Role of cognitive bias and skill in fruit machine gambling

1. Role of cognitive bias and skill in fruit machine gambling Professor Mark Griffiths International Gaming Research Unit mark.griffiths@ntu.ac.uk

2. THE STUDY IN CONTEXT (1988-1990) • Exploratory observational/semi-structured interview study of eight adolescent fruit machine gamblers (Griffiths, 1990a - JGS) • Semi-structured interview study of 50 adolescent fruit machine gamblers (Griffiths, 1990b - JGS; 1990c - JGS) • Case studies of adolescent fruit machine gamblers (Griffiths, 1991- BJA; 1993 - JGS) • Longitudinal observational study of adolescent gamblers in amusement arcades (Griffiths, 1991- JCASP) • Postal study of ‘Parent of Young Gamblers’ members and their adolescent gambling children (Griffiths, 1993a - JGS) • Experimental study of cognition in fruit machine gamblers (Griffiths, 1994 - BJP) • Experimental study of arousal in fruit machine gamblers (Griffiths, 1993b - Add.Behs)

3. GLOBAL MODEL OF GAMBLING BEHAVIOUR(Griffiths, 2006; Parke & Griffiths, 2007)

4. Individual Characteristics(Griffiths & Delfabbro, 2001)

5. Situational Characteristics(Griffiths & Parke, 2003)Example: UK Amusement Arcade

6. Structural characteristics(Griffiths, 1993;1995; Parke & Griffiths, 2001; 2007)Example: Slot machines

7. COGNITIVE BIAS IN GAMBLING(Wagenaar, 1988) • “Gamblers are motivated by a way of reasoning, not by defects of personality, education or social environment” • “Gamblers gamble not because they have a bigger repertoire of heuristics but because they select heuristics at the wrong occasions”

8. STUDY’S MAIN HYPOTHESES • Hypothesis 1: • There would be significant differences in the thoughtprocesses (irrational verbalisations) between regular and non-regular gamblers • Hypothesis 2: • There would be no significant differences in the (skill-based) behaviours of regular and non-regular gamblers

9. METHOD • Quasi-experiment • Two groups of participants • IV = regular or non-regular gambler

10. PARTICIPANTS • 30 regular gamblers • 30 non-regular gamblers • Regular (29 males & 1 female; play at least once week) • Non-regular(15 males & 15 females; play once month or less) • Volunteer Sample • Mainly recruited via a poster

11. THE ‘SUBJECTIVE’ DVs (1) COGNITIVE ACTIVITY • Measured by ‘thinking aloud’ (2) PERCEPTION OF SKILL • Measured by post-experiment semi- structured interview

12. THE ‘OBJECTIVE’ (BEHAVIOURAL) DVs • Total number of plays in session • Total minutes of play in session • Play rate - Total plays per minute in session • End stake – total winnings • Total number of wins in session • Win rate (time) – time between wins • Win rate (plays) – number of plays between wins

13. PROCEDURE • In arcade each participant given £3 to gamble on machine that gave 30 free gambles • Objective:To stay on machine for 60 gambles • To break even and win back the £3 • If they achieved 60 gambles they could choose to keep the money or carry on gambling

14. CONTROLS • Participants played same machine‘FRUITSKILL’ • Randomly assigned to thinking aloud/non-thinking aloud • All recordings transcribed within 24 hours - Say everything that goes through your mind - Do not censor your thoughts - Keep talking continuously - Don’t have to speak in complete sentences

15. Behavioural FINDINGS: DV Non Regular NTA Non Regular TA Regular TA Regular NTA Total Plays 47.8 56.3 55.7 65.6 Total Time 8.4 8.5 11.5 9.9 Play Rate ** 6.5 7.5 5.3 8.4 End Stake 4.0 0 7.3 13.9 Win 6.1 8.0 8.3 6.0 Win rate - time 2.0 1.0 1.7 1.8 Win rate plays ** 12.5 7.5 8.0 14.6

16. MAIN RESULTS (Behavioural variables) • ANOVAs showed no significant differences on all variables except: • Regular gamblers stayed on the machine slightly longer (F(1,56) = 4.27, p=0.044) • Regular gamblers had a significantly higher play rate (F(1,56) = 7.96, p=0.007) • Non-regular gamblers who thought aloud had slightly more wins than any other group (F(1,56) = 5.09, p=0.028) • Regular gamblers who thought aloud had a significantly lower win rate than any other group (F(1,56) = 7.85, p=0.007)

17. MAIN RESULTS (Cognitive variables) • Content analysis of thinking aloud transcripts • 31 different categories (4 irrational, 27 rational) • Regular gamblers produced significantly more irrational verbalisations than non-regular gamblers (14% vs. 2.5%; p < 0.001) • Further analysis of transcripts revealed gamblers using a variety of heuristics (e.g., hindsight bias)

18. Content Analysis Examples of FINDINGS: DV Non-Regular gamblers (%) Regular gamblers  (%) Machine personification 1.14 7.54 ** Explaining away losses 0.41 3.12 * Referencing the ‘number’ system 0.90 2.64 ** Swearing at machine 0.08 0.60 * Referencing skill 1.47 5.34 * Verbalising confusion (statements/questions) 4.81 ** 13.24 ** 1.72 1.56

19. IRRATIONAL VERBALISATION • “This ‘fruity’ is not in a good mood” • “It wants its money back” • “Putting only a quid in ‘bluffs’ the machine” • “The machine thinks I am a F***wit” • “This machine won’t pay out happily”

20. CONCLUSIONS • Regular gamblers are slightly more skilful (e.g. knowing the reels and when to nudge) • Regular gamblers believe they are more skilful than they are • Gamblers know they will ‘lose’ but they play with money not for it (staying on the machine is the objective) • Regular gamblers make more irrational verbalisations demonstrating cognitive bias

21. APPLICATIONS? • May help to rehabilitate ‘gambling addicts’ through cognitive behavioural therapy • Can be used to help ‘problem gamblers’ change the way they think (recognise and change their cognitive bias) and behave • e.g. by playing back their irrational thinking

22. EVALUATION • Both quantitative and qualitative DVs • Validity the ‘thinking aloud method’?? • Reliability of content analysis?? • Biased sample (29 male regular gamblers) - does this matter?? • Ecological validity (level of realism) • Generalisability to other forms of ‘gambling’ (e.g horse racing, dice, roulette)

23. Thankyou! Any questions?