Symptoms of cybersex addiction can be linked to both approaching and avoiding pornographic stimuli: results from an analog sample of regular cybersex users (2015)

Jan Snagowski¹ and Matthias Brand^1,2,*

There is no consensus regarding the phenomenology, classification, and diagnostic criteria of cybersex addiction. Some approaches point toward similarities to substance dependencies for which approach/avoidance tendencies are crucial mechanisms. Several researchers have argued that within an addiction-related decision situation, individuals might either show tendencies to approach or avoid addiction-related stimuli. In the current study 123 heterosexual males completed an Approach-Avoidance-Task (AAT; Rinck and Becker, 2007) modified with pornographic pictures. During the AAT participants either had to push pornographic stimuli away or pull them toward themselves with a joystick. Sensitivity toward sexual excitation, problematic sexual behavior, and tendencies toward cybersex addiction were assessed with questionnaires. Results showed that individuals with tendencies toward cybersex addiction tended to either approach or avoid pornographic stimuli. Additionally, moderated regression analyses revealed that individuals with high sexual excitation and problematic sexual behavior who showed high approach/avoidance tendencies, reported higher symptoms of cybersex addiction. Analogous to substance dependencies, results suggest that both approach and avoidance tendencies might play a role in cybersex addiction. Moreover, an interaction with sensitivity toward sexual excitation and problematic sexual behavior could have an accumulating effect on the severity of subjective complaints in everyday life due to cybersex use. The findings provide further empirical evidence for similarities between cybersex addiction and substance dependencies. Such similarities could be retraced to a comparable neural processing of cybersex- and drug-related cues.

Approach-Avoidance Tendencies in Substance Dependencies

According to Breiner et al. (1999), tendencies to approach or avoid addiction-related stimuli are connected with cue-reactivity and craving, which are frequently investigated in addiction literature (for review see Tiffany and Wray, 2012). Cue-reactivity represents subjective and physiological responses to addiction-related cues (Drummond, 2001). A consensual definition of craving is still missing (for review see Skinner and Aubin, 2010). Craving is mostly referred to as a subjectively experienced urge to consume a drug (Sayette et al., 2000), while other approaches argue to additionally assess non-subjective craving reactions with physiological measures of cue-reactivity (Ooteman et al., 2006) or behavioral inclinations toward drug use (Marlatt, 1985; Buydens-Branchey et al., 1997). Further, neurophysiological theories refer to adaptations in the mesolimbic dopaminergic pathway due to repeated drug use and argue that craving might also occur as an unconscious urge to consume a substance, which is called “wanting” (e.g., Robinson and Berridge, 1993, 2001, 2008). However, cue-reactivity and craving seem to be related concepts (Drummond, 2001), while there is enough evidence to neglect a one-dimensional definition of craving (Skinner and Aubin, 2010).

Striving for a differentiated definition of craving, Breiner et al. (1999) proposed a multi-dimensional model for alcohol dependency focusing on the role of an evaluative space in an addiction-related decision situation. The evaluative space can be divided into the states approach, avoidance, ambivalence, and indifference. Approach and avoidance are competing action-tendency states. Approach is supposed to cause alcohol consumption whereas avoidance represents an oppositional process in which the urge to consume alcohol is suppressed. Further, ambivalence and indifference can be described as ambiguous states, which might be entered if the inclinations of the action-tendency states are balanced. In this context, ambivalence represents a high and indifference a low intensity of ambiguity. Breiner et al. (1999) argue that the state entered in an addiction-related decision situation depends on either positive or negative expectancies toward drinking, which are further influenced by historical (e.g., psychological and physiological predispositions) as well as current factors (e.g., positive or negative incentives). Positive expectancies thereby promote the state approach, while negative expectancies are likely to cause avoidance. Regarding different aspects of craving, approach is synonymous with an irresistible “wanting” and thereby able to elicit an automatic response. Contrary, avoidance is supposed to be a subjectively experienced process. Consequently, the approach/avoidance framework is in line with dual-process models highlighting the role of automatic and controlled processes for the development and maintenance of addictive behaviors (e.g., Bechara, 2005; Wiers and Stacy, 2006). A simplified overview of the approach/avoidance framework by Breiner et al. (1999), which we have transferred to cybersex addiction, is summarized in Figure ​Figure11.

 

FIGURE 1

Simplified overview of the approach/avoidance framework by Breiner et al. (1999) adapted to cybersex addiction. Straight lines represent inclinations which may elicit tendencies promoting cybersex use while dashed lines rather embody tendencies to avoid …

Approach-Avoidance Tendencies in Cybersex Addiction

Based on the theoretical approach/avoidance framework by Breiner et al. (1999) and the suggested similarities between Internet addiction and substance dependencies it is plausible to assume comparable patterns in individuals with tendencies toward cybersex addiction. Regarding cue-reactivity and craving in cybersex addiction, studies already provided preliminary evidence for such similarities (Brand et al., 2011; Laier et al., 2013a). These studies indeed indicated that individuals with tendencies toward cybersex addiction showed both cue-reactivity and an increase of subjective craving when confronted with pornographic pictures. Further, sexual stimuli are known to induce neural activations which are similar to those induced by drug-related cues and are theoretically also able to promote adaptions in the mesolimbic dopaminergic pathway (Georgiadis and Kringelbach, 2012). Moreover, Laier and Brand (2014) recently proposed a theoretical framework for cybersex addiction which displays some similarities with the model by Breiner et al. (1999). For instance, the historical factors suggested by Breiner et al. (1999; e.g., person’s characteristics, past reinforcement, physiological reactivity) are in line with influences of specific predispositions toward sex as well as the suggested role of gratification proposed by Laier and Brand (2014). Further, Laier and Brand (2014) suggest a mediating role of cybersex use expectancies on the use of cybersex, which can be compared with the role of expectancies in the model by Breiner et al. (1999).

Regarding existing evidence for approach/avoidance tendencies in cybersex addiction, Schiebener et al. (2015) performed a study in which participants had to execute different tasks in a multitasking paradigm. These tasks were related to one of two picture sets, whereas the first picture set contained neutral and the second one contained pornographic pictures. Participants were instructed to execute all different tasks in equal measure, while they could switch autonomously between tasks and picture sets. The deviation from the optimal set balance was taken as dependent variable, indicating either a preference to work on the neutral or the pornographic set. Using this measure, the authors found a quadratic relationship between tendencies toward cybersex addiction and the deviation from set balance, which means that individuals with high tendencies toward cybersex addiction either preferred to work on the pornographic (approach) or on the neutral (avoidance) set. In contrast, participants with low tendencies toward cybersex addiction did not prefer to work more on one of the picture sets. Since the multitasking paradigm used by Schiebener et al. (2015) was not explicitly designed to measure tendencies to approach or avoid pornographic stimuli, it seems plausible to use a standard approach/avoidance paradigm in order to deeper investigate this phenomenon.

