THERAPISTS' ASSESSMENT AND TREATMENT OF INTERNET INFIDELITY CASES

In this study, we investigated through an Internet survey of 508 practicing marriage and family therapists which treatment decisions varied by gender of the client and background variables of therapists. The subjects responded to several typical Internet infidelity scenarios. We varied the gender of the person initiating the infidelity for half of one sample. We also asked the family therapy participants to respond to how they might assess and treat each presenting problem. They also evaluated problem severity, prognosis of the case, number of sessions necessary for treatment, and the extent to which a therapist would focus individually or relationally. Results indicate that there were differences in how therapists assessed and treated clients based on client gender, therapists' age, therapists' gender, how religious therapists reported they were, and the extent of therapists' personal experience with infidelity.

Technology can be used for both good and bad. With one of its features being relational and communicative (Mantovani, 2001), one of the apparent misuses involves its role in faci�tating Internet infidelity. Partners and spouses are able to meet and develop relationships with people from other ethnic groups, genders, professions, and countries discreetly, at any time of the day and at a relatively low cost (Cooper & Griffin-Shelley, 2002). Unfortunately, there is no consensus on the treatment of Internet infidelity and the prevalence of this problem for couples is increasing rapidly.

There are a plethora of treatment models for infidelity (Blow & Hartnett, 2005). Internet infidelity treatment frameworks have also been proposed (Atwood & Schwartz, 2002; Hertlein & Piercy, 2005; Shaw, 1997). Yet Nelson (2000) found that there is presently no consensus around the treatment of Internet infidelity. As a result, couple therapists may be more apt to rely on other factors to treat these cases. Some of these factors influencing treatment might be personal values and biases, putting clients at risk for inappropriate differential treatment.

This study was designed to identify treatment methods clinicians report and the impact of certain factors on a clinician's view of Internet infidelity treatment. We were interested in learning how therapists treat several types of Internet infidelity. We also wanted to know if these decisions about treatment were influenced by the gender of the unfaithful partner. Finally, we examined the extent to which, if at all, therapists' proposed intervention strategies varied according to the participants' (a) gender, (b) age, and (c) religiosity.

Internet Access and Usage for Romantic Purposes

Millions of people access the World Wide Web on a daily basis (U.S. Department of Commerce, 2002). Recent investigations find that throughout this large population of Internet users, a wide variety of users are accessing the Internet for sexual purposes (Cooper, 1997; Schnarch, 1997). These purposes include (but are not limited to) viewing pornography, placing personal ads, engaging in cybersex, educating oneself about sex, and developing sexual interpersonal relationships. Users are inundated with sexually explicit information on the web in the form of spam within email accounts, advertisements, and a multitude of search engines (Spink, Koricich, Jansen, & Cole, 2004).

Since the Internet has become a part of everyday life to most people, communicating online has made it easy for people to begin interpersonal relationships (Hatala, Milewski, & Baack, 1999; McCown, Fischer, Page, & Homant, 2001; Merkle & Richardson, 2000; Underwood & Findlay, 2004). Parks and Roberts (1996) report 26.3% of Internet users who state they have established at least one interpersonal relationship with someone online indicate a romantic relationship with someone online. Other studies indicate that this number might be higher. One survey found that 40% of U.S. adults report they use email to flirt with someone to whom they are attracted (PEW Internet and American Life Project, 2006). Cooper, Scherer, and Mathy (2001) found that approximately 20% of all Internet users report engaging in some sexual activity online.

Despite the rising Internet usage for romantic purposes, scholars have focused more on demographics regarding sexual behavior online and have only recently begun to explore the specific phenomenon of Internet infidelity (Hatala et al., 1999). In a summary of the research related to relationships and technology, Hertlein and Webster (2008) found the research could be summarized into two groups: exploratory and descriptive studies. The exploratory studies were those which were interested in how online relationships develop, understanding the nature of the relationships, and focusing on the effects these relationships have on the lives of the participants (see, for example, Boies, Cooper, & Osborne, 2004; Underwood & Findlay, 2004). The descriptive studies sought to understand perceptions of online relationships (see, for example, Whitty, 2003, 2005).

Merkle and Richardson (2000), as well as other scholars (Hertlein & Piercy, 2006) propose that Internet infidelity constitutes the same sense of betrayal as found in traditional infidelity scenarios. Additionally, meeting others online for romantic purposes is becoming more userfriendly. For example, online personal advertisement websites typically request users at the time of ad placement to select relationship status from a drop-down menu, which often includes the category "married but looking."

Social norms might also influence the prevalence of Internet infidelity. Buunk and Bakker (1995) found that women who know other women engaging in infidelity were more likely to engage in infidelity themselves. If Internet use is becoming more and more part of the social norm, it may be considered more acceptable to meet people online and engage in sexual or romantic relationships. In other words, the more Internet use becomes a norm, and the more Internet infidelity becomes a norm, Internet infidelity will continue to increase in its occurrence. As these rates increase, it is imperative that couple and family therapists identify effective treatment strategies.

Further, online romances are characterized by emotional components. Greenfield (1999) investigated Internet behavior in nearly 18,000 individuals and found that 41% of the sample reported intense intimacy while online. Rather than judgments on physical attractiveness or physical sexuality, persons involved in online relationships are connected by a sense of how close they feel to someone based on self-description and communication (Cooper & Sportolari, 1997). Cooper and Griffin-Shelley (2002) identified three characteristics about the Internet that make it suitable for sexual behavior, known as the "Triple A" engine: the Internet is accessible from a variety of locations, it is affordable, and users can choose to be anonymous, revealing as much or as little information as desired. Emotional infidelity, or building an intimate emotional relationship with another person to the exclusion of one's primary partner, can flourish via the Internet.

