"So why did you do it?": Explanations provided by Child Pornography Offenders

Hannah L. Merdian¹, Nick Wilson², Jo Thakker¹, Cate Curtis¹, Doug P. Boer^1
1 University of Waikato, Hamilton, New Zealand
² Department of Corrections, Hamilton, New Zealand

[Sexual Offender Treatment, Volume 8 (2013), Issue 1]

Abstract

Aim/ Background: There has been some professional discussion surrounding the "function" of child pornography offending for the individual offender: Why is someone looking at child pornography and what needs are met with this behaviour? The limited research in this area has mainly focused on information extracted from interview transcripts, which is likely biased by the offender's situation at the time.
Material/Method: Using a computerised online survey, child pornography offenders (with and without contact sex offences against minors) were questioned anonymously about their motivations to start viewing child pornography material.
Results: Thematic Analysis lead to the development of a topical map that revealed four main themes in their offence motivations: (1) no (direct) explanation provided, (2) initial triggers of their child pornography consumption, (3) emotional reasons, and (4) sexual reasons. These themes were analysed with regards to their subthemes, their relationship with each other and to the offending behaviour. Offender characteristics differed between the thematic groups. In addition, "pure" child pornography offenders were more likely than offenders with contact victims to provide more than one explanation for their behaviour. Individuals with contact offences were found more likely to admit a sexual interest in children.
Conclusions: These findings point to the value of a motivation-based offender typology. For treatment providers, the assessment of offence motivations appears as a promising source about an individual's risks and needs.

Key words: child pornography, child sexual exploitation material, sex offending, offence motivations, offender typology, cognitive distortions

Accounts of individuals using the internet to sexually offend have become increasingly prevalent. In this paper, the focus is on men who have engaged in the viewing, possession, distribution, or trading of online child sexual exploitation material, more commonly referred to as child pornography (CP). One of the most prevalent questions in the work with child pornography offenders (CPOs) is the assessment of risk, predominantly their proclivity to cross-over to contact sex offending (Webb, Craissati, & Keen, 2007). To date, it appears that only a subgroup of CPOs will continue to engage in contact sex offending; in the most recent meta-analysis including a combined sample of 2,630 online offenders, Seto, Hanson, and Babchishin (2011) reported that 3.4% of online offenders were found to reoffend with another CP offence, while only 2% reoffended with a contact sex offence. Given the majority of CPOs thus appear to present as low risk, it makes conceptual and economic sense to focus resources on the identification and treatment of this offender subgroup with the highest risk to commit a contact offence.

Taxonomic approaches to broaden the existing knowledge base and to identify assessment and treatment needs of an offender type are conceptually well grounded in the research tradition on offender populations. In the Handbook of Sexual Assault, Knight and Prentky (1990) stated that "understanding the taxonomic structure of a deviant population is the keystone of theory building and the cornerstone of intervention" (p. 23). Indeed, their typology for adult sex offenders, the Massachusetts Treatment Centre Typology, has found wide application in the treatment of rapists in terms of the identification of specific needs and risks of each offender type (Reid, Wilson, & Boer, 2011). Based on existing CPO typologies (e.g., Carr, 2006; Hartmann, Burgess, & Lanning; 1984; Krone, 2004, 2005; McLaughlin, 2000), Merdian, Curtis, Thakker, Wilson, and Boer (2011) developed a conceptual classification model for the assessment of CPOs, pointing to the individual's offence motivation as a distinguishing factor in the assessment process, stating that understanding an offender's motive (or motives) can provide insight into the needs he intends to meet with his behaviour and thus the meaning of CP for the individual.

The meaning an individual attributes to his CP consumption has been described elsewhere as functions of CP (for example, see Taylor & Quayle, 2003). In a series of interviews with convicted consumers of online CP, Taylor and Quayle identified six principal functions of child pornographic material: Whereas the majority of offenders had viewed the images for sexual arousal, some users reportedly had gained satisfaction from the actual collection process rather than the content of the images. Others stated that they had used the images mainly to foster online social contacts with other adults interested in child sexual abuse material. Some participants reported that these activities were means of escaping their real life problems. Finally, a few users conceptualised CP as a form of "therapy" that had had allowed them to explore their sexual preferences but reportedly prevented them from progressing to contact child sexual abuse. The six functions originally identified by Taylor and Quayle guided other research projects (e.g., Caple, 2008; Sheldon & Howitt, 2007; Surjadi, Bullens, van Horn, & Bogaerts, 2010), and additional motivations have been identified (see Table 1). In a recent study, Seto, Reeves, and Jung (2010) screened transcripts from police and clinical interviews conducted with a sample of 84 CPOs, analysing the explanations they had provided for their offending. Again, they found a variety of motives: While most people in both samples reported to be motivated by a sexual interest in children (46% in the police-sourced and 38% in the clinical sample), the offenders also named accidental access, curiosity, or addiction to pornographic material as reasons for their behaviour and, less frequently, indiscriminate sexual interests, "internet addiction", or their general interest in collecting activities. Six percent in each sample reported that they had used CP as a substitute for contact offending. It was noted that 36% in the police sample and 68% in the clinical sample provided three or more motives for their behaviour. Seto et al. acknowledged the offenders' situation and the interviewers' influence as potential biases to the study. Nevertheless, based on the outcomes of these studies, it seems valid to conclude that CPOs as a group display a wide variety of motivations for their behaviour and that the individual offender may have more than one explanation for his behaviour. It is further acknowledged that the functions of CP for the individual may change over time (Seto et al., 2010; Surjadi et al., 2010), requiring specific offender management strategies according to the functions they assign to the material.

