upon “perceived control” as the defining criterion for PE has other problems as well. Men who have ejaculation latencies that exceed several minutes, for whatever reason, may have less need to be concerned about “control” over the timing of their I ejaculation. Therefore, control may become an issue for men only when latencies are short and thus interfere with sexual satisfaction.

Theoretical Approaches to Premature Ejaculation

It is not our intention to provide an exhaustive review of the many theoretical and empirical approaches to the etiology of PE, as these have recently been discussed in a number of comprehensive reviews (see Grenier & Byers, 1995; Metz et al., 1997). Rather, we briefly present some of the recent concepts and ideas with two goals in mind: to engender in the reader an appreciation for the broad range of putative factors that conceivably impact ejaculatory latency and control, and to set the stage for our discussion of the psychophysiological investigation of PE.

Psychological Considerations

Psychological/behavioral explanations for premature ejaculation abound. And although these approaches tend to be theoretically rich, many have been empirically arid. Most that predate the 1970s are intrapsychic in nature (see Kilmann & Auerbach, 1979) and generally not considered mainstream by most sexologists today. However, this should not be construed as devaluation of the importance of a man’s sexual history and relationships (e.g., anger and hostility toward the partner: Kaplan, 1974; Salzman, 1972) in the development of dysfunctional response patterns. Rather, such individualized historical factors are simply difficult to investigate within the context of an experimental analysis, which aims primarily at establishing group rather than individual differences. Two more recent psychological approaches (not intrapsychic) have gained sufficient foothold to generate experimental investigation and to impact current treatment strategies for PE: one deals with behavioral changes that influence the ejaculatory process; the second with the interrelated dimensions of anxiety and arousability.

Those taking behavioral approaches toward PE have, to some extent, been less concerned with delineation of cause and effect than with identifying strategies that may lessen the likelihood of its occurrence. For example, Zilbergeld (1992) suggested that men can learn to increase their control over ejaculation by making behavioral and positional adjustments during intercourse. Such strategies, combined with cognitive efforts aimed at monitoring sexual arousal and ejaculatory inevitability, appear to help many men cope successfully with PE (McCarthy, 1994). Somewhat implicit in this approach is the idea that PE reflects the lack of learning effective control over the ejaculatory process (Kaplan, 1974; Masters & Johnson, 1970). However, no empirical evidence details what, if any, learning/behavioral experiences might account for PE in the first place, although speculation regarding a conditioning process during adolescence has been suggested.

Anxiety has long been associated with sexual dysfunctions, and PE is no exception (Cooper, 1969; Hale & Strassberg, 1990; Masters & Johnson, 1970, Williams, 1984). Depending on varying situations and dysfunctions, states of anxiety are known both to enhance and to inhibit arousal (Beck, Barlow, Sakheim, & Abrahamson, 1987; Rowland & Heiman, 1991). Regarding premature ejaculation, anxiety presumably increases sympathetic dominance and, given the role of the sympathetic system in ejaculation, facilitates the reflex. This theoretically appealing explanation has garnered only qualified or minimal empirical support (Kockott, Feil, Ferstl, Aldenhoff, & Besinger, 1980; Strassberg, Mahoney, Schaugaard, & Hale, 1990). In addition, a number of theorists argue that anxiety may be secondary to the dysfunctional response rather than its actual cause, particularly in men with erectile insufficiency (Cooper, 1969; Gospodinoff, 1989; Strassberg et al., 1990). Finally, a major premise of this idea has never been adequately addressed: Given the presumed role of the parasympathetic system in achieving erection (Brindley, 1983; Carati, Creed, & Keogh, 1987), one might predict a slower rate of penile tumescence in PE men (who are under sympathetic dominance) than functional controls, particularly in view of reports that PE men show less sympathetic nervous system depression than controls during erection (Ertekin, Colakoglu, & Altay, 1995). Such an expectation has thus far not been confirmed (Spiess, Geer, & O’Donohue, 1984). It may be, of course, that anxiety does not affect ejaculatory mechanisms through local autonomic function in the genital region, but acts more centrally to depress the descending inhibitory influence on the reflex. Such ideas remain to be investigated.

An alternative conceptualization of PE was offered by Kaplan (1974) who suggested that because of their anxiety, such men are less cognizant of their sexual arousal. In underestimating their state of (hyper)arousal, these men ejaculate prior to expectation. The oftcited (Spiess et al., 1984), though unconfirmed (Rowland et al., 1991), negative correlation between PE and frequency of sexual activity suggests several mechanisms might be at work. The first is based on the finding that sexual arousal habituates to repeated stimulation (O’Donohue & Geer, 1985; Over & Koukounas, 1995) but that this habituation is less likely to occur in PE men because of their lower overall frequency of activity (LoPiccolo & Stock, 1986). The second is that, because of their lower sexual activity and their shortened ejaculatory latency, PE men may have had fewer opportunities to learn adequate control over the ejaculatory process (Kaplan, 1974). As with the anxiety hypothesis, the arousal theory has received minimal empirical support, and the studies that have been undertaken have not provided convincing evidence for its involvement in PE, at least in the manner proposed here (Spiess et al., 1984; Strassberg, Kelly, Carroll, & Kircher, 1987).

Perhaps some of the current dissatisfaction with psychological approaches toward PE (e.g., Assalian, 1994) may be attributed to (a) the false assumption that intrapsychic theories or vaguely elaborated learning theories form the basis for most current psychological thinking about sexual dysfunction; (b) the fact that psychological explanations for PE run counter to the current medicalizing trend that encourages technologically appealing (and often economically opportune) solutions to sexual disorders that emulate “disease” conditions; and (c) the general perception that even the more recent psychological theories, such as those involving anxiety and arousal, have not been experimentally supported. Regarding this last point, the lack of support for such hypotheses may stem not so much from the inadequacy of the hypotheses themselves, but rather from inadequate testing of those hypotheses. Specifically, until recently the response of PE men had not been sufficiently differentiated from that of functional controls within a laboratory context to enable the experimental study of factors influential to the ejaculatory process. This, together with the lack of consistent definitions of relevant psychological constructs (e.g., anxiety) and the difficulty of testing such concepts, has undoubtedly hindered progress in this area.

