Home            |        AS             |        A2          |         Links       |

2. Gambling

DSM IV-TR (2000) considers gambling to be an impulsive disorder characterized

• A preoccupation with the behaviour
• Repeated and unsuccessful attempts to stop
• Continuing the habit despite losses and other costs

Generally there seems to be a fixed development:

1. Social gambling

2. Frequent gambling

3. Problem gambling

4. Pathological addiction

Each stage shows a greater commitment to the cause, both financially and in terms of time and demands.  More than half of all pathological gamblers steal to pay for their habit and many have alcohol-related issues.  Many also report suicidal tendencies and attempts.

Initiation

Genetic

Studies into gambling families tend to suggest that there are trends within families towards gambling problems.  Black et al (2006) found that the first degree relatives of pathological gamblers were themselves more likely to have gambling problems.  Ibanez et al (2003) put the concordance rate at about 20%.  As always such research fails to disentangle genetic and environmental factors and the likelihood that being exposed to gambling behaviour is also a likely trigger.  However, Wendy Slutske (2000) believes 64% of the variation in similarity can be attributed to genetic factors.

Slutske (2010) has looked at the concordance rates of MZ and DZ twins in relation to gambling addiction.  She studied nearly 5,000 individuals and found that the concordance rate for MZ twins was twice as high as the concordance rate for DZ twins.  However, as she herself points out, it isn’t possible to isolate the genetic effects from environmental and social.  Instead she describes a ‘perfect storm’ of inherited vulnerability combined with the presence of gambling role models in the form of parents.  Slutske’s is also important in highlighting similar risks for both men and women.  Previous research has tended to focus on men. 

Specific gene?

However, stronger evidence comes from specific gene studies, in particular the much greater prevalence of D2A1 variant in gamblers than in the general population (Bennett 2006).  However, given that there are gamblers without the gene and non-gamblers with it seems that this variant is neither necessary nor sufficient to be responsible for gambling behaviour.  It is highly unlikely that any one gene can be responsible for such a complex behaviour as an addiction. 

It’s also worth mentioning that genetics might not be a direct predisposing factor but may be having their influence indirectly through personality.  Pathological gamblers tend to be impulsive sensation seekers with a low boredom threshold.   

Behaviourist explanations

Mowrer’s two stage approach to explaining the acquisition and maintenance of a behaviour usually assumes acquisition via classical associations and SLT with operant explaining maintenance. 

Usually operant conditioning is unable to explain acquisition.  By definition a behaviour needs to be performed before it can be reinforced.  However, addiction is a little different.  Addiction does not occur from the outset, but rather follows a period of time in which the behaviour is performed.  Acquisition of the behaviour can therefore be explained traditionally using classical and SLT but there is a role for all three in explaining initiation of the addiction phase. 

Seeing others being rewarded for their gambling behaviour acts as a vicarious reinforcer and may trigger the desire in others.  Similarly, as the national lottery reminds us, you have to be in it to win it.  There is an association between buying your ticket or scratchcard and the possibility of that big win.  Having triggered the behaviour an early win will act as a powerful (operant) reinforcer that may then lead to addiction.    

Cognitive explanations

Perceptions are crucial.  Gamblers over-estimate their level of control in gambling situations and tend also to over-estimate their winnings and under-estimate their losses.  There is also the possibility of addictive personality (considered later) which includes risk-taking, sensation seeking and impulsivity.  Risk taking however only seems to be common in high-risk gambling situations such as casinos, suggesting that we need to distinguish motives depending on the type of gambling. 

Maintenance and Relapse

Chemical

Dopamine is released following a win and results in the ‘buzz’ we associate with pleasure and excitement.  So as with all other addictions, gambling appears to trigger activity in the brain’s mesolimbic system (MLS).  Interestingly there also seems to be a corresponding increase in noradrenaline activity, usually associated with increased arousal and stress (the sympathetic nervous system).

Crucially there appears to be a difference between brain activity and social gamblers when compared to pathological.  Social gamblers get the buzz whilst gambling, whereas mere anticipation of gambling can trigger mesolimbic activity in the pathological. 

Withdrawal from gambling produces very similar physical effects to withdrawal from drugs, again suggesting similar brain mechanisms and chemistry are involved.  Typically withdrawal involves insomnia, loss of appetite, muscle weakness, increased breathing, racing pulse and even the chills associated with ‘cold turkey.’

Meyer et al (2004) measured heart rate and salivary cortisol levels in ten male Blackjack players.  The increases in both were similar to mild but long lasting stress suggesting an increased level of arousal.  Perhaps it is this increase in sympathetic activity that acts as the reinforcer. 

Chase and Clark (2010) studied 24 regular gamblers.  They completed a computerized gambling game based on slot machines whilst their mid-brains were scanned using fMRI.  Increased activity in the mesolimbic system resulted from wins but crucially also from near misses.  In addition, the higher the level of gambling severity the greater the response recorded.  Chase and Clarke conclude that pathological gamblers get the greatest hit from a near miss.  Makers of fruit machines use this idea of ‘clustering’ to create high numbers of near-miss scenarios which increase the perception of win likelihood and decrease the perception of randomness. 

