Abstract
Background and aims
Theories posit that the combination of external (e.g. cue exposure) and internal (e.g. attention biases) factors contributes to the development of game craving. Nevertheless, whether different components of attentional biases (namely, engagement bias and disengagement bias) play separate roles on game craving has not been fully elucidated. We aimed to examine the associations between two facets of attentional biases and game craving dynamics under a daily life setting.
Methods
Participants (110 regular internet game players) accomplished the modified attentional assessment task in the laboratory, after which they entered a 10-day ecological momentary assessment (EMA) to collect data on their momentary game craving and occurrence of game-related events at five different time points per day.
Results
We found that occurrence of game-related events was significantly associated with increased game craving. Moreover, attentional disengagement bias, instead of engagement bias, bore on the occasional level variations of game craving as moderating variables. Specifically, attentional disengagement bias, not engagement bias, was associated with a greater increase in game craving immediately after encountering a game-related event; however, neither attentional engagement bias nor disengagement bias was associated with the craving maintenance after a relatively long period.
Discussion and conclusions
The present study highlights the specific attentional processes involved in game craving dynamics, which could be crucial for designing interventions for attentional bias modification (ABM) in Internet Gaming Disorder (IGD) populations.
Online games have become popular, especially since online multiplayer games that serve social and recreational purposes emerged (Nuyens et al., 2016). They have become an indispensable part of many people's lives as a popular pastime. However, overindulgence could lead to players developing Internet gaming disorder (IGD), “characterized by persistent gaming and functional impairment in multiple areas of life” (King & Delfabbro, 2018, p. 17). Several researchers have recently advocated IGD as an essential public health issue worthy of attention (King, Koster, & Billieux, 2019; Stein et al., 2018). In line with this, the Diagnostic and Statistical Manual for Mental Disorders, Fifth Edition (DSM-5; APA, 2013), includes IGD as a condition for further study in Section 3. Furthermore, in 2018, the World Health Organization officially recognized IGD as an addictive disorder in the International Classification of Diseases, 11th Revision (ICD-11; WHO, 2018). Identifying potential psychological mechanisms underlining IGD development is of great significance theoretically and practically.
Dual process model of addiction (Brand, 2022; Wiers, Gladwin, Hofmann, Salemink, & Ridderinkhof, 2013) highlights the combination of strengthened cravings to cues and weakened reflective processes as the defining feature in addiction. Craving, often defined as the subjective, intense desire or urge to initiate addictive behaviors (Giuliani & Berkman, 2015; Lavallee, 2020), serve a crucial role in drug addictions (APA, 2013), as well as behavioral addictions (Dong et al., 2018; Potenza et al., 2003; Wegmann, Muller, Trotzke, & Brand, 2021). For instance, previous studies have consistently found heightened game craving in individuals with IGD exposed to game-related cues (Bargeron & Hormes, 2017; Dong, Dong, et al., 2021; Shin et al., 2018). Additionally, game craving was positively correlated with problematic game playing (Moretta & Buodo, 2018), general psychopathology (i.e., depression, anxiety, and stress), decreased life satisfaction, and impulsivity (Bargeron & Hormes, 2017). In a recent international Delphi study, game craving, though not included in the ICD-11 and DSM-5 definitions, was considered an additional relevant diagnostic criterion with sufficient prognostic value (Castro-Calvo et al., 2021).
Diverse paradigms have been developed in laboratory settings to explore game craving. The most widely used ones usually involve cue exposure procedures (e.g., presenting participants with game-related pictures and sounds; Dong, Dong, et al., 2021; Niu et al., 2016). However, artificial stimuli and strict laboratory settings far from real-life contexts did not fully reveal the ephemeral and fluctuant nature of craving dynamics (Enkema, Hallgren, Bowen, Lee, & Larimer, 2021). Fortunately, mobile technologies, such as ecological momentary assessment (EMA; Stone & Shiffman, 1994), have permitted real-time data collection in natural contexts and therefore offer a solution to numerous methodological barriers to emotion and addiction research (Mun et al., 2021). The feasibility and validity of the EMA have been repeatedly established in individuals with different addiction syndromes (see review from Serre, Fatseas, Swendsen, & Auriacombe, 2015). Further, they are well suited to explore the craving state as a function of various determinants. Repeated within-day assessments capture fluctuations in craving dynamics and inform researchers about prospective relationships (Shiyko & Ram, 2011).
Theories posit that a combination of external and internal factors contributes to game craving and addiction development (Brand, Young, Laier, Wölfling, & Potenza, 2016, 2019). Game craving has been conceptualized as an outcome of classical conditioning (Heuer, Mennig, Schubö, & Barke, 2021). Specifically, when environmental cues (e.g., the sight of game advertisements and the sound of game characters) have been repeatedly associated with playing games, they become conditioned stimuli that elicit gaming urges. This explanation is supported by Dong et al. (2021) and Shin et al. (2018). They found that exposure to both real and virtual gaming environments increased subjective craving among problematic Internet game players. Recent neurocognitive studies have also investigated potential behavioral and neural mechanisms of cue reactivity and craving in IGD populations (see review from Fauth-Bühler & Mann, 2017). The up-to-date findings demonstrated the involvement of both the ventral and dorsal striatum (Dong, Dong, et al., 2021), which give preliminary support for parallels among IGD, pathological gambling (Crockford, Goodyear, Edwards, Quickfall, & el-Guebaly, 2005) and substance-use disorders (Skinner & Aubin, 2010).
Additionally, theories postulate that internal factors, such as attentional bias, are closely associated with the development of craving (Brand et al., 2016, 2019). Attentional bias is the preferential attention of addiction-related stimuli, making it difficult to deal with the current task (Field & Cox, 2008) and having a common neural mechanism with craving (Hester & Luijten, 2014). The celebrated incentive sensitization theory of addiction (Robinson & Berridge, 2008) purports that persistent addictive behaviors can cause “sensitization” or hypersensitivity to addiction-related stimuli. This incentive-sensitization produces attentional bias toward those stimuli, leading to enhanced craving and addiction motivation. Although theoretically crucial in addiction, investigation of attentional bias in individuals with IGD has been scarce, producing mixed evidence: some have found evidence of attentional bias toward gaming-related stimuli using classical paradigms, such as Stroop or dot-probe tasks (Jeromin, Nyenhuis, & Barke, 2016; Lorenz et al., 2013; Metcalf & Pammer, 2011), whereas others have not (Van Holst et al., 2012; Zhang et al., 2016). This may be because of the conceptual distinction between facilitated attentional engagement with gaming-related stimuli (reflecting a disproportionate tendency for attention to be captured by gaming-related stimuli), and impaired attentional disengagement from gaming-related stimuli (reflecting an excessive predilection for attention to be held by gaming-related stimuli). Specifically, Heuer et al. (2021) found that IGD gamers demonstrated longer reaction times than casual gamers. Further, they demonstrated increased sustained posterior contralateral negativity (SPCN) amplitude to computer-related stimuli in a visual search task, indicating that the disengagement of attention from addiction-related stimuli was impaired in IGD individuals. Contrastingly, the initial deployment of attention to addiction-relevant stimuli was relatively intact.