Also disappointing has been the lack of updating and refinement of psychological perspectives to incorporate recent developments in cognitive and motivational (affective) psychology. Application of such concepts has permeated the discussion of erectile dysfunction (Barlow, 1986; Cranston-Cuebas & Barlow, 1990; Dekker & Everaerd, 1988), yet the study of PE has suffered from lack of comparable input. Contemporary psychological theorizing about PE is practically nonexistent, though recent researchers have at least begun to address this deficiency (Strassberg et al., 1990; Rowland et al., 1996).

Biological Approaches

In contrast to the dearth of research on psychological/behavioral approaches, researchers using biological approaches have offered a number of recent findings that may contribute to a broader understanding of PE. Some studies investigating the biological domain have focused on patients with primary PE or on other defined subsets of PE patients. The rationale is that their ejaculation latency is independent of developmental and situational factors and therefore more likely to represent an inherently biological phenomenon rather than a dysfunction acquired through pathophysiology or psychologically related events.

As stated earlier, biological approaches have concentrated on aberrations in the afferent and efferent processes involved in the ejaculatory reflex. Beginning with the afferent system, penile receptor thresholds may be lower in men with PE, although contradictory evidence surrounds this issue (Gospodinoff, 1989; Rowland, Haensel, Blom, & Slob, 1993; Xin, Chung, Choi, Seong, & Choi, 1996). The extent to which such differences in sensitivity, should they exist, contribute to ejaculatory latency is also unclear, as some men with very high sensitivity show no inclination toward PE, and some with very low sensitivity (e.g., men with concomitant PE and ED) exhibit PE. Interestingly, hypogonadism seems to be associated with penile hypersensitivity (Burris, Gracely, Carter, Sherins, & Davidson, 1991), yet no evidence supports the existence of shortened latencies in these men. Furthermore, although reducing penile sensitivity through topical anesthetics reportedly alleviates the PE (Berkovitch, Keresteci, & Koren, 1995; Xin, Choi, Seong, & Choi, 1995), it does not necessarily follow that hypersensitivity is, therefore, the root of the problem. Interference with any of the multiple steps involved in the ejaculatory reflex is likely to retard the ejaculatory process.

Evoked potentials (an averaged electrophysiological response measured from the somatosensory cortex) resulting from electrical stimulation of the dorsal penile nerve show greater amplitude in men with PE than functional controls, suggesting augmented central representation of penile stimulation in these men (Fanciullacci, Colpi, Beretta, & Zanollo, 1988). In addition, some evidence suggests involvement of efferent (motor) striate muscle response in rapid ejaculation. EMG response at the perineal level is heightened and the bulbocavernosal response is shortened in PE men (Colpi et al., 1986; Gospodinoff, 1989; Vignoli, 1978). Such findings would suggest hypersensitivity and/or hyperexcitability in the ejaculatory reflex of men with PE, conditions that presumably could interfere with the normal developmental/learning process of voluntary control (e.g., Gospodinoff, 1989; Kaplan, 1974). Although these findings provide interesting inroads in the study of the physiological involvement in PE, significant problems remain. Some of these findings have not been replicated (Fanciulla et al., 1988; Kim et al., 1992). And, presuming they are eventually reliable, they nevertheless rest on the untested assumptions that these biological mechanisms comprise fairly immutable aspects of the ejaculatory reflex and therefore exert significant impact on variation in or control over the latency to ejaculation.

The Psychophysiological Approach to the Study of Premature Ejaculation

The psychophysiological study of sexual response in general, and PE in particular, incorporates both a conceptual orientation and a methodological approach. Conceptually, the psychophysiological perspective emphasizes the bidirectional and interactive relationships between biological and psychological functioning as illustrated in Figure 1. In this respect, the psychophysiological study of sexual response is not unlike other areas of investigation that involve both a strong psychological and somatic component, as in the studies of emotion, pain, and addictions. The psychophysiological approach may be applied in any number of different ways to the study of sexual dysfunction. It may, for example, emphasize the psychosomatic nature of sexual problems, focusing on the ways in which psychological interpretations affect physiological responding. Or it may stress the cyclic nature of sexual responding in that functional or dysfunctional genital response within a given stimulus context has the potential to affect future perceptions, cognitions, and emotions about sexual situations. For example, a sexually dysfunctional outcome, once interpreted as a “sexual failure” by the individual, is likely to interfere with future erectile and ejaculatory responses. Such failures impact on self-efficacy, and although perhaps not part of the original etiological blueprint, may become an important factor in maintaining or exacerbating the dysfunction.

Also important to the psychophysiological approach is its application to the study of sexual dysfunction at multiple levels. At a very broad level, PE has been characterized as a condition of somatic vulnerability which interacts with varying psychological states, including anxiety (Strassberg et al., 1990). One can extend this approach to a more detailed level of analysis. Take, for example, the common complaint of penile hypersensitivity by PE men, sometimes offered as an explanation for their short ejaculation latencies. Rather than a cause itself, the conclusion of penile hypersensitivity may represent a post hoc accounting of the repeated experiences these men have with uncontrolled, rapid ejaculation. Thus, researchers examining this issue should employ unbiased methods of determining penile thresholds, because just the expectation of higher sensitivity may influence their subjectively assessed thresholds (Rowland, Cooper, Haensel, & Slob, 1997). In addition they should attempt to specify how and whether penile sensitivity has a direct bearing on psychological processes such as heightened sexual arousal. A similar approach might be applied to reported hyperexcitability of the bulbocavernosal response in PE men. Investigators should ask why the usual psychological strategies that inhibit the ejaculatory response seem ineffective in these men. Although the specific questions will vary, they share the common feature that in each instance, the potential bearing of one domain upon the processes of the other is taken into account.

Methodologically, sexual psychophysiology has typically been used to relate psychological stimuli and their consequent mental events to components of the sexual response, defined both in physiological and psychological terms. In a typical psychophysiological study (see Figure 2), sexual (psychological) stimulation (traditionally visual and/or audio) is presented under controlled conditions to the subjects and their penile and other physiological parameters are measured. During the poststimulus period, a variety of self-report measures are obtained from the subject regarding his assessment of the previous sexual experience. Because the study takes place within a laboratory setting, the experimenter can control the nature of the stimuli, remove or measure extraneous variables, monitor patterns of genital response, and obtain direct measurement of other aspects of the individual’s sexual response, including subjectively assessed arousal and concomitant cognitive and emotional responses. In carrying out a behavioral assessment of PE in this manner, the experimenter has greater potential to delineate relationships between psychological and physiological sexual processes, free of bias introduced by anamnestic reports and perceptual distortions emanating from the dysfunctional condition itself (see Rowland & Slob, 1995).

