Associations between compulsive internet use and the autism spectrum

https://doi.org/10.1016/j.rasd.2015.12.005Get rights and content

Highlights

  • Compulsive internet use (CIU) correlated with varied Facebook behaviors.

  • Non-social autistic traits accounted for the CIU and autistic trait relationship.

  • CIU may not develop from attempts to compensate for offline social challenges.

  • In a replication study, students with and without autism did not differ in CIU.

Abstract

The increasingly complex relationship between individuals and online environments has prompted research exploring the potential for compulsive Internet use (CIU) in varying populations. Identification of the determinants of CIU, such as offline social difficulties associated with autism spectrum disorder (ASD), could clarify whether individuals compensate for offline difficulties through online interactions. In Study 1, 597 undergraduates took an online survey measuring CIU, social symptoms of ASD, restricted interests and repetitive behaviors (RIRB), self-esteem, and Facebook behaviors. A relationship between CIU and autistic traits was attributable to non-social characteristics (RIRB) and a relationship was identified between RIRB and information-seeking behaviors. These results complicate the social compensation hypothesis by suggesting that non-social personality characteristics contribute more strongly to CIU than individuals’ attempts to compensate for offline social challenges. Study 2 matched a sample of students with ASD (n = 33) to neurotypical students (n = 33) and uncovered no significant differences in CIU scores between the matched samples. These results suggest that findings observed by measuring autistic traits in neurotypical samples may not replicate with autistic samples.

Introduction

People with autism spectrum disorder (ASD) may experience unique communicative benefits from the Internet (Gillespie-Lynch, Kapp, Shane-Simpson, Smith, & Hutman, 2014). At the same time, people with ASD (MacMullin, Lunsky, & Weiss, 2015) as well as people in the general population with heightened autistic traits (Finkenauer, Pollmann, Begeer, & Kerkhof, 2012; Romano, Osborne, Truzoli, & Reed, 2013; Romano, Truzoli, Osborne, & Reed, 2014) may be at heightened risk for compulsive Internet use (over-attachment to the Internet or to specific functions of the Internet). While the Internet affords people who face offline social challenges, such as ASD, shyness, and social anxiety, with opportunities to compensate for these challenges (Benford, 2008, McKenna and Bargh, 2000, Sheeks and Birchmeier, 2007), using the Internet to compensate for social challenges may contribute to compulsive Internet use (Caplan, 2005, Davis, 2001, Kim and Haridakis, 2009).

Associations between autistic traits and compulsive Internet use have been interpreted as evidence that people with heightened autistic-like social symptoms become exceptionally attached to the Internet because it provides opportunities for them to compensate for their offline social challenges (Finkenauer et al., 2012, Romano et al., 2013, Romano et al., 2014). However, ASD is defined not only by social difficulties, but also by restricted interests and repetitive behaviors (RIRB; American Psychiatric Association, 2013). In addition to using the Internet for social endeavors, people with highly focused interests, a common characteristic of ASD, may use the Internet to engage in their interests (Jordan & Caldwell-Harris, 2012). Both primarily social activities (e.g., social network site use) and primarily non-social activities (e.g., shopping, Internet surfing) have been associated with psychological dependence on the Internet (Kim & Haridakis, 2009; Kim, La Rose, & Peng, 2009; Kuss, Griffiths, & Binder, 2013; Li & Chung, 2006; Mazer & Ledbetter, 2012; van den Eijnden, Meerkerk, Vermulst, Spijkerman, & Engel, 2008). Therefore, observed associations between heightened autistic traits (more generally) and compulsive Internet use may be attributed to social challenges (e.g., difficulty with social interactions), as well as non-social challenges (e.g., restricted interests), or both. Autistic traits may predispose people toward using the Internet to investigate their (often non-social) focused interests. Indeed, parental reports suggest that individuals with ASD may be drawn to non-social uses of the Internet (MacMullin et al., 2015; Mazurek, Shattuck, Wagner, & Cooper, 2012; Mazurek & Wenstrup, 2013).