The existing research has provided an overview of the potential functions of CP offending and, due to its mostly qualitative nature, aimed to present veridical accounts of the offenders' individual experience. However, one shortcoming of the current research body is that the information is mostly drawn from interviews with the offenders. This approach is limited for at least two reasons. Firstly, the offenders' responses may be biased due to their current situation; For example, a pre-conviction police interview is likely to result in different responses from an individual than a post-conviction clinical assessment, with the former possibly being further influenced by third parties, such as lawyers. Secondly, every interview consists of an interactive exchange between the interviewee and interviewer and its outcomes are partly dependent on the interviewer's prompting and note-taking, as well as the interviewee's openness and verbal fluency. A more impartial account of the offender's explanations for their offending has so far been missing from the current research. The current research project was aimed at compensating these limitations and deducing offenders' potential motives from anonymous accounts. The central research question of this study was summarised into one simple, open-ended question: "Why do you think you started viewing child pornography?".

Data were collected via an anonymous computer survey; this medium was expected to strengthen the research findings in three ways. Firstly, an anonymous, non-interactive data collection process was employed to compensate for the limitations identified in interview-based research. Secondly, it allowed offenders to skip the research question, hence only participants willing to provide an explanation for their offending were assumed to have responded. Finally, participants were asked for their immediate thoughts to a question that was purposefully kept simple, which was intended to reduce cognitive efforts and thus attempts for impression management.

This study was integrated into a larger project aimed to compare CPOs, contact child sex offenders, and offenders with both offence types in areas identified as critical for an assessment of risk. The research question for the current study was presented only to individuals who admitted to CP consumption.

The main aim of this study was to deduce the various motivations for CP offending based on short narratives provided by the offenders. Resulting from the above review of the existing literature, two main confirmatory research hypotheses were identified:

1. As a group, offenders present a range of reasons for their CP offending, with at least one of whom being related to sexual satisfaction.
2. As individuals, at least some of the offenders present more than one explanation for their offending.

The current research design also allowed for the analysis of a number of risk-related items from the complete data corpus in conjunction with these statements, including demographics or criminal history. Thus, a third, investigative research hypothesis was added:

3. Systematic relationships between offence motivation and risk-related constructs can be identified.

Method

Participant Recruitment

Individuals were eligible for this study if they were of at least 18 years of age, were of male gender, had a sufficient understanding of English reading and writing, had no intellectual impairment that affected the person's ability to make an informed decision about participation and to understand the test material. Participants were recruited from both community sex offender treatment centres and prison settings throughout New Zealand.

Procedure and Stimulus Material

Each participant was asked to respond to a survey on portable computers. The participants were self-guided through the questions while a bar on the side of the screen indicated their progress through the survey. Overall, the survey consisted of 211 items; the current research question was presented towards the the end of the survey.

Data Analysis

A mixed-method analysis was employed for this study. Given the small number of participants and the uneven sub-sample sizes, descriptive information was analysed using non-parametric methods of group comparisons. For the two confirmatory research questions, a qualitative approach using thematic analysis (TA) was employed. Derived from content analysis, TA is a well-established method to identify both apparent and latent themes in a text (Joffe, 2011). As Fereday and Muir-Cochrane (2006) summarised, TA is "is a form of pattern recognition within the data, where emerging themes become the categories for analysis" (p. 82). TA is also appropriate for rather limited narratives, such as open-ended responses to questionnaire items (see Saunders & Byrne, 2002). A number of systematic guidelines for undertaking TA are available (e.g., see Attride-Stirling, 2001; Joffe, 2011). The current study followed the step-by-step guide presented by Braun and Clarke (2006), describing six phases of TA: (1) Familarising with data, (2) Generating initial codes, (3) Identifying themes, (4) Reviewing themes, (5) Defining and naming themes, and (6) Producing the report. As part of stage four, an independent rater was involved in coding the data and any differences were discussed until agreement was reached.

For the exploratory research question, a descriptive analysis of the relationship between the identified themes and the following variables was conducted: level of education, level of perceived personal stress, non-sexual criminal history, length of treatment attended to date, consumption of deviant pornography other than CP, sexual interest in minors (including contact offences, online grooming of minors, visit to child-lover websites, victim preferences in CP material, declared sexual arousal to CP), online networking with other offenders, length of CP consumption, type, content, and range of CP material, and time investment in CP. In addition, participants completed the Abel and Becker Cognition Scale (ABCS; Abel, Becker, Cunningham-Rathner, Rouleau, Kaplan, & Reich, 1984) and some selected items from Children & Sexual Activities (Howitt & Sheldon, 2007). Thus, the endorsement of cognitive distortions according to these scales defined the last variable tested in its relationship to the explanations provided.

Results

Participants

Overall, the research question was presented to 22 CPOs and 17 offenders who also had a contact sex offence with a child victim (mixed offenders; MOs). Personal information about the participants is summarised in Table 2. Overall, the participant groups appeared fairly similar across all variables, with the exception of four items. CPOs were significantly more likely than MOs to have ever been diagnosed with a mental health problem (50% vs. 17.6%; Fisher's Exact Test, p(1-sided) > .05) and there was a trend for CPOs to have been more likely to have viewed pornography displaying acts involving urination /defecation (86.4% vs. 58.8%; Fisher's Exact Test, p(1-sided) = .057) and to have engaged in online offending, such as illegal downloading (59% vs. 52.9%; Fisher's Exact Test, p(1-sided) = .056). On the other hand, MOs were highly significantly more likely to be in prison at the time of data collection (4.6% vs. 58.8%; Fisher's Exact Test, p(1-sided) = .00) and reported significantly more convictions as part of their criminal history (Mdn = 1 vs. Mdn = 2 across 5 possible areas; U = 114, z = -2.215; p(1-sided) < 0. 5; r = -.35), mainly based on a significant difference in their usage of weapons (27% vs. 44.8%; Fisher's Exact Test, p(1-sided) <.05). The main focus of this study was to deduce the various motivations for CP offending based on short narratives provided by the offenders.