Psychophysiological Research on Premature Ejaculation

Initial psychophysiological studies on PE provided preliminary assessments of psychological processes purportedly involved in ejaculation latency. For example, contrary to the expectation of hyperarousal in PE men, several researchers (Kockott et al., 1980; Speiss et al., 1984) reported no differences between PE men and controls on genital and subjective arousal to erotic visual stimuli. More recently, Strassberg et al. (1987) have demonstrated that self-assessments of sexual arousal were not less accurate for PE men than controls, refuting the idea that these men misjudge their level of sexual excitation. Although PE men exhibited shorter latencies to ejaculation than controls in a laboratory setting, such differences were not related to interpersonal stress or self-reported anxiety about the sexual situation (Strassberg et al., 1990). Although such findings may argue against a role for arousal/anxiety constructs in the etiology and/or maintenance of PE, before discarding these hypotheses entirely, it should be determined whether these researchers actually provided an adequate test of the relevant hypotheses. For example, in some studies (Kockott et al., 1980; Spiess et al., 1984), only visual erotic stimulation was used, whereas a more appropriate test might have included penile tactile stimulation (e.g., Strassberg et al., 1990). In addition, men did not reach high levels of sexual arousal in two of these studies: They neither ejaculated in the laboratory nor were queried about any aspects of their ejaculatory threshold during sexual arousal.

In our own experimental study of PE (Rowland, Cooper, Slob, & Houtsmuller, 1997), we began by developing a protocol whereby (a) ejaculatory response could be studied under laboratory conditions, and (b) the response of PE men could be distinguished from that of functional controls. In developing such a methodology, the relevant stimuli and response parameters could be identified and implemented, and other (nongenital) aspects of the sexual response (affective and cognitive) could be monitored. Furthermore, once the response of the dysfunctional group has been reliably discriminated from functional men in the laboratory, a controlled investigation of the dysfunction is possible. Specifically, the parameters surrounding the dysfunction could be systematically manipulated by introducing new variables (biological or psychological) into the situation that might further influence either by inhibiting or facilitating-various aspects of the ejaculatory response.

Laboratory Study and Analysis of Premature Ejaculation

As mentioned earlier, previous psychophysiological studies of PE may have been limited by the lack of inclusion of stimuli and response measures specific to the ejaculatory response. To address these shortcomings, we included in our study of PE (Rowland, Cooper, Slob, & Houtsmuller, 1997), in addition to visual erotica, penile tactile stimulation and subjective and objective assessments of pending or actual ejaculation. In our analysis, the effects of two independent variables, dysfunctional status (PE men, PE men with coexisting ED, and functional men) and type of sexual stimulation (erotic visual stimulation, or erotic visual stimulation combined with penile vibrotactile stimulation) were assessed on erectile and ejaculatory response.

The results revealed several novel characteristics of the sexual response of men with PE. Prior to the psychophysiological session, all subjects selected an intensity of vibrotactile penile stimulation that they found “most pleasant.” This intensity level was then used in conjunction with visual erotica during the psychophysiological lab session. One might anticipate a preference for lower stimulus intensity among PE men compared with controls, if such men do in fact have penile hypersensitivity. Interestingly, the intensity level chosen by PE men did not differ from that of functional controls, indicating that maximal pleasure is derived from about the same tactile intensity for both groups. This general finding, however, needs to be examined in men with different subtypes of PE separately, a procedure that was not implemented in our analysis.

From the laboratory session itself, two notable patterns in genital response were evident. First, in examining maximum penile response over the session (see Figure 3a), penile stimulation had greater impact in men with PE than functional controls. Specifically, functional controls responded with equivalent penile response, whether or not penile stimulation was included. In contrast, PE men (both men with just PE and men with coexisting ED) responded with stronger genital response when vibrotactile stimulation was added to erotic visual stimulation. This finding verifies a long-standing supposition regarding the particular relevance of penile stimulation to overall arousal in PE men. Furthermore, it appears consistent with the findings of Fanciulla et al. (1988) that sensory stimulation of the penis has greater cortical representation (electrophysiologically), and perhaps meaning, in PE men than controls. Second, under the stronger stimulus condition of both visual and penile stimulation, maximum penile response of both PE and control groups was about the same, but penile response averaged over the entire session was much less in the PE than the control group (see Figure 3b). Thus, although PE men showed a maximum level of genital response similar to functional controls, their average penile response across the session was only about half that of the comparison group. This unexpected finding was related to the PE men’s greater fluctuation in penile response, with less time at maximum than controls, over the stimulus session. The wavering penile response in these men is strikingly similar to the fluctuating penograms reported in a subgroup of PE men by Kim et al. (1992). Whether such variation in response results from coexisting erectile insufficiency in some men, or whether it signifies intentional strategies on the part of these men to actively reduce arousal when it becomes intense, has yet to be determined. Actual ejaculation, as well as self-assessed measures of impending ejaculation, differed across PE and control groups. Among men with primary PE, about 50% ejaculated during the combined visual and penile stimulation. Among men with PE and coexisting ED, about 42% ejaculated. In contrast, only about 8-12% of functional controls ejaculated. Not surprisingly, we found self-assessed “closeness to ejaculation” increased in both PE and functional men when penile stimulation was added to the visual stimulation (compared with visual erotic stimulation alone). However, when subjects were asked about the control they felt over their ejaculation, functional men showed no change between the erotic video alone versus combined stimulation of erotic video and vibrotactile stimulation, whereas PE men reported a significant decrease in control during the combined visual and penile stimulation (see Figure 3c). Exactly what factors contribute to this sense of decreasing control in men with PE (e.g., lack of adequate sensory feedback, although some have argued that this is not an issue in men with PE [Kockott et al., 1980; Spiess et al., 1984; Strassberg et al., 1987]; and/or a lack of efficacy of cognitive/attentional strategies in delaying ejaculation) is not clear and continues to be the subject of ongoing investigation in our laboratory. Interestingly, when the PE group was further divided into those with primary PE and those with coexisting ED, we found that the latter group reported greater loss of control over ejaculation than men with PE alone. Although the sample sizes were small in this analysis, this pattern calls into question the assumption that men with coexisting PE and ED may ejaculate rapidly because they fear losing their erection (e.g., Buvat et al., 1990; Kaplan, 1974).