Based on the aforementioned findings and related theories, we tentatively propose two possible pathways through which external cues and internal attentional biases contributes to game craving development: the reactivity model (MacLeod, Rutherford, Campbell, Ebsworthy, & Holker, 2002) and the perseverance model (Joormann, Yoon, & Zetsche, 2007). Attentional biases to game-related information (especially for impaired attentional disengagement) contribute to prolonged and perseverative game-related information processing (Koster, De Lissnyder, Derakshan, & De Raedt, 2011). These, in turn, may amplify the emotional impact of game cue exposure regarding acute craving change (reactivity model) and/or persistent high craving maintenance (perseverance model).
In sum, the present study aimed to examine the factors contributing to daily game craving dynamics and test specific models in a college sample who regularly played Internet games. First, we compared general game craving, attentional biases, and gaming behaviors among Internet gamers with different levels of IGD tendencies. We hypothesized that Internet gamers with higher levels of IGD would have higher levels of game craving, demonstrate altered attentional biases (especially in attentional disengagement), and spend more time playing Internet games compared with lower IGD tendency gamers. Subsequently, multilevel modeling was used to examine the specific role of attentional biases and everyday game cues in developing game cravings. We hypothesized that both the reactivity and perseverance model were valid in the current study.
Method
Participants and procedures
Participants were 110 (39 female and 71 male; Mage = 19.05 years; SD = 0.93 years) undergraduate students who played the game of Glory of Kings1 (Tencent Company) at the Zhejiang University, China.
First every participant provided a signed informed consent after being briefed of the purpose of the study. Later, they completed the dichotomous 9-Item Internet Gaming Disorder Scale (Lemmens, Valkenburg, & Gentile, 2015), the Internet Gaming Addiction Scale (revised from the Internet Addiction Scale; Young, 1999), the Craving Questionnaire (Weiss, Griffin, & Hufford, 1995), and the modified attentional assessment task (Rudaizky, Basanovic, & MacLeod, 2014). Beginning on the following day, participants took part in 10 days of EMA sampling (Including 6 working days and 4 weekends). After completing all the sessions, participants were debriefed and were given monetary compensation based on their EMA compliance. The sample size meets the guideline of 30 individuals for 30 observations (the 30/30 rule; e.g., Hox, 2010). Other studies indicated that the 30/30 rule can achieve a statistical power greater than 0.80 to detect a medium-to-large fixed effect (Mathieu, Aguinis, Culpepper, & Chen, 2012).
Measures
Questionnaires
9-Item Internet Gaming Disorder Scale (IGD-9)
IGD-9 (Lemmens et al., 2015) is an 9-item online gaming disorder tendency scale developed based on the 9 diagnostic criteria (including preoccupation, withdrawal, tolerance, unsuccessful attempts to control, loss of other interests, continued excessive use despite psychosocial problems, deceiving regarding online gaming, escape, and functional impairment) formulated in DSM-5 (APA, 2013). Respondents need to rate all items with either NO (0) or YES (1). The item scores are summed to arrive at a total score, with higher scores indicating higher online gaming disorder tendency (range 0–9). If the total score is 5 points or above, the respondent is believed to meet the diagnostic criteria of IGD (APA, 2013). The internal consistency of the IGD-9 in the present study was acceptable (α = 0.65).
Internet Gaming Addiction Scale (IGA)
The Internet Gaming Addiction Scale (IGA) is revised from the Internet Addiction Scale developed by Young (1999). The original scale provides a framework for assessment of specific situations or problems that have been caused by computer overuse to facilitate subsequent treatment planning. This scale consists of 20 items and respondents need to rate all items on a 5-point scale, ranging from 1 (never) to 5 (always). The item scores are summed to arrive at a total score (range 20–100), with higher scores indicating higher Internet Addiction tendency. We replace the word “Internet” for the phrase “Internet gaming” to measure the level of Internet Gaming Addiction in this study. If the total score is 50 points or above, the respondent is believed to meet the moderate to severe level of IGD (Young, 1999). The internal consistency of the IGA in the present study was good (α = 0.90).
Craving Questionnaire (CQ)
The Craving Questionnaire is revised from the Cocaine Craving Scale developed by Weiss et al. (1995), which is designed to assess five dimensions of craving: 1) immediate intensity, 2) intensity during the previous 24 h, 3) frequency, 4) reactivity of craving to game-related cues, and 5) likelihood of playing if in an environment with high online game availability. This scale consists of 5 items, and each item is rated on a 10-point scale, ranging from 1 (not at all) to 10 (very much so). The item scores are summed to arrive at a total score, with higher scores indicating higher craving trait. The internal consistency of the CQ in the present study was good (α = 0.88).
Modified attentional assessment task
We adapted attentional engagement bias and disengagement bias assessment task, developed by Rudaizky et al. (2014), to enable the independent assessment of each facet of selective attention. In this task, each trial begins with the displaying of two 200 × 150 pixels white rectangle outlines, on both sides of the screen. An 80 × 60 pixels red square outline appears in one of these white rectangles with equal frequency, signaling the location where the cue probe will briefly appear. Meanwhile, participants are required to fixate their attention at the red square. After 1,000 ms, a 20 pixels red line will appear within the red square outline, with equal probability of being horizontal or vertical. The cue probe will remain for only 200 ms, therefore its orientation can only be perceived if participants have already attended to the cued location. Immediately thereafter an image pair appears, with one image filling each of the two white rectangles. One image is abstract and the other representational.2 The representational image appears with equal frequency either in the distal location from that where the participant is already attending (attentional engagement bias assessment trials) or in the same location as the participant is already attending to (attentional disengagement bias assessment trials). This image display lasts for either 500 ms or 1,000 ms, with equal probability to prevent participants from fixed response tendency. But in the present study, we emphasized on the trials with a duration of 500 ms only because they can better reflect the automatic attentional bias mode (Grafton & MacLeod, 2016). The screen then clears, and a target probe immediately appears in either one of these two screen regions, with equal probability. Again this is a 20 pixels red line, either horizontally or vertically. Participants are required to indicate whether the orientation of this target probe matched that of the cue probe, which was the case on 50% of trials. Participants respond by pressing “j” or “f” key on the keyboard to respectively indicate that the probes match or mismatch. Reaction times (RTs) to make this probe discrimination decision are recorded and used to calculate the relevant indices of attentional bias (see below). After a 1,000 ms interval, the next trial begins. An example trial is presented in Fig. 1.