Measuring Approach/Avoidance Tendencies

One way to assess tendencies to approach or avoid addiction-related stimuli is the Stimulus-Response-Compatibility Task (SRC; Field et al., 2006). During the SRC a manikin figure has to be moved toward and away from addiction-related cues in two separated blocks by using a standard keyboard. The difference between the mean reaction times (RTs) recorded in the two blocks is thereby supposed to reflect the relative inclination to either approach or avoid addiction-related cues. Several studies using the SRC showed stronger tendencies to approach than to avoid addiction-related stimuli in smokers (Mogg et al., 2003), regular cannabis users (Field et al., 2006), as well as heavy alcohol and cannabis users (Field et al., 2008; Cousijn et al., 2012). Regarding relationships between subjective craving and tendencies to approach or avoid addiction-related stimuli, outcomes have been inconsistent regarding whether these relationships may be linear or quadratic (Field et al., 2005a,b; Schoenmakers et al., 2008). As an extension of the SRC, Rinck and Becker (2007) introduced the Approach-Avoidance-Task (AAT), which includes physical movement to enhance effects of approaching and avoiding pictorial stimuli. By using a joystick, participants have to pull stimuli presented on a computer screen toward themselves (approach) or push them away (avoidance) from themselves. Originally, the AAT was designed to investigate fear-related behaviors (Rinck and Becker, 2007). Later, since competing tendencies to approach or avoid addictive behaviors are supposed to be essential in addiction-related decision situations (Breiner et al., 1999), modified versions of the AAT were used in studies regarding smoking (Wiers et al., 2013), a heavy use of cannabis (Cousijn et al., 2011, 2013) and alcohol dependency (e.g., Wiers et al., 2009; Sharbanee et al., 2013, 2014). In this context, most experimental studies found linear relationships between addictive behaviors and the tendency to approach addiction-related stimuli. However, in accordance with the dual-process models of addiction (Bechara, 2005; Wiers and Stacy, 2006), there is also empirical evidence for the assumption that addicted individuals might also show tendencies to avoid addiction-related stimuli, e.g., as a consequence of computerized avoidance training programs (Wiers et al., 2011; Eberl et al., 2013a,b). Moreover, Spruyt et al. (2013) found that abstaining alcohol-dependent individuals had, compared to matched controls, avoidance tendencies in an SRC, while relapse rates were positively associated with the strength of avoidance tendencies.

Aims and Hypotheses

The aim of the current study is to investigate whether approach/avoidance tendencies might be mechanisms underlying cybersex addiction. While relying on the theoretical framework by Breiner et al. (1999) as well as the results provided by Schiebener et al. (2015), we expect to find that individuals with high tendencies toward cybersex addiction either show approach or avoidance tendencies toward pornographic stimuli. Additionally, low tendencies toward cybersex addiction should go along with balanced tendencies to approach or avoid pornographic stimuli. On an operationalized level, the relationship between approach/avoidance tendencies and cybersex addiction is supposed to be not linear but quadratic. Moreover, it is assumed that there will be neither a linear nor a quadratic relationship between tendencies to approach or avoid neutral stimuli and tendencies toward cybersex addiction. Further, since sensitivity toward sexual arousal as well as problematic sexual behavior were shown to promote the development and maintenance of cybersex addiction (Laier and Brand, 2014), we hypothesize that a combination of approach/avoidance tendencies toward pornographic pictures and a high problematic sexual behavior/sensitivity toward sexual excitation should have an accumulating effect on the severity of subjective complaints in everyday life due to the use of cybersex activities.

Participants

In the present study a total of 123 heterosexual male participants were examined (M_age = 23.79 years, SD = 5.10). The mean age of first cybersex use was 15.61 (SD = 4.01) years. On average, participants used cybersex sites 3.66 (SD = 3.52) times per week, while spending M_time = 22.25 (SD = 14.22) minutes per visit. Only participants of legal age (at least 18 years-old) were recruited. Recruitment was done through local advertisements at the University of Duisburg-Essen (Germany) and online platforms. It was stated in the advertisements that explicit pornographic material would be presented. Students could collect credits, non-student participants were paid €10 for participation. All participants gave written informed consent prior to the experiment and were debriefed at the end of the study. The study was approved by a local ethics committee.

Measures

Approach-Avoidance-Task

Participants performed a modified version of the AAT (Rinck and Becker, 2007), in which pictures presented on a computer screen either had to be pulled toward (approach) or pushed away (avoidance) from their body with a joystick. Every single trial had to be manually started by the participant by pressing a button on the joystick, while the joystick had to be in the default position. Following a 500 ms inter-trial interval (ITI), a pictorial cue was presented. Due to the joystick movement, an implemented zooming feature increased (pull-movement) or decreased (push-movement) the size of the cue. In accordance to Rinck and Becker (2007), the joystick had to be moved ∼30° in one direction in order to terminate the trial. Further, a logarithmic growth function was used to increase or decrease cue size in order to allow participants to experience changes in cue size as immediate reactions to their joystick movements. All cues had an initial size of 700 × 500 pixel and were presented on a 15.6 inch screen. Due to moving the joystick ∼30° to one direction, the cue size changed to a maximum of 2100 × 1500 pixel (pull-movement), respectively a minimum of 233 × 166 pixel (push-movement). At the end of each trial, another 500 ms ITI was presented. The participants’ RTs were recorded in each trial. Similar to previous studies, the stimuli were separated into addiction-related and neutral cues (Wiers et al., 2009, 2013; Cousijn et al., 2011). As neutral cues, 40 pictures of the International Affective Picture System (IAPS; Lang et al., 2008) were used. Pictures showed one or two persons in neutral situations. As addiction-related cues we used 40 pornographic pictures out of four categories, which Laier et al. (2013a) identified as being sexually arousing for heterosexual men (heterosexual intercourse as vaginal sex and fellatio, homosexual intercourse between two women in the form of tribadism and oral sex). Additionally, five neutral and five pornographic pictures, which were not taken for the experimental trials, were used in the practice trials. Overall, the AAT and the pornographic picture rating used different pornographic cues.

During instruction, participants completed 30 practice trials, which were separated into four rounds (push, pull, porn-push/neutral-pull, porn-pull/neutral-push). After each round, participants were informed about the amount of correct reactions and could decide to repeat the round. The experimental trials were divided into four blocks with 80 trials each, resulting in a total of 320 trials. Every stimulus was presented once during one block in a semi-random order (maximally three stimuli of the same category were allowed to appear in a row). Participants were randomly assigned to one of two experimental conditions, which differed with respect to the instruction in the first block (porn-push/neutral-pull or porn-pull/neutral-push). In the following blocks the instruction was inverted. The experimental condition was counterbalanced across participants. By separating the type of instruction (direct vs. indirect), previous studies used different versions of the AAT. Versions with direct instructions (e.g., Rinck and Becker, 2007) included two stimulus categories, while indirect AATs (e.g., Wiers et al., 2009) used more than two stimulus categories and instructed participants to push or pull the joystick dependent on the picture format (horizontal vs. vertical). Thus, indirect AATs represent task-irrelevant designs, while direct AATs embody task-relevant paradigms. In this study, a task-relevant AAT was used, since a meta-analysis by Phaf et al. (2014) could not provide evidence for an advantage of task-irrelevant versions.