Range of Explanations provided for CP Offending

It was hypothesised that as a group, offenders present a range of reasons for their CP offending, with at least one of whom being related to sexual satisfaction. Figure 1 depicts the topical map of offence motivations; it includes a separate thematic tree, referring to those offenders who did not provide an explanation for their offending (n = 7). Although only three participants actually withheld a response, others provided a reply failing to answer the question. For example, Case 5447 stated: "Stress leading to depression leading to failed relationship leading to more stress and anxiety and increased depression", missing a causal link back to the research question. Case 5351's response stood out - rather than explaining his reasons for viewing CP, he described his reasons for not continuing to view: "i only looked on one occassion and never lookoked again as i was too scared of detection (…) i didn't get hooked but could have easily [sic]".

Figure 1: Thematic map of offenders' explanations for their child pornography usage, resulting from thematic analysis following the guidelines from Braun and Clarke (2006)

Overall, three main themes emerged during the analysis of the offenders' responses: (1) Emotional explanations, (2) Sexual explanations, and (3) Explanations referring to initial triggers of the behaviour. Figure 2 depicts the themes according to the frequency of their endorsement. Overall, the range of themes presented confirms the first hypothesis.

Figure 2: Thematic map of offense motivations, weighted according to frequency of their appearance. Larger circles represent higher rates of being named by the participants.

Initial trigger

Within this category, participants explained why they had initially viewed CP but failed to report what motivated them to continue. Four participants claimed accidental exposure ("found it by accident when looking for other porn", Case 5098; "by pure chance accessed a site which contain [sic] it", Case 5248) while five participants referred to "curiosity" ("I was curious and just wanted to have a look.", Case 5351). It appeared that for some participants, curiosity was the mediating factor between initial exposure and a sexual interest in minors that developed or manifested itself with prolonged exposure while others seemed to use it as a synonym for a paedophilic interest. For example, Case 5224 explained: "Curiosity to see the development of the young girl into the young woman.i [sic] collected photo's of girls posing on their own."

Sexual explanation

Two subthemes were identified, concerning one's sexual identity and progression from legal material.

Sexual identity. The offenders' responses concerning sexual identity appeared to belong to two distinct groups. For one group of the offenders (n = 7), CP seemed to be a replacement for an adequate sexual object, not limited to the depicted content; as Case 5443 described, "Being sexually impotent lead me to search out new means of attaining an erection, like most visual subjects I viewed this did not help me at all". For three offenders, CP consumption was based specifically on the lack of an appropriate adult sex partner-as Case 5164 described: "Lack of sexual experience and the belief that children because of their own lack of experience wouldn't reject me."

For the majority of offenders amongst all subgroups (n = 13), CP consumption was based on their sexual attraction to minors: "my sexual attraction was only towards children at the time" (Case 5101).

Progression from legal material. For nine participants, their CP offending appeared to be the result of prolonged exposure and potential desensitation to legal pornography. Some participants provided fairly detailed responses of their journey:

The gradual escalation from normal adult material to more extreme material(dehumanising) after first accessing the internet,  that I used it to cope with emotional and stressful situations. Followed by viewing younger and younger woman, girls and preteen, i.e. child modeling [sic] and cartoons showing extreme adult and other abusive subject matter. (Case 5164)

Again, some of the responses clearly linked back to a developing sexual interest in children, based on increasing exposure to the material.

Emotional explanation

A group of offenders pointed to their emotional rather than sexual experience when viewing CP, mostly using CP as a way of making themselves feel better.

Detached/ passive emotions. Case 5092 described how the financial need to support his drug usage overpowered any reflective thoughts on the material he was dealing with.

Positive emotions. The majority of respondents (n = 15) referred to the positive emotions that CP consumption elicited in them; viewing CP became a source of relief and a means to escape negativity, resulting from depression (e.g., Case 5096, Case 5447), anxiety or stress (e.g., Case 5099, Case 5162, Case 5250, Case 5164), sexual frustration (Case 5452), and emotional drought ("loneliness, wish to be loved." Case 5099). For some offenders, the positive side of CP was in the shock-value of the depicted material. For instance, Case 5102 stated: "long story but i needed a distraction and nothing else worked cos I was not too bothered about it i.e. booze. CP was, in my view, very disturbing and t acted as the best distraction [sic]".

For others, the viewing process went beyond the relief aspect, triggering feelings of "being in control". Case 5098 described his experience:

Shocked initially but then was intrigued enough to go back. I think I carried on as a way to control some positive/ exciting feelings in a world that I felt I was drowning in due to stress from relationships, work, aprenting [sic] etc. – essentially I wasnt [sic] coping and this was a means to receive a positive feeling I could control.

The feeling of being in control or being powerful was crucial to the experience of many offenders. Some offenders explicitly referred to their own sexual abuse story; Case 5093 stated: "it made me feel good to see it happen to someone els it remined me on how it felt when it wos hapaning to me" [sic]. For others, the control-aspect seemed to refer to their perceived superiority towards censorship online. Case 5444 illustrated: "Because I'd always had an interest and because it was forbidden, yet easy to find."

Case 5061's response stood out: "Confusion around my sexuality and feelings of hate." While most of the offenders chose emotional labels giving them a passive role (e.g., a victim of depression, stress, sexual abuse), Case 5061 described his experience with the active emotion, "hate". It is not clear from the short data snippet if he refers to self-hatred or feelings of hatred towards others; however, his response read as if viewing CP was used to aid in his sexual identity confusion and to deal with his negative emotions, again supporting the link to a CP as a source of positive emotions.