Example of a trial on the modified attentional assessment task. Trial displayed is a disengagement bias assessment trial. Attentional focus is initially anchored proximally to the game-related image
Citation: Journal of Behavioral Addictions 11, 4; 10.1556/2006.2022.00085
Engagement bias index = (Cue probe distal to game-related image in game-related/abstract image pair: RT for target probe distal to game-related image minus RT for target probe proximal to game-related image) minus (Cue probe distal to neutral image in neutral/abstract image pair: RT for target probe distal to neutral image minus RT for target probe proximal to neutral image). A higher score on this engagement bias index will represent selectively enhanced shifting of attention towards initially unattended distal images when these are game-related rather than neutral.
Disengagement bias index = (Cue probe proximal to game-related image in game-related/abstract image pair: RT for target probe distal to game-related image minus RT for target probe proximal to game-related image) minus (Cue probe proximal to neutral image in neutral/abstract image pair: RT for target probe distal to neutral image minus RT for target probe proximal to neutral image). A higher score on this disengagement bias index will represent a heightened tendency for attention to be held in the locus of initially attended proximal images when these are game-related rather than neutral.
In the present study, the task contained a total of 256 trials. In the task, each image pair was presented twice, left or right position was counterbalanced across participants, and the order of presentation of the images was randomized. Participants were instructed to respond as quickly as possible while maintaining accuracy. After every 64 trials participants were instructed to take a self-timed break. Before the formal experiment, participants performed 10 practice trials, including 10 pairs of neutral images (different from the neutral images in the formal experiment). Participants entered the formal experiment after the correct rate of practice trials reached 75%.
Ecological Momentary Assessment (EMA)
We used a mobile phone-based EMA to collect data on participants' momentary internal and external experiences (Takano & Tanno, 2011). Participants received 5 messages on their own mobile phones between 9:40 and 22:30 each day (the five time points are 9:40, 13:00, 16:30, 19:10, 22:30, to stagger the participants' class schedule). Each message contained a URL for an online questionnaire (see Table 1). When participants received the e-mail, they were required to access and complete the questionnaire within 15 min. The EMA sampling continued over 10 consecutive days, including 4 weekend days.
Test items included in Ecological Momentary Assessment (EMA)
| EMA items | Response options |
| Craving intensity: How strong do you want to play online games now? | Rating 0–100 |
| Anxiety intensity: How anxious you are now? | Rating 0–100 |
| Game thoughts episode: Since last questionnaire, how many times did you experience urges to play online games? | ( ) Times |
| Have you played any online games? | Yes(1)/No(0) |
| How long did you play? | ( ) Minutes |
| What online game did you play | (The name of online games) |
| Game-related event occurrence: Since last questionnaire, did something happen that remind you of online games? | Yes(1)/No(0) |
| a Gaming duration: How much time did you spend playing online games, in total today? | ( )hours |
Note. aGame duration was assessed in the last questionnaire (22:30) in each day.
Statistical analyses
Attentional biases indices
For attentional bias indices, the mean response latencies observed under the different task conditions were used to calculate the engagement bias and disengagement bias indices of attentional preference for game-related information for each participant. In computing these indices, response latencies from incorrect trials were excluded. In addition, outliers were identified using a 99% confidence level meaning scores that were greater than 2.58 standard deviations from the participant's mean latency for that experimental condition were excluded (in total 7% of trials were excluded from the analysis).
EMA data preparation
During the 10-day EMA sampling, we excluded data from five participants: one made errors on more than 30% of the trials in the attentional assessment task, and 4 participants did not complete EMA seriously whose responses were all 0.
Multilevel modeling
To test the association between attentional biases and craving reactivity and perseverance, we estimated two multilevel models, wherein game craving intensity at time t was predicted a) by game-related event at time t and attentional biases (i.e., the reactivity model; cf. Bylsma, Taylor-Clift, & Rottenberg, 2011; Peeters, Berkhof, Rottenberg, & Nicolson, 2010), and b) by game-related event at time t-1 and attentional biases (i.e., the perseverance model; cf. Iijima, Takano, & Tanno, 2018). Note that game-related event at time t refers to an event that happened in the interval between the last (time t-1) and current (time t) occasions.
The γ200 coefficient indicates the main effect of the game-related event occurrence, and γ201 reflects the cross-level interaction effect between attentional biases and game-related event occurrence.
The perseverance model was the same as the reactivity model, except for the timing of game-related event: the model included negative event at time t-1 instead of at time t as a predictor.
Therefore, the reactivity model tested an immediate craving change in response to a game-related event that occurred in the interval between time t-1 and t, whereas the perseverance model captured a persistent process following a game-related event that occurred during the interval from time t-2 to t-1. This conceptualization of the craving perseverance drew on emotion dynamic research by Koval et al. (2015), wherein perseverance was operationally defined as the difference in effect between time t and t-1, when an event was reported at time t-1 (i.e., occurring since the previous occasion). All multilevel modeling analyses were conducted using HLM Version 6.08.
Ethics
This study was approved by the Ethics Committee of the department of psychology and behavioral science of the Zhejiang University.