To analyze AAT data, median RT scores were calculated since medians are less vulnerable with respect to RT outliers than mean scores (Palfai, 2006; Wiers et al., 2009; Cousijn et al., 2013). RTs < 200 ms, > 2000 ms as well as RTs from false responses were discarded. An error rate >25% led to a complete exclusion from the data analysis. For each participant a compatibility effect score (Rinck and Becker, 2007) for both the pornographic (pornographic approach/avoidance score) and the neutral (neutral approach/avoidance score) stimulus category was calculated by subtracting the median pull from the median push RT (median RT push – median RT pull). According to Rinck and Becker (2007, p. 110), the compatibility effect score represents the “relative strength of approach and avoidance tendencies” while positive values indicate approach (median RT push > median RT pull) and negative values avoidance (median RT push < median RT pull) tendencies. The basic idea of these scores is that compatible trials (e.g., approach pornographic pictures) lead to faster RTs compared to incompatible trials (e.g., avoid pornographic pictures). Moreover, the pornographic approach/avoidance score is the main dependent variable, while the neutral approach/avoidance score represents a control variable, since approaching and avoiding neutral stimuli should not be connected to other dependent variables such as tendencies toward cybersex addiction.

Additionally, an overall effect score (overall RT score) was calculated by subtracting the median RT for all neutral stimuli from the median RT for all pornographic stimuli (median RT porn – median RT neutral). While the movement direction in specific trials is not taken into account for this measure, negative values indicate that participants were faster to respond to pornographic stimuli (median RT porn < median RT neutral), while positive values point toward slower RTs for pornographic stimuli (median RT porn > median RT neutral). Thus, the overall RT score is equivalent to the assessment of indirect attentional biases in substance use disorders (Field and Cox, 2008; Coskunpinar and Cyders, 2013; Field et al., 2014) than to measuring approach/avoidance tendencies with respect to stimulus type (pornographic vs. neutral). In analogy to substance dependency research, positive values of the overall RT score indicate the existence of an attentional bias toward pornographic pictures (slower RTs to pornographic compared to neutral stimuli). A general overview of all dependent variables of the AAT is summarized in Table ​Table11. The AAT was programmed using Presentation®software (Version 16.5, www.neurobs.com).

 

Table 1

Calculation and interpretation of AAT scores.

Short- and Long-Term Measurements

The instruments used in the current study can be separated into short-term (pornographic picture rating, craving Δ sexual arousal/masturbation, AAT) and long-term measurements (s-IATsex, HBI, SES). In this context, short-term measurements refer to reactive (immediate) responses, which might be influenced by environmental factors, such as preceding cybersex consumption. In contrast, long-term measurements rather resemble individual characteristics, which are supposed to remain stable over a longer period of time.

Statistical Analyses

Data analysis was done using IBM, SPSS Statistics Version 22.0. Relationships between two variables were analyzed with Pearson correlations. Differences between two variables were evaluated with one sample t-tests. Effect sizes are reported according to Cohen (1988) using Pearson’s r (r = 0.10, small; r = 0.30, medium; r = 0.50, large) and Cohen’s d (d = 0.20, small; d = 0.50, medium; d = 0.80, large). Quadratic relationships between two variables were evaluated using curve-linear regression analyses. Further, interactions between two variables as predictors of a single dependent variable were analyzed with hierarchical moderated regression analyses (all predictors centralized; Cohen et al., 2003). The significance level for all statistical tests was p = 0.05. Further, in order to check whether variables violated the assumption of normality, skewness, and kurtosis are reported in Table ​Table22. According to West et al. (1995), skewness < | 2.00| and kurtosis < | 7.00| indicate that a variable is distributed normally. Here, all variables used for curve-linear and moderated regression analyses fulfilled these criteria (s-IATsex craving, HBI loss of control, SES, pornographic/neutral approach/avoidance score). However, in case other variables, which were used for further calculations, violated the assumption of normality, parametric tests were applied nonetheless, since it was shown that the parametric statistical methods are robust against this violation (Rasch and Guiard, 2004).

 

Table 2

Mean values of the s-IATsex, HBI, SES, pornographic picture rating and subjective ratings of sexual arousal as well as need to masturbate and AAT scores.

Pornographic Picture Rating

A one-sample t-test was calculated to compare the ratings for heterosexual and homosexual pictures, t(122) = 32.79; p < 0.001; d = 4.11, indicating that heterosexual pictures were rated as significantly more sexually arousing. Regarding the evaluation of sexual arousal and the need to masturbate before (t1) and after (t2) the pornographic picture rating, two t-tests for dependent samples revealed higher subjective sexual arousal, t(122) = -9.05; p = 0.001; d_z = 0.85, and a higher need to masturbate, t(122) = -7.30; p < 0.001; d_z = 0.61, at t2 compared to t1 (for mean values see Table ​Table22). These results indicate that due to watching pornographic pictures, participants experienced a state of being sexually aroused before starting the AAT. This is of particular importance since sexual arousal and the need to masturbate are operationalized as craving measures, which are supposed to be linked to tendencies to approach or avoid pornographic stimuli.

Approach-Avoidance-Task

Descriptively, the pornographic approach/avoidance score (M = -1.09, SD = 72.64) and the neutral approach/avoidance score (M = -56.91, SD = 55.03) had negative mean values. These results indicate a mean tendency for avoiding both pornographic and neutral stimuli in the AAT, while this effect was stronger for neutral stimuli, t(122) = 8.52; p < 0.001; d = 0.87. In contrast, the overall RT score (M = -37.79, SD = 42.74) had a negative mean value, which indicates that participants on average had no attentional bias toward pornographic stimuli (such an attentional bias would be reflected by slower RTs for pornographic pictures and therefore a positive mean overall RT, which is not the case, since we observed faster RTs for pornographic compared to neutral stimuli).

Correlations between the AAT scores and selected variables are summarized in Table ​Table33. Regarding the pornographic and the neutral approach/avoidance score there were no significant correlations with other measures. However, the overall RT score correlated significantly with sensitivity toward sexual excitation, the HBI loss of control scale as well as the craving Δ sexual arousal and the craving Δ need to masturbate score.

 

Table 3

Bivariate correlations between the AAT scores and selected variables.

