Neuroscience research fails to support claims that excessive pornography consumption causes brain damage

Dear Sir,

In their editorial on excessive pornography use, Hilton and Watts[3] offer some interesting neuroscience perspectives on their conceptualization of pornography problems as an addictive disorder. They highlight several parallels between dysregulated pornography consumption and other maladaptive behaviors, some of which are viewed as addictions. Although we believe these parallels are worthy of scientific inquiry, Hilton and Watts offered little, if any, convincing evidence to support their perspectives. Instead, excessive liberties and misleading interpretations of neuroscience research are used to assert that excessive pornography consumption causes brain damage. We wish to clarify what the research actually does suggest with several accompanying illustrations.

First, Hilton and Watts assert a “postulate” that “all addictions create, in addition to chemical changes in the brain, anatomical and pathological changes” which they state results in cerebral dysfunction. Depending on how addiction is defined, this is either well supported (e.g., brain atrophy arising from the neurotoxicity of alcohol) or wholly speculative as in the case of pornography consumption. A number of studies are cited in support of their position but the interpretation of the findings requires us to assume that cortical atrophy due to some type of excess (cocaine, obesity, or pedophilia) is universal and similarly distributed, and therefore the type of excess is irrelevant. Many of the studies cited merely compare groups on brain density scans in cross-sectional designs and inferences about causality cannot be made. For example, their citation of a 2007 study of pedophilia[11] that used correlational data is reported as evidence that “sexual compulsion can cause physical, anatomic change in the brain.” Even if such atrophy could be shown in relation to excessive pornography consumption, how much atrophy would actually be necessary before it would functionally impair (e.g., brain damage severe enough to cause behavioral dysfunction) a given individual? The notion that cerebral atrophy assessed through imaging is assumed to be synonymous with brain damage and therefore evidence of an addictive process is a perspective fraught with problems. For example, it is well established that cerebral atrophy occurs progressively as part of normal aging and if such a correlation is considered to be evidence of an addictive process then all of us are “addicted” to growing old. Illustrating a related concern, the imaging study of Miner and colleagues[5] cited by Hilton and Watts does little to support neuroscientific perspectives on “pornography addiction” given that the majority of the sexually compulsive patient sample had a history of alcohol abuse or dependence and no provisions were made to control for patients with adult ADHD. As a result, it is difficult to determine whether cortical differences and performance on measures of impulsivity in the study were related to hypersexuality, substance misuse, or other pathology already known to be associated with frontal deficits and executive control. Most importantly, the Miner study did not report that any of the subjects had problems specifically with excessive pornography use. Collectively, references to neuroimaging studies by Hilton and Watts are unsupportive of their assertion that excessive pornography consumption parallels other maladaptive behavioral patterns such as substance-related disorders or causes significant atrophy in the brain leading to behavioral dysfunction. Even the authors of these studies refrain from drawing such inferences. For example, Franklin et al, state “… this study cannot address the etiology of the structural abnormalities. The observed differences may be related to preexisting dysfunction, either environmentally or genetically determined, or a result of the effects of chronic cocaine assault.”[2]

Hilton and Watts seem intent on skewing findings from the studies they cite to support their perspectives rather than evaluating several plausible explanations for the various results reported by study investigators. For example, several explanations exist for the finding of lower density frontal matter in the 2006 study[6] on obese subjects including dysregulation of insulin or leptin resistance often found in obese individuals. It is also notable that even if the lower density in prefrontal matter of obese subjects, compared to healthy lean controls, was actually a result of atrophy (which this study was not designed to demonstrate), should it be interpreted as evidence demonstrating “visible damage in a natural endogenous addiction” as Hilton and Watts assert? They ignore the possibility that the grey matter differences and any possible frontal neurodegeneration could have predated obesity in the subjects or been an influence of the genetic or biological precipitating risk factors. Indeed, the most parsimonious explanation of the data cited is that frontal deficits may be a risk factor, i.e., preexisting and leading to the poor decision making and excessive indulgence characteristic of each clinical condition. This appears to be a preferred explanation of Schiffer et al,[11] who—contrary to Hilton and Watts‘ interpretation–hypothesize that early neurodevelopment leads to the brain differences, which serve as a risk factor for the pedophiles they studied.