Results
Group differences in attentional bias indices and EMA summary
Descriptive statistics are presented in Table 2. Generally, the current sample did not meet the diagnostic criteria in IGD-9 scale (M = 3.41, SD = 1.93), whereas they reach the moderate level of IGD in IGA scale (M = 52.41, SD = 11.99). In order to determine whether high and low disorder tendency online gaming participants differed in terms of sample characteristics, attentional bias indices, and EMA descriptive statistics, participants first were divided into different IGD tendency groups based on their IGD-9 and IGA scores (high IGD tendency: 50 and above in IGD-9, as well as 50 and above in IGA, n = 30; low IGD tendency: below 5 in IGD-9, as well as below 50 in IGA, n = 32; APA, 2013; Young, 1999). Results demonstrated that two groups did not differ in terms of age (P > 0.050) and gender (P > 0.050). High IGD tendency group showed significantly higher scores on IGD-9, IGA, and CQ (all P < 0.001) compared with low IGD tendency group (see Table 2).
Main sample characteristics, attentional bias indices, and EMA summary
| Group | t (p) | 95% CI | All (N = 105) M (SD) | ||
| Variable | d Low disorder tendency (n = 32) M (e SD) | f High disorder tendency (n = 30) M (SD) | |||
| Main Sample Characteristics | |||||
| Age | 19.03 (0.86) | 19.17 (1.26) | 0.49 (0.621) | [−0.41, 0.68] | 19.08 (0.94) |
| Gender (male) | 12 | 21 | 69 | ||
| a IGD-9 | 1.88 (1.07) | 5.83 (0.95) | 15.36 (0.000) | [3.44, 4.47] | 3.41 (1.93) |
| b IGA | 38.44 (7.53) | 63.70 (8.42) | 12.47 (0.000) | [21.21, 29.31] | 52.41 (11.99) |
| c CQ | 20.59 (6.14) | 29.50 (7.61) | 5.09 (0.000) | [5.40, 12.41] | 25.36 (7.99) |
| Attentional Bias Indices | |||||
| Engagement bias index | 19.01 (73.25) | 1.30 (126.33) | −0.67 (0.506) | [-70.97, 35.53] | 18.07 (108.79) |
| Disengagement bias index | −37.61 (99.78) | 25.35 (75.96) | 2.78 (0.007) | [17.68, 108.23] | 2.28 (108.93) |
| EMA Summary | |||||
| Game Craving - Person Mean | 18.22 (15.31) | 27.21 (18.42) | 2.10 (0.040) | [0.41, 17.58] | 24.70 (17.10) |
| Anxiety–Person Mean | 26.25 (19.70) | 31.80 (16.20) | 1.21 (0.232) | [-3.64, 14.75] | 28.02 (16.95) |
| Thoughts to play games between two occasions | 0.54 (0.35) | 1.11 (1.12) | 2.66 (0.012) | [0.13, 0.99] | 0.82 (0.74) |
| Time on playing games between two occasions/min | 12.28 (10.51) | 18.12 (13.22) | 1.93 (0.058) | [-0.21, 11.89] | 15.32 (12.44) |
| Time on playing games a day h−1 | 1.18 (0.96) | 2.44 (1.95) | −3.19 (0.003) | [-2.47, −0.62] | 1.65 (1.44) |
Note. aIGD-9 = 9-Item Internet Gaming Disorder Scale; b IGA = Internet Gaming Addiction Scale; c CQ = Craving Questionnaire; d Low disorder tendency: below 5 in IGD-9, as well as below 50 in IGA; e SD = standard deviation; f High disorder tendency: 50 and above in IGD-9, as well as 50 and above in IGA.
For attentional bias indices, the independent-samples t test showed significant higher attentional disengagement bias scores in the high IGD tendency group compared to the low IGD tendency group (M = 25.35, SD = 75.96 vs. M = −37.61, SD = 99.78, P = 0.007, Cohen's d = 0.71). However, two groups did not differ in terms of attentional engagement bias (M = 1.30, SD = 126.33 vs. M = 19.01, SD = 73.25, P > 0.050, Cohen's d = 0.17). In addition, low and high disorder tendency online gaming participants did not differ in terms of accuracy (M = 0.93, SD = 0.04 vs. M = 0.94, SD = 0.04, P > 0.050, Cohen's d = 0.25). This possibly suggested that the higher disorder tendency online gaming participants displayed difficulty in disengaging from initially proximal game-related images.
For EMA descriptive statistics, we computed person-mean game craving, anxiety, number of thoughts to play games, and time on playing games, and made comparisons between two groups. High IGD tendency group generally demonstrated higher levels of game craving (M = 27.21, SD = 18.42 vs. M = 18.22, SD = 15.31, P = 0.040, Cohen's d = 0.53), had more thoughts to play online games between two occasions (M = 1.11, SD = 1.12 vs. M = 0.54, SD = 0.35, P = 0.012, Cohen's d = 0.69), and spent more time on playing online games each day (M = 2.44, SD = 1.95 vs. M = 1.18, SD = 0.96, P = 0.003, Cohen's d = 0.82). Two groups did not differ in terms of general anxiety (M = 31.80, SD = 16.20 vs. M = 26.25, SD = 19.70, P > 0.050, Cohen's d = 0.31). Correlations between attentional bias and craving dynamics3 parameters can be found in Supplementary materials.
Attentional disengagement bias moderates craving reactivity but not perseverance
Next, we examined the specific associations between attentional biases (engagement and disengagement respectively) and craving reactivity and perseverance. To test the enhanced reactivity hypothesis, we estimated a multilevel model wherein temporary game craving at time t was predicted by occurrence of a game-related event (from time t-1 to t), attentional biases, and their interaction, while controlling for game craving at the previous assessment occasion (time t-1) and current anxious level (time t). The estimated fixed effects are presented in Table 3. Game-related event occurrence (from time t-1 to t) was significantly associated with increased game craving at time t, as a main effect. Furthermore, this main effect was qualified by the interaction with attentional disengagement bias (t = 2.91, P = 0.004), but not engagement bias (t = 0.40, P > 0.050). To further illustrate the pattern of this interaction, a post hoc analysis was used to test the simple slopes of game-related event for individuals with relatively higher and lower levels of attentional disengagement bias (M ± 1SD; Preacher, Curran, & Bauer, 2006). The conditional effect of game-related event occurrence on game craving was greater for individuals with higher levels of attentional disengagement bias (B = 16.58, SE = 1.65, t = 10.04, P < 0.001) than for those with lower levels of bias (B = 8.48, SE = 1.70, t = 4.99, P < 0.001). These findings suggest that attentional disengagement bias, not engagement bias, serves as a moderator that amplifies immediate craving elevation in response to a game-related event, which supports the hypothesis that attentional disengagement bias is associated with increased game craving reactivity to game-related cues.