Curve-Linear Regression Analysis

To test whether the relationship between the pornographic approach/avoidance score and the s-IATsex factor craving was not linear but quadratic, a curve-linear regression analysis was calculated. In a first step the pornographic approach/avoidance score was entered but did not significantly explain s-IATsex craving variance, R² = 0.003, F(1,122) = 0.33, p = 0.567, which indicates that no linear relationship between the two variables exists in the data. In a second step the squared pornographic approach/avoidance score was included, which led to a significant explanation of 23.7% of the s-IATsex craving variance, ΔR² = 0.234, F(1,122) = 18.80, p < 0.001. This estimated curve (see Figure ​Figure22) indicates that individuals with a high s-IATsex craving tended to either show approach (positive approach/avoidance values) or avoidance (negative approach/avoidance values) inclinations with respect to pornographic stimuli. Further regression values are summarized in Table ​Table44.

 

FIGURE 2

Relationship between the compatibility effect score for pornographic pictures (pornographic approach/avoidance score) and the s-IATsex factor craving.

 

Table 4

Values of the curve-linear regression analysis with the s-IATsex factor craving as dependent variable.

As a manipulation check, a second analysis was calculated to investigate the relationship between the s-IATsex craving and the neutral approach/avoidance score. Here, no significant quadratic relationship could be found (p = 0.239).

Moderated Regression Analyses

To investigate the relationship between sensitivity toward sexual excitation (SES), tendencies to approach or avoid pornographic stimuli (pornographic approach/avoidance score), and tendencies toward cybersex addiction, a hierarchical moderated regression analysis with the s-IATsex factor craving as dependent variable was calculated (all variables centralized; Cohen et al., 2003). In the first step, the SES explained 13.5% of the s-IATsex craving variance, F(1,121) = 18.83, p < 0.001. In the second step, the pornographic approach/avoidance score led to a significant increase of variance explanation, ΔR² = 0.029, ΔF(2,120) = 4.19, p = 0.043. In the third step, the interaction of the SES and the pornographic approach/avoidance score led to a significant increase of variance explanation, ΔR² = 0.044, ΔF(3,119) = 6.62, p = 0.011. Overall, the regression model was significant and explained 20.8% variance of the s-IATsex craving, F(3,122) = 10.41, p < 0.001.

To investigate the observed moderation effect in more detail, simple slopes were analyzed (see Figure ​Figure3A3A). The slope of the regression line representing approach tendencies (1 standard deviation above the mean) was not significantly different from zero, t = 1.71, p = 0.090. In contrast, the slope of the regression line representing avoidance tendencies (1 standard deviation below the mean) was significantly different from zero, t = 5.50, p < 0.001, indicating that a high SES, accompanied by avoidance tendencies resulted in a high s-IATsex craving score. When using tendencies to approach or avoid neutral stimuli (neutral approach/avoidance score) as moderator, no significant interaction could be found (p = 0.196).

 

FIGURE 3

Graphical illustration of simple slopes with respect to the interactions between the compatibility effect for pornographic pictures (pornographic approach/avoidance score) and (A) the sensitivity toward sexual excitation (SES) as well as (B) problems …

A second model was calculated to investigate the relationship between the control component of problematic sexual behavior (HBI loss of control), tendencies to approach or avoid pornographic stimuli (pornographic approach/avoidance score), and tendencies toward craving components in cybersex addiction. In the first step, the HBI loss of control explained 22.2% of the s-IATsex craving variance, F(1,121) = 34.52, p < 0.001. In the second step, the pornographic approach/avoidance score did not lead to a significant increase of variance explanation, ΔR² = 0.017, ΔF(2,120) = 2.70, p = 0.103. In the third step, the interaction of the HBI loss of control and the pornographic approach/avoidance score led to a significant increase of variance explanation, ΔR² = 0.037, ΔF(3,119) = 6.02, p = 0.016. Overall, the regression model was significant while explaining 25.7% variance of the s-IATsex craving, F(3,122) = 15.10, p < 0.001. Further values for both moderated regression analyses are summarized in Table ​Table55.

 

Table 5

Values of the moderated regression analyses with s-IATsex factor craving as dependent variable.

Similar to the first model, the simple slopes were analyzed (see Figure ​Figure3B3B). The slope of the regression line representing approach tendencies (1 standard deviation above the mean) was significantly different from zero, t = 2.85, p = 0.005. The slope of the regression line representing avoidance tendencies (1 standard deviation below the mean) was also significantly different from zero, t = 6.14, p < 0.001, indicating that both approach and avoidance toward pornographic pictures, accompanied by a high HBI loss of control resulted in a high s-IATsex craving score. Similar to the first moderated regression analysis, using tendencies to approach or avoid neutral stimuli (neutral approach/avoidance score) as moderator showed no significant interaction (p = 0.166).

Additionally, to investigate whether the HBI loss of control scale, the SES, and the pornographic approach/avoidance score have an accumulating effect on tendencies toward cybersex addiction, a linear regression analysis with the s-IATsex factor craving as dependent variable was calculated. In the first step, the HBI loss of control explained 22.2% of the s-IATsex craving variance, F(1,121) = 34.52, p < 0.001. In the second step, the SES led to a significant increase of variance explanation, ΔR² = 0.052, ΔF(2,120) = 2.63, p = 0.004. In the third step, the pornographic approach/avoidance score led to a significant increase of variance explanation, ΔR² = 0.024, ΔF(3,119) = 4.47, p = 0.037. Overall, the regression model was significant and explained 30.1% variance of the s-IATsex craving, F(3,122) = 17.04, p < 0.001. Further regression values are summarized in Table ​Table55.

Relationships between Actual Cybersex use and Addiction-Related Measurements

In order to investigate possible relationships between actual cybersex use and measurements related to cybersex addiction, several additional correlations were calculated. There were positive relationships between the s-IATsex factor craving and both the frequency of weekly cybersex use (r = 0.227, p = 0.011) and the average time spent on cybersex sites during one visit (r = 0.198, p = 0.028). However, no significant relationships could be found between the frequency of weekly cybersex use and HBI loss of control (r = 0.136, p = 0.133), SES (r = 0.119, p = 0.190) as well as craving Δ sexual arousal/masturbation and AAT scores (all ps > 0.400). Similarly, there were no significant relationships between the average time spent on cybersex sites during one visit and HBI loss of control (r = 0.025, p = 0.781), SES (r = 0.161, p = 0.076) as well as craving Δ sexual arousal/masturbation and AAT scores (all ps > 0.500).

Future Directives

Future studies could further aim on expanding the focus of interest by including positive and negative expectancies as possible predictors for approach/avoidance tendencies in analogy to the proposed framework by Breiner et al. (1999). Thus, positive expectancies are supposed to promote inclinations to approach addictive behaviors, while negative expectancies could suppress such urges and lead to avoidance reactions. In the context of cybersex addiction, cybersex use expectancies could have a similar influence on approach/avoidance tendencies since it was already shown that Internet use expectancies are connected with Internet addiction (Brand et al., 2014a). Besides the existence of competing approach/avoidance inclinations, such expectancies could thereby explain which inclinations might be dominant in an addiction-related decision situation.

