Introduction

Gambling is illegal in Japan. However, gambling opportunities are abundant and more than half Japanese participate, because certain gambling industries are allowed as “amusement” or “public goods”. The former is an electronic gambling machine (EGM) called ‘pachinko and pachislot’, which is the most popular gambling form in Japan. It is not classified as gambling due to the prize exchange. The latter is a public-management gambling, namely state-regulated betting, such as horse racing, boat racing, and lotteries, etc. Thus, the social status of gambling in Japan is ambiguous.

Approximately 2.2% of the Japanese population has been estimated to have gambling problems (South Oaks Gambling Screen, SOGS 5+) (Ministry of Health, Labor & Welfare Japan, 2022). Gambling motivation has been found to play a dominant role in the progression of problem gambling (Blaszczynski et al., 2008; Daoud, 2017; Sundqvist et al., 2016). For successful interventions, it is important to reveal the nature of motivations associated with problem gambling (Carruthers et al., 2006).

In this paper, the term ‘‘problem gambling,” which is less specific than ‘‘gambling disorder’’ or “pathological gambling,” was adopted to emphasize a dimensional approach such as motivation-matched intervention, as opposed to a categorical approach (Shinaprayoon et al., 2017). Therefore, this study discussed “problem gambling” without examining whether individuals satisfied specific criteria.

Previous research has identified motivations that could lead to problem gambling using various gambling motivation scales such as the four-factor Gambling Motives Questionnaire-Financial (GMQ-F) scale (Dechant, 2014). The GMQ-F scale factors are negative reinforcement (to alleviate negative emotions), positive reinforcement (to induce emotions), social connections (to build relationships with others), and economic factors (to gain more money). Some researchers maintain that three motivations; negative reinforcement (escape/coping), positive reinforcement (enhancement/excitement), and economic factors (money motives) are associated with problem gambling (Barrada et al., 2019; Francis et al., 2015; Hearn et al., 2021; Wardle et al., 2011). Specifically, escape motives often lead to risky behavior in women, whereas enhancement motives often lead to risky behavior in men (Sundqvist et al., 2016).

There are other important motivations in addition to the above four. Some studies have revealed that problem gamblers play for challenges, control beliefs (competent), social interaction (sociality or to impress others), or apathy (without a purpose or reason, that is, amotivation) (Griffiths, 1990; Wood & Griffiths, 2007). Therefore, scales with four factors are insufficient for systematically assessing the association between gambling motivation and problem gambling.

The Modified Gambling Motivation Scale (MGMS), which was revised from the Gambling Motivation Scale (GMS) based on learning theory, has more factors and items than subsequent scales (Chantal et al., 1994; Karli, 2008; Komoto et al., 2022; Shinaprayoon et al., 2017; Wu & Tang, 2011). The GMS has a seven-factor structure with a seven-point Likert scale comprising: knowledge, accomplishment (to feel competent), excitement, socialization, social recognition (to feel important and increase self-esteem), monetary gain, and amotivation. In the GMS, social interaction and negative reinforcement (coping and relaxation) were combined into “socialization.” In contrast, the MGMS consists of six factors and has a new factor, namely “intellectual challenge,” instead of knowledge and accomplishment.

Studies using the GMS and MGMS have revealed that amotivation, accomplishment, and social recognition are associated with problem gambling (Chantal et al., 1994; Komoto et al., 2022; Marmurek & Cooper, 2023; Shinaprayoon et al., 2017; Wu & Tang, 2011).

The Japanese version of the MGMS (J-MGMS) with a dichotomous scale not only has the original six-factor structure but also a new four-factor structure. The four-factor structure is useful for motivation-matched interventions to discover adequate substitute behaviors (Komoto et al., 2022). In a previous study that used the J-MGMS with a four-factor structure, only amotivation was significantly associated with gambling problems in a linear regression analysis (Komoto et al., 2022). However, the J-MGMS with six factors was not analyzed in this previous study.

Generally, gambling motivation is considered to have different characteristics, depending on age. First, in adolescents and early adults, self-identity is often unstable because of negative self-image. Therefore, they are more ambivalent than older gamblers (Komoto et al., 2017) and ideal self-images brought about by winning are used to raise feelings of self-value. Namely, younger gamblers tend to gamble for winning money, socialization, accomplishment, and the “buzz’’ (i.e. impression made to others) (Felsher et al., 2004). Compared with younger gamblers, older gamblers tend to have more free time and disposable income for gambling (McNeilly & Burke, 2002). Therefore, their dominant motivation is to use gambling as a pastime. Furthermore, at their age, they may experience declines in social communication, separations from spouses or friends, tendencies to withdraw socially, and chronic disease, which could attract them to gambling as a momentary time killer from the empty hours ahead (McNeilly & Burke, 2000). Gambling is also a convenient pastime for busy middle-aged workers. Consequently, the gambling motivations of older adults may be simpler and more uncomplicated than that of younger people. The difference in gambling motivation between early and late adults is evidently not negligible. Therefore, differences between early and late adults must be analyzed to assess the association between gambling motivation and problem gambling.