Affective Response to Sexual Stimulation in Men With Premature Ejaculation

An important aspect of the psychophysiological approach is that of integrating physiological and psychological responses. Thus, as part of our laboratory investigation of premature ejaculation, we have been interested in expanding the response repertoire to include more than genital response and simple subjective assessment of sexual arousal. Accordingly, after each sexual stimulus presentation, subjects were asked to rate how they felt on a 30-item adjective checklist that conveyed arousal, positive affect, and negative affect (see Table 1). By assessing affective response at this point within the experimental session, we were assessing the subject’s emotional response to the sexual experience in the laboratory rather than experimentally manipulating the subject’s emotional state, a process which itself is known to affect sexual arousal (Rowland, 1996a).

Cluster analysis was undertaken to pool items to which subjects responded consistently in the same way. Independent of the erotic stimulus conditions, items generally fell into two major clusters: The first consisted of items conveying positive emotional states and states of sexual arousal, whereas the second was comprised of items indicating negative emotion. This pattern of clustering was apparent in functional men, men with ED, and men with PE. In contrast, men with coexisting PE and ED (PE+ED) exhibited a different pattern of item clustering such that several items assessing sexual arousal grouped with items of negative emotion (Table 1). Thus, in most men, a positive emotional state was experienced as a component of sexual arousal, or may have been elicited by the sexual arousal. However, men with coexisting PE and ED showed dissociation between sexual arousal and positive emotion, perhaps because these men, once they became aroused, were concerned that their erections might not last. Alternatively, these men may have been venting the negative feelings (e.g., frustration) that result from high subjective sexual arousal in the absence of strong genital response (caused by erectile problems), further exacerbated by the tendency to terminate the sexual experience prematurely with ejaculation. Whatever the explanation, such affective patterns warrant further examination as they afford insight into potential differences among PE subtypes.

As reliability coefficients for the two item clusters, referred to as positive arousal and negative emotion, were generally high, a composite score for each of the clusters was generated for each man by calculating the mean of the items comprising the scale. As dependent measures to sexual stimulation, these psychoaffective scales revealed marked differences between the PE and PE+ED groups. Specifically, direct penile stimulation, whether used alone or in conjunction with visual sexual stimulation, augmented positive arousal in the PE+ED group, but not in the PE group, suggesting that these groups differ in their response to, and perhaps need for, tactile stimulation to the penis. In contrast, the PE alone group exhibited low positive emotion to penile stimulation alone, a condition that might resemble the physical stimulation associated with rapid ejaculation during sexual interactions with a partner.

The second psychoaffective construct, negative emotion, was not endorsed strongly by any group; nevertheless, the PE and PE+ED groups also diverged on this dimension. PE-alone men showed higher negativity to all stimulus conditions than either PE+ED men or functional controls. Thus, perhaps more so than men with any other dysfunction, men with primary PE may be most negatively disposed to the sexual situation. Although we did not go so far as to identify the particular negative emotion that was most salient (e.g., embarrassment, anxiety), the higher negative emotion in our PE sample is consistent with previous theorizing that PE men are more anxious than controls or ED men (Cooper, 1969; Munjack, Kanno, & Oziel, 1978). Such findings emphasize the value of assessing nongenital aspects of the sexual response, and further, suggest different treatment approaches for these two groups of PE men. For example, for men with PE alone, cognitive-behavioral techniques aimed at reducing hyperresponsivity to tactile stimulation may be combined with an attempt to reduce negative emotion and anxiety. For men with PE+ED, realignment of positive emotional responses with arousal, combined with sexual pleasuring exercises and stronger genital stimulation, might yield better results.

The Direction of Psychophysiological Research on Premature Ejaculation

Due to time and cost, psychophysiological assessments such as those described are unlikely to comprise a standard clinical strategy in the diagnosis of PE (Ruff & St. Lawrence, 1985). By using this approach, however, we have been able to underscore basic differences between PE men and functional controls. Our preliminary analysis lends support to waning theoretical perspectives (e.g., higher anxiety or negative emotional response in PE men); it corroborates existing suppositions and findings about the response of PE men (e.g., importance of penile stimulation), yet questions a number of previously offered explanations (e.g., PE men with ED ejaculate quickly out of fear of losing their erections). It has also provided fertile ground for future research on ejaculatory dysfunctions (e.g., penile response patterns, psychoaffective response). Furthermore, this methodology will enable sexologists to revisit questions regarding the relationship between arousal/anxiety and ejaculatory response that, due to previous limitations in methodology, may not have been addressed adequately. In general, psychophysiological analysis provides an avenue for the systematic study of novel variables, biological or psychological, that might influence ejaculation latency. Regarding this last point, for example, we (Haensel, Rowland, Kallan, & Slob, 1996) and others (R. C. Rosen, personal communication, June, 1996) are now successfully using this methodology to evaluate the effect of pharmacological agents on ejaculation latency in PE men and controls.

Psychophysiological Analysis of Treatment Strategies

Causes and Treatments of Premature Ejaculation-Not Necessarily Related

There are at least as many treatments for PE as there are theories about its etiology, with each theoretical perspective spawning its own therapeutic approach. Yet to assert a causal link between theory and treatment, even though the treatment may be effective, is not necessarily justified. It is often possible to correct a problem without treating its cause, a lesson poignantly demonstrated by the behaviorists in the 1960s and 1970s, and a strategy that typifies the current biomedical approach to the treatment of many behavioral and physical problems ranging from severe depression (e.g., antidepressant agents) to erectile dysfunction (e.g., penile prostheses, injection of vasoactive substances). A similar perspective can undoubtedly be applied to the treatment of PE-any strategy that increases ejaculation latency and/or perceived ejaculatory control, whether biological or psychological, is likely to contribute to the alleviation of the dysfunction, independent of its relationship to the cause. Thus, it is not surprising that pharmacological agents that delay ejaculation through local autonomic action on genital response may be used with some degree of success for PE with varied etiologies.