Moreover, it could be beneficial to investigate if competing neural networks are involved in approach/avoidance reactions. In this context, studies already showed parallels to the dual-process model by Bechara (2005) since different neural networks were preliminary shown for approach (nucleus accumbens, medial prefrontal cortex) and avoidance (amygdala, dorsolateral prefrontal cortex) in alcohol addicted individuals (Ernst et al., 2012; Wiers et al., 2014b). By strengthening this finding, Schlund et al. (2011) reported balanced activations of these networks for approach/avoidance behavior in healthy individuals. Moreover, it could be shown that cognitive bias modification programs reduced approach/avoidance related activations in the medial prefrontal cortex and in the amygdala (Wiers et al., 2014a, 2015). Based on these results, it seems plausible to assume that the AAT is able to measure both approach- and avoidance-biases. Consequently, future studies should address the investigation of neural correlates connected with approach/avoidance tendencies in cybersex addiction in order to strengthen the current studies’ findings. Moreover, both substance dependency and cybersex addiction research could benefit from the application of sophisticated analyses methods (e.g., bin-analyses). Thus, such methods could provide more evidence for the assumption that the AAT assesses both approach- and avoidance-biases.

Thinking further, previous studies mostly investigated linear relationships between approach/avoidance tendencies and addiction-related measurements, whereas such an approach might not cover the complexity of addictive behaviors. However, the most prominent belief seems to be that only approach-biases are connected to the development and maintenance of addictive behaviors, although this assumption is not entirely supported by existing findings. For instance, some studies reported approach-biases in individuals with problematic substance use (e.g., Cousijn et al., 2012), whereas avoidance tendencies were found in abstinent or treatment-seeking subjects (Eberl et al., 2013a). Moreover, Wiers et al. (2013) found approach-biases in smokers, but not in ex-smokers. Further, relationships between approach/avoidance tendencies and addiction-related measurements such as subjective craving or relapse rates are inconsistent since both positive (e.g., Mogg et al., 2005) as well as negative associations (e.g., Cousijn et al., 2011; Spruyt et al., 2013) have been reported. Therefore, it seems plausible to assume that not only approach, but also avoidance tendencies might be important factors within addictive behaviors. Thus, curve-linear regression analyses, which allow the analysis of both inclinations in a single model, might not only be beneficial for investigating cybersex addiction, but also for studying approach/avoidance tendencies in other behavioral addictions or substance dependencies.

At last, it might be useful to investigate to which extent approach/avoidance tendencies influence the development and maintenance of cybersex addiction. Here, longitudinal study designs could be beneficial. Moreover, such an approach seems plausible since the results of the current study suggested the AAT to measure effects due to long-term cybersex use, while more research is necessary to justify this assumption.

Limitations

First of all, it has to be noted that the curve-linear regression analysis used to test the assumed quadratic relationship between approach/avoidance tendencies and craving-related symptoms of cybersex addiction may rather be considered as an exploratory method. Moreover, the results are not emphasizing a perfect quadratic relationship. Therefore, the findings have to be interpreted with caution and need to be replicated. Nonetheless, these results at least point toward a non-linearity of the relationship between approach/avoidance tendencies and cybersex addiction. Since we only included heterosexual male participants, our results can hardly be generalized to women or homosexual individuals. Further, the majority of the sample consisted of regular cybersex users, while a minority reported subjective symptoms in everyday life due to their cybersex use. Although, the investigation of disorders with analog samples offers many benefits (Abramowitz et al., 2014), our findings cannot be entirely transferred to a clinical population since none of the participants were diagnosed as being addicted to cybersex. Therefore, future studies could benefit from investigating individuals in a clinical setting, though it has to be noted that missing diagnostic criteria can make it difficult to compare a cybersex-addicted patient group with a control group in the classical way. However, such an approach could be useful because the AAT could also be used for cognitive bias modification training (Wiers et al., 2011) in cybersex addiction treatment.

We thank Dr. Christian Laier and Dr. Johannes Schiebener for their valuable contributions to the study. They helped us significantly with conducting the experiment and improving the manuscript. Additionally, we thank Michael Schwarz for his appreciated assistance regarding the implementation of the AAT.