Purpose

This study aimed to examine the role of motivation in problem gambling in Japan. Additionally, the differences between early and late adults concerning gambling motivations and their association with problem gambling were examined. These insights can guide the development of tailored interventions for problem gambling.

Hypothesis

First, younger gamblers are more ambivalent than older gamblers and are expected to be more motivated to gamble than older gamblers. Second, young gamblers with problem gambling expect to increase their self-image, namely, social recognition. Third, older gamblers with problem gambling are expected to gamble as a pastime, namely, through amotivation. Finally, Japanese gamblers with self-determinant (autonomous) motivations, such as excitement and socialization, are not expected to progress to gambling problems.

Methods

Participants

Candidates were recruited in July 2019 through the online recruitment website ASMARQ Co. Ltd. and provided their written informed consent to participate. First, a survey was completed by 1,145 individuals who had gambled within the past six months, were residents of the Tokyo metropolitan area, and were aged 20 years or older. Next, 160 participants were randomly selected out of the 1,145 candidates, ensuring approximately equal numbers of male and female respondents for each age group (from 20’ to 60’ years), namely, 32 participants (16 males and 16 females) for each group.

Measures

Modified Gambling Motivation Scale Japanese Version (J-MGMS)

We used a dichotomous J-MGMS format with a six-factor structure (Komoto et al., 2022). Only the dichotomous format of the J-MGMS had good factor validity (RMSEA = 0.063; SRMR = 0.072).

The six factor-structure of the J-MGMS consisted of “intellectual challenge” (e.g., ‘‘I enjoy learning new strategies’’ and “I feel competent when I gamble”), excitement (e.g., ‘‘It is exciting to gamble’’), socialization (e.g., ‘‘It is the best way to relax’’), social recognition (e.g., ‘‘It makes me feel important’’), monetary gain (e.g., ‘‘I play for money’’) and amotivation (e.g., ‘‘I play for money, but I sometimes wonder what I get out of gambling’’).

The internal consistencies (α value) of the six-factor structure of J-MGMS were excellent; total scores (0.960), intellectual challenge (0.893), excitement (0.865), socialization (0.861), social recognition (0.880), monetary gain (0.892) and amotivation (0.888). According to the self-determined theory (SDT), which is the fundamental theory for MGMS, these six factors are ordered from the most autonomous (self-determined or intrinsic) motivations to the least autonomous (non-self-determined or extrinsic) motivations (Chantal et al., 1994). The most autonomous motivation is an intellectual challenge, followed successively by excitement, socialization, social recognition, and monetary gain. The least autonomous motivation is amotivation.

Scoring System for the J-MGMS

Participants completed 28 items (four or eight items per factor) using a dichotomous scale (1 = no, 2 = yes). Higher scores indicate a more intense motivation to gamble for a specific reason or in general. Each subscale score ranges from 4 to 8, although the intellectual challenge score ranges from 8 to 16. This scoring allows researchers to assess the systematic psychological structure of the motivation to gamble.

The South Oaks Gambling Screening Japanese Version (J-SOGS)

The modified Japanese version of the SOGS (J-SOGS) (Kido & Shimazaki, 2007) was used to assess participants’ gambling-related problems in the previous year, because this study targeted not a clinical sample, but a population sample. The scale was developed based on old DSM-III criteria for pathological gambling. Participants rated 20 items on a dichotomous scale (0 = no, 1 = yes) to assess their risk of gambling problems resulting in a score between 0 and 20.

Demographic Questionnaire

The demographic questionnaire measured sex (male or not), age (early adults: age 20–29 or late adults: 30–69 years), present marital status (married or not), present occupation (working or not), the most dominant gambling type (pachinko and pachi slot or others), and gambling frequency (less than once a week or at least once a week). In Japan, Pachinko and Pachi-slots are the most popular forms of gambling. They are categorized as non-strategic gambling and are played individually in parlors using arcade game-like machines.

Procedure

The recruitment and selection of participants were performed according to ASMARQ’s ethical rules. All participants were provided with information about the study, and informed consent was obtained. This study was approved by the Ethical Review Board for Research of Yoshino Hospital. The study was performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.

Statistical Analysis

Data entry and statistical analysis were performed using Microsoft Excel® 2021 MSO. Pearson correlation analysis was used to assess the correlations between the J-MGMS and the six-factor scores, demographic variables, and J-SOGS scores. Each correlation level was judged by correlation coefficient values as Weak (-0.3), Moderate (0.4–0.6), or Strong (0.7-) (Akoglu, 2018).