Estimated fixed effects of the multilevel model predicting momentary game craving (at Time t) for Reactivity and Perseverance models
| Variables | Reactivity model | Perseverance model | ||||
| Coefficient (b SE) | t | P | Coefficient (SE) | t | P | |
| Person-level predictors | ||||||
| Intercept | 13.54 (1.85) | 7.34 | 0.000 | 17.64 (1.89) | 9.35 | 0.000 |
| Gender | 3.34 (2.56) | 1.31 | 0.195 | 5.70 (2.62) | 2.17 | 0.032 |
| a IGD-9 | 0.76 (0.78) | 0.97 | 0.334 | 0.88 (0.76) | 1.16 | 0.251 |
| Occasion-level predictors | ||||||
| Game craving (t-1) | 0.07 (0.03) | 2.68 | 0.007 | 0.09 (0.03) | 3.05 | 0.002 |
| Game-related events (t-1 to t) | 13.83 (1.30) | 10.61 | 0.000 | — | — | — |
| Game-related events (t-2 to t-1) | — | — | — | −0.33 (0.72) | −0.46 | 0.648 |
| Current anxious mood | 0.14 (0.03) | 4.26 | 0.000 | 0.16 (0.03) | 4.72 | 0.000 |
| Cross-level interaction | ||||||
| Disengage × Game-related events (t-1 to t) | 0.03 (0.01) | 2.91 | 0.004 | — | — | — |
| Engage × Game-related events (t-1 to t) | 0.01 (0.01) | 0.40 | 0.690 | — | — | — |
| Disengage × Game-related events (t-2 to t-1) | — | — | — | 0.00 (0.01) | 0.04 | 0.967 |
| Engage × Game-related events (t-2 to t-1) | — | — | — | 0.00 (0.01) | 0.27 | 0.786 |
Note. 4,981 observations across 105 participants were used for model estimation.
a IGD-9 = 9-Item Internet Gaming Disorder Scale;
b SE = standard error.
To test the enhanced perseverance hypothesis, we estimated a similar multilevel model, in which game craving at time t was predicted by the occurrence of a game-related event (time t-2 to time t-1), attentional biases, and their interaction. Table 3 also illustrates the estimated results of this perseverance model, which showed non-significant interaction between attentional biases (both engagement and disengagement) and game-related events (all P > 0.050). These null interactions suggest that both attentional engagement and disengagement bias may bear no influence on the maintenance of craving for a delayed time point.
Discussion
Although theoretically important, game craving has been investigated in a few studies on Internet game players. The present study examined the external and internal factors that contributed to daily game craving dynamics using EMA in a college sample who regularly played Internet games. Our main results illustrate that the occurrence of game-related events4 was significantly associated with increased game craving. More importantly, we explored the moderating function of attentional biases as an internal factor bore on occasional level variations in game craving. Specifically, attentional disengagement bias (not engagement bias) was associated with a more significant increase in game craving immediately after encountering a game-related event (i.e., the reactivity model). However, neither attentional engagement nor disengagement biases were associated with craving maintenance after a relatively long period (i.e., the perseverance model).
Researchers have claimed that attentional disengagement probably distinguishes individuals with an addiction history from those without and can contribute to the maintenance of addictive behaviors (e.g., problematic gambling and substance abuse; Heuer et al., 2021; Hudson, Olatunji, Gough, Yi, & Stewart, 2016). Consistently, we found that individuals with relatively high IGD tendencies had more impaired disengagement of attention from addiction-related stimuli than individuals with low IGD tendencies. This result indicated that attentional disengagement may be a potential vulnerability factor helping to differentiate between different patterns of Internet game use (Brand, 2022). Based on multilevel model analyses, this study also demonstrated that although encountering game-related events amplified momentary craving in all game players, individuals with higher levels of attentional disengagement bias exhibited more significant instant changes. This result supports the distinctive role of attentional disengagement toward addiction-related information on craving, which has not been investigated in regular Internet game players. According to the cognitive processing model (Franken, Stam, Heniks, & van den Brink, 2003), continued attentional processing of addiction-related stimuli may contribute to explicit desire thinking (see review from Brandtner, Antons, Cornil, & Brand, 2021), which is believed to be an essential part of the craving experience that leads to excessive memory bias and fuels the strength of the craving experience. In support of this potential mechanism, we found a significant positive correlation between attentional disengagement bias indices and the number of thoughts about playing online games (person mean) during EMA sampling (r = 0.265, P = 0.006; see Supplementary materials). Conversely, it is also possible that prolonged attentional processing of addiction-related stimuli contributes to the impairment of cognitive control (Dong, Dong, et al., 2021), which may explain the difficulties in controlling gaming cue-elicited cravings. Actually, intensive investigations have demonstrated deficits of self-control in problematic internet gaming use (Brand, 2022). However, the exact mechanism warrants further study.
The present study highlights the specific attentional processes involved in game craving dynamics, which could be crucial for designing interventions for attentional bias modification (ABM, Chia & Zhang, 2020) in IGD populations. We suggest that future studies design effective intervention strategies targeting the components of disengagement. Nevertheless, several limitations should be considered when interpreting the results of this study. First, the present study only focused on two driving factors (namely external cues and internal attentional biases) contributing to game craving. It would be more theoretically-elaborated to consider the self-control process as counterpart, as illustrated in the dual process model of addiction (Brand, 2022; Wiers et al., 2013). Investigations into the dynamic interplays of driving and control processes will produce a more comprehensive understanding of craving development. Second, the definition of reactivity and perseverance depends mainly on the assessment interval of the EMA paradigm. Replications using different timeframes are warranted before coming to a consensus. Third, a single-item visual analogue scale was used to assess the momentary craving. Therefore, we were unable to explicitly distinguish between reward and relief cravings which were motivationally different. Actually, several integrative models have posit craving to have different (sub)dimensions associated with corresponding subtypes of addictive behaviors (Glöckner-Rist, Lémenager, Mann, & PREDICT Study Research Group, 2013; Heinz et al., 2003). Further studies are encouraged to resolve this issue. Forth, the multilevel model analyses included all participants with varying levels of IGD tendency (although IGD-9 scores were controlled for in the model). Future studies should investigate whether Internet game players with or without IGD diagnoses demonstrate different patterns of results.
