There is an increasing number of neuroimaging studies using visual sexual stimuli (VSS, Figure 1A). VSS are often used as pleasant, arousing stimuli that have an intrinsic positive value (see Wierzba et al., 2015). Brain reactivity triggered by VSS is often interpreted within popular theoretical frameworks describing learning processes or motivated behavior such as Reinforcement Learning (Sutton and Barto, 1998; Botvinick et al., 2009) or Incentive Salience Theory (Robinson and Berridge, 1993; Berridge, 2012). Importantly, these theories make a key distinction between conditioned stimuli (CS) and unconditioned stimuli (UCS), which are related to reward anticipation/wanting vs. reward consumption/liking, respectively. Accordingly, it is important to make explicit whether VSS play a role of CS or UCS, i.e., whether they are incentive cues predicting an upcoming reward, or whether they are rewarding by themselves. This issue has been surprisingly overlooked in past studies, despite its important implications. We reviewed 40 human studies published between 2013 and 2016, using VSS in combination with neuroscience methods (fMRI, EEG, ERP, PET, MEG or TMS; Figure 1B):

The Incentive Salience Theory framework, proposed by Robinson and Berridge (1993), distinguishes two basic components of motivated behavior—“wanting” and “liking”. The latter is directly linked to the experienced value of the reward (UCS), while the former is related to the expected value of the reward, often carried by a predictive cue (CS). Studies on substance and gambling addiction show that learned cues (CS) related to addiction evoke increased responses in the ventral striatum as well as increased motivated behavior (i.e., shorter reaction times, higher accuracy) among addicted individuals, while responses to the reward itself remain unchanged or undergo blunting over time (Berridge, 2012; Robinson et al., 2015).

Thus, the conceptualization of VSS as cues or rewards in experimental designs is not just a semantic debate, because it has important consequences for the interpretation of neuroimaging results. One important consequence is on the emerging field of neuroscientific research on compulsive sexual behaviors (CSB; Love et al., 2015; Kraus et al., 2016a,b; Figure 1). A central question in this field is whether CSBs (such as excessive pornography consumption Gola et al., 2016a,b) share common brain mechanisms with widely studied substance and behavioral addictions (Love et al., 2015; Gola and Potenza, 2016; Gola et al., 2016c; Kraus et al., 2016b). Depending on how VSS are conceptualized, different predictions can be formulated. If one assumes that VSS play a role of cue, then increased ventral striatal reactivity among subjects with CSB (in comparison with controls) would speak in favor of the addiction hypothesis, while under the assumption that VSS play a role of reward, it is the opposite result (decreased ventral striatal reactivity) that would speak in favor of the same hypothesis. Therefore we feel that it is important to address the question of whether VSS should be considered as cues (CS) or rewards (UCS) in human studies. Here we present our own perspective, hoping that it will initiate a scientific discussion on this topic.

To answer this question we think it is important to distinguish the meaning of VSS in real life vs. in the laboratory setting (Figure 2). In many real life situations, VSS such as the naked body of a sexually attractive partner increase sexual arousal and lead to approach behaviors initializing dyadic sexual activity and ending with orgasm (Georgiadis and Kringelbach, 2012; Gola et al., 2015a). In this case, we argue that VSS play a role of cue (CS), while orgasm plays the role of (primary) reward (UCS). The reasoning is similar in most cases of solitary sexual activity. Most common VSS are pornographic videos or photos (cue/CS), which increase sexual arousal, and lead to masturbation ending with orgasm (reward/UCS). In contrast, during laboratory experiments, subjects are usually not allowed to initiate any sexual activity (such as masturbation) and natural UCS—orgasm—is unavailable. Even if subjects would be allowed to masturbate during the study, laboratory conditions are far less comfortable than the usual context of pornography consumption or dyadic sexual activity. Thus, individuals participating in laboratory experiments do not expect any other reward than being exposed to VSS. Therefore, we posit that in laboratory setting VSS play a role of reward (UCS; Figure 2). The conceptualization of VSS as rewards in the context of laboratory experiments comes with several predictions. Among healthy subjects we should observe: (1) experience of pleasure while watching VSS, possibly accompanied by genital reaction; (2) reward-related brain activity correlated with these pleasurable feelings in response to VSS; (3) a willingness to exert effort to view VSS similarly as for other rewarding stimuli such as money; and (4) conditioning of cues (CS) predictive of VSS. Below we review evidence supporting these predictions.

In studies collecting hedonic ratings of VSS, subjects consistently report that watching VSS is a subjectively pleasurable experience when these match the subjects’ sexual preferences (Chivers and Bailey, 2005; Rupp and Wallen, 2009; Jacob et al., 2011; Wierzba et al., 2015). In addition, these hedonic ratings have been shown to be accompanied by genital reactions as measured by penile plethysmography in male participants (Stoléru et al., 1999; Redouté et al., 2000; Ferretti et al., 2005). Erectile reaction among males takes some time so it is easier to observe it with long-lasting VSS such as videos or long presentations of pictures (Ferretti et al., 2005), however even brief presentations of static sexual photographies are related with subjective pleasure and arousal (Ferretti et al., 2005; Wierzba et al., 2015).