To evaluate motivational predictors independently associated with problem gambling, a multivariate analysis was performed using a linear regression model. In this model, the dependent variable was the total J-SOGS score. The explanatory variable was the average score of the six factors of the J-MGMS.

The differences between early adults and late adults were investigated by a χ2-test or t-test. Like the analysis of all participants, Pearson’s correlation analysis and multivariate regression analysis using a linear regression model were performed for each group.

Multicollinearity leads to incorrect regression results. Therefore, variance inflation factor (VIF) was used as a diagnostic tool for multicollinearity. If VIF is greater than 10, the specified factor must be removed because of multicollinearity (Daoud, 2017). Statistical significance was set at p < 0.05.

Results

Demographic Description

The sample consisted of 160 participants (50% men and 50% women). Of these early adults (under 30) comprised 20.0%, participants who were married comprised 66.8%, those with occupations comprised 91.9%, participants who dominantly played Pachinko & Pachi-slot comprised 53.1%, and frequent players (at least once a week) comprised 45.0%.

The only significant demographic difference between the early and late adults was marital status (p = 0.002). Moreover, early adults had significantly more gambling problems than late adults (p < 0.001) (Table 1).

Table 1 Demographic description
Full size table

Motivation Description

Excitement is the most common motivation for gambling. The ratio of the number of yes answers to the excitement factor for all participants (average % yes for excitement) was 65.0%. The average percentage of yes responses for each factor was as follows: 48.3% (amotivation), 41.5% (monetary), 35.3% (intellectual challenge), and 34.2% (socialization). Social recognition was the least common motivation for participating (18.2%). Moreover, early adults had significantly more gambling motivation than late adults (p < 0.001–0.013) (Table 2).

Table 2 Japanese version of the modified gambling motivation scale description
Full size table

Correlation Analysis

In early adults, gambling frequency (the number of frequent gamblers) and all six factors of the J-MGMS had a significant positive correlation with the J-SOGS (r = 0.317–0.418). Additionally, amotivation was moderate (Table 3).In early adults, two factors of the J-MGMS, intellectual challenge and social recognition, had moderately significant correlations with J-SOGS scores (r = 0.434–0.559) (Table 4). In late adult pachinko and pachi slot players, gambling frequency (frequent gambler), and almost all factors of the J-MGMS, except social recognition, had a significant positive correlation with the J-SOGS (r = 0.193–0.403). However, only amotivation was moderate (Table 5).

Table 3 Association between demographic variables, motivation factors and South Oaks Gambling Screening score in total participants
Full size table
Table 4 Association between demographic variables, motivation factors, and South Oaks gambling screening score in early adult group
Full size table
Table 5 Association between demographic variables, motivation factors, and South Oaks gambling screening score in late adult group
Full size table

Multivariate Analysis

Linear regression analysis revealed that only amotivation was significantly associated with SOGS scores in all participants. This variable accounted for 17.0% of J-SOGS scores (adjusted R2 = 0.170) (Table 6).In early adults, only social recognition was significantly associated with the SOGS scores. This variable accounted for 29.0% of J-SOGS scores (adjusted R2 = 0.290) (Table 7).In late adulthood, only amotivation was significantly associated with J-SOGS scores. This variable accounted for 15.6% of J-SOGS scores (adjusted R2 = 0.156). Social recognition was nearly significant, although negative (p = 0.0503; standardized partial regression coefficient=-0.220) (Table 8). None of the VIF values were greater than 10; therefore, no variables needed to be removed because of strong multicollinearity.

Table 6 Result of the linear regression analysis with South Oaks gambling screening scores pertaining to the six-factor Japanese version of the modified Gambling motivation scale in total participants
Full size table
Table 7 Result of the linear regression analysis with South Oaks Gambling Screening scores pertaining to the six-factor Japanese version of the modified gambling motivation scale in early adult participants
Full size table
Table 8 Result of the linear regression analysis with South Oaks gambling screening scores pertaining to the six-factor Japanese version of the modified gambling motivation scale in late adult participants
Full size table

Discussion

This study used the J-MGMS to investigate gambling motivations and their association with problem gambling among Japanese gamblers. Additionally, differences in gambling motivation between early and late adults were investigated.

Confirming previous studies, early adults are more often engaged in problem gambling than late adults are (Buth et al., 2017; Hing & Russell, 2020). In addition, unstable self-identity and delayed discounting and impulsivity (the tendency to act rashly) were identified in young, disordered gamblers (Steward et al., 2017). Furthermore, both positive and negative urgency is a mediator of severity levels in young problem gamblers (Blaszczynski & Nower, 2002). Generally, greater gambling motivations lead to gambling problems (Clarke, 2008; Shinaprayoon et al., 2017). Additionally, this study revealed that early adults were more motivated to gamble than late adults.