• Abramowitz J. S., Fabricant L. E., Taylor S., Deacon B. J., Mckay D., Storch E. A. (2014). The utility of analogue studies for understanding obsessions and compulsions. Clin. Psychol. Rev. 34 206–217. 10.1016/j.cpr.2014.01.004 [PubMed] [Cross Ref]
• APA. (2013). Diagnostic and Statistical Manual of Mental Disorders, 5th Edn Washington DC: APA.
• Bechara A. (2005). Decision making, impulse control and loss of willpower to resist drugs: aneurocognitive perspective. Nat. Neurosci. 8 1458–1463. 10.1038/nn1584 [PubMed] [Cross Ref]
• Brand M., Laier C., Pawlikowski M., Schächtle U., Schöler T., Altstötter-Gleich C. (2011). Watching pornographic pictures on the Internet: role of sexual arousal ratings and psychological-psychiatric symptoms for using Internet sex sites excessively. Cyberpsychol. Behav. Soc. Netw. 14 371–377. 10.1089/cyber.2010.0222 [PubMed] [Cross Ref]
• Brand M., Laier C., Young K. S. (2014a). Internet addiction: coping styles, expectancies, and treatment implication. Front. Psychol. 5:1256 10.3389/fpsyg.2014.01256 [PMC free article] [PubMed] [Cross Ref]
• Brand M., Young K., Laier C. (2014b). Prefrontal control and Internet addiction: a theoretical model and review of neuropsychological and neuroimaging findings. Front. Hum. Neurosci. 8:375 10.3389/fnhum.2014.00375 [PMC free article] [PubMed] [Cross Ref]
• Breiner M. J., Stritzke W. G. K., Lang A. R. (1999). Approaching avoidance. A step essential to the understanding of craving. Alcohol. Res. Ther. 23 197–206. 10.1023/A:1018783329341 [PubMed] [Cross Ref]
• Buydens-Branchey L., Branchey M., Fergeson P., Hudson J., Mckernin C. (1997). Hormonal, psychological, and alcohol craving changes after m-chlorophenylpiperazine administration in alcoholics. Alcohol. Clin. Exp. Res. 21 220–226. 10.1111/j.1530-0277.1997.tb03753.x [PubMed] [Cross Ref]
• Carpenter D. L., Janssen E., Graham C. A., Vorst H., Wicherts J. (2010). “The sexual inhibition/sexual excitation scales-short form SIS/SES-SF,” in Handbook of Sexuality-Related Measures eds Fisher T. D., Davis C. M., Yarber W. L., Davis S. L., editors. (Abingdon, GB: Routledge; ) 236–239.
• Cash H., Rae C. D., Steel A. H., Winkler A. (2012). Internet addiction: a brief summary of research and practice. Curr. Psychiatry Rev. 8 292–298. 10.2174/157340012803520513 [PMC free article] [PubMed] [Cross Ref]
• Cohen J. (1988). Statistical Power Analysis for the Behavioral Sciences. Hillsdale, NJ: Lawrance Earlbaum Associates.
• Cohen J., Cohen P., West S. G., Aiken L. S. (2003). Applied Multiple Regression/Correlation Analysis for the Behavioral Science. Mahwah, NJ: Lawrence Erlbaum.
• Coskunpinar A., Cyders M. A. (2013). Impulsivity and substance-related attentional bias: a meta-analytic review. Drug Alcohol Depend. 133 1–14. 10.1016/j.drugalcdep.2013.05.008 [PubMed] [Cross Ref]
• Cousijn J., Goudriaan A. E., Ridderinkhof K. R., Van Den Brink W., Veltman D. J., Wiers R. W. (2012). Approach-bias predicts development of cannabis problem severity in heavy cannabis users: results from a prospective FMRI study. PLoS ONE 7:e42394 10.1371/journal.pone.0042394 [PMC free article] [PubMed] [Cross Ref]
• Cousijn J., Goudriaan A. E., Wiers R. W. (2011). Reaching out towards cannabis: approach-bias in heavy cannabis users predicts changes in cannabis use. Addiction 106 1667–1674. 10.1111/j.1360-0443.2011.03475.x [PMC free article] [PubMed] [Cross Ref]
• Cousijn J., Snoek R. W. M., Wiers R. W. (2013). Cannabis intoxication inhibits avoidance action tendencies: a field study in the Amsterdam coffee shops. Psychopharmacology 229 167–176. 10.1007/s00213-013-3097-6. [PubMed] [Cross Ref]
• Davis R. A. (2001). A cognitive-behavioral model of pathological Internet use. Comput. Hum. Behav. 17 187–195. 10.1016/S0747-5632(00)00041-8 [Cross Ref]
• Döring N. M. (2009). The Internet’s impact on sexuality: a critical review of 15 years of research. Comput. Hum. Behav. 25 1089–1101. 10.1016/j.chb.2009.04.003 [Cross Ref]
• Drummond D. C. (2001). Theories of drug craving, ancient and modern. Addiction 96 33–46. 10.1046/j.1360-0443.2001.961333.x [PubMed] [Cross Ref]
• Eberl C., Wiers R. W., Pawelczack S., Rinck M., Becker E. S., Lindenmeyer J. (2013a). Approach bias modification in alcohol dependence: do clinical effects replicate and for whom does it work best? Dev. Cogn. Neurosci. 4 38–51. 10.1016/j.dcn.2012.11.002 [PubMed] [Cross Ref]
• Eberl C., Wiers R. W., Pawelczack S., Rinck M., Becker E. S., Lindenmeyer J. (2013b). Implementation of approach bias re-training in alcoholism. How many sessions are needed? Alcohol. Clin. Exp. Res. 38 587–594. 10.1111/acer.12281 [PubMed] [Cross Ref]
• Ernst L. H., Plichta M. M., Dresler T., Zesewitz A. K., Tupak S. V., Haeussinger F. B., et al. (2012). Prefrontal correlates of approach preferences for alcohol stimuli in alcohol dependence. Addict. Biol. 19 497–508. 10.1111/adb.12005 [PubMed] [Cross Ref]
• Field M., Cox W. M. (2008). Attentional bias in addictive behaviors: a review of its development, causes, and consequences. Drug Alcohol Depend. 97 1–20. 10.1016/j.drugalcdep.2008.03.030 [PubMed] [Cross Ref]
• Field M., Eastwood B., Bradley B., Mogg K. (2006). Selective processing of cannabis cues in regular cannabis users. Drug Alcohol Depend. 85 75–82. 10.1016/j.drugalcdep.2006.03.018 [PubMed] [Cross Ref]
• Field M., Kiernan A., Eastwood B., Child R. (2008). Rapid approach responses to alcohol cues in heavy drinkers. J. Behav. Ther. Exp. Psychiatry 39 209–218. 10.1016/j.jbtep.2007.06.001 [PubMed] [Cross Ref]
• Field M., Marhe R., Franken I. H. A. (2014). The clinical relevance of attentional bias in substance use disorders. CNS Spectr. 19 225–230. 10.1017/S1092852913000321 [PubMed] [Cross Ref]
• Field M., Mogg K., Bradley B. P. (2005a). Alcohol increases cognitive biases for smoking cues in smokers. Psychopharmacology 180 63–72. 10.1007/s00213-005-2251 [PubMed] [Cross Ref]
• Field M., Mogg K., Bradley B. P. (2005b). Craving and cognitive biases for alcohol cues in social drinkers. Alcohol Alcohol. 40 504–510. 10.1093/alcalc/agh213 [PubMed] [Cross Ref]
• Georgiadis J. R., Kringelbach M. L. (2012). The human sexual response cycle: brain imaging evidence linking sex to other pleasures. Progr. Neurobiol. 98 49–81. 10.1016/j.pneurobio.2012.05.004 [PubMed] [Cross Ref]