Compulsive Internet use is defined by a preoccupation with online activities, difficulty reducing the amount of time spent online, and symptoms of withdrawal when separated from the Internet or specific sites on the Internet (Meerkerk, van den Eeijnden, Vermulst, & Garretsen, 2009). It is associated with heightened depression, lower self-esteem, and lower levels of perceived social support (e.g., Armstrong, Phillips, & Saling, 2000; Davis, 2001, Meerkerk et al., 2009, Thorsteinsson and Davey, 2014). Researchers use a variety of terms to refer to characteristics of compulsive Internet use including Internet addiction, Internet dependence, pathological Internet use, and problematic Internet use (Meerkerk et al., 2009). These terms reflect differences in theoretical orientations rather than clear behavioral differences (Kim & Haridakis, 2009). For coherency, we describe all terms for this construct as compulsive Internet use in this report.

Despite emerging evidence that autistic traits (more generally) are associated with compulsive Internet use in the general population (Finkenauer et al., 2012, Romano et al., 2013, Romano et al., 2014), prior research has not examined whether social symptoms and/or non-social symptoms (RIRB) drive the association. Prior studies that demonstrated associations between autistic traits and compulsive Internet use measured autistic traits in the general population with a standardized measure (autism quotient: Hoekstra et al., 2011) and reported that the severity of autistic traits correlated positively with compulsive Internet use (Finkenauer et al., 2012, Romano et al., 2013, Romano et al., 2014), but did not relate to the frequency of Internet use (Finkenauer et al., 2012).

Finkenauer et al. (2012) sampled married couples (n = 390) in the Netherlands. Participants completed the Autism Quotient, items assessing their frequency of Internet use, and the Compulsive Internet Use Scale (Meerkerk et al., 2009). Although the frequency of Internet use was unrelated to autistic traits, compulsive Internet use was positively associated with heightened autistic traits. In a subsequent study, researchers documented a positive relationship between autistic traits and compulsive Internet use in a non-clinical sample of 60 undergraduates (Romano et al., 2013).

Replicating this link, Romano et al. (2014) sampled 90 university students, divided into groups with high vs. low autistic traits, and found higher levels of compulsive Internet use in the high autistic trait group. While both autistic traits and anxiety were positively associated with compulsive Internet use, compulsive Internet use was less common among people with heightened autistic traits who also reported heightened anxiety. The authors interpreted this finding as evidence that anxiety alters the functions that the Internet serves for individuals with heightened autistic traits. This finding also suggests that anxiety and autistic traits may contribute to compulsive Internet use in unique ways. For example, generalized anxiety may mirror social anxiety, shyness, and loneliness (e.g., Caplan, 2002, Caplan, 2005, Davis, 2001, Engelberg and Sjöberg, 2004, Kim and Haridakis, 2009) in predisposing people toward using the Internet for social compensation, whereas autistic traits may predispose people toward using the Internet to investigate their focused interests. However, due to inherent limitations of the Autism Quotient, researchers in the three aforementioned studies were unable to determine the type of autistic traits (social or non-social) that were associated with compulsive Internet use.

Consequently, a primary aim of the studies in this report was to examine whether autistic-like social symptoms and/or RIRB were associated with compulsive Internet use in a general population sample of college students without ASD (Study 1) and in a sample of college students with ASD (Study 2) by using the Social Responsiveness Scale (Constantino & Gruber, 2012), which is a well-validated measure of autistic traits that distinguishes between social symptoms and RIRB. Given that associations between autistic traits and compulsive Internet use in the general population have led to concerns that compulsive Internet use may be elevated among individuals with ASD. The primary aim of Study 2 was to examine if college students with ASD report heightened levels of compulsive Internet use relative to their non-ASD peers.

A recent study suggests that compulsive Internet use may indeed be elevated among youth and young adults with ASD. MacMullin et al. (2015) recruited an online sample of parents of individuals with and without ASD (age range 6–21 years: M = 12). Parents of individuals with ASD reported that their child spent more time online and showed higher levels of compulsive Internet use relative to parents of individuals without ASD. Parents of individuals with ASD were more likely to report that they felt their child’s use of electronics was negatively impacting their child’s life and that their child learned fewer social functions of the Internet and used these functions at later ages, when compared to parental reports of individuals without ASD. However, no ASD-related differences in parent-reported frequency of social Internet use were observed. MacMullin et al. (2015) concluded that youth and young adults with ASD might be at heightened risk for negative impacts of Internet use in comparison to their peers without ASD as a result of social and non-social difficulties associated with ASD.