Because of the structure of the Internet, lovers engage in many behaviors, forcing couples, therapists, and society to expand the definition of what is considered infidelity behavior. Internet infideUty behaviors are found on a continuum from emotionally connecting with another partner to physically having sexual involvement with someone over the Internet to the exclusion of your primary partner (Merkle & Richardson, 2000). Internet infidelity is becoming so insidious that recently a court in Belgium ruled that erotic chat with someone online other than one's primary partner was sufficient grounds for a divorce (Erotic Chat Grounds for a Divorce?, 2005). Therapists need more skills to negotiate these diffuse boundaries.

Biases in Treatment

Although therapists may strive to base their therapy on scientific research, the human element is also at work. According to Aponte (1985), "Therapy is shaped by the interaction of the personal values of therapists with the values of societal institutions . . ." (p. 324). Values direct our actions and allow us to interpret and judge social phenomena, but also affect the therapeutic process (Aponte, 1985). In fact, it is the value system of the therapist that defines the problem for treatment as well as the course of treatment: Aponte (1985) also states that "Although therapists are often unconscious of these aspects of human functioning in their assessments, they nevertheless operate through them in the biases inherent in the therapies they use ..." (p. 326). In other words, biases come through in our work because we are shaped by society and our family background.

Differential treatment has been an area of concern in the psychotherapy literature (Stabb, Cox, & Harber, 1997). There is no guarantee that therapists will be impervious to the influence of biases, especially when the presenting problem is a sexuality concern. Unchecked biases bleed into our understanding, interpretation, and treatment strategies in several areas, ranging from gender to sexual addiction. In the area of sexuality, Hecker, Trepper, Wetchler, and Fontaine (1995) found that therapists viewed single persons engaging in intercourse with multiple partners as more pathological (more likely to be a sex addict) than married persons engaging in the same amount and type of sexual activity with their monogamous partner. Similarly, male therapists who reported high levels of religiosity rated clients in the vignettes as more pathological than therapists who reported lower levels of religiosity (Hecker et al., 1995).

Therapists may also view cases differently based on a client's gender. In a study examining client gender and perceived therapeutic resistance, Korner and Goldberg (1996) found that therapists attributed different perceptions of resistance based on client gender. As infidelity research previously considered gender to be the most consistent predictor of infidelity behavior (Seal, Agostinelli, & Hannett, 1994; men were believed to be more likely to engage in infidelity primarily due to greater opportunity), this belief might influence infidelity treatment. The attempt to connect social background variables with sexual behavior, known as "sexual demography," is used a great deal in attempting to understand infidelity (Liu, 2000). Some factors under investigation for their relationship to infideUty include age (Atkins, Baucom, & Jacobson, 2001), religiosity (Liu, 2000; Treas & Giesen, 2000), and education (Wyatt, Peters, & Guthrie, 1998a, 1998b). With such information, therapists may view cases that do not fit within this mold as less traditional and alter treatment strategies.

Purpose

The research sought to answer two broad sets of questions. The first asked, "What do therapists do when they assess and treat Internet infidelity?" The second question asked, "Do social background characteristics of the therapist or client influence therapists' treatment decisions?" Specifically, do marriage and family therapists' assessment and treatment decisions in cases of Internet infidelity change depending on the gender of the identified client? Also, do the assessment and treatment decisions made in cases of Internet infidelity vary when examined in terms of therapist social background variables, such as age, gender, religion, marital status, and whether they report they have been impacted by infidelity in their lives?

Operational Definition of Internet Infidelity

Therapists struggle with the definition of Internet infidelity, and cannot agree on a way to treat it, or even decide if it is something to be treated (Glass, 2001; Nelson, 2000). Some couples have developed relationship understandings that accommodate multiple sexual/romantic partners. For the purpose of this study, Internet infidelity is defined as a romantic or sexual contact facilitated by Internet use that is seen by at least one partner as an unacceptable breach of their marital contract of faithfulness. The breach is seen as sufficiently severe for the couple to enter therapy.

METHODS

Design

In this study, we used the Internet to distribute the surveys and collect data for Phase 1 of the research. The Internet as a way to conduct research has several advantages over more traditional means. It is accessible to many, is cost effective, and provides a form of interaction that does not exist with telephone or mail surveys. Researchers can access a sample from a wide geographic region. Furthermore, the research can be conducted quickly, and the computer can assist in the analysis and interpretation of data (Dillman, 2000).

Phase 1 Methodology

Participants. We solicited the participation of 2,687 clinical members of the American Association for Marriage and Family Therapy (AAMFT). Of the 2,327 surveys that potentially reached the intended participants, 508 members responded to the survey (21.8%). Their names were randomly selected from a list of 15,000 names listed on http://www.TherapistLocater.net. There were no restrictions as far as age, ethnicity, or any other factors.

The analysis included a total of 241 (48.49%) male therapists and 256 (51.61%) female therapists, with seven participants not reporting their sex. Participant ages ranged from 20 to 78 years, with a mean age of 51.24 years. The majority of the sample was Caucasian (93.5%). The remainder was composed of 2.2% African American, 0.4% Asian, 1.2% Hispanic, 1.0% Native American, and 1.6% who identified themselves as "Other" such as Arab-American, Asian-Caucasian, Biracial, White-Hispanic, etc. Most of the participants reported that they were married (n = 321, 64.2%). Overall, 3.2% (n = 16) of the sample reported they were single and never married; another 1 1 .22% (n = 56) were single after a divorce or an annulment; approximately 1.0% (n = 5) were widowed and 1.2% (n = 6) reported they were widowed but remarried. Another 1.0% were in other long-term relationships (n = 5), 0.4% (n = 2) were separated, and 0.8% (n = 4) reported a domestic partnership. Most respondents had completed their master's degree (58.6%, ? = 293). Of all participants, 37% (n = 185) completed a doctoral degree; 27.5% (n = 138) completed a PhD; 4% (n = 20), 0.8% (n = 5), 0.4% (n = 2), 4.4% (n = 22) reported completion of an Ed.D., a Psy.D, an M.D., and a D.Min, respectively. "Other" was composed of Ed.S., Th.D., R.N., Advanced Degree in MFT, CAGS, and other. There were also two participants who indicated that they were doctoral candidates, and were collapsed into the master's degree participant group.

