Problematic Internet use,reward sensitivity and decision making

The concept of Internet addiction has been controversial (e.g. Starcevic, 2013). One of the key issues has been a variety of problematic online behaviours (i.e. excessive and harmful gaming, gambling, sexual activities, shopping and social networking) encompassed by the term. Another issue is a lack of valid and widely accepted criteria for distinguishing between problematic Internet use and bona fide addiction to the activities performed on the Internet. Despite these problems, Internet addiction has become a popular, if not fashionable, term. As a result, the literature on Internet addiction has grown exponentially.

An article by Seok et al. (2015) is a good example of this literature. It reports that individuals with Internet addiction exhibited functional changes in the brain regions involved in reward, conflict monitoring and cognitive control. These individuals had characteristics similar to those found in people with substance addiction: increased reward anticipation and decreased ability to realistically appraise the consequences of reward-driven behaviour and to control such behaviour. These characteristics influence the decision-making process so that risk and hazards are overlooked and the authors suggest that risky decision making might be a ‘biological risk factor for Internet addiction’.

Considering that the design of the study was cross-sectional, the characteristics of individuals with Internet addiction cannot point to the direction of causation. In other words, risky decision making may constitute a predisposition to Internet addiction just as it could be a consequence of Internet addiction. At best, risky decision making can be regarded as a correlate of Internet addiction.

Risky decision making is not specific for Internet addiction, as it is also a feature of other addictions and certain forms of psychopathology (e.g. severe personality disorders). Seok et al. (2015) included in their study only participants without ‘DSM-IV Axis I disorders’ and other addictions, so risky decision making in those with Internet addiction could be attributed to Internet addiction. However, the authors did not report whether they attempted to ascertain the presence of personality disorders or relevant personality traits (e.g. high novelty seeking and low harm avoidance, impulsivity) that might be associated with risky decision making. Another important limitation of the study is that individuals with ‘pure’ Internet addiction (i.e. without any other psychopathology) are rarely encountered; it is questionable whether findings based on a sample of such individuals could apply to most people with Internet addiction.

The Seok et al. (2015) study was conducted in a small sample (N = 15) and most participants with Internet addiction exhibited problematic online gaming/gambling. Others reported problematic online chatting, watching of online pornography and pursuit of ‘online entertainment’. Besides confirming that Internet addiction is not a homogeneous and uniform entity (e.g. Pawlikowski et al., 2014), this heterogeneity of activities is itself problematic because of the variety of pursued rewards and the consequences of such pursuits. The nature of the reward is important, considering that changes in reward sensitivity have been conceptualised as one of the key features of addiction (Volkow et al., 2010). In fact, a heightened sensitivity to reward, i.e., a pursuit of gains on a gambling task, along with decreased sensitivity to the negative consequences of pleasurable pursuits, i.e., ignoring the losses on the gambling task, have been demonstrated in individuals with ‘Internet addiction disorder’, most of whom had ‘Internet gaming addiction’ (Dong et al., 2013). Such contrasting sensitivities to reward and punishment might help explain continued engagement with online activities despite their negative consequences.

Looking at the mean scores on the instruments used by Seok et al. (in press) to identify individuals with Internet addiction, it is clear that a number of those in the group with Internet addiction in fact scored in the range of ‘mild Internet addiction’, possibly suggesting a pattern of excessive, but not necessarily pathological or addictive, Internet use. This highlights the problem of unclear boundaries between high, non-pathological Internet engagement and problematic Internet use; lumping the two patterns of Internet use together and treating them as a homogeneous entity is not justified.

In conclusion, the work of Seok et al. (2015) expands our understanding of the reward and control dynamics and the attendant decision making in individuals deemed to exhibit Internet addiction. At the same time, this work demonstrates some of the main limitations of the literature on Internet addiction. The way forward is to first disentangle various Internet-related activities that are pursued excessively and to one’s detriment and better discriminate them from their normal variants. The rewards, faulty control and risky decision making that may be relatively specific for these behaviours should then be studied separately, ideally within a longitudinal, prospective design while controlling for the potentially confounding variables such as certain types of psychopathology and personality traits. This might lead to a breakthrough in our understanding of the various problematic behaviours mediated by the Internet.

See Research by Seok et al., 49(10): 923–932.