In some instances, effective therapeutic approaches, devoid of any a priori rationale, may actually engender a post hoc theoretical explanation for the dysfunction. Such thinking is exemplified by suggestions that because monoamine-altering pharmacological agents are effective in the treatment of PE, then the cause of the dysfunction is likely to be biological, and more specifically, related to monoamine imbalance (Assalian, 1994). Or, applying such logic to a finer level of analysis-because topical anesthetics which dull penile sensitivity delay ejaculation, the cause of the PE must be related to penile hypersensitivity (Xin et al., 1996). Such conclusions are not logically supported, and so without evidence, remain little more than speculation.

What is obvious is that various treatment procedures have been tried, and claims of success have ensued. These therapeutic procedures include psychoanalytic and psychosocial approaches that address problems within the sexual dyad (Levine, 1992; McCarthy, 1994); cognitive/behavioral approaches that modify physiological response through cognitive control, penile manipulation (squeeze technique: Masters & Johnson, 1970) and positional adjustments or changes in the mechanics of intercourse (Kaplan, 1989; LaPera & Nicastro, 1996; Zilbergeld, 1992); and pharmacotherapy used to inhibit local reaction to stimulation (e.g., topical penile anesthetics: Berkovitch et al., 1995) or to depress ejaculatory function through psychotropic drugs (see Balon, 1996, for a review). Combinations of these approaches have also been employed with moderate success, with most sexologists recognizing the need to address PE at multiple levels: relational, behavioral, and physiological (e.g., Levine, 1992; Rosen, 1995). As with the treatment of any dysfunction, therapy tailored to meet the multiple needs of the individual is likely to be most successful (Althof, 1995).

The Need to Assess Treatment Strategies

A number of critical pieces of information, however, are lacking from current treatment scenarios. These include assessment of (a) how effective the treatment is, (b) for whom it works best, (c) how it works, and (d) whether there are other consequences (positive or negative) that occur as the result of the PE treatment. Considering the first of these concerns, few well-controlled studies have been conducted in which the quantitative and qualitative effects of treatment have been carefully assessed. Most researchers have used simple preand posttreatment designs without control groups. Those studies using more sophisticated designs are often pharmacological-not surprising in view of the fact that most pharmacological effects are readily reversible, thereby enabling use of controlled trials that incorporate placebos, double blinding, and treatment crossover. Many psychological treatments, especially those that alter sexual relationship dynamics, are not reversible, and therefore are not amenable to crossover designs; those treatments that can be implemented and subsequently withdrawn (e.g., squeeze technique) suffer from lack of subject-blind treatment conditions. Nevertheless, even well-known research strategies designed to overcome these limitations (e.g., multiple baseline measures prior to the inception of treatment) are seldom implemented in clinical research on PE.

Even less attention has been given to the characterization of the PE patient most likely to benefit from a particular treatment strategy, to the mechanisms through which treatment effects change (whether it is specific to the treatment or related to more general factors, such as relationship improvement), and whether other positive or negative consequences accrue from the PE treatment itself (e.g., changes in relationship dynamics resulting from pharmacological treatment of PE).

Psychophysiological Analysis of the Effectiveness of Treatment: An Illustration

Although multiple research strategies are available to address questions such as those above, the psychophysiological laboratory approach again provides a powerful tool for addressing these questions. This approach enables the researcher to assess treatment efficacy under controlled conditions with minimal interference from factors that might otherwise confound results (e.g., spontaneous recovery, the patient’s own perceptual biases). At the same time, other issues related to the treatment may be systematically examined, including the type of patient most likely to benefit, the possible mechanisms through which the treatment works, and whether other consequences ensue from the treatment.

Recently we (Haensel et al., 1996) incorporated a psychophysiological component into an investigation of the delaying effects of clomipramine on ejaculation in men with PE (Althof et al., 1995; Assalian, 1988; Segraves, Saran, Segraves, & Maguire, 1992). Using a double-blind, placebo-controlled crossover design, we confirmed under laboratory analysis that clomipramine (25 mg taken 12-24 hr prior to testing) does indeed retard ejaculation in men with PE, and further, that this effect was limited to men with primary PE. Those premature ejaculators having coexisting erectile dysfunction did not experience significant improvement under clomipramine. These findings were evident from both the daily diaries kept by patients and from the laboratory analysis of changes in ejaculatory response. However, laboratory measurements revealed several response patterns that were not apparent from data extracted from the patients’ daily records of sexual activities and response. Specifically, in response to erotic stimulation, all groups reported lower sexual arousal and estimated erectile response under clomipramine compared with placebo (Rowland, 1996b) (see Figure 4). In addition, under clomipramine their actual erectile response was slightly diminished, and their penile response was significantly lower during nocturnal penile tumescence (NPT) testing. These findings may explain not only why the drug affords no beneficial effects to men with coexisting ED (as these men’s sexual potency is further inhibited by the drug), but may also provide clues as to how drugs such as clomipramine exert their effect on ejaculation, a process that until now has been largely a matter of speculation. Part of clomipramine’s effect may be mediated through diminished subjective arousal and hence lower erectile response, processes which increase the distance to the ejaculatory threshold. Such effects would not be surprising given that many agents that retard ejaculation are primarily used as antidepressants and thus are known for their ability to moderate emotional response and arousal. If, indeed, clomipramine affects ejaculatory latency by attenuating arousal (rather than or besides through some local autonomic or somatic effect in the genital region), then the use of complementary strategies (e.g., cognitive/affective) that achieve the same endpoint of lowered arousal might be worthy of consideration. One might find that, given the lack of specificity of most pharmacological agents and the increasing concern regarding side effects of such psychotropic drug use, psychological strategies aimed at reducing arousal combined with occasional “on-demand” use of pharmacological agents that ensure immediate results might prove more flexible and specific, and thus have greater potential for success in the long run.