We are open to the notion that frontal impairment might make people vulnerable to a variety of over-indulgences, which can subsequently lead to substance dependence, maladaptive coping patterns, poor judgment, impulsivity or emotional disturbance, which people may seek to escape by turning to problematic behaviors, such as the case with many pathological gamblers. However, given the lack of studies designed to infer causality, we find it difficult to readily assume the converse — that these diverse dysfunctional behaviors lead to common frontal dysregulation or any cortical atrophy worthy of mention. Admittedly, a causal mechanism strikes us as more likely when substances are involved (e.g., cocaine, high blood sugar, or high lipid levels damaging brain cells), but such causation is speculative for non-substance activities such as pornography use despite that likelihood that the sexual response cycle activated by pornography consumption also activates endogenous neurochemical reactions in the brain. If we consider that most people eat several times a day, are Hilton and Watson suggesting that the somewhat elevated activity of “eating behavior” is sufficiently different in obese persons to cause brain pathology? Similarly, would they argue that a “runner's high” from extensive exercise leads to brain damage? The parameters of what constitutes pattern, excess, cognitive reward, and the like need to be more clearly explicated and then studied within pornography users.

We are in agreement with Hilton and Watts that the study of executive deficits and frontostriatal systems in patients with dysregulated pornography use or hypersexual behavior is worthy of investigation. Using the proposed DSM-5 criteria for Hypersexual Disorder (HD), our research team has conducted two such studies that have yielded mixed findings. In one study, using neuropsychological self-report measures in a sample of hypersexual men (including those with excessive pornography problems), we found some evidence that executive deficits may exist in this population.[9] However, when actual performance was assessed on neuropsychological tests sensitive to frontal deficits common in executive dysfunction, no differences were found between hypersexual patients and healthy controls.[8] We interpreted these findings to support our theory that hypersexuality, including excessive pornography use, is a context specific phenomena which is expressed when triggered by a sexual cue or another stimuli, that when activated, is paired with sexual behavior (e.g., a learned behavior arising in response to dysphoric mood or stress such as been proposed in the current DSM-5 criteria for HD). Regardless, the current literature on excessive pornography use and hypersexuality diverge in many regards from those found in studies among patients seeking help for addictive disorders such as chemical dependency or among patients with impulse control problems such as pathological gamblers. Furthermore, our research on psychological profiles of hypersexual men, including those with pornography problems, failed to find evidence of elevations on addiction indices, but instead found characteristics common in populations with obsessive tendencies.[7] These findings suggest that hypersexual patients with pornography problems may represent a distinct population and grouping these patterns of behavior with other addictive disorders constitutes a premature conclusion that lacks empirical support.

Hilton and Watts perspectives on pornographic activation of dopaminergic transmission in mesolimbic pathways of the nucleus accumbens, prefrontal cortex, and other brain regions associated with the pleasure reward system does not offer meaningful insights given the variety of activities that engage this system. Watching the NCAA basketball play-offs will likely lead to similar neurochemical processes for many individuals. Some of us may even experience negative consequences in relation to viewing the play-offs and we may be willing to sacrifice important tasks in exchange for TV time. A few may even feel unable to resist the urge to view information online about the play-offs while at work despite possible violations of corporate policies about appropriate Internet use in the workplac. Are we to conclude that such patterns of behavior constitute an addictive disorder, given their potential relationship to activating dopaminergic transmission in the mesolimbic pathways? Alternatively, we prefer to clarify that substantial evidence suggests that dopamine release in these regions is not associated with a reward mechanism per se, but rather, it is part of an arousal process that alerts the brain to the presence of new or novel stimuli in the internal or external environment and such stimuli is not always associated with potential rewards.[10] Subsequently, any release of dopamine in these brain regions in response to pornography exposure may very well be due to the novelty of the pornographic stimuli and would likely occur for individuals naïve to erotic content as well as seasoned consumers of such material. Regardless, it does not provide readers with any evidence that excessive pornography use is an addictive disorder.

It was unclear to us, and perhaps some of your readers, why Hilton and Watts elected to reference literature about increased ΔFosB in the nucleus accumbens in copulating laboratory rats. These hypersexual rats were engaged in relational sexual activities with female partners, not in autoeroticism in response to sexually-provocative stimuli. Although the rodent study is interesting, we dispute the notion that it is analogous to humans excessively masturbating to pornography and thus the generalizations of the results cited by Hilton and Watts are questionable. Furthermore, the degree of ΔFosB induction in the nucleus accumbens in response to the natural rewards (e.g., sex) was significantly less than that observed in studies of drug rewards suggesting possible differences, not similarities, between drug addiction and sexual activity. Additionally, the significance of ΔFosB in the accumbens appeared to be limited in its effects where sexually naïve rats required fewer intromissions for ejaculation. Notably, cellular changes associated with increased ΔFosB are also found in cells exposed to a wide variety of stimuli unrelated to pleasure or reward behaviors. For example, stressors, sensory stimuli involved in learning, and evoked memory have been associated with such changes.[4] Given the fact that there are no human studies on ΔFosB in patients with excessive pornography problems and generalizing research from animal studies in order to provide evidence of biological parallels between addictive disorders and pornography problems is once again, speculative not scientific.