Conclusion
The present study demonstrated that though encountering game-related events amplified momentary game craving for all internet game players, individuals with higher levels of attentional disengagement bias exhibit more intense elevations in game craving. The component of attentional disengagement distinctively contributes to game craving dynamics, which could be crucial for designing interventions for attentional bias modification in IGD populations.
Funding sources
This research was supported by the following funds: Humanities and Social Sciences Projects of Ministry of Education of China (18YJC190032); Basic Public Welfare Research Project of Zhejiang Province, China (LGF21C090003); Fundamental Research Funds for the Central Universities (2021FZZX001-06); National Natural Science Foundation of China (31200786); National Natural Science Foundation of China (32071044).
Authors’ contribution
Yucheng Zhou, Yanling Zhou, Jifan Zhou, Mowei Shen, and Meng Zhang designed the study and wrote the protocol as a team. Yucheng Zhou, Yanling Zhou, Jifan Zhou, and Meng Zhang conducted literature searches and wrote reviews of prior researches. Yucheng Zhou, Yanling Zhou, and Meng Zhang collected data. Yucheng Zhou, Yanling Zhou, and Jifan Zhou conducted the statistical analysis. Yucheng Zhou and Yanling Zhou wrote the first draft of the manuscript and all authors conducted to and have approved the final manuscript.
Conflict of interest
The authors declare that they have no conflicts of interest.
Supplementary materials
Supplementary data to this article can be found online at https://doi.org/10.1556/2006.2022.00085.
References
American Psychiatric Association (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.
Bargeron, A. H., & Hormes, J. M. (2017). Psychosocial correlates of internet gaming disorder: Psychopathology, life satisfaction, and impulsivity. Computers in Human Behavior, 68, 388–394. https://doi.org/10.1016/j.chb.2016.11.029.
Brand, M. (2022). Can internet use become addictive? Science, 376(6595), 798–799. https://doi.org/10.1126/science.abn4189.
Brandtner, A., Antons, S., Cornil, A., & Brand, M. (2021). Integrating desire thinking into the I-PACE model: A special focus on internet-use disorders. Current Addiction Reports, 8(4), 459–468. https://doi.org/10.1007/s40429-021-00400-9.
Brand, M., Wegmann, E., Stark, R., Müller, A., Wölfling, K., Robbins, T. W., & Potenza, M. N. (2019). The interaction of person-affect-cognition-execution (I-PACE) model for addictive behaviors: Update, generalization to addictive behaviors beyond internet-use disorders, and specification of the process character of addictive behaviors. Neuroscience and Biobehavioral Reviews, 104, 1–10. https://doi.org/10.1016/j.neubiorev.2019.06.032.
Brand, M., Young, K. S., Laier, C., Wölfling, K., & Potenza, M. N. (2016). Integrating psychological and neurobiological considerations regarding the development and maintenance of specific internet-use disorders: An interaction of person-affect-cognition-execution (I-PACE) model. Neuroscience and Biobehavioral Reviews, 71, 252–266. https://doi.org/10.1016/j.neubiorev.2016.08.033.
Bylsma, L. M., Taylor-Clift, A., & Rottenberg, J. (2011). Emotional reactivity to daily events in major and minor depression. Journal of Abnormal Psychology (1965), 120(1), 155–167. https://doi.org/10.1037/a0021662.
Castro‐Calvo, J., King, D. L., Stein, D. J., Brand, M., Carmi, L., Chamberlain, S. R., … Billieux, J. (2021). Expert appraisal of criteria for assessing gaming disorder: An international delphi study. Addiction (Abingdon, England), 116(9), 2463–2475. https://doi.org/10.1111/add.15411.
Chang, Y., Chang, K., Hou, W., Lin, C., & Griffiths, M. D. (2021). Internet gaming as a coping method among schizophrenic patients facing psychological distress. Journal of Behavioral Addictions, 9(4), 1022–1031. https://doi.org/10.1556/2006.2020.00081.
Chia, D. X. Y., & Zhang, M. W. B. (2020). A scoping review of cognitive bias in internet addiction and internet gaming disorders. International Journal of Environmental Research and Public Health, 17(1), 373. https://doi.org/10.3390/ijerph17010373.
Crockford, D. N., Goodyear, B., Edwards, J., Quickfall, J., & el-Guebaly, N. (2005). Cue-induced brain activity in pathological gamblers. Biological Psychiatry (1969), 58(10), 787–795. https://doi.org/10.1016/j.biopsych.2005.04.037.
Dong, G. H., Dong, H., Wang, M., Zhang, J., Zhou, W., Du, X., & Potenza, M. N. (2021). Dorsal and ventral striatal functional connectivity shifts play a potential role in internet gaming disorder. Communications Biology, 4(1), 1–9. https://doi.org/10.1038/s42003-021-02395-5.
Dong, G., Wang, M., Zheng, H., Wang, Z., Du, X., & Potenza, M. N. (2021). Disrupted prefrontal regulation of striatum-related craving in internet gaming disorder revealed by dynamic causal modeling: Results from a cue-reactivity task. Psychological Medicine, 51(9), 1549–1561. https://doi.org/10.1017/S003329172000032X.
Dong, G., Zheng, H., Liu, X., Wang, Y., Du, X., & Potenza, M. N. (2018). Gender-related differences in cue-elicited cravings in internet gaming disorder: The effects of deprivation. Journal of Behavioral Addictions, 7(4), 953–964. https://doi.org/10.1556/2006.7.2018.118.
Enkema, M. C., Hallgren, K. A., Bowen, S., Lee, C. M., & Larimer, M. E. (2021). Craving management: Exploring factors that influence momentary craving-related risk of cannabis use among young adults. Addictive Behaviors, 115, 106750–106750. https://doi.org/10.1016/j.addbeh.2020.106750.
Fauth-Bühler, M., & Mann, K. (2017). Neurobiological correlates of internet gaming disorder: Similarities to pathological gambling. Addictive Behaviors, 64, 349–356. http://doi.org/10.1016/j.addbeh.2015.11.004.
Field, M., & Cox, W. M. (2008). Attentional bias in addictive behaviors: A review of its development, causes, and consequences. Drug and Alcohol Dependence, 97(1), 1–20. https://doi.org/10.1016/j.drugalcdep.2008.03.030.
Franken, I. H. A., Stam, C. J., Heniks, V. M., & van den Brink, W. (2003). Neurophysiological evidence for abnormal cognitive processing of drug cues in heroin dependence. Psychopharmacology, 170(2), 205–212. https://doi.org/10.1007/s00213-003-1542-7.