Questions regarding other positive or negative effects resulting from the treatment of PE have essentially been ignored by researchers examining this dysfunction. In our analysis, we began preliminary investigation of such effects, having noted several changes in self-reported affective response resulting from pharmacotherapy. Under clomipramine treatment, responses on a presession questionnaire indicated significant improvement in the PE man’s relationship with his sexual partner. During the laboratory session itself, correlated with increased latency to ejaculation under clomipramine were increases in both positive affect and sensuality during penile stimulation. Interestingly, overall feelings of negative emotion remained unchanged by clomipramine treatment. Such findings suggest that positive emotional response within the sexual situation may be tied closely to perceived successful performance (delay of ejaculation), and further, that it may change as a consequence of improved performance. In contrast negative emotional response in PE men may be only weakly related to sexual performance-the higher negative emotion found in PE men in our study may reflect a persistent trait in these subjects rather than a “state” response to sexually stressful situations, a concept consistent with findings from previous studies indicating high anxiety in men with primary PE. Clearly, the way in which changes in sexual response interact with affective reactions has significant implications for treatment strategies for men with PE.

Concluding Remarks

Psychophysiological analysis has only just begun to delve into the many issues surrounding premature ejaculation; its potential for in situ assessment of factors that impact ejaculatory latency and control, orgasmic quality, and the mechanisms underlying treatment strategies, remains largely untapped. There is, for example, a need to continue basic research that identifies differences in the physiological, behavioral, and psychological responses of PE and functional men. At the same time, there is ample cause to investigate more specific questions, such as whether physiological response characteristics that distinguish PE men from functional controls (e.g., stronger somatosensory evoked potentials or shorter bulbocavernosal reflex latencies) undergo change following various biological or psychological treatments.

Although it is evident from the preceding discussion that the psychophysiological approach to the study of PE has much to offer, it is not without its limitations. Methodologically, it is time-intensive for participants, and therefore recruitment of volunteers may be difficult and subject to self-selection. And it is labor-intensive for researchers conducting the study, often resulting in small sample studies that reduce statistical power. Conceptually, the issue of external validity remains to be adequately addressed. We have found that data obtained through psychophysiological laboratory analysis tend to parallel those obtained through patient’s daily diaries (Haensel et al., 1996). Nevertheless, the important issue of whether results obtained from solitary sexual experiences under laboratory conditions advance the understanding of premature ejaculation within the context of an actual sexual encounter, where the consequence of unsuccessful performance is substantially augmented, needs further clarification. Finally, relationship and sociocultural factors, both critical to healthy sexual response, defy quantitative laboratory analysis. The evaluation of their contributions to PE may best be left to social scientists whose methodological tools are more suited to analysis of historical and other endogenous factors that resist scientific control.


The authors express gratitude to Donald Strassberg, Ph.D., and Forrest Vance, Ph.D., for their thoughtful comments on the manuscript.


AHLENIUS, S., LARSSON, K., SVENSSON, L., HJORTH, S., & CARLSSON, A. (1981). Effects of a new type of 5-HT receptor agonist on male rat sexual behavior. Pharmacology, Biochemistry, & Behavior, 15, 785-792.

ALTHOF, S. E. (1995). Pharmacologic treatment of rapid ejaculation. Psychiatric Clinics of North America, 18, 85-94.

ALTHOF, S. E., LEVINE, S. B., CORTY, E. W., RISEN, C. B., STERN, E. B., & KURIT, D. M. (1995). A double blind crossover trial of clomipramine for rapid ejaculation in 15 couples. Journal of Clinical Psychiatry, 56, 402-407.

AMERICAN PSYCHIATRIC ASSOCIATION. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. ASSALIAN, P. (1988). Clomipramine in the treatment of premature ejaculation. The Journal of Sex Research, 24, 213-215.

ASsN, P. (1994). Premature ejaculation: Is it really psychogenic? Journal of Sex Education & Therapy, 20, 1-4.

BALON, R. (1996). Antidepressants in the treatment of premature ejaculation. Journal of Sex & Marital Therapy, 22, 85-96.

BANCROFT, J. (1989). Human sexuality and its problems. Edinburgh: Churchill Livingstone.

BARLOW, D. H. (1986). Causes of sexual dysfunction: The role of anxiety and cognitive interference. Journal of Consulting and Clinical Psychology, 54, 140-148. BECK, J. G., BARLOW, D., SAKHEIM, D. K., & ABRAHAMSON, D. J. (1987). Shock threat and sexual arousal: The role of selective attention, thought content, and affective states. Psychophysiology, 24, 165-172.

BERKOVITCH, M., KERESTECI, A. G., & KOREN, G. (1995). Efficacy of prilocaine-lidocaine cream in the treatment of premature ejaculation. Journal of Urology, 154, 13601361.

BIESTER, R. J., & HOWARDS, S. S. (1988). Failure of seminal emission and retrograde ejaculation. In E. Tanagho, T. Lue, & R. McClure (Eds.), Contemporary management of impotence and infertility (pp. 285-290). Baltimore: Williams & Wilkins.

BRINDLEY, G. S. (1983). Physiology of erection and management of paraplegic infertility. In T. B. Hargreave (Ed.), Male infertility (pp. 261-279). New York: Springer-Verlag.

BURRIS, A. S., GRACELY R., CARTER, S., SHERINS, J., & DAVIDSON, J. M. (1991). Testosterone therapy is associated with reduced tactile sensitivity in human males. Hormones & Behavior, 25,195-205.

BWAT, J., BWAT-HERBAUT, M., LEMAIRE, A., MARCOLIN, G., & QUITTELIER, E. (1990). Recent developments in the clinical assessment and diagnosis of erectile dysfunction. Annual Review of Sex Research, 1, 265-308.

CARMICHAEL, M., HUMBERT, R., DEEN, J., PALMISANO, G., GREENLEAF, W., & DAVIDSON, J. M. (1987). Oxytocin increase in human sexual response. Journal of Clinical Endocrinology and Metabolism, 64, 27-31.

CAVALLINI, G. (1995). Alpha-1 blockade pharmacotherapy in primitive psychogenic premature ejaculation resistant to psychotherapy. European Urology, 128, 126-130.

CARATI, C. J., CREED, K. E., & KEOGH, E. J. (1987). Autonomic control of penile erection in the dog. Journal of Physiology, 384, 525. COLPI, G. M., FANCIULLACCI, G., BERETTA, G., NEGRI L., & ZANOLLO, A. (1986). Evoked sacral potentials in subjects with true premature ejaculation. Andrologia, 18, 583-586.