Overall, 20% of the therapists indicated that they used an integrative stance, followed by 16% who reported that they used a solution-focused perspective. Another 11% of the total respondents indicated they used systemic therapies, followed by 6.2% who stated they used experiential therapies. Approximately 9.25% reported they used social constructionist; 5.3% used structural; 4.9% used behavioral; 3.34% used a variety of other modalities; 4.133% used cognitive and/or cognitive behavioral; 2.76 reported they were eclectic, with the same amount reporting they used meta-frameworks; 3.74% identified themselves as psychodynamic; 3.15% reported they were strategic; and 0.59% reported they were influenced primarily by EFT.

Participants also reported their level of religiosity. Approximately 43.3% of the total number of participants (n - 215) reported that they were very religious; 40.73% (n = 202) reported they were somewhat religious, and the remaining 16.13% (n = 80) reported that they were not at all religious.

Procedures. The AAMFT clinical member participants completed an inventory with vignettes and Likert-scale items. We emailed potential participants a letter requesting their participation. In this letter, we requested that they read an informed consent form and respond to a short demographic questionnaire. Participants were instructed to go to a website (they were randomly sent to Website 1 or 2), read each of three Internet infidelity involvement vignettes, and respond to Likert-type statements immediately following each vignette. The vignettes in Website 2 were the same as in Website 1, except that the clients engaging in Internet infidelity were identified as male in Website 1 and females in Website 2. Half of the respondents were randomly assigned to Website 1, half to Website 2.

These three basic vignettes were similar to the ones Nelson (2000) provided in his Internet infidelity treatment study. They represent three of the most common Internet infidelity situations. In Vignette 1, one individual discovers that his or her partner has been exchanging flirtatious emails with someone else, which brings the couple to therapy. Vignette 2 described a situation where one partner met an individual online and began a physical relationship, including intercourse. As a result, this couple came to therapy. Finally, the couple in Vignette 3 came to therapy because one partner was caught downloading pornography by the other. The vignettes also were consistent with those identified by Whitty (2003), who found that people consider that there are three types of Internet infidelity situations - sexual, emotional, and pornography.

The assessment and treatment items to which the therapists responded included the following:

1. How serious is the presenting problem?

2. To what degree is the presenting problem normal?

3. How damaging is the problem to the relationship?

4. To what degree is the identified client a sex addict?

5. What is the prognosis for the marital relationship?

6. How many sessions would you estimate successful therapy to take?

7. What course of treatment would you take?

8. How likely would you be to focus on individual issues?

9. How likely would you be to focus on environmental issues?

10. How likely would you be to focus on relational issues?

11. How likely would you be to focus on crisis management?

Data analyses. Results for the Likert-scale items were statistically analyzed through t-tests, MANOVA, and regression. The dependent variables were the assessment decisions of the therapists (e.g., prognosis estimates and the perceived seriousness of the problem), while the primary independent variables included the gender of the client and the gender, age, and religiosity of participants. The Bonferroni test was used to minimize the possibility of alpha error.

Phase 2 Methodology

Participants. Phase 2 involved phone interviews with selected participants to get their perspective on the findings from Phase 1. Participants included eight clinical members of the AAMFT. We performed a stratified random sample in selecting five male and five female participants from those who had already completed the initial survey and had volunteered to take part in a follow-up phone interview. There were no restrictions as far as age, ethnicity, or any other factors other than gender, as previously mentioned.

Procedures. Each interview lasted approximately 30-45 min. We asked participants to elaborate on their thinking related to the findings as well as to their assessment, perceptions, and treatment decisions in Internet infidelity cases.

Data analysis. The interview data allowed us to provide rich illustrations of the participants' thinking behind various treatment decisions and perceptions of client behavior. The results for the interviews were analyzed qualitatively through analytic induction and constant comparison (Strauss & Corbin, 1990). We audiotaped the interviews with the participants. The audiotapes were transcribed. After transcription, we read the data and looked for themes in the statements by the therapists that emerged from the data. Once themes were identified, we looked for categories across the themes.

Achieving rigor. We attempted to assume the credibility of our qualitative findings in a number of ways, including our use of triangulation and member checks (Anfara, Brown, & Mangione, 2002). We compared interview responses to previously completed survey responses. We also shared our results with the participants (member checks) and asked for their reflections (Merriam, 1998). We used rich, thick description in presenting our findings to further support the credibility and transferability. We also outlined a clear, replicable audit trail around what we did and why we did it, again to support the rigor, credibility, and trustworthiness of our qualitative analyses and results.

RESULTS

Descriptive Information

Reported in Table 1 are the mean, standard deviations, and other descriptive information for assessment and treatment items for the vignettes.1

In general, family therapists did view male and female clients differently. For example, male clients who engaged in Internet infidelity behavior were more likely to be viewed by therapists as having a higher degree of sex addiction than their female counterparts. This finding was consistent across all three vignettes, t = 6.49, p = .000, t = 4.36, p = .000, and t = 3.25, p = .001 for Vignettes 1, 2, and 3, respectively. Therapists were also more likely to focus treatment individually for male identified clients when the presenting problem for the couple was viewing pornography, t = 3.306, p = .001. Likewise, female clients viewing pornography was rated as less typical, t = -3.349, p = .001.