• Griffiths M. D. (2005). A ‘components’ model of addiction within a biopsychosocial framework. J. Subst. Use 10 191–197. 10.1080/14659890500114359 [Cross Ref]
• Jović J., Đinđić N. (2011). Influence of dopaminergic system on Internet addiction. Acta Med. Medianae 50 60–66. 10.5633/amm.2011.0112 [Cross Ref]
• Kafka M. P. (2010). Hypersexual disorder: a proposed diagnosis for DSM-V. Archiv. Bexual Behav. 39 377–400. 10.1007/s10508-009-9574-7 [PubMed] [Cross Ref]
• Koo H. J., Kwon J.-H. (2014). Risk and protective factors of Internet addiction: a meta-analysis of empirical studies in Korea. Yonsei Med. J. 55 1691–1711. 10.3349/ymj.2014.55.6.1691 [PMC free article] [PubMed] [Cross Ref]
• Kuss D. J., Griffiths M. D. (2011). Internet sex addiction: a review of empirical research. Addict. Res. Theory 20 111–124. 10.3109/16066359.2011.588351 [Cross Ref]
• Kuss D. J., Griffiths M. D., Karila L., Billieux J. (2014a). Internet addiction: a systematic review of epidemiological research for the last decade. Curr. Pharm. Des. 20 4026–4052. 10.2174/13816128113199990617 [PubMed] [Cross Ref]
• Kuss D. J., Shorter G. W., Van Rooij A. J., Griffiths M. D., Schoenmakers T. M. (2014b). Assessing Internet addiction using the parsimonious Internet addiction components model – A preliminary study. Int. J. Ment. Health Addict. 12 351–366. 10.1007/s11469-013-9459 [Cross Ref]
• Kuss D. J., Shorter G. W., Van Rsooij A. J., Van De Mheen D., Griffiths M. D. (2014c). The Internet addiction components model and personality: establishing construct validity via a nomological network. Comput. Hum. Behav. 39 312–321. 10.1016/j.chb.2014.07.031 [Cross Ref]
• Laier C., Brand M. (2014). Empirical evidence and theoretical considerations on factors contributing to cybersex addiction from a cognitive-behavioral view. Sex. Addict. Compulsivity 21 305–321. 10.1080/10720162.2014.970722 [Cross Ref]
• Laier C., Pawlikowski M., Brand M. (2014). Sexual picture processing interferes with decision-making under ambiguity. Arch. Sex. Behav. 43 473–482. 10.1007/s10508-013-0119-8 [PubMed] [Cross Ref]
• Laier C., Pawlikowski M., Pekal J., Schulte F. P., Brand M. (2013a). Cybersex addiction: experienced sexual arousal when watching pornography and not real-life sexual contacts makes the difference. J. Behav. Addict. 2 100–107. 10.1556/JBA.2.2013.002 [PubMed] [Cross Ref]
• Laier C., Schulte F. P., Brand M. (2013b). Pornographic picture processing interferes with working memory performance. J. Sex. Res. 50 37–41. 10.1080/00224499.2012.716873 [PubMed] [Cross Ref]
• Lang P. J., Bradley M. M., Cuthbert B. N. (2008). International Affective Picture System (IAPS): Affective Ratings of Pictures and Instruction Manual. Gainesville, FL: University of Florida.
• Marlatt G. A. (1985). “Cognitive factors in the relapse process,” in Relapse Prevention: Maintenance Strategies in the Treatment of Addictive Behaviors eds Marlatt G. A., Gordon J. R., editors. (New York, NY: Guilford Press; ) 128–200.
• Meerkerk G.-J., Van Den Eijnden R. J. J. M., Garretsen H. F. L. (2006). Predicting compulsive Internet use: it’s all about sex! Cyberpsychol. Behav. 9 95–103. 10.1089/cpb.2006.9.95 [PubMed] [Cross Ref]
• Mogg K., Bradley B., Field M., De Houwer J. (2003). Eye movements to smoking-related pictures in smokers: relationship between attentional biases and implicit and explicit measures of stimulus valence. Addiction 98 825–836. 10.1046/j.1360-0443.2003.00392.x [PubMed] [Cross Ref]
• Mogg K., Field M., Bradley B. P. (2005). Attentional and approach biases for smoking cues in smokers: an investigation of competing theoretical views of addiction. Psychopharmacology 180 333–341. 10.1007/s00213-005-2158-x. [PubMed] [Cross Ref]
• Montag C., Bey K., Sha P., Li M., Chen Y.-F., Liu W.-Y., et al. (2015). Is it meaningful to distinguish between generalized and specific Internet addiction? Evidence from a cross-cultural study from Germany, Sweden, Taiwan and China. Asia Pac. Psychiatry 7 20–26. 10.1111/appy.12122 [PubMed] [Cross Ref]
• Olsen C. M. (2011). Natural rewards, neuroplasticity, and non-drug addictions. Neuropharmacology 61 1109–1122. 10.1016/j.neuropharm.2011.03.010 [PMC free article] [PubMed] [Cross Ref]
• Ooteman W., Koeter M. W. J., Vserheul R., Schippers G. M., Van Den Brink W. (2006). Measuring craving: an attempt to connect subjective craving with cue reactivity. Alcohol. Clin. Exp. Res. 30 57–69. 10.1111/j.1530-0277.2006.00019.x [PubMed] [Cross Ref]
• Palfai T. P. (2006). Activating action tendencies: the influence of action priming on alcohol consumption among male hazardous drinkers. J. Stud. Alcohol. Drugs 67 926–933. 10.15288/jsa.2006.67.926 [PubMed] [Cross Ref]
• Pawlikowski M., Altstötter-Gleich C., Brand M. (2013). Validation and psychometric properties of a short version of Young’s Internet Addiction Test. Comput. Hum. Behav. 29 1212–1223. 10.1016/j.chb.2012.10.014 [Cross Ref]
• Phaf R. H., Mohr S. E., Rottevel M., Wicherts J. M. (2014). Approach, avoidance, and affect: a meta-analysis of approach-avoidance tendencies in manual reaction time tasks. Front. Psychol. 5:378 10.3389/fpsyg.2014.00378 [PMC free article] [PubMed] [Cross Ref]
• Rasch D., Guiard V. (2004). The robustness of parametric statistical methods. Psychol. Sci. 2 175–208.
• Reay B., Attwood N., Gooder C. (2013). Inventing sex: the short history of sex addiction. Sex. Cult. 17 1–19. 10.1007/s12119-012-9136-3 [Cross Ref]
• Reid R. C., Garos S., Carpenter B. N. (2011). Reliability, validity, and psychometric development of the Hypersexual Behavior Inventory in an outpatient sample of men. Sex. Addict. Compulsivity 18 30–51. 10.1080/10720162.2011.555709 [Cross Ref]
• Rinck M., Becker E. (2007). Approach and avoidance in fear of spiders. J. Behav. Ther. Exp. Psychiatry 38 105–120. 10.1016/j.jbtep.2006.10.001 [PubMed] [Cross Ref]
• Robinson T. E., Berridge K. C. (1993). The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res. Rev. 18 247–291. 10.1016/0165-0173(93)90013-p [PubMed] [Cross Ref]
• Robinson T. E., Berridge K. C. (2001). Incentive-sensitization and addiction. Addiction 96 103–114. 10.1080/09652140020016996 [PubMed] [Cross Ref]
• Robinson T. E., Berridge K. C. (2008). The incentive sensitization theory of addiction: some current issues. Philos. Trans. R. Soc. B Biol. Sci. 363 3137–3146. 10.1098/rstb.2008.0093 [PMC free article] [PubMed] [Cross Ref]
