One of the proudest moments of my academic career was when my 1994 study on the role of cognitive bias in slot machine gambling published in the British Journal of Psychology was introduced as a compulsory study that all ‘A’ [Advanced] Level students on the OCR syllabus have to learn about here in the UK. Today’s blog looks at that 1994 study in context.
I began a PhD on the psychology of slot machines back in 1987 and spent the first three or four months reading everything I could about how psychological research methods had been used to study this relatively new area of research. As a PhD student, the paper that really inspired me was a pioneering study by Anderson and Brown (also published in the British Journal of Psychology in 1984). Up until the mid-1980s almost all of the experimental work on the psychology of gambling had been done in laboratory settings and the question of ecological validity was something that I had great concerns about. I didn’t want to study gamblers in a psychology laboratory, I wanted to examine them in the gambling environments themselves. Anderson and Brown studied the role of arousal in gambling and used heart rate measures as an indicator of arousal. They found that regular gamblers’ heart rates increased significantly by around 23 beats per minute (compared to baseline resting levels) when they were gambling in a casino but when doing the same activity in a laboratory setting there was no significant increase in heart rate. To me, this perhaps explained why previous studies on arousal during laboratory gambling had failed to find significant heart rate increases above baseline levels.
Anderson and Brown claimed that Skinnerian reinforcement theory couldn’t account for the phenomenology of addictive gambling (especially relapse after abstinence). As a result of their ecologically valid experimental study, Anderson and Brown postulated a theoretical model centred upon individual differences in cortical and autonomic arousal in combination with irregular reinforcement schedules. They argued for a neo-Pavlovian model in which arousal played a central role in the addiction process. According to Anderson and Brown this model accounts for reinstatement after abstinence and allows for the maintenance of the behaviour by internal mood/state/arousal cues in addition to external situation cues. I found this theoretical perspective too restrictive and believed that gambling addiction was a more complex process and was the consequence of a combination of a person’s biological/genetic predisposition, their psychological make-up (personality, attitudes, beliefs, expectations, etc.), and the environment they were brought up in. This is what most people would now recognize as a biopsychosocial perspective that runs through much of my subsequent writing and research. Added to this, I passionately believed there were other important factors at play including the situational factors of where the activity took place such as the design of the gambling environment, and the structural features of the activity itself such as the speed of play and ambient factors like lights, colour, noise and music.
My 1994 study found that regular gamblers produced significantly more irrational verbalisations that non-regular gamblers. (The ethics committee wouldn’t let me use non-gamblers as they didn’t want participants to be introduced to gambling via a university research study!). One of the most observations in my study was that regular gamblers personified the machine and often treated the machine as if it was a person. They attributed thought processes to it and would talk to it as if it could actually hear them. Another of the more interesting observations concerned ‘the psychology of the near miss’ (or more accurately. ‘the near win’). I noticed that when I used the ‘thinking aloud method’ as a way of gaining direct cognitive access to what gamblers were thinking as they played a slot machine, regular gamblers often explained away their losses and changed clear losing situations into near winning ones. On a cognitive level gamblers weren’t constantly losing, they were constantly nearly winning, and this, I argued, was both psychologically and physiologically rewarding for them. (I also did a study where I measured gamblers’ heart rates in an amusement arcade where, like Anderson and Brown I found regular gamblers had significantly increased heart rates when compared to baseline resting levels).
Anyone reading my 1994 paper will instantly spot what appears to be a major limitation of the study – the fact that there was no inter-rater reliability in the coding of the verbalisations that I transcribed. Could this be (as some have argued) the Achilles Heel of the study? I have argued that in the context of this study having a second rater might have added a confounding variable in itself. Another rater wouldn’t have had the time with the data that I had and wouldn’t have been there at the time of the experiment. In short, ‘not being there’ would have been a great disadvantage to a second coder as they would not have understood the context in which various verbalizations were made. I transcribed each tape straight after each trial so that I could remember the context of everything that was said by each player. I would also add that this was one study that was done in conjunction with lots of others simultaneously (the details of which are provided below).
The work of Paul Delfabbro in Australia built on my idea of analysing gamblers within session and postulated that gambling is maintained by winning and losing sequences within the operant conditioning paradigm (i.e., that the only rewards and reinforcers in gambling are purely monetary). I then argued in response to that paper (in a 1999 issue of the British Journal of Psychology) that Delfabbro’s contribution was too narrow in its focus in that they had taken no account of the 'near miss' in relation to operant conditioning theory and that there may be other reinforcers that play a role in the maintenance process (such as physiological rewards, psychological reward and social rewards). I also argued that gambling was biopsychosocial behaviour and should therefore be explained by a biopsychosocial account.
My 1994 study showed that gamblers could be studied in real-life contexts and that useful data could be collected. It also showed the complexity of gambling and that gamblers could turn apparently objective outcomes (i.e., losing) into ones that were highly subjective (i.e., near winning ones). I also showed that this had implications for treatment and that maybe these cognitive biases could be used by psychologists as a way of ‘re-educating’ gamblers through some kind of ‘cognitive correction’ technique. I should also point out that this one experimental study was one small part of a much bigger jigsaw. What I mean by this is that my 1994 shouldn’t be seen in isolation but read along with my simultaneous observational studies of arcade gamblers, my other experimental studies, my semi-structured interview studies, surveys, and my case studies. All of these studies as a whole were featured in my first book (Adolescent Gambling, published in1995).