In this study, excitement (positive enhancement) was the most common motivation for gambling, as has been shown in previous studies (Auer & Griffiths, 2023). However, among the six motivation factors, excitement was the least associated with problem gambling. Additionally, excitement was far from problem gambling in the linear regression analysis of all participants (standardized partial regression coefficient: -0.0444). Therefore, excitement may be a key motivation for preventing problem gambling among Japanese gamblers.

Contrary to this Japanese finding, most American-European studies revealed that intrinsic motivations such as excitement and coping were associated with problem gambling (Shinaprayoon et al., 2017; Wu & Tang, 2011). Previous studies in East Asia, including this study, have revealed that amotivation is an independent predictor of problem gambling (Komoto et al., 2022; Wu & Tang, 2011). This difference may be caused by cultural differences concerning collectivism versus individualism; namely, East Asians, including Japanese gamblers, are easily influenced by surrounding social relationships and forget their individual, original reasons for gambling (i.e., amotivation occurs) (Komoto et al., 2022).

Another noteworthy finding of this study is that the association between amotivation and problem gambling is limited to late adulthood. Older adults often have decreased self-control owing to diminished executive function, which is a possible contributor to pathological gambling (von Hippel et al., 2009). This opaque mood leads to aimless gambling – that is, gambling with amotivation. Therefore, for older adults, gambling with amotivation may often lead to problematic gambling behaviors. However, social recognition is not a problem motivation, but an anti-problem motivation in older adults. This motivation may prevent a fall into gambling through amotivation by providing older adults with a modest aim.

In contrast, amotivation was found not to be an independent predictor of problem gambling in early adults. In other words, amotivation does not play a dominant role in the progression of problem gambling. In general, early adults are more ambivalent or uncertain than late adults (Komoto et al., 2017) but this study found that early adults had more motivation than late adults. Accordingly, equivocal amotivation does not always lead to problem gambling in early adults who regularly gamble. For them, pastimes with amotivation may be a popular motivation for habitual behaviors, not limited to gambling.

In early adulthood, intellectual challenges and social recognition are moderately associated with problem gambling. Social recognition was found in this study to be an independent predictor of problem gambling. Therefore, intellectual challenges may lead to problem gambling through social recognition. In a study of university students, pathological gamblers gamble, not for the needs gained from the act of gambling itself, but for extrinsic reasons (Felsher et al., 2004). They gamble to satisfy their ego needs or increase their self-esteem (social recognition). In this young sample, intrinsic motivations, such as stimulation (excitement), had a weaker association with pathological gambling than extrinsic motivations. Another study revealed that problem gamblers were motivated to gamble for self-esteem, such as feeling important in the eyes of others (social recognition) (Walters, 1994). Individuals with a weak sense of self-worth may be focused on the momentary boost in self-esteem brought about by a win (Ryan & Deci, 2000). Therefore, motivation for social recognition in early adulthood often becomes more compulsive than in late adulthood because of a negative self-image. For these individuals, chasing losses may be equivalent to increasing self-esteem.

From the above, in this study three important findings were revealed:

  1. 1)

    Self-determined (autonomous) motivations, such as excitement (positive reinforcement) and socialization (negative reinforcement), do not contribute to progress in problem gambling.

  2. 2)

    Two non-self-determining (non-autonomous) motivations are important predictors of problem gambling: social recognition in early adulthood and amotivation in late adulthood.

  3. 3)

    Social recognition is a dichotomic and paradoxical motivation in the progress of gambling problems.

These results should be understood in the context of the limitations of this study. First, this study employed self-reported data with some biases. Second, the data excluded individuals age 70 or older. Retired people in their 70s may have different motivations for gambling, such as battling boredom and loneliness. (Guillou et al., 2019). Third, the SOGS does not rate pathological gambling. Therefore, the characteristic motivations for pathological gamblers, such as self-medication, were not identified in this study. Forth, the data were analyzed using cross-sectional methods, which made it difficult to conclude a causal relationship between motivation and other variables. Future longitudinal research should investigate the causality between gambling motivation and problem gambling. Fifth, other potential psychosocial factors affect gambling motivations such as emotion regulation (Marchica et al., 2020). Future studies should investigate these additional factors.

Despite these limitations, this study revealed that non-self-determined motivations, such as social recognition in early adults and amotivation in late adults, more strongly influenced the progression of gambling problems than self-determined motivations, such as excitement and socialization, which are predictive motivations for gambling problems in Western cultures. By identifying the motivations associated with the progression of problem gambling at different ages, this study has significant implications for the development of tailored interventions to prevent problem gambling in Japan. Clinicians should, for example, take cognizance of the motivational differences between Western and Eastern cultures and between gamblers in early and late adulthood.