Number of Explanations Provided by Individual Offenders

Once the final themes had been identified, it was determined how many themes individual participants included in their responses. It was hypothesised that at least some of the offenders present more than one explanation for their offending. Overall, the majority of participants referred to a singular theme (54%), 36% included two themes, and 10% named three themes. None of the respondents referred to more than three themes in their response. . Thus, the second hypothesis was confirmed. In addition, it became apparent that CPOs were more inclined to provide more than one theme in their responses (41% one theme, 45% two themes, 14% three themes) than MOs (71% one theme, 23% two themes, 6% three themes).

Relationship between Self-reported Themes and Offence Characteristics

It was hypothesised that systematic relationships between offence motivation and risk-related constructs can be identified. The theme "detached emotion" was excluded from this analysis given that only one participant endorsed this theme. Based on the low number of participants, the high number of variables and the similarity between participants (given that participants with multiple explanations were duplicated), this analysis is purely descriptive and shall be considered for exploratory purposes only. Overall, this analysis revealed some systematic relationships between risk-related variables and the thematic groups, thus confirming the third hypothesis.

Initial trigger

Members who endorsed this theme had the highest median years of education, reported the lowest stress levels, and were the least likely to have visited a child-lover website. Although they possessed CP material with the lowest severity in terms of content, offenders had a high variety of material types (e.g., photos, videos). Even though this theme was clearly linked to a sexual attraction to children, only half of the offenders reported they got sexually aroused by the material. About half of the offenders had engaged in online networking with other offenders.

Overall, these findings support the notion that the offenders have not been deeply involved in their CP consumption. However, for at least some offenders there is a strong connection to sexual satisfaction with child material. The high percentage of offenders who engaged with other offenders online may indicate that other online users were their first point of contact to access CP. The high variety of material may indicate that these individuals conducted unfiltered searches or received a mixed variety from other offenders as potential material distributors.

Source of positive emotions

This theme was more strongly endorsed by CPOs than MOs. Offenders who referred to this theme stated the highest stress levels and the lowest ability to cope with their stressors in comparison to all other offenders. Nearly 90% in this group reported they get sexually aroused by the material. All members had also viewed other forms of deviant pornography and a fifth had visited child-lover websites in addition to their CP consumption. Eighty percent reported a specific victim preference in their material. This theme was linked to lack of an appropriate sexual object and progression from legal material.

In sum, the high stress levels and the perceived low stress coping abilities indicate that CP was used as a coping mechanism, however, with a strong sexual connotation. It is apparent that for the majority of offenders in this thematic group, CP is a source of sexual satisfaction, for some even to the extent of developing specific preferences. On the other hand, sex as a coping mechanism is not necessarily restricted to CP in this group as evidenced in the consumption of other extreme forms of pornography. Alternatively, CP may have been the final target in their search for sexual satisfaction within other deviant material.

Progression from other material

This theme was also more strongly endorsed by CPOs than MOs. About half of the offenders reported high stress levels, and nearly 80% stated that they have low capacities to deal with stress. One third of the offenders had a criminal history (non-sexual crimes) and all of the offenders had also consumed other types of deviant pornography. Regarding CP, nearly 90% stated they were sexually aroused by the material and about two thirds reported specific victim preferences. Overall, from all thematic groups, offenders belonging to this group reportedly had invested the highest amount of time per week in the search and sorting of their CP material. Finally, while the other groups had mostly used digital pictures and digital videos, this offender group had used child pornographic narratives with a higher frequency. In terms of their internet usage, two thirds of the offenders had engaged with others on the internet, more than a fifth had visited child-lover websites, and ten percent of the offenders had used the internet to groom minors, which is the highest percentage of all members belonging to a thematic group. This theme was clearly related to usage of CP as a source of positive emotions.

Overall, this theme shared some similarities with the previous theme in that CP, used as a source of sexual satisfaction, acts as a potential stress reliever. However, for offenders belonging to this thematic group, CP had been approached via progression through other forms of pornography, which may still be used. In addition, the sexual aspect appears to be more dominant in this group, given the considerable time investment offenders make in their material, the high frequency of non-visual material, and the cross-over to direct contact with a minor. The high online engagement with other offenders may be based on using this network as a way to access more (and more deviant) material.

Sexual attraction to minors

This theme was endorsed by a very high percentage of MOs. Members in this group displayed no elevated stress levels and reported average confidence in their coping mechanisms. About 80% of offenders had a non-sexual criminal history, the highest percentage of members of any thematic group. Offenders in this group had also spent the highest amount of time in treatment. About half of the offenders had visited child-lover websites and about 80% had engaged with other offenders online, again the highest percentage of all groups on these variables. More than 90% of offenders had also consumed other types of deviant pornography, and one offender had engaged in online grooming of minors.

With regards to CP, only about 60% of offenders reported that they got sexually aroused by the material, although nearly 80% reported a specific victim preference in their material. They reported to have possessed CP wirh the most extreme content in comparison to the other offender groups and endorsed the second-highest level of cognitive distortions amongst all offenders.

This theme appears to be the antidote to the previous two themes. Here, CP is consumed to satisfy the offender's sexual attraction to minors. The low arousal profile of offenders and the high percentage of contact offenders suggest that CP is just one aspect in the general contact-driven profile of these offenders. This group may be more likely to endorse violent and/or sadistic thoughts and behaviours, given the highly explicit and violent content level they admitted to, the need for other deviant pornography, as well as the high general criminal activity observed in this group. Their sexual deviancy may be related to the generally increased level of cognitive distortions within this group. The high networking activity of this group may thus be related to finding social support within a marginalised group and normalising one's cognitive distortions.