COOPER, A. (1969). Clinical and therapeutic studies in premature ejaculation. Comprehensive Psychiatry, 10, 285-295.

COOPER, A. J., CERNOVKSY, Z. Z., & COLUSSI, K. (1993). Some clinical and psychometric characteristics of primary and secondary premature ejaculators. Journal of Sex & Marital Therapy, 19, 277-288.

CRANSTON-CUEBAS M., & BARLow, D. (1990). Cognitive and affective contributions to sexual functioning. Annual Review of Sex Research,119-161. DAVIDSON, J. M. (1980). The psychobiology of sexual experience. In J. M. Davidson, & R. J. Davidson (Eds.), The psychobiology of consciousness (pp. 271-332). New York: Plenum.

DEGRoAT, W. C., & BooTH, A. M. (1980). Physiology of male sexual function. Annals of Internal Medicine, 92, 329-331.

DEKKER, J., & EVERAERD, W. (1988). Attentional effects on sexual arousal. Psychophysiology, 25, 45-54.

ERTEEN, C., COLAKOGLU, Z., & ALTA, B. (1995). Hand and genital sympathetic skin potentials in flaccid and erectile penile states in normal potent men and patients with premature ejaculation. Journal of Urology, 153, 76-79.

FANCIULLACCI, F., COLPI, G. M., BERETTA, G., & ZANOLLO, A. (1988). Cortical evoked potentials in subjects with true premature ejaculation. Andrologia, 20, 326-330. FERNANDEZ-GUASTI, A., ESCALANTE, A., AHLENIUS, S., HILLEGAART, V., Hz LARSSON, K. (1992). Stimulation of 5-HT1A and 5-HT1B receptors in brain regions and its effects on male rat sexual behavior. European Journal of Pharmacology, 210, 121-129.

GOSPODINOFF M. L. (1989). Premature ejaculation: Clinical subgroups and etiology. Journal of Sex & Marital Therapy, 15, 130-134.

GRENIER, M. A., & BYERS, S. (1995). Rapid ejaculation: A review of conceptual, etiological, and treatment issues. Archives of Sexual Behavior, 24, 447-472.

GRENIER, M. A., & BYERS, S. (1997). The relationships among ejaculatory control, ejaculatory latency, and attempts to prolong heterosexual intercourse. Archives of Sexual Behavior, 26, 27-48.

HAENSEL, S., MOS, J., OLIVER, B., & SLOB, A. K. (1991). Sex behavior of male and female Wistar rats affected by the serotonin agonist 8-OH-DPAT. Pharmacology, Biochemistry, & Behavior, 40, 221-228.

HAENSEL, S. M., ROWLAND, D. L., KALL,AN, K., & SLOB, A. K. (1996). Clomipramine and sexual function in men with premature ejaculation and controls. Journal of Urology, 156, 1310-1315.

Hnr.ATA, Z., & MUNGER, B. L. (1986). The neuroanatomical basis for the protopathic sensibility of the human glans penis. Brain Research, 371, 205-230. HALE, V. E., & STRASSBERG, D. S. (1990). The role of anxiety on sexual arousal. Archives of Sexual Behavior, 19, 569-581.

HAWTON, K. (1992). Sex therapy research: Has it withered on the vine? Annual Review of Sex Research, 3, 49-72.

KAPLAN, H. S. (1974). The new sex therapy. New York: Brunner/Mazel. KAPLAN, H. S. (1989). How to overcome premature ejaculation. New York: Brunner/Mazel.

KEDIA, K. (1983). Ejaculation and emission: Normal physiology, dysfunction, and therapy. In R. Krane, M. Siroky, & I. Goldstein (Eds.), Male sexual dysfunction (pp. 3754). Boston: Little, Brown.

KILMANN, P. R., & AUERBACH R. (1979). Treatment of premature ejaculation and psychogenic impotence: A critical review of the literature. Archives of Sexual Behavior 8, 81-100.

KIM, Y. C., CHOI, H. K., & LHO, Y. S. (1992). Penile hemodynamics in premature ejaculation. International Journal of Impotence Research, 4, 31-36.

KOCKOTT, G., FEIL, W., FERSTL, R., ALDENHOFF, J., & BESINGER, U. (1980). Psychophysiological aspects of male sexual inadequacy: Results of an experimental study. Archives of Sexual Behavior, 9, 477-493.

KUHR, C. S., HEIMAN, J., CARDENAS, D., BRADLEY, W., & BERGER, R. E. (1995). Premature emission after spinal cord injury. Journal of Urology, 153, 429-431. LAPERA, G., & NICASTRO, A. (1996). A new treatment for premature ejaculation: The rehabilitation of the pelvic floor. Journal of Sex & Marital Therapy, 22, 22-26. LAUMANN, E., GAGNON, J., MICHAEL, R., & MICHAELS, S. (1994). The social organization of sexuality: Sexual practices in the United States. Chicago: University of Chicago Press.

LEVINE, S. (1992). Helping men to control ejaculation. In S. Levine (Ed.), Sexual Life: A clinician’s guide (pp. 90-106). New York: Plenum. LOPICCOLO, J., & STOCK, W. (1986). Treatment of sexual dysfunction. Journal of Consulting and Clinical Psychology, 54, 158-167.

MADAKASIRA, S., & ST. LAWRENCE, J. (1997). Premature ejaculation: Assessment and treatment. Bailliere’s Clinical Psychiatry, 3, 91-112. MARSON, L., & MCKENNA, K. E. (1992). A role for 5-hydroxytryptamine in descend

ing inhibition of spinal sexual reflexes. Experimental Brain Research, 88, 313-320. MASTERS, W., & JOHNSON, V. (1970). Human sexual inadequacy. Boston: Little, Brown.

McCARTHY, B. (1988). Cognitive-behavioral strategies in the treatment of early ejaculation. In S. Leiblum, & R. Rosen (Eds.), Principles and practices of sex therapy: Update for the 1990’s (pp. 141-167). New York: Guilford.

McCARTHY, B. (1994). Etiology and treatment of early ejaculation. Journal of Sex Education & Therapy, 20, 5-6.