To answer the question of whether assessment and treatment decisions made in cases of Internet infidelity vary when examined in terms of therapist social background variables, such as age, gender, religion, marital status, and whether they report they have been impacted by infidelity in their lives, a MANOVA was performed using the general linear model function via SPSS, considering the social background characteristics as covariates in the model (age, sex, religiosity, marital status, and impact of infidelity) and the 33 assessment and treatment items (11 per vignette, 3 vignettes) as the dependent variables.

Therapist's age. There were two variables across the three Internet infidelity vignettes in which age of the therapist appeared to be related to the assessment and treatment variables. While holding all other variables constant, age was significantly related to how many sessions were prescribed in Vignette 1, F= 8.206, p = .004, and the amount that treatment should focus on environmental issues in Vignette 2, F= 11.297, p = .001. That is, the younger the therapist, the fewer sessions the therapist would prescribe and the more likely he or she would be to focus on environmental issues in treatment.

We ran separate regression analyses for the above items. Results indicated that when considering the other social background characteristics, the younger the therapist, the more likely he or she would be to focus on environmental issues (t = -3.487, ? = .001), such as moving the computer to another room. Although significant, age only adds 3.1% or the variance explaining a therapist's decision to focus on environmental issues in Vignette 2 (see Table 2).

Therapist's gender. Gender of the therapist did influence assessment and treatment in a few circumstances. Specifically, in Vignette 1 (meeting online and having an affair), there were differences between male and female therapists in how much treatment should focus on connecting the present problem to larger processes within the couple's interaction (F = 8.254, p = .004). In Vignette 3 (about viewing pornography), the number of sessions prescribed also differed based on therapist gender (F = 9.346, p = .002), with male therapists prescribing fewer sessions than did female therapists.

After a separate regression analysis, we found female therapists were more likely to connect the presenting problem to larger processes within the couple's relationship when the presenting problem was that of a partner talking online to someone but not engaging in intercourse with that person, t = -2.556, ? = .011, R2 = .017 (see Table 3).

Therapist's religiosity. How religious therapists believed they were appeared to predict their assessment and treatment decisions. In Vignette 1 (meeting online and having an affair), the degree of therapist religiosity affected how sex addicted they believed the client was (F = 16.003, p = .000), as well as how much focus should be placed on environmental issues (F = 14.961, p = .000). Specifically, more religious therapists were more likely to label the behavior as an addiction and were also more likely to suggest changes in the environment, such as moving the computer from the basement into the living room. In Vignette 3 (the viewing of pornography), a therapist's religiosity predicted not only his or her assessment of the client as a sex addict (F = 20.811, p = .000; again, with the more religious therapists more likely to label the behavior as a sex addiction), but also the assessment of problem severity (F = 16.307, p = .000) and damage to the relationship (F = 13.579, p = .000, respectively), again, with the more religious therapists seeing higher severity and damage. Religiosity also affected the extent to which therapists focused on individual and environmental issues in treatment, such as removing the computer from the room (F = 9.464, p = .002, F = 20.204, p = .000, respectively). Again, the more religious the therapist, the more likely he or she was to advocate environmental changes.

We ran separate regression analyses to understand how therapists' religiosity affected their assessment and treatment of Internet infidelity cases. We found that reUgiosity of the therapist predicted in the same direction or above how serious therapists viewed both the online chatting and pornography viewing presenting problems, t = -4.062, p = .000, R^sup 2^ = .022, and t = 3.797, p = .000, R^sup 2^ = .029, respectively. Further, we found that therapists who view themselves as more religious place greater focus on individual issues when treating a couple presenting with the pornography issue, t = -3.384, p = .001, R^sup 2^ = .023. More religious therapists also placed a greater focus on environmental issues in treatment in cases where the presenting problems are with viewing pornography (t = -4.701, p = .000, R^sup 2^ = .043) and online chatting with someone other than one's spouse (t = -3.723, p = .000, R^sup 2^ = .027). Likewise, more religious therapists rate clients as having a greater degree of sex addiction when the presenting problems are either flirtatious online chatting with another, t = -4.4354, p = .000, R^sup 2^ = .039, or viewing pornography, t = -4.711, p = .000, R^sup 2^ = .045 (see Tables 4 and 5).

Therapist's own infidelity experiences. There were two assessment items over the three vignettes that appeared to be related to whether the responding therapists indicated that infidelity had affected them personally. In the first vignette, whether infidelity affected therapists personally influenced the degree to which they believed the presenting problem was typical (F = 5.048, p = .025). In the third vignette (related to viewing pornography), this experience related to how many sessions the therapist believed treatment in viewing pornography online would take, F = 3.939, p = .048. That is, those therapists with personal infidelity experience predicted fewer sessions.

Subsequent regression analyses indicate similarly, that therapists who reported their partner's infidelity as having a negative impact on them viewed the flirtatious chatting vignette as less typical, t = 2.529, p = .013, R^sup 2^ = .056 (see Table 6). Further, the more negatively a therapist rated his or her experience with his or her partner's infidelity, the less that therapist was likely to focus on individual issues for the flirtatious chatting presenting problem (t = -2.763, p = .007, R^sup 2^ = .071) and for the intercourse with an online partner vignette (t = -2.873, p = .005, R^sup 2^ = .078). The more negatively a therapist's own infidelity affected him or her, the less likely he or she was to report that the pornography vignette was typical, t = -2.274, p = .025, R^sup 2^ = .043.

Qualitative Findings

Characteristics of the Internet. Internet as a vehicle. Therapists described the Internet as another technology people can use to be unfaithful if they want to be. Places where people used to go to satisfy these needs might include strip clubs, purchasing pornographic material, etc. The advent of the Internet, they say, is one more way to achieve sexual gratification.