Lack of appropriate sexual object

This group displayed no elevated stress levels; however, 70% reported a perceived lack of coping skills. About 30% had a non-sexual criminal history but overall, they had the lowest amount of time spent in treatment from all offenders. All offenders in this group had consumed other types of deviant pornography alongside CP. Indeed, only 70% reported sexual arousal to CP, but about 60% still had clear victim preferences. They had consumed a wide range of CP types, however remained on a relatively low level of explicitness. They also had the lowest average length of consumption. Only a small amount of offenders (14.3%) had visited child-lover websites. In addition, they endorsed the least cognitive distortions in comparison to the other offender groups.

Overall, this theme clearly reflects their search for an arousing sexual object. There is no sense of progression or need within their CP consumption, which may indicate that CP is not the sole source of their sexual satisfaction. Their cognitive scores reveal an understanding of the inappropriate nature of this behaviour. Their low perception of coping skills is interesting, suggesting some potential for sexual activity or pornography to act as a catalyst for stress.

No (direct) explanation provided

This theme was endorsed by a high percentage of MOs. These offenders had the lowest education of all offenders. They also had the lowest percentage of offenders with a non-sexual criminal history (14.3%) and the lowest consumption rate of deviant pornography other than CP. They were the least likely to network with other offenders online in comparison to the other thematic groups. Only about 60% reported they got sexually aroused by the material. On the other hand, this group had the highest consumption rate of child-lover websites and about 60% expressed clear victim preferences. Certainly noteworthy, this group reported the highest length of consumption in comparison to the other thematic groups. However, given that they also reported CP magazines (pre-dating internet times) as a common type in their collection, these findings may be a reflection of age rather than preference. Individuals in this group usually possessed only one type of CP (e.g., digital images) but varied in terms of their explicitness. They spent the lowest amount of time per week with their CP collection in comparison to the other offenders. However, this group displayed the highest level of cognitive distortions in comparison to all thematic groups.

In a way, it appears that this group encompasses both very serious and less serious offenders. CP is clearly not their main focus, nor their main source of sexual satisfaction. Two findings are especially noteworthy: They high level of cognitive distortions as well as their high length of consumption, which could both be cause and effect to each other. It is surprising that these offenders, who as a group view children as sexual objects, cannot provide an explanation or justification for their CP consumption. Although their lower education may act as a mediating factor, the inability to verbalise one's intention to offend may indicate a higher level of risk as evidenced in their cognitive distortions or persistence of offending.

Discussion

Offenders provided a variety of motivations for their CP offending, only one of whom explicitly referred to sexual attraction to minors. The identified offence motivations confirmed some of the themes presented in previous research. As in existing studies (e.g., Taylor & Quayle, 2003; Seto et al., 2010), a sexual attraction to minors clearly appears to be the main motive for viewing CP. However, in contrast to the previous studies, none of the offenders reasoned that they would use the material as a substitute for contact offending. In addition, even though some of the MOs had engaged in activities that bore similarity to their CP offending (e.g., filming their sexual activities with a minor victim), none of them indicated that they had used their material for victim grooming purposes or as part of their contact sex offending. There were other functions identified in previous research that were not reported by this sample, for example CP as a pure collective (Taylor & Quayle, 2003; Seto et al., 2010) or as an expression of internet addiction (Seto et al., 2010).

In this study, participants commonly provided more than one offence motivation, especially amongst CPOs. Some noteworthy relationships between the motivational themes and other variables were identified, for example the explicitness of CP material. Not only do these results confirm the findings by Seto et al. (2010) and Surjadi et al. (2010), they also provide further support for Seto et al.'s (2010) claim that offender's motivational themes may indicate differential treatment needs. The value of a motivation-based typology for CPOs further lies in the indicative risk variables identified in this study, namely the motivational differences between contact and non-contact offenders, the identification of cognitive distortions regarding children as sexual objects, and the role of online social networking with other offenders.

The current study revealed some differences between offenders with and without child contact sex offences. While CPOs more frequently reported to use CP as a source of positive emotions and as the result of a progression through other forms of pornographic material, MOs were more likely to admit a sexual attraction to minors or to provide no (direct) explanation for their CP offending (see Figure 3).

Figure 3: Thematic map of offence motivations displaying differences between offender-types, weighted according to frequency of their appearance. Larger circles represent higher rates of being named by the participants. CPOs are represented in orange, MOs in green colour.

In addition, CPOs were more likely to provide more than one explanation for their offending. MOs' stronger endorsement of paedophilic interest can be explained with internal attribution processes. According to Kelley's (1967) Covariation Model, internal attributions are based on repeated observations of behaviours across multiple situations. Given that for MOs, a sexual interest in minors was consistently displayed in a variety of contexts (i.e., online and offline), it follows that offenders with a contact sex offence are more likely to attribute their CP consumption to an internal trait, namely a sexual interest in children. The internal attribution hypothesis finds empirical support in Seto, Cantor, and Blanchard's (2006) study. They compared 100 CPOs, 178 contact child sex offenders, 216 rapists, and 191 non-offenders in terms of their sexual attraction to children, retrieved from clinical interviews and the outcomes of a phallometric assessment. Diagnostic criteria for paedophilia (using a paedophile index score) were met by 61% of CPOs, 35% of contact sex offenders, 13% of adult sex offenders, and 22% of non-offenders. Seto et al. concluded from these findings that CP consumption "is a valid diagnostic indicator of paedophilia" (p. 613), stronger than contact sexual behaviour towards a child. However, within the group of CPOs, 43% also had a history of contact sex offending against a child and this group achieved the strongest arousal pattern towards minors. Thus, this outcome suggests that at least for a subgroup of CP users, namely MOs, CP may be an expression of a more persistent sexual preference for minors that, by presenting as an internalised trait, corresponds with the diagnostic criteria for pedophilia. Consequently, CPOs who admit a sexual interest in minors as their main motivation to consume CP may thus present a higher risk to progress to contact sex offending or to have undisclosed historic contact sex offences. However, this conclusion requires a causal relationship between offence motivations and offender type, which can only be assumed based on the current data set and research design.

