Expressing Gambling-Related Cognitive Biases in Motor Behaviour: Rolling Dice to Win Prizes

Supplementary Material; v1; 23th October 2012

Expressing gambling-related cognitive biases in motor behaviour: Rolling dice to win prizes

Matthew S.M. Lim1, Henrietta Bowden-Jones2 & Robert D. Rogers1

1Department of Psychiatry, University of Oxford, Warneford Hospital, Headington, Oxford, OX3 7JX

2Department of Medicine, Imperial College London, SW7 2AZ

Address for correspondence:

Robert Rogers

Department of Psychiatry,

Warneford Hospital

Oxford, OX3 7JX

Tel: 44 (1865) 613103

Fax: 44 (1865) 793101

E-mail:

Data Analysis

Fig. S1 Mean proportion of decisions to play as a function of probability of winning.

Table S1. Participant-level regression (entry method) of decisions to roll a simulated die to win monetary prizes, or forgo a buy-in, against gamble features in 60 gamblers.

a Value to buy-in (£0, £20, £40, £60 or £80); value of bets (£20, £40, £60, £80 or £100); probability of winning (0.16, 0.33, 0.50, 0.66 or 0.83); value of winning numbers (small - 1,2 or 1,2,3 or 1,2,3,4 or 1,2,3,4,5; large - 5,6 or 4,5,6 or 3,4,5,6 or 2,3,4,5,6); game phase (5 consecutive phases of 25 plays each).
Table S2. Group-level regression of decisions to roll a computer-simulated die to win monetary prizes, or forgo a buy-in, against demographic characteristics, gambling cognitions and participation, and psychological state measures in 60 gamblers

a R2 = .20 (p< .01); +p < .10; GRCS - Gambling-related cognitions scale (Raylu and Oei 2004).

Fig. S2 Mean roll-time (logarithm milliseconds) as a function of subscales on the Gambling-related Cognitions Scale (Raylu and Oei 2004): (a) illusions of control; (b) near-miss chasing; and (c) interpretive bias cognitions, in 60 gamblers.

Supplementary data analysis of gamblers scoring less than 36 on the GRCS

To be sure that associations between die-roll times and gambling cognitions were not artefacts of low scores on the illusions of control and interpretive bias sub-scales in some of our gamblers, we repeated the analysis, but removed all participants who scored lower than the mean GRCS total score of 36 reported in a comparable sample (Raylu and Oei 2004). This left a restricted sample of 47 regular gamblers.

Die roll-times. Table S3 shows the results of the group-level regression of gamblers' roll-times against the demographic characteristics, gambling participation and gambling-related cognitions of our sample. These results were mostly consistent with those obtained with the full sample. Controlling for age and gambling losses (all |t|s< 1.62), gamblers who reported stronger illusions of control allowed the die to roll for longer than gamblers with weaker illusions of control (t(46)= 3.13, p< .01). Gamblers who tended to chase following near-misses in commercial gambling settings showed the same behaviour (t(46)= 2.06, p< .05). Finally, higher interpretative bias scores were associated with significantly shorter die rolls (t(46)= -3.19, p< .01).

Table S3. Group-level regression of die-roll times against demographic characteristics, gambling frequency and cognitive biases in 47 gamblers with GRCS scores greater than 36 (Raylu and Oei 2004) (see main text for details).

a R2 = .31 (p< .01); +p < .10; *p < .05; **p < .01; Chasing traits - Chasing Questionnaire (CHQ; O'Connor and Dickerson 2003); GRCS - Gambling-related cognitions scale (Raylu and Oei 2004).

Fig. S3 Associations between die-roll times (logarithm milliseconds) and value of prizes as a function of: (a) score on the CHQ measure of chasing near-misses (O'Connor and Dickerson 2003) and (b) score on the gambling expectancy subscale of the GRCS (Raylu and Oei 2004)

References

O'Connor, J., & Dickerson, M. (2003). Definition and Measurement of Chasing in Off-Course Betting and Gaming Machine Play. Journal of Gambling Studies, 19(4), 359-386.

Raylu, N., & Oei, T. P. S. (2004). The Gambling Related Cognitions Scale (GRCS): development, confirmatory factor validation and psychometric properties. Addiction, 99(6), 757-769.

1