Many studies have shown that passive VSS viewing evokes ventral striatum activity (Arnow et al., 2002; Stark et al., 2005; Sabatinelli et al., 2007; Demos et al., 2012; Georgiadis and Kringelbach, 2012; Stoléru et al., 2012; Wehrum-Osinsky et al., 2014). It is difficult to assess whether striatal activity reflects cue related wanting or reward related liking in these studies given that the ventral striatum is known to respond to both appetitive cues (CS) and rewards (UCS; Flagel et al., 2011; Liu et al., 2011) However, the observed correlation between striatal activity and hedonic ratings triggered by VSS in various studies (Walter et al., 2008; Sescousse et al., 2010, 2013b) favors the hypothesis that VSS act like rewarding stimuli. In this respect, VSS play a similar role as monetary rewards: they activate similar brain areas including the ventral striatum, and trigger comparable hedonic reactions and motivated behaviors (Sescousse et al., 2010, 2013b, 2015). The main difference is that VSS are primary rewards (i.e., they have an intrinsic and innate reward value), whereas money is a secondary reward (whose value is learned by exchange against other rewards). This difference leads to a partially different mapping onto the brain reward system, and different strengths of activation (Sescousse et al., 2010, 2013b, 2015).

Even though most studies using VSS have used passive viewing paradigms, a few investigations have employed more advanced experimental designs aiming to measure participants’ willingness to exert effort towards VSS. In a series of studies, we have used a modified version of the monetary incentive delay task (Knutson et al., 2001) to include VSS (Sescousse et al., 2010, 2013a, 2015; Gola et al., 2015b, 2016c). In this task subjects see two types of cues that are predictive of either VSS or monetary gains. These cues are followed by a discrimination task in which subjects have to press the correct button (out of two) within a time limit of 1 s. The receipt of a monetary gain or VSS is partly dependent on their performance on this task, such that reaction times can be interpreted as an indirect measure of the motivation to obtain these rewards. Importantly, cues predicting VSS elicit similar reaction times as those predicting monetary rewards, demonstrating that participants are willing to exert effort to view VSS, and that their motivation is similar for both rewards (Sescousse et al., 2010). This willingness to exert effort, which is a hallmark of reward (Thorndike, 1965), has been observed in other studies using effort (but also delay) discounting paradigms with VSS (Prévost et al., 2010). In addition, we have shown that individual differences in the effort exerted for money vs. VSS is strongly correlated with the relative brain activity evoked by corresponding cues in the ventral striatum (Sescousse et al., 2015; Gola et al., 2016c). This precise fine-tuning of brain activity and reaction times by VSS predicting cues further confirms that VSS have intrinsically rewarding properties.

Finally, recent studies have shown that abstract CS (such as colorful patterns or dots) associated with VSS maintain their incentive salience even when they are not predictive of VSS anymore (Banca et al., 2016; Klucken et al., 2016). In the study by Banca et al. (2016), abstract visual patterns acquired positive predictive value (CS+) or neutral predictive value (CS−) by being repetitively paired with VSS or neutral stimuli, respectively. In the following phase of the experiment, subjects had to make choices between those CS and novel abstract stimuli, while both CS were now paired with increased chances of monetary gains (but not VSS anymore). Despite both CS having equal chances of leading to monetary gains, CS+ were chosen more frequently than CS− on average (mostly by subjects with CSB), demonstrating the strong rewarding properties of VSS.

As we have shown above, there is a consistent body of evidence supporting our view that in laboratory settings VSS play a role of reward rather than cue. Moreover, even in everyday life VSS do not always play a role of cue for sexual activity and orgasm. Long before the development of photography people have liked art such as sculptures and paintings depicting nudity. Perhaps (similarly to modern times) this type of art was a source of pleasure rather than cue for sexual activity. In the era of photography people showed willingness to pay for pictures and videos with erotic and pornographic content, then internet technology provided everyone with easy and free access to a whole variety of VSS (Cooper, 1998). Perhaps most of contemporary VSS (such as internet pornography) play a role of cue for solitary or dyadic sexual activity, but in some cases VSS are sought after for themselves, again demonstrating their intrinsic rewarding value. A good example in everyday life are calendars with erotic pictures, that people buy and expose in their workplace or at home. Similarly, the popularity of strip clubs, in which people are willing to pay to watch nude dancers with whom they are not allowed to engage in sexual activity, illustrate the potency of VSS as hedonic stimuli (Figure 2).

Based on the above arguments, we argue that VSS play a role of reward—rather than cue—in most experimental setups in which sexual activity and climax experience are unavailable. As we outlined above, viewing VSS is a pleasurable experience that people are willing to work and wait for (Prévost et al., 2010), and activates the same brain reward regions as monetary gains (Sescousse et al., 2010, 2013a, 2015; Gola et al., 2015b, 2016c). Moreover neutral stimuli associated with VSS through Pavlovian conditioning acquire incentive value (Sescousse et al., 2010, 2013a, 2015; Banca et al., 2016; Gola et al., 2016c; Klucken et al., 2016). This conceptualization of VSS as rewards rather than cues calls for the re-examination—and possibly re-interpretation—of the results reported in earlier studies defining VSS as cues. Certainly it may have a strong impact on the interpretation of neuroimaging studies investigating neurobiological similarities between CSB and addiction; for instance, based on the popular Incentive Salience Theory framework, one would expect opposite ventral striatal reactivity for VSS depending on whether they are conceptualized as a cue or reward (as an example of such ambiguous interpretation see: Prause et al., 2015, 2016; see also Gola, 2016 for discussion). If in most of experimental setups, as we argue, VSS play a role of reward, then diminished (rather than increased) ventral striatal reactivity to VSS in individuals with problematic pornography use (Gola et al., 2016a) would speak in favor of the addiction hypothesis (Robinson et al., 2015). We would expect this to be accompanied by increased ventral striatum activations for CS that are predictive of VSS, as well as increased effort or shorter reaction times to gain access to these VSS. In future studies, we hope that the role played by VSS in the specific protocols that are used will receive increased attention, and that appropriate interpretations of results will be made accordingly.