Offence-supportive cognitions have traditionally been linked to the commission of contact sex offences, and it is yet to determine if they are aetiological components, as temporarily activated cognitions or stable attitudes, or if they are post-hoc explanations to justify one's actions in hindsight, consciously or unconsciously (see Blumenthal, Gudjonsson, & Burns, 1999; Gannon & Polaschek, 2006; Maruna & Mann, 2006). In this study, endorsement of cognitive distortions was generally low. However, there were some noteworthy differences between members of the thematic groups. Whereas participants who reported lack of an appropriate sexual object were the least likely to endorse cognitive distortions, offenders who admitted a sexual attraction to minors and offenders who provided no (direct) explanation for their offending reported the highest scores. However, considering the offender type distribution across these themes, the relationship between offence motivation and cognitive distortions may also be explained by offender types. Indeed, MOs (Mdn = 63; max. 145) reported significantly higher scores on the cognitive distortion scale than CPOs (Mdn = 41.5), U = 114.5, z = -2.1, p < .05, r = -.33. Extending the attribution theory presented above, CPOs may indeed be less likely to endorse cognitive distortions due to the situational limitations of their offending. Alternatively, CPOs may endorse qualitatively different cognitive distortions from offenders with contact victims, and may thus appear as less distorted on conventional measures that are not validated on non-contact sex offenders.

There has been some recognition that paedophile online newsgroups play an important role in validating and normalizing such cognitive distortions (Taylor & Quayle, 2003). With the exception of the offenders who had not provided any (direct) explanation, the majority of participants across all thematic groups engaged in online networking with other users of CP. The information flow within the community of CP users has previously been found to play a significant role in the exchange and trading of the material (Quayle & Taylor, 2002). These communities also facilitate the establishment of contacts to other users with deviant sexual interests (Beech, Elliott, Birgden, & Findlater, 2008). As Taylor and Quayle (2005) pointed out, for many of their offending participants, these online relationships often replace unsatisfactory relationships in the offline world and provide important social support. In a case study of a 33 year old online sex offender, Quayle and Taylor (2001) described how important, nearly therapeutic, the notion of kinship in his newsgroup was for this particular offender—but also the significant role the paedophile community played in aggravating his offending behaviour.

There is a growing body of research confirming the importance of social support for the offending behaviour of CPOs. In their CPO typology, Merdian et al. (2011) identified that increased networking with other offenders was representative of heightened seriousness of one's offending; this was expressed in increased trading activity beyond collecting CP, possession of images with more deviant content, as well as enhanced knowledge about security measures to protect oneself. McLaughlin (2000) found that in his sample of CP collectors, a movement from static to more dynamic online locations (which represents a shift in increased networking activities) was related to a move from collecting to distribution of CP. Finally, Carr (2006) identified that a more serious engagement in CP offending was related to a tendency to more secure internet applications, a reduction in opportunistic behaviour (such as chatting in a broader social forum), and increased networking with other offenders. As Carr pointed out, status within the online community is often defined by the rarity of the provided material. It is thus not uncommon for producers to deliberately withhold some pictures of a thematic series to increase its value (Hesselbarth & Haag, 2004). Reportedly, some offenders in Carr's sample felt encouraged to self-produce highly sought-after material in order to raise their status. It thus appears that online social involvement with other offenders as well as the particular role of the social networking can function as mediator variables to behaviour that may be a direct indicator of risk, such as increased access to deviant material or increased security awareness.

The above findings support the importance of identifying offence motivations in the assessment and treatment of CPOs, as a way of identifying higher risk offenders. This research confirmed that not only do offenders express differential needs according to their offence motivation but that the explanations provided may be indicative of different stages of the offending process.

Limitations

This study presented with a number of limitations, such as the self-selection of participants, the transparency of the survey, and the small sample size, which limited the methodological choices available. Most MOs were institutionalised at the time of data collection while nearly all CPOs resided in the community, which introduced a systematic difference in the method of data collection. Even though presence of the researcher or other participants are potential biases to the survey responses, these differences in data collection were necessary due to differing requests by participating institutions and were accepted in order to increase study participation.

Conclusion

The current study confirmed the value of a motivational typology as a move towards more focused and more economic risk assessment of CPOs, for example based on a separation between sexual and non-sexual offence motivations. The outcomes confirm the heterogeneity of CPO subgroups that may require specific assessment and treatment plans, depending on their unique offence pathways. The above discussion has identified some differences between CP consumers with and without contact victims, concerning their offence motivations as well as their level of cognitive distortions. However, these findings can only be considered as a first step in identifying a relationship between offenders' risks/needs and their offence motivations, and more, and more rigorous, research is needed to develop a motivational-based typology of CPOs and to integrate these findings into a comprehensive understanding of the risk presented by CPOs, both regarding reoffending and cross-over to contact offending.

Author Note

Hannah L. Merdian, Cate Curtis, Jo Thakker, Douglas P. Boer, School of Psychology, University of Waikato, Hamilton, New Zealand; Nick Wilson, Department of Corrections, Hamilton, New Zealand; Hannah L. Merdian is now at the School of Psychology, University of Lincoln, UK; Douglas P. Boer is now at the School of Psychology, University of Canberra, AUS.
The research leading to these results received funding from the University of Waikato, Hamilton, New Zealand.