The degree to which MacMullin et al. (2015) findings align with the aforementioned literature linking autistic traits to compulsive Internet use among adults in the general population is difficult to ascertain as the studies involving non-clinical samples used adult self-reports whereas MacMullin et al. (2015) used parental reports about youth and young adults (mean age 12 years). In addition, MacMullin et al. (2015) used different recruitment strategies for parents of individuals with ASD and parents of individuals without ASD. While parents of individuals with ASD were recruited from autism organizations using a flier that described potential costs and benefits of computer use, parents of individuals without ASD were recruited from a Qualtrics panel of paid survey-takers. These different recruitment strategies leave open the possibility that parents of children with ASD participated in the study primarily out of concern about their child’s computer use while parents of children without ASD participated for different reasons. Reliance on an online sample of parents of individuals with and without ASD also made it impossible to verify ASD classifications. In order to address these limitations, a primary aim of the second study described in this report was to evaluate self-reported compulsive Internet use among college students with and without ASD. College students are believed to be especially vulnerable to compulsive Internet use due to the impulsivity associated with emerging adulthood, nearly continuous access to the Internet, and expectations that students use the Internet for coursework (e.g., Kandell, 1998).

People with ASD are often socially isolated (Howlin, Goode, Hutton, & Rutter, 2004) and thus may use the Internet for social compensation. Indeed, people with ASD have characterized the Internet as a liberating environment where they can interact with others on a more equal basis (Benford, 2008) partially due to perceptions of increased control over when, how, and with whom one interacts online (Gillespie-Lynch et al., 2014).

Individuals with social-communicative challenges such as ASD may attempt to use social network sites such as Facebook to compensate for their social difficulties (Kuss & Griffiths, 2011; Steinfield, Ellison, & Lampe, 2008). Indeed, both shyness and loneliness are associated with heightened Facebook use (Balci and Golcu, 2013, Orr et al., 2009). Adults with ASD report using social network sites to connect with others (Burke, Kraut, & Williams, 2010; Mazurek, 2013). These studies lend support to the social compensation hypothesis–that online environments provide individuals who face offline social challenges with opportunities to compensate for these difficulties (Kraut et al., 1998, Kraut et al., 2002).

On the other hand, individuals with ASD have cited the paucity of consistent social feedback (e.g., the relative absence of nonverbal cues) as a significant disadvantage of computer-mediated communication (Burke et al., 2010). In offline environments, individuals with ASD may struggle to identify social norms and engage effectively with others in ways that respect those norms, which can result in difficulties initiating and maintaining friendships. Although limited research has examined whether offline difficulties associated with ASD improve, worsen, or remain the same on social network sites, some people with ASD find it difficult to interpret social norms on Facebook (Burke et al., 2010). Similarly, adults with ASD report heightened enjoyment of certain Internet functions, such as blogs and discussion boards, but report less enjoyment of social network sites when compared with their peers (Gillespie-Lynch et al., 2014).

Research also suggests that many individuals with ASD find the social isolation they experience offline recapitulated on social networks (Burke et al., 2010), with offline social challenges potentially exacerbated through online interactions (van den Eijnden, Meerkerk, Vermulst, Spijkerman, & Engel, 2008). If individuals with ASD struggle to identify and adhere to offline social norms, such as the give-and-take that occurs during conversation, they may struggle even more when trying to use the reduced nonverbal cues available online to interpret online norms.