My work into the role of cognitive bias in gambling and gambling addiction also led to me studying behavioural addictions more generally. Since I finished my PhD I have branched out and carried out research (among others) into videogame addiction, Internet addiction, sex addiction, work addiction, and exercise addiction. Many psychologists don’t view excessive behaviour as an addiction, but for me gambling is the 'breakthrough' addiction. I have argued that when gambling is taken to excess it can be comparable to other more recognised addictions like alcoholism. If you accept that gambling can be a genuine addiction, there is no theoretical reason why other behaviours when taken to excess cannot be considered potentially addictive if ‘gambling addiction’ exists.
A key difference between excessive use and addiction is the detrimental effects (or lack of) that arise as a result of that behaviour. When people are addicted to a behaviour that becomes the single most important thing in their life, they compromise everything else in their life to do it. A person’s job/work, personal relationships and hobbies are severely compromised. The basic difference between an excessive healthy enthusiasm and an addiction is that healthy enthusiasms add to life – addictions take away from it. This is very much a (non-psychological) lay view, but there is a lot of truth in it.
I am the first to admit that my 1994 study when taken in isolation is hardly up there with the ‘classic’ studies of Freud, Watson, Skinner or Milgram. However, as part of two decades of other research into gambling and other potentially excessive behaviours I would like to think I have had an influence in my field. Only time will tell.
References and further reading
Allegre, B., Souville, M., Therme, P. & Griffiths, M.D. (2006). Definitions and measures of exercise dependence, Addiction Research and Theory, 14, 631-646.
Anderson, G. & Brown, R.I.F. (1984). Real and laboratory gambling, sensation seeking and arousal. British Journal of Psychology, 75, 401-410.
Delfabbro, P. & Winefield, A.H. (1999). Poker machine gambling: An analysis of within-session characteristics. British Journal of Psychology, 90, 425-439.
Griffiths, M.D. (1990). The acquisition, development and maintenance of fruit machine gambling. Journal of Gambling Studies, 6, 193-204.
Griffiths, M.D. (1991). The observational study of adolescent gambling in UK amusement arcades. Journal of Community and Applied Social Psychology, 1, 309-320.
Griffiths, M.D. (1991). Fruit machine addiction: Two brief case studies. British Journal of Addiction, 85, 465.
Griffiths, M.D. (1993). Fruit machine gambling: The importance of structural characteristics. Journal of Gambling Studies, 9, 101-120.
Griffiths, M.D. (1993b). Tolerance in gambling: An objective measure using the psychophysiological analysis of male fruit machine gamblers. Addictive Behaviors, 18, 365-372.
Griffiths, M.D. (1993). Pathological gambling: Possible treatment using an audio playback technique. Journal of Gambling Studies, 9, 295-297.
Griffiths, M.D. (1993). Factors in problem adolescent fruit machine gambling: Results of a small postal survey. Journal of Gambling Studies, 9, 31-45.
Griffiths, M.D. (1993). Fruit machine addiction in adolescence: A case study. Journal of Gambling Studies, 9, 387-399.
Griffiths, M.D. (1994). The role of cognitive bias and skill in fruit machine gambling. British Journal of Psychology, 85, 351-369.
Griffiths, M.D. (1995a). The role of subjective mood states in the maintenence of gambling behaviour, Journal of Gambling Studies, 11, 123-135.
Griffiths, M.D. (1995b). Adolescent gambling. London: Routledge.
Griffiths, M.D. (1999). The psychology of the near miss (revisited): A comment on Delfabbro and Winefield. British Journal of Psychology, 90, 441-445.
Griffiths, M.D. (2005). A “components” model of addiction within a biopsychosocial framework. Journal of Substance Use, 10, 191-197.
Griffiths, M.D. (2008). Diagnosis and management of video game addiction. New Directions in Addiction Treatment and Prevention, 12, 27-41.
Griffiths, M.D. & Delfabbro, P. (2001). The biopsychosocial approach to gambling: Contextual factors in research and clinical interventions. Journal of Gambling Issues, 5, 1-33. Located at: http://www.camh.net/egambling/issue5/feature/index.html.
Griffiths, M.D. & Parke, J. (2003). The environmental psychology of gambling. In G. Reith (Ed.), Gambling: Who wins? Who Loses? pp. 277-292. New York: Prometheus Books.
Parke, J. & Griffiths, M.D. (2006). The psychology of the fruit machine: The role of structural characteristics (revisited). International Journal of Mental Health and Addiction, 4, 151-179.
Parke, J. & Griffiths, M.D. (2007). The role of structural characteristics in gambling. In G. Smith, D. Hodgins & R. Williams (Eds.), Research and Measurement Issues in Gambling Studies (pp.211-243). New York: Elsevier.