References

1. Abel, G. G., Becker, J. V., Cunningham-Rathner, J., Rouleau, J. L., Kaplan, M., & Reich, J. (1984). Treatment manual: The treatment of child molesters. Atlanta, US: Emory University.
2. Aftab, P. (2000). The parent's guide to protecting your children in cyberspace. New York, NY: McGraw-Hill.
3. Attride-Stirling, J. (2001). Thematic networks: an analytic tool for qualitative research. Qualitative Research, 1(3), 385-405. doi: 10.1177/146879410100100307
4. Beech, A. R., Elliott, I. A., Birgden, A., & Findlater, D. (2008). The internet and child sexual offending: A criminological review. Aggression and Violent Behavior, 13, 216-228. doi:10.1016/j.avb.2008.03.007
5. Blumenthal, S., Gudjonsson, G., & Burns, J. (1999). Cognitive distortions and blame attribution in sex offenders against adults and children. Child Abuse and Neglect, 23(2), 129-143.
6. Braun, V. & Clarke, V. (2006) Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101. doi: 10.1191/1478088706qp063oa
7. Calcetas-Santos, O. (2001). Child pornography on the internet. In C. A. Arnaldo (Ed.), Child abuse on the internet: Ending the silence (pp. 57-60). Oxford, UK: UNESCO Publishing/ Berghahn Books.
8. Caple, T. (2008). A comparison of the characteristics and motivations of abusing and non-abusing CP offenders (Unpublished doctoral dissertation). James Cook University, Townsville, AUS.
9. Carr, A. (2006). Internet censorship offending: A preliminary analysis of the social and behavioural patterns of offenders (Doctoral dissertation). Retrieved from http://epublications.bond.edu.au/theses/carr/
10. Choo, K. R. (2009).Online child grooming: A literature review on the misuse of social networking sites for grooming children for sexual offences. AIC Reports: Research and Public Policy Series, 103. Retrieved from www.aic.gov.au/publications/current%20series/rpp/100-120/rpp103.aspx
11. Davis, R. A. (2001). A cognitive-behavioural model of pathological internet use. Computers in Human Behavior, 17(2), 187-195. Retrieved from www.elsevier.com/locate/comphumbeh
12. Durkin, K. & Bryant, C. (1999). Propagandising pederasty: A thematic analysis of the online exculpatory accounts of unrepentant paedophiles. Deviant Behavior: An Interdisciplinary Journal, 20(2), 103-127. doi: 10.1080/016396299266524
13. Fereday, J. & Muir-Cochrane, E. (2006). Demonstrating rigor using thematic analysis: A hybrid approach of inductive and deductive coding and theme development. International Journal of Qualitative Methods 2006, 5(1). Retrieved from http://ejournals.library.ualberta.ca/index.php/IJQM/article/view/4411
14. Foley, T. P. (2002). Forensic assessment of internet child pornography offenders. In B. K. Schwartz (Ed.), The sex offender: Current treatment modalities and systems issues (Vol. IV, pp. 1-18). Kingston, NJ: Civic Research Institute
15. Gannon, T. A. & Polaschek, D. L. L. (2006). Cognitive distortions in child molesters: A re-examination of key theories and research. Clinical Psychological Review, 26, 1000-1019. doi: 10.1016/j.epr.2005.11.010
16. Hartmann, C. R., Burgess, A. W., & Lanning, K. V. (1984). Typology of collectors. In A. W. Burgess & M. Lindequist Clark (Eds.), CP and sex rings (pp. 93-109). Toronto, CA: Lexington Books.
17. Hill, A., Briken, P., & Berner, W. (2006). Pornographie im Internet: Ersatz oder Anreiz fuer sexuelle Gewalt? [Pornography on the internet: Substitute or stimulus for sexual violence?]. In Stiftung Deutsches Forum fuer Kriminalpraevention (Ed.), Internet-Devianz [Internet deviancy] (pp. 113-134). Berlin, Germany: Bundesverwaltungsamt
18. Holt, T. J., Blevins, K. R., & Burkert, N. (2010). Considering the paedophile subculture online. Sexual Abuse: A Journal of Research and Treatment, 22(1), 3-24. doi: 10.1177/1079063209344979
19. Howitt, D. & Sheldon, K. (2007). The role of cognitive distortions in paedophilic offending: Internet and contact offenders compared. Psychology, Crime, and Law, 13(5), 469-486. doi: 10.1080/10683160601060564
20. Itzin, C. (1997). Pornography and the organization of intrafamilial and extrafamilial child sexual abuse: Developing a conceptual model. Child Abuse Review, 6(2), 94-106. Retrieved from http://search.ebscohost.com.ezproxy.waikato.ac.nz/login.aspx?direct=true&db=aph&AN=11842836&site=ehost-live
21. Jenkins, P. (2003). Beyond tolerance: Child pornography on the internet. New York, NY: New York University Press
22. Joffe, H. (2011). Thematic analysis. In D. Harper & A. R. Thompson, Qualitative research methods in mental health and psychotherapy: A guide for students and practitioners (pp. 209-223). Chichester, UK: John Wiley & Sons.
23. Kelley, H. H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation, Vol. 15 (pp. 192-238). Lincoln: University of Nebraska Press.
24. Knight, R. A. & Prentky, R. A. (1990). Classifying sexual offenders: The development of corroboration of taxonomic models. In W. L. Marshall, D. R. Laws, & H. E., Barbaree (eds.), Handbook of sexual assault (pp. 23-52). New York, NY: Plenum.
25. Krone, T. (2004, July). A typology of online CP offending. Trends and issues in Crime and Criminal Justice, 279. Retrieved from www.aic.gov.au/publications/tandi2/tandi279.pdf