All of the aforementioned hypotheses have been conceptualized in terms of potential influences of offline social behaviors on Internet use. However, people with ASD may be particularly drawn to non-social uses of the Internet (Mazurek et al., 2012, Mazurek and Wenstrup, 2013). A tendency to use the Internet for non-social purposes by individuals with ASD might underlie differences between shy people and those with ASD in their enjoyment of social network sites. Although shy people typically maintain fewer friends on Facebook than their peers, they use Facebook more than their peers (Orr et al., 2009). In contrast, people with ASD may enjoy social network sites less than others (Gillespie-Lynch et al., 2014). Shy individuals may be drawn toward online social opportunities to compensate for offline social difficulties, whereas people with ASD may be drawn to more interest-based online media that provide opportunities to engage their focused interests.

The studies in this report examined associations between social and non-social (RIRB) autistic traits and compulsive Internet use. Study 1 examined associations between autistic-like social symptoms, RIRB, compulsive Internet use, and specific behaviors on Facebook to (1) clarify whether social difficulties or RIRB contribute to compulsive Internet use, and (2) examine associations between specific ways of connecting with others on Facebook and autistic traits. If observed associations between overall autistic traits and compulsive Internet are attributable to autistic-like social symptoms in particular, compulsive Internet use may arise from attempts to use the Internet to overcome social challenges, thereby supporting the social compensation hypothesis. Alternatively, an association between RIRB and compulsive Internet use would suggest that people with heightened RIRB may be susceptible to compulsive Internet use due to the rich opportunities available online to explore their focused interests.

Social difficulties were expected to impact specific social networking behaviors on Facebook, such as how users maintain contact with known others, initiate contact with others through (multiple) friend requests, and use blocking to discontinue contact with others. Associations between more autistic-like social symptoms and less maintenance of contact with known others on Facebook would support the hypothesis that offline social isolation is recapitulated online. Positive associations between social symptoms and the use of Facebook to initiate contact with unknown others would support the social compensation hypothesis. Positive associations between social difficulties and repeated friend requests would support the hypothesis that offline social difficulties may be exacerbated online.

Study 2 aimed to (1) clarify whether college students with ASD reported more compulsive Internet use than a matched-sample of peers without ASD, and (2) determine if associations between autistic traits and compulsive Internet use replicated among college students with ASD. We expected compulsive Internet use to be heightened among college students with ASD based on MacMullin et al. (2015) research documenting heightened parent-reported compulsive Internet use among a younger sample of individuals with ASD with a mean age of 12 years.

Section snippets

Participants

College students were recruited from a Psychology Department subject-pool at a large, urban public university. Students were enrolled in psychology courses, and participated for research participation credits or extra credit; the majority of students accessing the subject pool were general education students enrolled in Introductory Psychology. Participants accessed the study through SurveyMonkey where they provided informed consent and demographic information, and answered questions about

Results

Table 1 reports descriptive statistics for the predictor and outcome variables in this study. As preliminary analyses, zero-order correlations were computed between the continuous measure of compulsive Internet use, Facebook connection strategies, friending behaviors on Facebook, self-esteem, gender, and autistic traits (social symptoms, RIRB; see Table 2). Due to non-normality of compulsive Internet use scores when assessed as a continuous measure and based upon prior research that has

Discussion

Study 1 replicated prior research linking autistic traits with compulsive Internet use in the general population (Finkenauer et al., 2012, Romano et al., 2013, Romano et al., 2014), but extended this line of research by demonstrating that this association is attributable to non-social characteristics (e.g., RIRB) rather than to social difficulties associated with ASD. This finding also extends parental reports that individuals on the spectrum may be drawn to non-social uses of the Internet (

Participants

A total of 66 undergraduate college students participated in Study 2. Students with ASD (N = 33; 21 men, 17 women) were recruited from a mentorship program conducted at a university Center for Student Accessibility (CSA; N = 24) and from the online psychology subject pool at the same large, urban university (N = 9). The 9 students with ASD from the subject pool were initially recruited for Study 1, but excluded to ensure a non-clinical sample. The researchers did not interact with these students and

Results

Consistent with other matching studies (e.g., Ozonoff, Pennington, & Rogers, 1990), we confirmed differences in autistic traits along three of the five dimensions of the SRS-A between ASD and controls. The group difference for the RIRB dimension was marginally significant (Table 5).