Accessibility. To engage in infidelity on the Internet was described by one therapist as "so easy" because of its accessibility. Another therapist exemplified this when he stated:

... the Internet issue is ... introducing things to couples sort of what I call the Pandora's Box syndrome, where now all of a sudden, things that were not possible are all of a sudden, readily accessible ... they never had that opportunity before because society in a sense would create a buffer ... it's easier to get more involved much quicker, much deeper ... Especially, for example ... things that men would have never even thought about seeing, you know, ... (they) wouldn't go to the seedy side of the city and sit and watch happen ... it's much broader in society ... (but now they can easily, and in private).

Therapists' Conceptualizations of Internet Infidelity

Betrayal. Betrayal was something that was central to the definitions of conventional and Internet infidelity. In part, betrayal was characterized by a breach of the marital contract. Although sexual contact with another was generally considered a breach of the marital contract, sexual contact with another person did not have to be in the picture for Internet behavior to be considered betrayal. One participant clarified this when he stated:

... infidelity happens if someone feels betrayed in a relationship ... I think that Internet infidelity is the same thing. And that ... if one party in a partnership feels betrayed because of someone else's Internet activity, flirtatiousness or beyond, then that's infidelity for them.

Online relationships and sexual behaviors stemming within these relationships also contain an element of secrecy (Cooper, Delmonico, & Burg, 2000a, 2000b). Cooper et al. (2000a, 2000b) found that individuals who kept their online activity secret from others also displayed more sexually compulsive behavior.

Betrayal also disrupts one's "couplehood." One participant exemplified this point when he stated:

But, it's not so much the sexual contact that makes any difference. I think it's the secrecy and betrayal of trust and the splitting of something called emotional specialness ... the betrayal of specialness is the pivotal point.

Nature of infidelity. The spectrum of Internet infidelity can include a wide variety of behaviors. On one end of the spectrum may be spending time on the computer rather than one's primary relationship, while at the other end may be the physical meeting and subsequent intercourse of two people who met online. Some behaviors that are considered infidelity by one couple may not be considered infidelity or problematic by another couple. One therapist stated:

... one spouse may consider chatting in a chat room with anybody infidelity, where the other person may not have that script at all as infidelity. One partner, for example, may see presenting sexual play on the Internet as infidelity where the other one just considers it no different than buying a Playboy magazine ... what have been the boundaries or understandings, either spoken or unspoken about ... sexual context or content in the relationship?

Deficit model. Shaw's (1997) model for the treatment of Internet infidelity outlines several factors that make couples more vulnerable to the event of Internet infidelity. One of the vulnerabilities she identifies is a lack of connection with another partner or lack of ability to communicate about problems in the relationship. Additionally, Cooper, McLoughlin, and Campbell (2000) assert that an individual's use of cybersex can, in some cases, be a symptom of a deeper relational problem, often relating to intimacy, dependency, and a feeling of abandonment.

The participants interviewed in the present study also consistently identified this vulnerability factor during the interviews. Overall, therapists saw Internet infidelity as a symptom of a larger problem in the primary couple's relationship. Areas of symptomology that therapists referred to in the present study include a breakdown in communication, a breakdown in boundaries, and a breakdown in commitment. In understanding infidelity in this manner, therapists often included a piece in their treatment that addressed what "led up" to the affair. One therapist describing his treatment of infidelity cases stated: "Then we work on boundaries, what was missing, what was needed, how to change their way of communication so that they can begin to look at what each other's needs are and try to meet them." Specific to Internet infidelity treatment, he later stated: "... we work on trying to figure out what need he felt . . . and if his wife can't meet it, so the Internet is trying to meet it." Another therapist, in discussing her treatment strategies, indicated that romance might be missing in the primary relationship: "(It is important to) explore with the wife what she was getting from those conversations that she wasn't getting from her everyday ordinary existence."

Such strategies are consistent with the treatment procedures outlined by Young (2006). Specifically, Young describes how to improve communication within the relationship, including using "I" language, and empathetic listening. She also suggests addressing the underlying issues (similar to the deficit model that many therapists cited in the present study) and addressing any rationalizations.

"Stepping out of the relationship." Therapists identified the nature of Internet infidelity as one in which an individual "stepped out" of the relationship. This terminology was similar in many of the interviews. "Stepping out" could be emotional, physical, or a combination of both.

DISCUSSION

The Internet adds new dimension to the assessment and treatment of infidelity cases. The results of the present study illustrate that therapists who are more religious tended to view Internet infidelity as that which might require a greater individual focus. Age and gender of the therapists also affected how therapists treated cases, with younger therapists more likely to alter environmental issues and female therapists more likely to focus on couple processes in treatment. Further, therapists also viewed men as more likely to be sex addicted, while women engaging in the same behavior were viewed as atypical. Therapists also made differential decisions in treatment based on client needs, client gender, client health, and their personal theoretical orientation.

Therapists' Biases in Assessment and Treatment of Internet Infidelity

As Stabb et al. (1997) report, differential treatment is an area of concern in clinical work, as biases can affect how one develops treatment strategies for the presenting problem. Of the areas in which there were differences in treatment for Internet infidelity cases, more often these differences were reflected in treatment as therapists prescribed further emphasis on the individual rather than systemic factors, particularly when they saw Internet infidelity more pathologically (as an addiction, or as atypical, or as more severe).

Although Stabb et al. (1997) speak to the problem of differential treatment, they do not mention differential assessment. There were several instances in the present study where therapists demonstrated differences in their assessment, but these differences in assessment did not appear to translate into differentiated treatment.