A final concern related to the perspectives of Hilton and Watts is the lack of clarity about what is meant by the term addiction. Our research team, along with others, have reported elsewhere[7–10] on various aspects of hypersexuality and excessive pornography consumption that diverge from commonly held ideas regarding persons addicted to substances.[11] The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR)[1] completely avoids the term, instead referring to Substance-Related Disorders. By design, problematic behavioral patterns are treated elsewhere in the DSM. In the forthcoming DSM-5, a new category of HD is under consideration for possible inclusion and our research team is currently conducting an independent DSM-5 field trial of the proposed HD criteria to determine the validity of the construct the whether it can reliably be diagnosed. Hilton and Watts statement on this matter may be somewhat misleading to your readers. They state that the forthcoming DSM-5 “contains in this new addition the diagnosis of HD, which includes problematic, compulsive pornography use.” As a point of clarification, the decision whether or not to include HD as a disorder has not yet been made, but it is worth noting that the definitions under study intentionally do not include allusions to addictions, compulsions, or obsessions. Thus, although for some it is common to speak of pornography addiction or other sexual addiction, the lack of convergence of findings has led a growing number to take a more modest and careful position, wherein the connections to other classes of excessive behavior patterns are still under study. Further, delineation of what constitutes an addiction has no agreed-upon standard. Thus, it becomes particularly problematic that Hilton and Watts made no effort to clarify what definition they use and why the term as they use it applies to the participants in the studies they cite such as references to obese subjects as having a “natural endogenous addiction,” even though subjects were screened to be free of psychiatric disorders, including eating disorders.

Despite our criticism of their work, we are encouraged that Hilton and Watts have made an attempt to bring increased awareness to patients exhibiting problems with excessive pornography consumption. We agree, and have published findings demonstrating, that such patterns of behavior have been associated with numerous negative consequences including attachment ruptures in romantic relationships, loss of employment, and psychological distress. Yet much remains to be learned about patients seeking help for hypersexual behavior and excessive pornography problems. Neuroscience has the potential to offer meaningful contributions to our understanding of this phenomenon but such research is lacking at this time. The tone and content of the Hilton and Watts article misleads readers to believe there is strong and convincing evidence based on neuroscientific research that excessive pornography problems constitute an addictive disorder causing brain abnormalities and cortical atrophy paralleling those found in substance abuse. Such assertions are speculative and unsupported by the studies cited by Hilton and Watts. Even if future research substantiates such claims, it is highly unlikely that such results will be generalized to all patients with excessive pornography problems given the consistent finding of heterogeneity in the characteristics of this population. We believe that addiction models may limit our understanding of this population and likely offer too simplistic a view of the vast array of complex issues encountered by patients with hypersexuality and pornography problems. In the interim, current research offers little support for conceptualizing excessive pornography problems as an addictive disorder. Research on tolerance or withdrawal, genetic associations, and neuroimaging in hypersexual patients with pornography problems are non-existent at this time. Although excessive pornography problems are part of the current proposed criteria for classification of HD in the forthcoming DSM-5, the field trial results have not been published and it is unclear whether such classification is valid or if it can be reliability diagnosed. Although the perspectives of Hilton and Watts may be appealing to some, we caution your readers in using their article to support or substantiate excessive pornography use as an addictive disorder based on the findings they attribute to neuroscience research. Collectively, their errors are egregious and detract from, rather than support, serious hypotheses for future research. In our own work with these patients, at least for those who seek treatment, the frequently attendant dysfunction in occupational, social, and other important activities, is sufficiently negative on its own, creating true dysfunction and significant clinical distress. We see no reason to exaggerate the known risks by suggesting that excessive pornography consumption leads to brain damage or other neuropathology. Admittedly, some are prone to dismiss pornography use of any kind as a natural outgrowth of human sexuality; however, those who study and work with these extreme cases are well aware of the difficulties encountered by these individuals, including their sense of frustration about the inability to reduce or stop their problematic behaviors despite negative consequences. We look forward to future work offering empirically derived perspectives on these conditions, including the associated neurological correlates, but preferably insights that remain within the scope of what the research data supports.