‘Behavioural addiction’ and ‘selfitis’ as constructs – The truth is out there! A reply to Starcevic et al. (2018)

The article by Starcevic et al. (2018) made a number of assertions concerning my research. Some of the assertions made were arguably unfair, misguided and/or not stated in context.

The construct of ‘selfitis’

Starcevic et al. noted that there has been a trend ‘to medicalize problematic behaviours’ (p. 1) and used the example of ‘selfitis’ to make their point. The way the article was written it would appear to the naïve reader that I and my co-author (Janarthan Balakrishnan) had coined the term ‘selfitis’. For instance, the article by Starcevic et al. cites our paper in specific reference to the following assertion:

Instead of labelling an excessive and sometimes dangerous practice of taking selfies a ‘selfie addiction’, this behaviour was conceptualised as an inflammation-like selfitis. (Balakrishnan and Griffiths, 2017)

This sentence clearly gives the impression that we conceptualised ‘selfitis’ and that our conceptualisation was that it was ‘inflammation-like’. However, we made it very clear to readers in the very first paragraph of our paper that the concept of ‘selfitis’ originally started a hoax claiming that the ‘disorder’ was to be included in the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders. The original hoax report defined selfitis as ‘the obsessive compulsive desire to take photos of one’s self and post them on social media as a way to make up for the lack of self-esteem and to fill a gap in intimacy’ which we again made clear in the second sentence of our paper. The two studies in our paper were exploratory and merely set out to examine whether there were individuals who were ‘obsessive selfie-takers’. In many parts of their article, Starcevic et al. appear to insinuate that our paper equates ‘selfitis’ with ‘selfie addiction’. For instance, they wrote,

Interestingly, the components of selfitis that were identified (environmental enhancement, social competition, attention seeking, mood modification, self-confidence and subjective conformity) have practically nothing in common with behavioural addiction … Therefore, selfitis appears to be a construct that is very different from ‘selfie addiction’, and its purported link with compulsivity also seems tenuous. (p. 1)

The six components comprising selfitis in our psychometric tool (Selfitis Behavior Scale [SBS]) were correctly reported but at no point in our paper did we ever say that ‘selfitis’ was a behavioural addiction. What we did write was that (1) ‘selfitis is a new construct in which future researchers may investigate further in relation to selfitis addiction and/or compulsion’ (p. 8) and (2) ‘the qualitative focus group data from participants strongly implied the presence of “selfie addiction” although the SBS does not specifically assess selfie addiction’ (p. 11). They also noted that our published paper on selfitis

… did not go unnoticed by the media, always ready to exploit everything that is ‘novel’ and sensational. Thus, one newspaper reported that selfitis, ‘the obsessive need to post selfies’, was a ‘genuine mental disorder’ and quoted one of the authors of the aforementioned article that the existence of selfitis appeared to be confirmed. (p. 2)

While it is true that our study did not go unnoticed by the media (and was reported in hundreds of worldwide news stories), only one newspaper journalist ever interviewed me about the study and at no point either in our published paper or in any conversations with the broadcast media did we ever say that ‘selfitis’ was a mental disorder. Our paper simply concluded that obsessive selfie-taking was a condition that appears to exist and made the observation that selfitis has ‘psychological consequences (which may be both positive and negative)’ (p. 12). In fact, we talked about the positive aspects of selfitis throughout the discussion section of our paper. In short, I would like it to be made clear that (1) we did not coin the term ‘selfitis’, (2) we have never anywhere in published print (academic papers or the print media) claimed selfitis is a mental disorder, (3) we have never claimed selfitis is a behavioural addiction and (4) we have never equated ‘selfitis’ with ‘selfie addiction’.

The construct of ‘behavioural addiction’

Starcevic et al. also claimed in their article that the term ‘behavioural addiction’ is ‘vague, misused and applied to an exceptionally wide variety of activities’ (p. 1). I would argue that the far from being ‘vague’, behavioural addiction has clearly been defined as any addiction that does not involve the ingestion of a psychoactive substance (Griffiths, 2005). I agree that it is sometimes misused and I have written dozens of populist articles on my personal blog pointing this out (https://drmarkgriffiths.wordpress.com). However, I totally disagree that behavioural addiction has been applied to an ‘exceptionally wide variety of activities’. As I noted in a recent paper, ‘Very few of the thousands of leisure activities that individuals engage in have ever been written about in terms of addiction in peer-reviewed scientific papers’ (Griffiths, 2017: 1719). Starcevic et al. would be hard pushed to name more than about 20 leisure activities that have ever been empirically examined as a possible behavioural addiction. Of the five activities named by Starcevic et al. in an attempt to show the behavioural addiction is being misused, three of them were actually just sub-types of more widely researched behavioural addictions (i.e. stock market addiction is a sub-type of gambling addiction, study addiction is a sub-type of work addiction and dance addiction is a sub-type of exercise addiction) as made clear in my papers on these topics.

Starcevic et al. also noted that Kardefelt-Winther et al. (2017) ‘recently made an effort to reach a consensus, promote conceptual rigour and avoid misuse by proposing an open (modifiable) definition of behavioural addiction’ (p. 1). More specifically, they provided four exclusion criteria and argued that behaviours should not be classed as a behavioural addiction if:

1. ‘The behaviour is better explained by an underlying disorder …

2. The functional impairment results from an activity that … is the consequence of a willful choice …

3. The behaviour … [results in] insignificant functional impairment or distress for the individual.

4. The behaviour is the result of a coping strategy’ (p. 1710).

No-one researching behavioural addictions would disagree with the third exclusion criterion because genuine behavioural addictions always comprise significant functional impairment/distress for the individual. However, I would point out that if the remaining criteria were applied to substance abuse, few substance users would ever be classed as addicted (Griffiths, 2017). More specifically, I have noted that three of the exclusion criteria proposed by Kardefelt-Winther et al. (2017) are simply untenable:

… it is proposed that any behaviour in which functional impairment results from an activity that is a consequence of wilful choice should not be considered an addiction. I cannot think of a single addictive behaviour that when the person first started engaging in the behaviour (e.g., drinking alcohol, illicit drug-taking, gambling) was not engaged in wilfully … Also, not being classed as an addiction if the behaviour is secondary to another comorbid behaviour (e.g., a depressive disorder) or is used as a coping strategy again means that some other substance addictions (e.g., alcoholism) would not be classed as genuine addictive behaviours using such exclusion criteria because many substance-based addictions are used as coping strategies and/or are symptomatic of other underlying pathologies. (Griffiths, 2017: 1718–1719)

I never look at a behaviour and claim that it cannot be potentially addictive. Using my own operational criteria for addiction (i.e. salience, conflict, tolerance, withdrawal, mood modification, and relapse; Griffiths, 2005), very few individuals would be classed as being addicted to activities such as sex, work, exercise or gaming. However, if there is evidence of all core components of addiction in activities that others believe should not be pathologised, I would not ignore such evidence if they caused significant functional impairment/distress for the individuals concerned.