Research like this is scientific, measurable and replicable.  However, it is also artificial in that there is no pressure from social situations and the gamblers are not in real-life win/loss situations. 

Recent research into lottery and scratchcard use throws up some interesting differences.  Scratchcards are high in ‘event frequency’ meaning there is a very short period of time between paying and winning so the behaviour can be repeated quickly.  This gives a quick hit and creates physiological arousal.  Lottery on the other hand appeals to a ‘sense of fantasy’ and the thought of winning millions.  Young scratchcard users (12 to 15 years) are more likely to smoke, drink alcohol and take illegal drugs and are more likely to play fruit machines.  Does this indicate an addictive personality or is it due to poverty or boredom? 

The medical model clearly employs some familiar behaviourist terms such as reinforcement and association. 

Schedules of reinforcement

The sections in italics on continuous and interval schedules is for background information only.  Variable ratio schedules tend to be employed in gambling. 

Continuous schedule

Operant conditioning is based on the idea of a behaviour reinforced is likely to be repeated.  However, the behaviour does not need to be reinforced every time it is performed for the learning to take place.  In fact if a rat were to be fed every time it pressed a lever (continuous reinforcement) it would very soon become full and lose interest.  Drinks machines operate on this schedule.  Put money into a drinks machine you expect to get something out every time.  If you don’t you stop the behaviour straight away… it becomes extinguished.  If you want to maintain a behaviour, in the absence of reward, for example as is needed in casinos, bingo halls, amusement arcades, you need to adopt a more irregular schedule of reinforcement. 

Other schedules can be based on time (interval) or number of behaviours performed (ratio). 

Interval schedules

A behaviour may be rewarded every 5 minutes providing the behaviour has occurred in that time.  This is called fixed interval.  Payment at the end of the month would be a human example.  Alternatively, reward may be on a variable interval schedule.  Reward may be after 5 minutes, or sometimes 15, perhaps sometimes 2 minutes etc.  This is less predictable and leads to slower extinction.  If after 5 minutes there is no reward the animal keeps pressing.  Perhaps reward may be after 25 minutes this time.

Ratio schedules

Time is no longer an issue.  In rat terms reward occurs after so many presses.  This may be every 10 presses (fixed interval) or it may be variable ratio. 

Operant conditioning of this sort is better at explaining maintenance as opposed to initial acquisition. 

However, it is effective at explaining both chemical and behavioural addiction via the process of reinforcement.  Chemical addiction can also be explained in terms of avoidance of the punishment of withdrawal.

The behaviourist model always emphasises the role of environmental factors in shaping our behaviour.  It has long been known that environment is crucial in relapse following treatment so the behaviourist approach has been one of the more successful approaches in the treatment of addictive behaviour.

As early as 1948, Wikler found that heroin addicts were far more likely to relapse when in a similar situation or with the same people as when indulging in the habit originally.

Shiffman (1996) asked former smokers to record when and where they relapsed and found it was always in situations were readily available and when with other smokers.

Evaluation of behaviourist approach

Operant conditioning is very good at explaining the maintenance of addictive behaviour and at its persistence via the deliberate use of variable ratio schedules of reinforcement.

However, classical conditioning is not so useful in explaining behavioural addictions.  If addiction is due to an association between gambling behaviour (for example) and winning, then why do addicts continue to gamble even after a very long losing streak?  Why doesn’t the behaviour become extinguished?

Cognitive model and Heuristics

Heuristics

A heuristic (like a schema or stereotype) is a mental shortcut that allows us to make judgments and decisions, with minimal thinking effort involved. 

Heuristics of addictive behaviour are perhaps easiest understood if we apply them to gambling.

Gambling is surely a perfect example of irrational thinking.  Its very premise seems to be built on a falsehood; the erroneous belief that an individual can beat the odds.  Griffiths suggests a number of such heuristics.  We shall consider a mere few of them:

Gambler’s fallacy seems to be an obvious starting point:  the idea that random events equal themselves out over time.  “I haven’t had a win for three months so it’s my turn soon.”  With the lottery, the idea that a number hasn’t been drawn for twelve weeks so it must come up this time.  

Availability bias: is in some respects the above heuristic in reverse.  The notion that because something has happened in the past it will occur again in the future.  Big winners on the lottery get oodles of coverage leading us to think it’s a common occurrence and hence likely to happen to us too.  In the early days of the lottery it soon became apparent that the number 44 was being drawn more than the others.  Result: everyone was picking the number 44!

Sunk cost bias: another possible explanation of the gambler’s fallacy.  Playing a game costs money, we expect something in return sooner or later.  Having made that initial investment and not had a return we feel obliged to continue so we don’t lose out.  Long term we could potentially lose a lot more!

