Prevention Messages for EGMs: Effects on Behaviours and Cognitions

EGMs and online gambling are the reputedly most damaging gambling type from a public health perspective. Pop-up messages are often used as a responsible gambling (RG) measure to prevent harm for these screen-based types of gambling. Despite some evidence of effectiveness in the literature for these messages, limitations persists, among which low ecological validity is of particular concern.

Indeed, gambling studies set in a controlled environment, usually a laboratory setting, often require participants to engage in a gambling task while wagering some form of virtual credits or money provided upfront by the researchers as a compensation for participation. Their behaviour is then assumed alike what would happen in a real gambling setting, even though gamblers are not risking their own money. Surprisingly, this assumption has not been subject to much scientific scrutiny. Moreover, some studies provide data which seems to contradict this assumption.

This study address this limitation by heightening the realism factor. This study starts with an intake interview by phone to evaluate eligibility, PGSI category, sociodemographic characteristics, gambling behaviours on EGMs (past 12 months), general level of fun while playing EGMs (past 12 months) and perceived self-control while playing EGMs. Participants are then asked to schedule a gambling session which is done in a laboratory on the university's campus. This laboratory replicates a typical bar, and is equipped with real EGMs. Participants are recruited under the false pretense of giving their opinion on the realism of the bar replica, and on the overall feeling of the gambling session in it. They are told that:

* They are free to gamble as much and for as long as they like. * They are allowed to take breaks. * Gambling is to be done with their own money. * The only compensation for participation are the potential winnings made while gambling. * Net winnings across their whole session are paid , but losses are real.

There is only one participant at a time in the laboratory for the gambling session. While in the bar replica, the participant is joined by a research assistant who plays the role of barmaid/barman. The assistant is there for general realism purposes, operating the cash register (Square terminal), and ensuring the general safety of the session. The bar replica has three EGMs, but only one can be played on and used to show the prevention pop-up messages. The other two EGMs are turned on for ambiance purposes, but have their money collector disabled to prevent participants playing on them or switching between EGMs mid-session. Limiting play to only one EGM also eases the recording of gambling behaviours and proper showing of the prevention messages. Participants are told the two "unusable" EGMs are awaiting to be serviced because of their money collector malfunction.

A second research assistant is located in an adjacent room. They are responsible for (a) recording all gambling behaviours using a computer connected to the EGM; and (b) operating the pop-up messages presentation schedule. A spy camera (disguised as a smoke detector) with a live-feed is used to observe gambling related behaviours that can't be recorded by the EGM's computer (e.g. taking a break).

The gambling session ends when the participant cashes out and unambiguously tell the research assistant they want to end their session. The participant is then brought to an office for a series of questionnaires which are, in order:

* Perception of realism of the bar replica and the gambling session in it. * Debriefing about the true goals of the study and validation of consent to participate in the study. At this point, participants will be able to guess in which group they were assigned. * Prevention pop-up messages recall (free and cued recall). * Cognitive and emotional response to prevention pop-up messages. * Evaluation of protocol credibility (i.e. if participant really believed they were gambling their own money and were really risking their own money).

Note that while the participant is told they can gamble for as long as they like, in reality there is a 2 hours time limit to the gambling session duration. This time limit is hidden from the participant. The gambling session starts with the first time money is inserted in the EGM by the participant (the "clock" is set at 0 hours) and ends with either the participant ending it of their own volition or upon reaching the time limit (2 hours). Breaks during gambling session are permitted (e.g. for going to the bathroom, for buying snacks from the barman/barmaid, etc.) and don't stop the "clock".

If the participant has not yet ended their gambling session on their own volition upon reaching the 2 hours time limit, they will be asked to take a small break to answer some questions. The existence of a time limit will be communicated to them during the debriefing. In the eventuality of a participant refusing to stop playing or wanting to rapidly quit the laboratory (e.g. because they are angry about losing money), the debriefing will be done earlier, before the questionnaire about their perception of realism of the bar replica and the gambling session in it. While not optimal, early debriefing is nevertheless considered acceptable in order to force the end of the gambling session or to prevent a participant from hastily quitting the study without receiving all the information to make an informed decision about their participation. There will be no more gambling done after debriefing.

Upon debriefing the participant will be reimbursed all money lost (if any).

The mains objectives are to test to what extent do RG pop-up messages affect:

1. Participants' gambling behavior (e.g. money betted, gambling session length, gambling intensity, etc.). 2. Participants' cognitions (e.g. thoughts elicited by messages, perceived effectiveness of message, etc.). 3. Participants' emotions (e.g. enjoyment of gambling, emotional response to messages, etc.).

Secondary objectives are to test:

1. to what extent are the main effects moderated by participants' characteristics (e.g., gender, age, education, PGSI category, etc.). 2. the feasibility of studies conducted in a laboratory setting with the use of real money (or using deception to make the participants believe they are gambling their own money). This objective is answered with four sub-questions:

* Was it possible to recruit enough participants for the study considering they were confronted with the prospect of loosing their own money by participating in the study and were not guaranteed any compensation other than what they could win on the EGMs? * Did the laboratory "bar" replicated well the overall "vibe" of a typical gambling venue with EGMs? * Did the participants found their gambling session in the bar replica as realistic as a real one? * Did the participants believed they were gambling their own money during the study (i.e. to what extent did the deception worked)?

• Perceived realism - Gambling location type most resembling the session in the laboratory [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - Session in the laboratory VS gambling on EGMs in a bar/restaurant [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - Similarities between session in the laboratory VS gambling on EGMs in a bar/restaurant [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - Differences between session in the laboratory VS gambling on EGMs in a bar/restaurant [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - What could be done to heighten realism of the session in the laboratory [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - Effect of using one own money on realism [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived realism - Effect of using one own money on realism (free) [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived correspondence between habitual gambling behaviour and gambling behaviour during the session in the laboratory [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Perceived correspondence between habitual gambling behaviour and gambling behaviour during the session in the laboratory (free) [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Protocol credibility [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Protocol credibility (free) [Time frame: Assessed at the end gambling session in the laboratory (Post-experiment phase). Session's length is decided by participant (but limited to 2 hours maximum).]
• Recruitment potential - Volunteers [Time frame: Assessed after recruitment for the study is ended (80 participants recruited) or 1 year after recruitment for the study has begun, whichever came first.]