METZ, M., PRYOR, J., NESVACIL, L., AnuzzAHAB, F., & KozNAR, J. (1997). Premature ejaculation: A psychophysiological review. Journal of Sex & Marital Therapy, 23, 3-23.

Mos, J., OLIVIER, B., BLOETJES, K., & PoTH, M. (1990). Drug-induced facilitation of sexual behavior in the male rat: Behavioral and pharmacological integration. In A. K. Slob & M. Baum (Eds.), Psychoneuroendocrinology of growth and development: Proceedings of an international symposium (pp. 221-232). Rotterdam: Medicom.

MUNJACK, D. J., KANNO, P. H., & OZIEL, L. J. (1978). Ejaculatory disorders: Some psychometric data. Psychological Reports, 43, 783-787.

O,DONOHUE, W. T., & GEER, J. H. (1985). The habituation of sexual arousal. Archives of Sexual Behavior, 14, 233-246.

OVER, R., & KOUKOUNAS, E. (1995). Habituation of sexual arousal: Product and process. Annual Review of Sex Research, 6, 187-223.

ROSEN, R. C. (1995). A case of premature ejaculation: Too little, too late? In R. Rosen & S. Leiblum (Eds.), Case studies in sex therapy. New York: Guilford. RowLAND, D. L. (1996a). The psychobiology of sexual arousal and behavior. In L. Diamant & R. McAnulty (Eds.), The psychology of sexual orientation, behavior, and identity (pp. 19-42). Westport, CT: Greenwood Press.

ROWLAND, D. L. (1996b, June). Steps toward understanding premature ejaculation. Paper presented at the meeting of the International Academy of Sex Research, Rotterdam, The Netherlands.

ROWLAND, D. L., COOPER, S. E., HAENSEL, S. M., & SLOB, A. K. (1997). Re: Penile sensitivity in men with primary premature ejaculation [Letter to the editor]. Journal of Urology, 158, 187-188.

ROWLAND, D. L., COOPER, S., & SLOB, A. K. (1996). Genital and psychoaffective response to erotic stimulation in sexually functional and dysfunctional men. Journal of Abnormal Psychology, 105, 194-203.

ROWLAND, D. L., COOPER, S. E., SLOB, A. K., & HOUTSMULLER, E. J. (1997). The study of ejaculatory response in the psychophysiological laboratory. The Journal of Sex Research, 34, 161-166.

RowLAND, D. L., HAENSEL, S. M., BLOM, J., & SLOB, A. K. (1993). Penile sensitivity in men with premature ejaculation and erectile dysfunction. Journal of Sex & Marital Therapy, 19, 189-197.

ROWLAND, D. L., & HEIMAN, J. R. (1991). Self-reported and genital arousal changes in sexually dysfunctional men following a sex therapy program. Journal of Psychosomatic Research, 35, 1-11.

RoWLAND, D. L., & HOUTSMULLER, E. J. (in press). 8-OH-DPAT interacts with sexual experience and testosterone to affect ejaculatory response in rats. Pharmacology, Biochemistry, & Behavior.

ROWLAND, D. L., & SLOB, A. K. (1992). Vibrotactile stimulation enhances sexual arousal in sexually functional men: A study using concomitant measures of erection. Archives of Sexual Behavior, 21, 387-400.

ROWLAND, D. L., & SLOB, A. K (1995). Understanding and diagnosing sexual dysfunction: Recent progress through psychophysiological and psychophysical methods. Neuroscience and Biobehavioral Reviews, 19, 201-209.

RUFF, G. A., & ST. LAWRENCE, J. S. (1985). Premature ejaculation: Past research progress, future directions. Clinical Psychology Review, 5, 627-639.

SALZMAN, L. (1972). Psychotherapy with patients with sexual disorders. In M. Hammer (Ed.), The theory and practice of psychotherapy with specific disorders (pp. 273301). Springfield, IL: Charles C Thomas.

SEGRAVES, R. T., SARAN, A., SEGRAVES, K., & MAGUIRE, E. (1992). Clomipramine versus placebo in the treatment of premature ejaculation: A pilot study Journal of Sex & Marital Therapy, 19, 198-200.

SPECTOR, I. P., & CAREY, M. P. (1990). Incidence and prevalence of the sexual dysfunctions: A critical review of the empirical literature. Archives of Sexual Behavior, 19, 389-403.

SPIESS, W. F., GEER, J. H., & O’DONOHuE (1984). Premature ejaculation: Investigation of factors in ejaculatory latency. Journal of Abnormal Psychology, 93, 242-245. STRASSBERG, D. S. (1994). A physiologically based model of early ejaculation: A solution or a problem? Journal of Sex Education & Therapy, 20, 215-217.

STRASSBERG, D. S., KELLY, M. P., CARROLL, C., & KIRCHER, J. C. (1987). The psychophysiological nature of premature ejaculation. Archives of Sexual Behavior, 16, 327336.

STRASSBERG, D. S., MAHONEY, J. M., SCHAUGAARD, M., & HALE, V. E. (1990). The role of anxiety in premature ejaculation. Archives of Sexual Behavior, 19, 251-257. VANDEREYCKEN, W. (1986). Towards a better delineation of ejaculatory disorders. Acta Psyciatrica Belgie, 86, 57-63.

VIGNOLI, G. C. (1978). Premature ejaculation: New electrophysiologic approach. Urology, II, 81-82.

WILLIAMS, W. (1984). Secondary premature ejaculation. Australian and New Zealand Journal of Psychiatry, 18, 333-340.

WILLIAMS, W. (1985). Anesthetic ejaculation. Journal of Sex & Marital Therapy, 11, 19-29.

XIN, Z., CHOI, Y, SEONG, D., & CHoI, H. (1995). Sensory evoked potential and effect of SS-cream in premature ejaculation. Yonsei Medical Journal, 36, 397-401.

XIN, Z., CHUNG, W., CHOI, Y., SEONG, D., & CHOI, H. (1996). Penile sensitivity in patients with primary premature ejaculation. Journal of Urology, 156, 979-981. ZILBERGELD, B. (1992). The new male sexuality. New York: Bantam.

Copyright Society for the Scientific Study of Sex 1997

Provided by ProQuest Information and Learning Company. All rights Reserved