Giuliani, N. R., & Berkman, E. T. (2015). Craving is an affective state and its regulation can be understood in terms of the extended process model of emotion regulation. Psychological Inquiry, 26(1), 48–53. https://doi.org/10.1080/1047840X.2015.955072.
Glöckner-Rist, A., Lémenager, T., & Mann, K., & PREDICT Study Research Group (2013). Reward and relief craving tendencies in patients with alcohol use disorders: Results from the PREDICT study. Addictive Behaviors, 38(2), 1532–1540. https://doi.org/10.1016/j.addbeh.2012.06.018.
Grafton, B., & MacLeod, C. (2016). Engaging with the wrong people: The basis of selective attention to negative faces in social anxiety. Clinical. Psychological Science, 4(5), 793–804. https://doi.org/10.1177/2167702615616344.
Heinz, A., Löber, S., Georgi, A., Wrase, J., Hermann, D., Rey, E. R., … Mann, K. (2003). Reward craving and withdrawal relief craving: Assessment of different motivational pathways to alcohol intake. Alcohol and Alcoholism, 38(1), 35–39. https://doi.org/10.1093/alcalc/agg005.
Hester, R., & Luijten, M. (2014). Neural correlates of attentional bias in addiction. CNS Spectrums, 19(3), 231–238. https://doi.org/10.1017/S1092852913000473.
Heuer, A., Mennig, M., Schubö, A., & Barke, A. (2021). Impaired disengagement of attention from computer-related stimuli in internet gaming disorder: Behavioral and electrophysiological evidence. Journal of Behavioral Addictions, 10(1), 77–87. https://doi.org/10.1556/2006.2020.00100.
Hox, J. J. (2010). Multilevel analysis: Techniques and applications (2nd ed.). https://doi.org/10.4324/9780203852279.
Hudson, A., Olatunji, B. O., Gough, K., Yi, S., & Stewart, S. H. (2016). Eye on the prize: High-risk gamblers show sustained selective attention to gambling cues. Journal of Gambling Issues, 2016(34), 100–119. https://doi.org/10.4309/jgi.2016.34.6.
Iijima, Y., Takano, K., & Tanno, Y. (2018). Attentional bias and its association with anxious mood dynamics. Emotion (Washington, D.C.), 18(5), 725–735. https://doi.org/10.1037/emo0000338.
Jeromin, F., Nyenhuis, N., & Barke, A. (2016). Attentional bias in excessive Internet gamers: Experimental investigations using an addiction Stroop and a visual probe. Journal of Behavioral Addictions, 5(1), 32–40. https://doi.org/10.1556/2006.5.2016.012.
Joormann, J., Yoon, K. L., & Zetsche, U. (2007). Cognitive inhibition in depression. Applied & Preventive Psychology, 12(3), 128–139. https://doi.org/10.1016/j.appsy.2007.09.002.
King, D., & Delfabbro, P. (2018). Internet gaming disorder: Theory, assessment, treatment, and prevention. Academic Press.
King, D., Koster, E., & Billieux, J. (2019). Study what makes games addictive. Nature (London), 573(7774), 346. https://doi.org/10.1038/d41586-019-02776-1.
Koster, E. H. W., De Lissnyder, E., Derakshan, N., & De Raedt, R. (2011). Understanding depressive rumination from a cognitive science perspective: The impaired disengagement hypothesis. Clinical Psychology Review, 31(1), 138–145. https://doi.org/10.1016/j.cpr.2010.08.005.
Koval, P., Brose, A., Pe, M. L., Houben, M., Erbas, Y., Champagne, D., & Kuppens, P. (2015). Emotional inertia and external events: The roles of exposure, reactivity, and recovery. Emotion (Washington, D.C.), 15(5), 625–636. https://doi.org/10.1037/emo0000059.
Lavallee, Z. (2020). Addictive craving: There's more to wanting more. Philosophy, Psychiatry, & Psychology, 27(3), 227–238. https://doi.org/10.1353/ppp.2020.0028.
Lemmens, J. S., Valkenburg, P. M., & Gentile, D. A. (2015). The internet gaming disorder scale. Psychological Assessment, 27(2), 567–582. https://doi.org/10.1037/pas0000062.
Lorenz, R. C., Krüger, J., Neumann, B., Schott, B. H., Kaufmann, C., Heinz, A., & Wüstenberg, T. (2013). Cue reactivity and its inhibition in pathological computer game players. Addiction Biology, 18(1), 134–146. https://doi.org/10.1111/j.1369-1600.2012.00491.x.
MacLeod, C., Rutherford, E., Campbell, L., Ebsworthy, G., & Holker, L. (20021965). Selective attention and emotional vulnerability: Assessing the causal basis of their association through the experimental manipulation of attentional bias. Journal of Abnormal Psychology, 111(1), 107–123. https://doi.org/10.1037/0021-843X.111.1.107.
Mathieu, J. E., Aguinis, H., Culpepper, S. A., & Chen, G. (2012). Understanding and estimating the power to detect cross-level interaction effects in multilevel modeling. Journal of Applied Psychology, 97(5), 951–966. https://doi.org/10.1037/a0028380.
Metcalf, O., & Pammer, K. (2011). Attentional bias in excessive massively multiplayer online role-playing gamers using a modified stroop task. Computers in Human Behavior, 27(5), 1942–1947. https://doi.org/10.1016/j.chb.2011.05.001.
Moretta, T., & Buodo, G. (2018). Autonomic stress reactivity and craving in individuals with problematic internet use. PloS One, 13(1), e0190951. https://doi.org/10.1371/journal.pone.0190951.
Mun, C. J., Finan, P. H., Epstein, D. H., Kowalczyk, W. J., Agage, D., Letzen, J. E., … Preston, K. L. (2021). Craving mediates the association between momentary pain and illicit opioid use during treatment for opioid‐use disorder: An ecological momentary assessment study. Addiction (Abingdon, England), 116(7), 1794–1804. https://doi.org/10.1111/add.15344.
Niu, G., Sun, X., Subrahmanyam, K., Kong, F., Tian, Y., & Zhou, Z. (2016). Cue-induced craving for internet among internet addicts. Addictive Behaviors, 62, 1–5. https://doi.org/10.1016/j.addbeh.2016.06.012.