Throughout this investigation, there were times in which therapists' social background characteristics influenced their assessment but not their treatment, influenced their treatment without influencing their assessment, or influenced their treatment and their assessment. This leads one to be curious about the degree to which assessment and treatment are practically linked when a therapist treats Internet infidelity. For example, the gender of the client, the gender of the therapist, and a therapist's age all were consistently related to differences in treatment, but not in assessment items. In the case of religiosity and the impact of personal infidelity experience of a therapist, there tended to be more assessment differences, but this did not necessarily translate into treatment. Future research may seek to understand under what circumstances treatment is (or should be) modified based on differential assessment.

As no agreed-upon assessment or treatment protocol exists for Internet infidelity, the current study points to assessment and decisions varying based on gender, religion, and even age biases. While Young, Cooper, Griffin-Shelley, O'Mara, and Buchanan (2000) proposed the ACE model for cybersex addiction is also a valuable model in treating online infidelity, the therapists in this study did not cite this model when discussing their treatment strategies. This can be seen as either disturbing or enlightening. From a worst-case perspective, the reader could conclude that therapist bias is alive and well, and may be playing too large a role in assessment and treatment decisions. On the other hand, we all know that our own socializations inform what we see and what we do in therapy. The confirmation of this fact in the present study can serve as a cautionary tale. A therapist should reflect on his or her own invisible beliefs, biases, and therapeutic guideposts. Perhaps the results of this study will support open discussion regarding our therapeutic differences where they arise, and which ones we should rethink. Further, it is possible that those who did not participate in the survey would have responded differently from those who did respond. For example, those who did not respond to the Internet survey may be those who feel less competent in their computer skills. Therefore, it is possible that the therapists who participated were qualitatively different from those who did not respond in their comfort level with technology, thus influencing the findings.

While the central issue in this research was Internet infidelity treatment, we cannot underestimate the importance of the operation of biases in therapy. One issue about biases in treatment is that they operate outside of the therapist's awareness. As aforementioned, those factors which operate outside of our awareness can be difficult to manage because we fail to realize they are there. Another issue is the therapist may be actively mindful of potential biases in the areas of diversity, gender, and age. This investigation demonstrates that biases which affect treatment are not restricted to the aforementioned large issues, but instead can come from small and subtle aspects of a therapist's life, such as his or her experience with technology, personal relationship history, or personal beliefs about the impact of the Internet on one's sexual behavior. Specific suggestions on how to address biases in therapy include seeking out and receiving regular supervision, particularly from a supervisor who is different enough from the supervisee to call attention to more of the subtle biases that may be operating in treatment. Another suggestion is to have the therapist identify ways in which his or her background can enhance the therapeutic process rather than detract from it. For example, therapists who have less experience with a computer are encouraged to use this as a way to join with their client and have their client teach them about the Internet rather than considering individuals who engage in online sexual behavior as more of an addict than other clients who engage in sexual behavior without the Internet.

The Complexity of Internet Infidelity

Hertlein and Piercy (2005) suggest that what might be behind some of the difference in treatment is that therapists find it difficult to define what exactly Internet infideUty is. As found in the present investigation, some couple and family therapists consider sex a primary criterion in the definition, while others believe that emotional intimacy defines Internet infidelity. Secrecy may also be seen as a critical component. Likewise, many also consider a breach of trust part of the definition of Internet infidelity. These observations are consistent with those made by traditional infidelity authors, such as Lusterman (1998) and Thompson (1984). Internet infidelity treatment might better be served by a framework that integrates traditional infidelity treatment frameworks and strategies which are specific to the Internet (e.g., Delmonico, Griffin, & Carnes, 2002).

The Changing Definition of Infidelity

Infidelity can be viewed on a continuum. In the past, infidelity was typically viewed as when one person engaged in intercourse with another person other than his or her partner (Johnson, 1972; Thompson, 1983). The Internet has demonstrated that communication can take many forms, some of which can violate the bounds of a couple's contract with one another. Some of the characteristics that Henline and Lamke (2003) found that infidelity behavior included are cybersex, sexual chatting, online dating/plans to meet online/emotional involvement with online contact, sexual interactions/flirting, betraying confidences of one's partner, and keeping secrets from one's partner. Many of the therapists in our follow-up qualitative interviews also beUeved that Internet infidelity combined many of these components. Likewise, they considered the definition of Internet infidelity to include any other relationship that depletes the primary dyad of energy and is facilitated by Internet use. This can include websites, online chat rooms, instant messengers, and other tools for sexual gratification accessed through computers and/or the Internet. Cooper and Griffin-Shelley (2002) define online sexual activities as

any activity (including text, audio, and graphic files) that involves sexuality for purposes of recreation entertainment, exploration, support, education, commerce, efforts to attain or secure sexual or romantic partners, and so on. (p. 3)

These activities can include downloading material, purchasing sexual material, sexually explicit conversations, and others (Cooper, Morahan-Martin, Mathy, & Maheu, 2002). Cooper, Griffin-Shelley, Delmonico, and Mathy (2001) performed a study of online sexual activities and found that most people who engage in online sexual activities do not report problems with their behavior. Those who did experience online sexual problems were more likely to be single and spent twice the time on the Internet as the sample. Further, those who were experiencing onUne sexual problems also reported less frequent sexual activities with their partner. Clearly, time on the Internet and reduced interaction with one's partner may be important variables in whether problems develop.

Of course, the definition of Internet infidelity will vary from couple to couple. The present study did not inquire as to whether the couple had their own definition of infidelity; instead, the therapists who were interviewed reported that they allow the couple to define it, but did not provide examples of what that definition might be. As each client defines infidelity in his or her own way, the assessment and treatment may be adjusted to best fit each client.