26. Krone, T. (2005, April). Does thinking make it so? Defining online CP possession offences. Trends and Issues in Crime and Criminal Justice, 299. Retrieved from http://www.aic.gov.au/documents/9/A/B/%7B9AB4DA8B-6EDB-401B-AC9E-C59EA93B1EE0%7Dtandi299.pdf
27. Kuhnen, K. (2007). Kinderpornographie und Internet: Medium als Wegbereiter fuer das (paedo-)sexuelle Interesse am Kind? [Child pornography and the internet: Medium as precursor for the (paedo-) sexual interest in children?]. Goettingen, Germany: Hogrefe.
28. Langevin, R. & Curnoe, S. (2004). The use of pornography during the commission of sexual offences. International Journal of Offender Therapy and Comparative Criminology, 48(3), 572-586. doi: 10.1177/0306624X03262518
29. Maruna, S. & Mann, R. E. (2006). A fundamental attribution error? Rethinking cognitive distortions. Legal and Criminological Psychology, 11, 155-177. doi: 10.1348/135532506X114608
30. McLaughlin, J. F. (2000). Cyber child sex offender typology. Knight Stick: Publication of the New Hampshire Police Association, 51, 39-42. Retrieved from http://www.ci.keene.nh.us/police/Typology.html
31. Merdian, H. L., Curtis, C., Thakker, J., Wilson, N., & Boer, D. P. (2011). The three dimensions of online CP offending. Journal of Sexual Aggression. [OnlineFirst Publication] doi:10.1080/13552600.2011.611898
32. Merdian, H. L., Thakker, J., Wilson, N., & Boer, D. P. (2011). Assessing the internal structure of the COPINE scale. Psychology, Crime & Law, [OnlineFirst Publication] doi: 10.1080/1068316X.2011.598158
33. Ost, S. (2009). Child pornography and sexual grooming: Legal and societal responses. Cambridge, UK: Cambridge University Press
34. Quayle, E., Erooga, M., Wright, L., Taylor, M., & Harbinson, D. (2006). Only pictures? Therapeutic work with internet sex offenders. Dorset, UK: Russell House Publishing.
35. Quayle, E. & Taylor, M. (2001). Child seduction and self-representation on the internet. CyberPsychology and Behavior, 4(5), 597-608. doi: 10.1089/109493101753235197
36. Quayle, E. & Taylor, M. (2002). Paedophiles, pornography and the internet: Assessment issues. British Journal of Social Work, 32(7), 863-875. doi: 10.1093/bjsw/32.7.863
37. Reid, S., Wilson, N. J., & Boer, D. P. (2011). Risk, needs, and responsivity principles in action: Tailoring rapist's treatment to rapist typologies. In D. P. Boer, R. Eher, & M. H. Miner, International Perspectives on the Assessment and Treatment of Sexual Offenders (pp. 287-297). Chichester, UK: Wiley-Blackwell.
38. Rettinger, L. J. (2000). The relationship between child pornography and the commission of sexual offences against children: A review of the literature (RR2000-5e). Department of Justice, CA: Research and Statistics Division. Retrieved from http://www.justice.gc.ca/eng/pi/rs/rep-rap/2000/rr00_5.html
39. Saunders, J. C. & Byrne, M. M. (2002). A thematic analysis of families living with schizophrenia. Archives of Psychiatric Nursing,16(5), 217-223. doi:10.1053/apnu.2002.36234
40. Seto, M. C., Cantor, J. M., & Blanchard, R. (2006). CP offenses are a valid diagnostic indicator of pedophilia. Journal of Abnormal Psychology, 115(3), 610-615. doi: 10.1037/0021-843X.115.3.610
41. Seto, M. C., Hanson, R. K., & Babchishin, K. M. (2011). Contact sexual offending by men with online sexual offenses. Sexual Abuse: A Journal of Research and Treatment, 23(1), 124-145. doi: 10.1177/1079063210369013
42. Seto, M. C., Reeves, L., & Jung, S. (2010). Explanations given by CP offenders for their crimes. Journal of Sexual Aggression, 16(2), 169-180. doi: 10.1080/13552600903572396
43. Sheldon, K. & Howitt, D. (2007). Sex offenders and the internet. Chichester, UK: West Sussex.
44. Surjadi, B., Bullens, R., van Horn, J., & Bogaerts, S. (2010). Internet offending: Sexual and non-sexual functions within a Dutch sample. Journal of Sexual Aggression, 16(1), 47-58. doi: 10.1080/13552600903470054
45. Taylor, M. & Quayle, E. (2003). CP: An internet crime. Hove, UK: Brunner-Routledge.
46. Taylor, M. & Quayle, E. (2005). Abusive images of children and the internet: Research from the COPINE project. In S. W. Cooper, R. J. Estes, R. J., A. P. Giardino, N. D. Kellogg, & V. I. Vieth (Eds.), Medical, legal and social science aspects of child sexual exploitation: A comprehensive review of pornography, prostitution, and internet crimes (pp. 257-275). St Louis, MO: GW Medical Publishing.
47. Taylor, M., Quayle, E., & Holland, G. (2001). Child pornography, the internet and offending. Sociologie et Sociétés, 32(2), 94-100.
48. Warden, N. L., Phillips, J. G., & Ogloff, J. R. P. (2004). Internet addiction. Psychiatry, Psychology, and Law, 11(2), 280-295. Retrieved from http://find.galegroup.com.ezproxy.waikato.ac.nz/itx/start.do?prodId=AONE
49. Webb, L., Craissati, J., & Keen, S. (2007). Characteristics of internet CP offenders: A comparison with child molesters. Sexual Abuse: A Journal of Research and Treatment, 19(4), 449-465. doi: 10.1007/s11194-007-9063-2.

Author address

Hannah Merdian
School of Psychology
University of Lincoln
Brayford Pool
Lincoln LN6 7TS, UK
hmerdian@lincoln.ac.uk