Representativeness bias: the tendency to confuse a representative sample with a random sample.  Clearly games like the lottery require a random drawing of numbers.  However, when we come to choose a random sample we tend to pick numbers that best represent the spectrum 1 to 49.  As a result we are likely to choose a single number, one from the twenties, one from the thirties etc.  The exception to this is the estimated thousands that pick 1,2,3,4,5,6!

Illusion of control: gamblers tend to over-estimate the amount of control they have.  With the lottery this is minimal, however, I suppose being able to chose your numbers provides some semblance of control.  The illusion of control is more likely with fruit machines which give the impression of control with features such as ‘nudge’ and ‘hold’ even though in practice very little skill is involved. 

As well as heuristics there is also a tendency by many gamblers to make it personal.  Gamblers will switch from one bandit to another, claiming the first one doesn’t like them.  Some think they can con the machine by only putting in £1 at the outset. 

Mark Griffiths (1994)

A study that supports many of the ideas above.

Griffiths carried out a natural experiment on a volunteer sample of 30 regular gamblers comparing them to a control group of 30 occasional gamblers.  Particiapnts answered a poster advertisement placed around a University campus.

Each participant was given £3 for which they got 30 gambling opportunities on a fruit machine.  Their aim was to win enough money to make a total of 60 gambles which means that had to break even and win back £3 from the money they had put in.  If they managed to achieve 60 gambles with the initial £3 stake they were given the choice of either keeping any of the winnings or carry on gambling.  

Throughout the procedure the participants were asked to ‘talk aloud’ so that their cognitive activity could be assessed.  Later they were each interviewed to gauge their perceived skill level. 

Findings:

Regular gamblers

·         See themselves as more skilful than non-gamblers (even though no such difference existed)

·         Made more irrational statements

·         Made more statements suggesting that the machine had a personality or moods

·         More likely to explain losses as near misses or even as near wins. 

Although the gamblers were more likely to break even after 60 gambles they were also five times more likely to stay on the machine until they had lost all of their money. 

Conclusions

Regular fruit machine gamblers perceive themselves as more skilful than they really are.  The study supports the argument that regular fruit machine users do use cognitive biases when gambling. However, he is cautious about whether such findings do explain that the difference between regular and non-regular gamblers.  Are the heuristics the underlying cause of irrational gambling or merely a symptom of a deeper personality defect?

Griffiths argues that knowledge of the heuristics gamblers use could be used to rehabilitate gamblers through cognitive behavioural modification. This would involve modifying the thought patterns of an individual in an attempt to moderate or stop their gambling.  

However, the measures of skill and perception used in this study are very subjective, based as they are, on self report techniques.  Also the self-selecting method of sampling ould not have gathered a representative cross-section of gamblers. 

Dickenson & Baron (2000) blame demand characteristics for the increased verbalizations believing they are attempts by gamblers to explain their behaviour to observers. 

Evaluation of the cognitive model

As is usually the case with cognitive explanations it’s difficult to disentangle cause and effect (chicken and egg if you will).  Research disagrees over what comes first the irrational thinking and heuristics (which the model assumes are causes of addiction) or the addiction, making the heuristics mere symptoms of the addiction. 

Think of similar problems we’ve seen before:

Does faulty perception of body image create anorexia or is it a symptom?

Does negative thinking cause depression or is it a symptom?

Assuming that the heuristics pre-date the addiction then how do they arise?  Why don’t we all develop this way of thinking?  What makes some people more susceptible than others?  Individual factors such as these seem best explained by the biological model and the possibility of some brains being more or less sensitive to dopamine and its rewarding effects.  This in turn being determined by our genes.  Or perhaps dispositional factors such as the fabled ‘addictive personality’ again predisposes some of us to all manner of addictive behaviours.

It would seem logical to conclude that games based on skill (or at least the perception of skill) would be more addictive given the cognitive explanation and its ideas of illusions of skill etc.  In games that are clearly random such as the lottery, the illusion of skill and control is going to be minimal in comparison to other forms of gambling such as cards (perhaps excluding pontoon).  Fruit machines employ what are called structural techniques such as nudges, ladders and near wins to make repeat gambling more likely. 

Evolutionary explanations of gambling

Perhaps taking risks is of evolutionary advantage.  But for those prepared to venture forth and explore we might still be hunter-gatherers dressed in antelope hide!  Those prepared to seek out new environments, strange new worlds (boldly going etc) in search of food and mates might well have been placed at an advantage.  More likely to thrive and more likely to attract a mate and reproduce. 

This might explain why it is younger men that are most attracted to the risky business of gambling.  Women of the same age becoming more cautious. 

When tested at the age of 11 to 15 both sexes are equally prepared to take a gamble.

However, during the 16 to 21 age bracket, the percentage of men increases and that of women decreases. 

However, this could just as easily be due to socialization and the way we are expected to behave!

Next page