DIRECTIONS FOR FUTURE RESEARCH

Researchers may want to further explore other circumstances in which assessment and treatment decisions are more vulnerable to social background characteristics. Social background characteristics that were biological (e.g., gender and age) in nature appeared to relate to treatment, whereas those characteristics that were psychological in nature (e.g., religiosity) were more likely to influence the assessments. Although this tentative result is based on the patterns from the present study, it may be an element of consideration for future researchers when exploring the impact of social background characteristics on treatment of infidelity cases. Future studies might also add a fourth vignette to depict masturbatory behavior as part of the infidelity. Another interesting question is whether the results would still hold up if compared with a control group given the same questionnaire and interview around three nonsexual issues. While the topic of the present investigation was sexual in nature, others interested in biases in therapy can replicate the study without the sexual content and draw appropriate comparisons.

There were several instances where more differences existed in assessment and treatment in the pornography vignettes than in the other two vignettes involving online communication. Another question for research might include the differences in the assessment and treatment of viewing pornography versus more interpersonally interactive behaviors. Additionally, future research might seek to understand under what circumstances Internet infidelity might be acceptable in the view of therapists and what impact this perception might have on treatment. Finally, future researchers might be interested in identifying the differences in how therapists see individual and relational treatment, using this information to further treatment frameworks related to the treatment of Internet infidelity.

[Reference]

REFERENCES

Anfara, V. A., Brown, K. M., & Mangione, T. L. (2002). Qualitative analysis on stage: Making the research process more public. Educational Researcher, 31, 28-38.

Aponte, H. J. (1985). The negotiation of values in therapy. Family Process, 24, 323-338.

Atkins, D. C., Baucom, D. H., & Jacobson, N. S. (2001). Understanding infidelity: Correlates in a national random sample. Journal of Family Psychology, 15, 735-749.

Atwood, J. D., & Schwartz, L. (2002). Cyber-sex: The new affair treatment considerations. Journal of Couple and Relationship Therapy, 1, 37-56.

Blow, A., & Hartnett, K. (2005). Infidelity in long-term committed relationships II: A substantive review. Journal of Marital and Family Therapy, 31, 217-233.

Boies, S. C., Cooper, A., & Osborne, C. (2004). Variations in Internet-related problems and psychosocial functioning in online sexual activities: Implications for social and sexual development of young adults. CyberPsychology & Behavior, 7, 207-230.

Buunk, B. P., & Bakker, A. B. (1995). Extradyadic sex: The role of descriptive and injunctive norms. Journal of Sex Research, 32, 313-318.

Cooper, A. (1997). The Internet and sexuality: Into the next millennium. Journal of Sex Education and Therapy, 22, 5-6.

Cooper, A., Delmonico, D. L., & Burg, R. (2000a). Cybersex users and abusers: New findings and implications. Sexual Addiction and Compulsivity: The Journal of Treatment and Prevention, 7, 5-29.

Cooper, A., Delmonico, D. L., & Burg, R. (2000b). Cybersex users, abusers, and compulsives: New findings and implications. In A. Cooper (Ed.), Cybersex: The dark side of the force (pp. 5-29). New York: Brunner-Routledge.

Cooper, A., & Griffin-Shelley, E. (2002). Introduction. The Internet: The next sexual revolution. In A. Cooper (Ed.), Sex and the Internet: A guidebook for clinicians (pp. 1-15). New York: Brunner-Routledge.

Cooper, A., Griffin-Shelley, E., Delmonico, D. L., & Mathy, R. M. (2001). Online sexual problems: Assessment and predictive variables. Sexual Addiction and Compulsivity, 8, 267-285.

Cooper, A., McLoughlin, I. P., & Campbell, K. M. (2000). Sexuality in cyberspace: Update for the 21st century. CyberPsychology & Behavior, 3, 521-536.

Cooper, A., Morahan-Martin, J., Mathy, R. M., & Maheu, M. (2002). Toward an increased understanding of user demographics in online sexual activity. Journal of Sex and Marital Therapy, 28, 105-129.

Cooper, A., Scherer, C., & Mathy, R. M. (2001). Overcoming methodological concerns in the investigation of online sexual activities. CyberPsychology & Behavior, 4, 437-447.

Cooper, A., & Sportolari, L. (1997). Romance in cyberspace: Understanding on-line attraction. Journal of Sex Education and Therapy, 22, 7-14.

Delmonico, D. L., Griffin, E., & Carnes, P. J. (2002). Treating online compulsive sexual behavior: When cybersex is the drug of choice. In A. Cooper (Ed.), Sex and the Internet: A guidebook for clinicians (pp. 147-167). New York: Brunner-Routledge.

Dillman, D. A. (2000). Mail and Internet surveys: The tailored design methods (2nd ed.). New York: Wiley.

Erotic Chat Grounds for a Divorce? (2005). Retrieved September 28, 2005, from http://news.yahoo.eom/s/nm/ 20050928/od_nm/belgium_adultery_dc

Glass, S. P. (2001). The trauma of infidelity: Research-based treatment. Paper presented at the annual meeting of the American Association for Marriage and Family Therapy, Nashville, TN.

Greenfield, D. N. (1999). Virtual addiction: Sometimes new technology can create problems. Retrieved September 28, 2005, from http://www.virtual-addiction.com/pdf/nature_internet_addiction.pdf

Hatala, M. N., Milewski, K., & Baack, D. W. (1999). Downloading love: A content analysis of Internet personal ads placed by college students. College Student Journal, 33, 124-129.

Hecker, L. L., Trepper, T. S., Wetchler, J. L., & Fontaine, K. L. (1995). The influence of therapist values, religiosity and gender in the initial assessment of sexual addiction by family therapists. American Journal of Family Therapy, 23, 261-272.

Henline, B. H., & Lamke, L. K. (2003). The experience of sexual and emotional online infidelity. Poster presented at the 65th annual conference of the National Council on Family Relations, Vancouver, BC.

Hertlein, K. M., & Piercy, F. P. (2005). A theoretical framework for defining, understanding, and treating Internet infidelity. Journal of Couple and Relationship Therapy, 4, 79-91.