Compulsive Internet use, measured as a continuous variable, was compared in the ASD and control samples using a paired-samples t-test. This analysis did not yield a significant difference between the ASD and control

Discussion

Although RIRB was marginally higher in the ASD group (as expected, based on ASD diagnostic criteria; American Psychiatric Association, 2013), Study 2 did not provide evidence that college students with ASD had heightened levels of compulsive Internet use when compared with a matched sample of peers without ASD. This finding contrasts with parent-reports of heightened compulsive Internet use among youth and young adults with ASD at a mean age of 12 years (MacMullin et al., 2015). Differences

Conclusions

The current research provides evidence that non-social characteristics of Internet users, such as their degree of restricted, focused interests and the tendency to engage in repetitive behaviors, may contribute more strongly to compulsive Internet use than social characteristics. The current findings are remarkably similar to findings from one of the first studies to examine Internet addiction: Shotton (1991) interviewed people who reported being dependent on the Internet and found that they

Conflict of interests

No competing financial interests exist.

Acknowledgements

The authors thank participants of this study and Ana Alvizurez, Erica Golin, and Marsha Dupiton for tremendous assistance in survey development and coding. The last author, who was the first author’s advisor, played a guiding role in study design and contributed substantially to the literature review and writing of the manuscript.

References (59)

  • C. Steinfield et al.

    Social capital, self-esteem, and use of online social network sites: a longitudinal analysis

    Journal of Applied Developmental Psychology

    (2008)
  • S. Wigham et al.

    The reliability and validity of the Social Responsiveness Scale in a UK general child population

    Research in Developmental Disabilities

    (2012)
  • American Psychiatric Association

    The Diagnostic and Statistical Manual of Mental Disorders: DSM 5

    (2013)
  • Ş. Balci et al.

    Facebook addiction among university students in Turkey: Selcuk University example

    Turkish Journal of Research

    (2013)
  • R. Bender et al.

    Multiple test procedures other than Bonferroni’s deserve wider use

    BMJ: British Medical Journal

    (1999)
  • P. Benford

    The use of Internet-based communication by people with autism. PhD Thesis

    (2008)
  • M. Burke et al.

    Social use of computer-mediated communication by adults on the autism spectrum

  • S. Byun et al.

    Internet addiction: Metasynthesis of 1996–2006 quantitative research

    CyberPsychology & Behavior

    (2009)
  • S.E. Caplan

    A social skill account of problematic Internet use

    Journal of Communication

    (2005)
  • J.N. Constantino et al.

    Social Responsiveness Scale

    (2012)
  • A.J. Ebsensen et al.

    Age-related differences in restricted repetitive behaviors in autism spectrum disorders

    Journal of Autism and Developmental Disorders

    (2008)
  • N.B. Ellison et al.

    Connection strategies: Social capital implications of Facebook-enabled communication practices

    New Media & Society

    (2011)
  • E. Engelberg et al.

    Internet use, social skills, and adjustment

    CyberPsychology & Behavior

    (2004)
  • C. Finkenauer et al.

    Brief report: Examining the link between autistic traits and compulsive Internet use in a non-clinical sample

    Journal of Autism and Developmental Disorders

    (2012)
  • N.W. Gelbar et al.

    Systematic review of articles describing experiences and supports of individuals with autism enrolled in college and university programs

    Journal of Autism and Developmental Disorders

    (2014)
  • K. Gillespie-Lynch et al.

    Intersections between the Autism Spectrum and the Internet: Perceived benefits and preferred functions of computer-mediated communication

    Intellectual and Developmental Disabilities

    (2014)
  • M. Griffiths

    Does Internet and computer addiction exist? Some case study evidence

    CyberPsychology and Behavior

    (2000)
  • R.A. Hoekstra et al.

    The construction and validation of an abridged version of the autism-spectrum quotient (AQ-Short)

    Journal of Autism and Developmental Disorders

    (2011)
  • P. Howlin et al.

    Adult outcomes for children with autism

    Journal of Child Psychology and Psychiatry

    (2004)
  • Cited by (19)

    View all citing articles on Scopus
    View full text