Problematic Use of the Internet (PUI) is defined by loss of control over internet-related preoccupations, urges and behaviours that result in marked distress and life impairment (Weinstein & Lejoyeux, 2015). PUI is encompasses various problematic online behaviours, such as gaming, gambling, pornography, entertainment streaming, and social networking (Fineberg et al., 2018). PUI affects approximately 7 % of the global population (Pan et al., 2020), though prevalence estimates range from 15 to 30 % amongst young adults and university students (Chi et al., 2016, Zhang et al., 2018), who appear to face increased vulnerability to developing problematic patterns of internet use. For individuals with PUI, online behaviours that were initially recreational and controlled become increasingly uncontrolled and maladaptive (Fineberg et al., 2018), with negative repercussions such as diminishing academic or professional performance, dysphoric mood states, impaired sleep, and interpersonal difficulties (Tokunaga, 2017).

Clinical classification of PUI remains contentious, with three distinctly different conceptualisations emerging from the literature (Tokunaga & Rains, 2016): the first views PUI as a behavioural addiction with features of withdrawal and tolerance (Kuss & Pontes, 2019); the second views PUI as an impulse-control issue underpinned by poor top-down control of internet-related impulses (Young, 1998); and the third views PUI as an artefact of difficulty building in-person social relationships and/or symptoms of depression and loneliness (Caplan, 2002). Whilst these accounts offer seemingly conflicting frameworks, they share two commonalities. Firstly, all three perspectives consider loss of inhibitory control – a central component of impulsivity – over internet-related urges to be a core feature of PUI (Tokunaga & Rains, 2016). Secondly, all three perspectives concur that PUI involves habitual repetition of online behaviours that oftentimes lack reward – namely, they emphasise a compulsive pattern of behaving (Spada, 2014). Thus, despite the lack of clinical consensus, the common denominator of these perspectives is that certain dimensions of impulsivity (actions that are rapid or unplanned, with insufficient regard for undesirable consequences; Fineberg et al., 2014) and compulsivity (maladaptive repetitive actions that persist despite undesirable consequences; Fineberg et al., 2014) may be the driving mechanisms of PUI. Consequently, testing both constructs may offer a potentially theoretically unbiased and dimensional avenue through which to understand the role of cognition in PUI.

Prominent theoretical models of impulsivity and compulsivity propose that these constructs are multidimensional, and dysfunction across dimensions of both constructs are transdiagnostic features of addictive and compulsive disorders (Cassú-Ponsatí et al., 2021, Fineberg et al., 2014, Rochat et al., 2018). While various theoretical accounts identify differing components of impulsivity (Stahl et al., 2014, Verdejo-Garcia et al., 2021), converging evidence indicates that impulsivity comprises at least three dimensions: disinhibition of pre-potent motor responses (response inhibition), disadvantageous or risky decision-making, and deficits in reward valuation (Cassú-Ponsatí et al., 2021, Fineberg et al., 2014, Lee et al., 2019a). Likewise, compulsivity is comprised of at least four dimensions: difficulty shifting behaviour in response to altered stimulus–response contingencies (cognitive flexibility), difficulty shifting between competing tasks or rules (set-shifting, a lower level of cognitive flexibility; Dajani and Uddin, 2015), difficulty disengaging attention from rewarding stimuli (attentional bias), and tendency towards repetitive habitual behaviours (habit formation; Fineberg et al., 2014, Lee et al., 2019a). Studies have predominantly focused on the role of impulsivity, with meta-analytic evidence indicating inhibitory control and decision-making deficits in individuals with PUI (Ioannidis et al., 2019). However, research into the role of compulsivity remains noticeably sparse. Research into behavioural aspects of compulsivity has shown that PUI can be predicted by compulsive features (e.g., checking compulsions, obsessive impulses) independent of impulsivity, and warrants further investigation into the role of cognitive compulsive drivers (Ioannidis et al., 2016). However, the few studies to investigate compulsive cognitive mechanisms have either examined a single dimension of compulsivity, such as attentional bias, in isolation with no consideration for the effects of other cognitive domains (Dong et al., 2014, Nikolaidou et al., 2019), or have investigated only a subgroup of PUI, such as role-playing online gamers (Metcalf & Pammer, 2011). Investigating multiple dimensions of both impulsivity and compulsivity simultaneously offers a more comprehensive understanding of the cognitive drivers PUI. Furthermore, research into the role of impulsivity and compulsivity in substance use disorders (Verdejo-Garcia et al., 2019) and gambling disorder (Mallorquí-Bagué et al., 2018) have yielded potential therapeutic targets and facilitated the implementation of novel treatment programs – likewise, investigating impulsive-compulsive mechanisms in PUI may reveal similarly optimal treatment targets.

Therefore, the current study aimed to apply a theory-driven model of impulsivity and compulsivity in individuals with PUI via comprehensive testing of multiple cognitive dimensions: response inhibition, risky decision-making, reward valuation, reward-related attentional bias, cognitive flexibility, and habit formation. Traditional cognitive tasks are typically repetitive, onerous for participants, and difficult to accurately administer outside of laboratory settings as they rely heavily on intrinsic motivation (Lumsden et al., 2016). To overcome this challenge, the study utilised gamified cognitive tasks, which facilitate participant engagement and motivation while maintaining psychometric validity and reliability (Lumsden et al., 2016, Lumsden et al., 2017). Self-report measures were also incorporated, as they may offer greater sensitivity to inter-individual differences (Enkavi et al., 2019b) Fig. 1. Self-report measures also test differing, complimentary dimensions of cognitive constructs (Eisenberg et al., 2019), enabling more comprehensive assessment of impulsivity and compulsivity than tasks alone. It was hypothesised that scores on cognitive task and self-report measures of impulsivity (response inhibition, risky decision-making, reward valuation) and compulsivity (reward-related attentional bias, cognitive flexibility, habit formation) would predict whether an individual has PUI.