Nuyens, F., Deleuze, J., Maurage, P., Griffiths, M. D., Kuss, D. J., & Billieux, J. (2016). Impulsivity in multiplayer online battle arena gamers: Preliminary results on experimental and self-report measures. Journal of Behavioral Addictions, 5(2), 351–356. https://doi.org/10.1556/2006.5.2016.028.
Peeters, F., Berkhof, J., Rottenberg, J., & Nicolson, N. A. (2010). Ambulatory emotional reactivity to negative daily life events predicts remission from major depressive disorder. Behaviour Research and Therapy, 48(8), 754–760. https://doi.org/10.1016/j.brat.2010.04.008.
Potenza, M. N., Steinberg, M. A., Skudlarski, P., Fulbright, R. K., Lacadie, C. M., Wilber, M. K., … Wexler, B. E. (2003). Gambling urges in pathological gambling: A functional magnetic resonance imaging study. Archives of General Psychiatry, 60(8), 828–836. https://doi.org/10.1001/archpsyc.60.8.828.
Preacher, K. J., Curran, P. J., & Bauer, D. J. (2006). Computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis. Journal of Educational and Behavioral Statistics, 31(4), 437–448. https://doi.org/10.3102/10769986031004437.
Robinson, T. E., & Berridge, K. C. (2008). The incentive sensitization theory of addiction: Some current issues. Philosophical Transactions of the Royal Society B, 363(1507), 3137–3146. https://doi.org/10.1098/rstb.2008.0093.
Rudaizky, D., Basanovic, J., & MacLeod, C. (2014). Biased attentional engagement with, and disengagement from, negative information: Independent cognitive pathways to anxiety vulnerability? Cognition and Emotion, 28(2), 245–259. https://doi.org/10.1080/02699931.2013.815154.
Serre, F., Fatseas, M., Swendsen, J., & Auriacombe, M. (2015). Ecological momentary assessment in the investigation of craving and substance use in daily life: A systematic review. Drug and Alcohol Dependence, 148, 1–20. https://doi.org/10.1016/j.drugalcdep.2014.12.024.
Shin, Y., Kim, J., Kim, M., Kyeong, S., Jung, Y. H., Eom, H., & Kim, E. (2018). Development of an effective virtual environment in eliciting craving in adolescents and young adults with internet gaming disorder. PloS One, 13(4), e0195677. https://doi.org/10.1371/journal.pone.0195677.
Shiyko, M. P., & Ram, N. (2011). Conceptualizing and estimating process speed in studies employing ecological momentary assessment designs: A multilevel variance decomposition approach. Multivariate Behavioral Research, 46(6), 875–899. https://doi.org/10.1080/00273171.2011.625310.
Skinner, M. D., & Aubin, H. (2010). Craving's place in addiction theory: Contributions of the major models. Neuroscience and Biobehavioral Reviews, 34(4), 606–623. https://doi.org/10.1016/j.neubiorev.2009.11.024.
Stein, D. J., Billieux, J., Bowden‐Jones, H., Grant, J. E., Fineberg, N., Higuchi, S., … Poznyak, V. (2018). Balancing validity, utility and public health considerations in disorders due to addictive behaviours. World Psychiatry, 17(3), 363–364. https://doi.org/10.1002/wps.20570.
Stone, A. A., & Shiffman, S. (1994). Ecological momentary assessment (EMA) in behavioral medicine. Annals of Behavioral Medicine, 16(3), 199–202. https://doi.org/10.1093/abm/16.3.199.
Takano, K., & Tanno, Y. (2011). Diurnal variation in rumination. Emotion (Washington, D.C.), 11(5), 1046–1058. https://doi.org/10.1037/a0022757.
Van Holst, R. J., Lemmens, J. S., Valkenburg, P. M., Peter, J., Veltman, D. J., & Goudriaan, A. E. (2012). Attentional bias and disinhibition toward gaming cues are related to problem gaming in male adolescents. Journal of Adolescent Health, 50(6), 541–546. https://doi.org/10.1016/j.jadohealth.2011.07.006.
Wegmann, E., Muller, S. M., Trotzke, P., & Brand, M. (2021). Social-networks-related stimuli Interferes decision making under ambiguity: Interactions with cue-induced craving and problematic social-networks use. Journal of Behavioral Addictions, 10(2), 291–301. https://doi.org/10.1556/2006.2021.00036.
Weiss, R. D., Griffin, M. L., & Hufford, C. (1995). Craving in hospitalized cocaine abusers as a predictor of outcome. The American Journal of Drug and Alcohol Abuse, 21(3), 289–301. https://doi.org/10.3109/00952999509002698.
Wiers, R. W., Gladwin, T. E., Hofmann, W., Salemink, E., & Ridderinkhof, K. R. (2013). Cognitive bias modification and cognitive control training in addiction and related psychopathology: Mechanisms, clinical perspectives, and ways forward. Clinical Psychological Science, 1(2), 192–212. https://doi.org/10.1177/2167702612466547.
World Health Organization (2018). The ICD-11 classification of mental and behavioural disorders: Diagnostic criteria for research. Geneva, Switzerland: World Health Organization, 2018.
You, X. (2022, May 18). Tencent Holdings Ltd.‘s first quarter of 2008 earnings. from https://xw.qq.com/amphtml/20220518A0BPE502 [Retrieved 7 June 2022].
Young, K. S. (1999). The research and controversy surrounding internet addiction. Cyberpsychology & Behavior, 2(5), 381–383. https://doi.org/10.1089/cpb.1999.2.381.
Zhang, J., Yao, Y., Potenza, M. N., Xia, C., Lan, J., Liu, L., … Fang, X. (2016). Effects of craving behavioral intervention on neural substrates of cue-induced craving in internet gaming disorder. NeuroImage Clinical, 12(C), 591–599. https://doi.org/10.1016/j.nicl.2016.09.004.
Glory of Kings is one of the most popular multiplayer online battle arena (MOBA) game in the world whose average daily active users in 2020 was 100 million (You, 2022).
The modified attentional assessment task contained 256 pictures, including abstract images, game-related images and neutral images. Half of the images (128) were “non-representational” images of an abstract nature. The other half images were “representational”. Among them, half of the images (64) were game-related while the other half (64) were neutral images. See more details in Supplementary materials.
Based on previous study (Iijima et al., 2018), we calculated three dynamics parameters from the EMA assessment of game craving. See more details in Supplementary materials.
Here, game-related events occurred in the interval between time t-1 and t.