• Sayette M. A., Shiffman S., Tiffany S. T., Niaura R. S., Martin C. S., Shadel W. G. (2000). The measurement of drug craving. Addiction 95 189–210. 10.1046/j.1360-0443.95.8s2.8.x [PMC free article] [PubMed] [Cross Ref]
• Schiebener J., Laier C., Brand M. (2015). Getting stuck with pornography? Overuse or neglect of cybersex cues in a multitasking situation is related to symptoms of cybersex addiction. J. Behav. Addict. 4 14–21. 10.1556/JBA.4.2015.1.5 [PMC free article] [PubMed] [Cross Ref]
• Schlund M. W., Magee S., Hudgins C. D. (2011). Human avoidance and approach learning: evidence for overlapping neural systems and experiential avoidance modulation of avoidance neurocircuitry. Behav. Brain Res. 225 437–448. 10.1016/j.bbr.2011.07.054 [PubMed] [Cross Ref]
• Schoenmakers T. M., Wiers R. W., Field M. (2008). Effects of a low dose of alcohol on cognitive biases and craving in heavy drinkers. Psychopharmacology 197 169–178. 10.1007/s00213-007-1023-5 [PMC free article] [PubMed] [Cross Ref]
• Sharbanee J. M., Hu L., Stritzke W. G. K., Wiers R. W., Rinck M., Macleod C. (2014). The effect of approach/avoidance training on alcohol consumption is mediated by change in alcohol action tendency. PLoS ONE 9:e85855 10.1371/journal.pone.0085855 [PMC free article] [PubMed] [Cross Ref]
• Sharbanee J. M., Stritzke W. G. K., Wiers R. W., Macleod C. (2013). Alcohol-related biases in selective attention and action tendency make distinct contributions to dysregulated drinking behaviour. Addiction 108 1758–1766. 10.1111/add.12256 [PubMed] [Cross Ref]
• Short M. B., Black L., Smith A. H., Wetterneck C. T., Wells D. E. (2011). A review of Internet pornography use research: methodology and content from the past 10 years. Cyberpsychol. Behav. Soc. Netw. 10 1–12. 10.1089/cyber.2010.0477 [PubMed] [Cross Ref]
• Skinner M. D., Aubin H.-J. (2010). Craving’s place in addiction theory: contributions of the major models. Neurosci. Biobehav. Rev. 34 606–623. 10.1016/j.neubiorev.2009.11.024 [PubMed] [Cross Ref]
• Spada M. M. (2014). An overview of problematic Internet use. Addict. Behav. 39 3–6. 10.1016/j.addbeh.2013.09.007 [PubMed] [Cross Ref]
• Spruyt A., De Houwer J., Tibboel H., Verschuere B., Crombez G., Verbanck P., et al. (2013). On the predictive validity of automatically activated approach/avoidance tendencies in abstaining alcohol-dependent patients. Drug Alcohol Depend. 127 81–86. 10.1016/j.drugalcdep.2012.06.019 [PubMed] [Cross Ref]
• Starcevic V. (2013). Is Internet addiction a useful concept? Aust. N. Z. J. Psychiatry 47 16–19. 10.1177/0004867412461693 [PubMed] [Cross Ref]
• Tiffany S. T., Wray J. M. (2012). The clinical significance of drug craving. Ann. N. Y. Acad. Sci. 1248 1–17. 10.1111/j.1749-6632.2011.06298.x [PMC free article] [PubMed] [Cross Ref]
• Weinstein A., Lejoyeux M. (2010). Internet addiction or excessive Internet use. Am. J. Drug Alcohol Abuse 36 277–283. 10.3109/00952990.2010.491880 [PubMed] [Cross Ref]
• West S. G., Finch J. F., Curran P. J. (1995). “Structural equation models with non-normal variables: problems and remedies,” in Structural Equation Modeling: Concepts, Issues and Applications ed. Hoyle R., editor. (Newbury Park, CA: Sage; ) 56–75.
• Wiers C. E., Kühn S., Javadi A. H., Korucuoglu O., Wiers R. W., Walter H., et al. (2013). Automatic approach bias towards smoking cues is present in smokers but not in ex-smokers. Psychopharmacology 229 187–197. 10.1007/s00213-013-3098-5 [PubMed] [Cross Ref]
• Wiers C. E., Ludwig V. U., Gladwin T. E., Park S. Q., Heinz A., Wiers R. W., et al. (2015). Effects of cognitive bias modification training on alcohol approach tendencies in male alcohol-dependent patients. Addict. Biol. [Epub ahead of print].10.1111/adb.12221 [PubMed] [Cross Ref]
• Wiers C. E., Stelzel C., Gladwin T. E., Park S. Q., Pawelczack S., Gawron C. K., et al. (2014a). Effects of cognitive bias modification training on neural alcohol cue reactivity in alcohol dependence. Am. J. Psychiatry [Epub ahead of print].10.1176/appi.ajp.2014.13111495 [PubMed] [Cross Ref]
• Wiers C. E., Stelzel C., Park S. Q., Gawron C. K., Ludwig V. U., Gutwinski S., et al. (2014b). Neural correlates of alcohol-approach bias in alcohol addiction: the spirit is willing but the flesh is weak for spirits. Neuropsychopharmacology 39 688–697. 10.1038/npp.2013.252 [PMC free article] [PubMed] [Cross Ref]
• Wiers R. W., Bartholow B. D., Van Den Wildenberg E., Thush C., Engels R. C. M. E., Sher K., et al. (2007). Automatic and controlled processes and the development of addictive behaviors in adolescents: a review and a model. Pharmacol. Biochem. Behav. 86 263–283. 10.1016/j.pbb.2006.09.021 [PubMed] [Cross Ref]
• Wiers R. W., Eberl C., Rinck M., Becker E. S., Lindenmeyer J. (2011). Retraining automatic action tendencies changes alcoholic patients’ approach bias for alcohol and improves treatment outcome. Psychol. Sci. 22 490–497. 10.1177/0956797611400615 [PubMed] [Cross Ref]
• Wiers R. W., Rinck M., Dictus M., Van Den Wildenberg E. (2009). Relatively strong automatic appetitive action-tendencies in male carriers of the OPRM1 G-allele. Genes Brain Behav. 8 101–106. 10.1111/j.1601-183X.2008.00454.x [PubMed] [Cross Ref]
• Wiers R. W., Stacy A. W. (2006). Implicit cognition and addiction. Curr. Dir. Psychol. Sci. 15 292–296. 10.1111/j.1467-8721.2006.00455.x [Cross Ref]
• Wölfling K., Beutel M. E., Koch A., Dickenhorst U., Müller K. W. (2013). Comorbid internet addiction in male clients of inpatient addiction rehabilitation centers: psychiatric symptoms and mental comorbidity. J. Nerv. Ment. Dis. 201 934–940. 10.1097/NMD.0000000000000035 [PubMed] [Cross Ref]
• Young K. S. (1998). Caught in the Net: How to Recognize the Signs of Internet Addiction – and a Winning Strategy for Recovery. New York, NY: John Wiley & Sons, Inc.
• Young K. S. (2008). Internet sex addiction: risk factors, stages of development, and treatment. Am. Behav. Scient. 52 21–037. 10.1177/0002764208321339 [Cross Ref]
• Young K. S., Pistner M., O’mara J., Buchanan J. (1999). Cyber disorders: the mental health concern for the new millennium. Cyberpsychol. Behav. 2 475–479. 10.1089/cpb.1999.2.475 [PubMed] [Cross Ref]