The relationship between Cyber-Dependent Crime and ASD needs further investigation

[Posted on 27 July, 2021, by Clare Allely]

Cybercriminals can use computers in three main ways in order to engage in their illegal activities online. These include: (1) carrying out malicious activities on other people's computers, including, for example, spreading viruses, data theft and identity theft (2) using computers to commit “conventional crimes” including: spam, fraud, illegal gambling, etc.; and (3) using computers to save stolen or illegal data. Numerous high-profile cases and published case studies have indicated that there may be some degree of relationship between autism spectrum disorder (ASD) and cybercrime - most notably, computer hacking.

One of the most high-profile examples is the British computer hacker Gary McKinnon (a Scottish systems administrator). McKinnon fought his extradition to the United States where the authorities were seeking to try him for gaining unauthorised access to a total of 97 government computers (United States Army, Navy, Department of Defense and the National Aeronautic and Space Administration, NASA) between 1 February 2001 and 19 March 2002. While he was hacking the government computers, he would install a suite of tools and deleted a range of data which resulted in a significant number of computers to shut down, become unworkable and/or become vulnerable to other hackers. He copied files onto his own computer. He also deleted log files on the computers in order to hide his activities (Freckelton, 2011). Having been first arrested on 19 March 2002 by the Hi-Tech Crime Unit, McKinnon was formally diagnosed with Asperger’s Syndrome on 23 August 2008 (Sharp, 2013). In 2012 he had his extradition blocked by the United Kingdom (UK) on humanitarian grounds. Critical to this appeal was his interim diagnosis of Asperger’s Syndrome and the associated risk of suicide if extradited [McKinnon v Secretary of State for the Home Department (UK)].

Computers can be particularly appealing to individuals with ASD. As recently highlighted by Higham, Girardi and Edwards (2021) the organised, logical and predictable structure of computer-based interaction has been proposed as being particularly appealing to individuals with ASD due to the ASD-specific traits. Some of the reasons that it is particularly appealing to individuals with ASD is the benefits it offers such as increased control of interaction, enhanced self-expression, self-esteem, empowerment, well-being, interest sharing and group discussion. Additionally, “the consistent structure of coding language, standardised terminology in forums and logical syntax-guided structure of the computer and the internet is concrete and there are boundless opportunities for solitary pursuit of preferred or special interest” (Ledingham & Mills, 2015, pp. 5).

Surprisingly, few studies have investigated ASD or autistic-like traits in relation to cyber-dependent crime. In one of the few studies to date, Payne and colleagues (2019) carried out an online survey to study the relationships between cyber-dependent crime and ASD, autistic-like traits, explicit social cognition and perceived interpersonal support. A total of 290 internet users (with no convictions or cautions for cybercriminal activity) took part in the study (194 male and 96 female, age range of 14 to 74 years, mean age was 24.24 years). A self-reported diagnosis of ASD was given in 23 of the 290 internet users who took part.  The University of Bath’s participant databases which include computer science students and alumni was used to recruit participants. In order to recruit individuals with advanced digital skills, the researchers contacted computer science students in local schools and also ‘Cyber Security Challenge’ (an organisation which promotes the development of cyber-skilled individuals). Findings revealed an association between higher autistic-like traits (as measured using the Autism-Spectrum Quotient, AQ, screening tool) and a greater risk of engaging in cyber-dependent crime. It was self-reported by 122 individuals (42%) that they had engaged in 333 cyber-dependent crimes. Interestingly, the researchers found that a significant proportion of the association between scores on the AQ and self-reported cyber-dependent criminal activity was mediated by advanced digital skills. These findings would indicate that there may be something about autistic-like traits beyond the ASD diagnostic criteria that relates to cyber-dependent criminal activity. The results indicated that a possible key factor for this is an association between autistic like traits and advanced digital skills. Participants who self-reported a diagnosis of ASD were less likely to report cyber-dependent criminal activity. It may be that an ASD diagnosis is protective against engaging in cyber-dependent criminal activity, which is a finding worthy of further investigation. Payne and colleagues acknowledge some of the limitations with their study. For instance, they state that there may be selection issues in relation to individuals who respond to an invitation to complete an online survey on this topic. It is unknown how many individuals did not respond to the invitations to take part in the study (therefore there is no way to determine a response rate). 

In an earlier study, Seigfried-Spellar and colleagues (2015) investigated the relationship between autistic-like traits and self-reported cyber deviancy in a sample of 296 college students and found mixed results. Of the 296 respondents, 179 (60%) self-reported that they had engaged in some type of cyber deviancy (such as hacking, cyberbullying, identity theft, and virus writing). Of the 60% who self-reported engaging in some type of cyber deviancy, 57% (n = 170) self-reported engaging in hacking behaviours, 13% (n = 38) self-reported engaging in identity theft, 23% (n = 66) self-reported engaging in cyberbullying and 8% (n = 23) engaged in virus writing. Findings revealed that, when compared to participants who self-reported less computer deviancy, individuals who self-reported engaging in all four types of computer deviant behaviour were significantly more likely to score high on the AQ total score. They were also found to exhibit more impairments with social skills, communication, and imagination. Also, there were no significant differences between participants who self-reported engaging in one type of cyber deviancy compared to those who reported engaging in two or three cyber deviancy types. Of the 296 participants who took part in the study only 275 students completed the AQ screening tool. Of the 296 university students, 179 (60%) reported that they had engaged in some form of computer deviant behaviour. However, only two students had clinically significant scores based on the AQ. Both of these students self-reported engaging in computer deviant behaviour, which was 1% of the participants who completed the AQ screening tool and also self-reported engaging in cyber deviancy (n = 161). Seigfried-Spellar and colleagues do acknowledge some of the limitations with their study. For instance, their sample is comprised of undergraduate students from a large, Southern university in the United States and as a result the findings are not representative of the entire population of computer deviants.

In sum, Payne and colleagues (2019) have recommended that there is a need for further research in this area which use larger samples in order to investigate the relationship between ASD and autistic-like traits and cyber-dependent crime compared with cyber-enabled crime. Also, the idea that a diagnosis of ASD may be a protective factor for engaging in cyber deviancy is only speculative at this stage. The reason for this perhaps being that a diagnosis of ASD may be generally protective in terms of "deviant development" in many different fields, whereas living with unrecognised marked autistic traits might be a mediator of deviant development. This is particularly the case given that in the few studies to date, the data is self-reported and diagnostic status was not independently verified. Moreover, in the study by Seigfried-Spellar and colleagues they included in their sample both cyber-enabled crimes such as cyberbullying and identity theft, as well as cyber-dependent crimes such as hacking and virus writing. As pointed out by Payne and colleagues, there may be important differences between cyber-dependent crime compared with cyber-enabled crime with respect to ASD or autistic-like traits. Some of the recommendations that Payne and colleagues have suggested include: focusing on autistic communities as well as those convicted of cyber-dependent and cyber-enabled crimes. Clearly there is an a very real need for further research in this area.

[This is a blog. The purpose of the blog is to provide information and raise awareness concerning important issues. All views and opinions expressed are those of the writer and not necessarily shared by the GNC.]