Hertlein, K. M., & Piercy, F. P. (2006). Internet infidelity: A critical review of the literature. The Family Journal, 14, 366-371.

Hertlein, K. M., & Webster, M. (2008). Technology, relationships, and problems: A research synthesis. Journal of Marital and Family Therapy, 34, 445-460.

Johnson, R. E. (1972). Attitudes toward extramarital relationships. Medical Aspects of Human Sexuality, 6, 168-191.

Korner, S., & Goldberg, E. H. (1996). Gender and the therapist-patient matrix: Gender bias or gender sensitivity? Psychotherapy in Private Practice, 15, 33-52.

Liu, C. (2000). A theory on marital sexual life. Journal of Marriage and the Family, 62, 363-374.

Lusterman, D. (1998). Infidelity: A survival guide. New York: MJF Books.

Mantovani, F. (2001). Networked seduction: A test-bed for the study of strategic communication on the Internet. CyberPsychology & Behavior, 4, 147-154.

McCown, J. A., Fischer, D., Page, R., & Homant, M. (2001). Internet relationships: People who meet people. CyberPsychology & Behavior, 4, 593-596.

Merkle, E. R., & Richardson, R. A. (2000). Digital dating and virtual relating: Conceptualizing computer mediated romantic relationships. Family Relations, 49, 187-192.

Merriam, S. B. (1998). Qualitative research and case study applications in education. San Francisco: Jossey-Bass.

Nelson, T. S. (2000). Internet Infidelity: A modified Delphi study. Unpublished doctoral dissertation, Purdue University.

Parks, M. R., & Roberts, L. D. (1996). 'Making MOOsic': The development of personal relationships on line and a comparison to their off-line counterparts. Journal of Social and Personal Relationships, 15, 517-537.

Pew Internet and American Life Project. (2006). Online dating. Retrieved August 20, 2008, from http://www. pewinternet.org/pdfs/PIP_Online_Dating.pdf

Schnarch, D. (1997). Sex, intimacy, and the Internet. Journal of Sex Education and Therapy, 22, 15-20.

Seal, D. W., Agostinelli, G., & Hannett, C. A. (1994). Extradyadic romantic involvement: Moderating effects of sociosexuality and gender. Sex Roles, 31, 1-22.

Shaw, J. (1997). Treatment rationale for Internet infidelity. Journal of Sex Education and Therapy, 22, 29-34.

Spink, A., Koricich, A., Jansen, B. J., & Cole, C. (2004). Sexual information seeking on web search engines. CyberPsychology & Behavior, 7, 65-72.

Stabb, S. D., Cox, D. L., & Harber, J. L. (1997). Gender-related therapist attributions in couples therapy: A preliminary multiple case study investigation. Journal of Marital and Family Therapy, 23, 335-346.

Strauss, A., & Corbin, J. M. (1990). Basici of qualitative research: Grounded theory procedures and techniques. Thousand Oaks, CA: Sage.

Thompson, A. P. (1983). Extramarital sex: A review of the research literature. Journal of Sex Research, 19, 1-22.

Thompson, A. P. (1984). Emotional and sexual components of extramarital relations. Journal of Marriage and the Family, 46, 35-42.

Treas, J., & Giesen, D. (2000). Sexual infidelity among married and cohabitating Americans. Journal of Marriage and the Family, 62, 48-60.

Underwood, H., & Findlay, B. (2004). Internet relationships and their impact on primary relationships. Behaviour Change, 21, 127-140.

U.S. Department of Commerce. (2002). A nation online: How Americans are expanding their use of the Internet. Retrieved September 28, 2005, from http://www.ntia.doc.gov/ntiahome/dn/nationonline_020502.htm

Whitty, M. T. (2003). Pushing the wrong buttons: Men's and women's attitudes toward online and offline infidelity. CyberPsychology & Behavior, 6, 569-579.

Whitty, M. T. (2005). The realness of cybercheating: Men's and women's representations of unfaithful Internet relationships. Social Science Computer Review, 23, 57-67.

Wyatt, G. E., Peters, S. D., & Guthrie, D. (1998a). Kinsey revisited, part I: Comparisons of the sexual socialization and the sexual behavior of white women over 33 years. Archives of Sexual Behavior, 17, 201-239.

Wyatt, G. E., Peters, S. D., & Guthrie, D. (1998b). Kinsey revisited, part II: Comparisons of the sexual socialization and the sexual behavior of black women over 33 years. Archives of Sexual Behavior, 17, 289-332.

Young, K. S. (2006). Online infidelity: Evaluation and treatment implications. Journal of Couple and Relationship Therapy, 5, 43-56.

Young, K. S., Cooper, A., Griffin-Shelley, E., O'Mara, J., & Buchanan, J. (2000). Cybersex and online infidelity: Implications for evaluation and treatment. Sexual Addiction and Compulsivity, 7, 59-74.

[Author Affiliation]

Katherine M. Hertlein

University of Nevada

Fred P. Piercy

Virginia Polytechnic Institute and State University

[Author Affiliation]

Katherine M. Hertlein, PhD, is an assistant professor in the Department of Marriage and Family Therapy, University of Nevada, Las Vegas; Fred P. Piercy, PhD, is a professor of family therapy and head of the Department of Human Development, Virginia Tech.

We acknowledge Megan Webster for her assistance in preparation of this article and dedicate this article to Al Cooper and Shirley Glass. This research was supported by the AAMFT Graduate Student Research Award and the James D. Moran III Dissertation Award.

Address correspondence to Katherine M. Hertlein, Department of Marriage and Family Therapy, University of Nevada, Las Vegas, 4505 Maryland Pkwy., Box 453045, Las Vegas, Nevada 89154-3045; E-mail: katherine.hertlein@unlv.edu