Aristotle called it akrasia. The puzzle of self-defeating behaviours. Why do we persist in doing things we know are against our interests? Things that we say we do not want to do.
Why is it when one leaves the beach at Yuurrgubarraalbigu for the five-hour drive back to Cairns one swears there will be no stopping at the KFC in the main street of Mareeba, four hours away? The recollection of how ill the last indulgence left one is vivid. What was irresistibly finger-licking good before the fatal capitulation soon turns into bilious recrimination. The regret starts before the indulgence is complete.
What was firm resolve back at Lakeland Downs starts dissolving down the straight at Biboohra and by the time you zoom past the “Welcome to Mareeba: More than 300 sunny days a year” sign, the Devil has seized the steering wheel. The cues are unavoidable, the final being the Colonel’s beneficent but evil smile saying in his Kentucky drawl: “You thought you were going to dodge me n-n-native, but where I come from black folks eat fried chicken and watermelon and I’m gonna make every one of you n-n-natives down under join ‘em!”
So under compulsion, natives stuff the red and white buckets into the overhead lockers heading north to Cape York and the Torres Strait, filling the plane with the alluring if faintly nauseating perfume of fried chicken.
The layman may call it gluttony, but the ancient Greeks called it akrasia. Weakness of will.
The scientific literature on food compulsions increasingly points to common hedonic mechanisms underlying obesity and drug addiction. Whether alcohol, drugs or fatty food, a wide range of disorders remind us no matter how much we may intellectually appreciate the larger costs associated with certain indulgences, such compulsions are immune to reason. The smoker is aware that smoking kills, but even the highest medical awareness of future costs is no match.
There are two ways to look at the apparent inability of addicts to resist temptation. The old disease model championed by Alcoholics Anonymous and others emphasise that addicts are powerless over the addiction.
In recent decades, theories of addiction have burgeoned around the role of choice. Addicts seem not completely bereft of choice. The literature on addictions is divided between those who urge a visceral view of addiction and those who say that we make choices around our addictions.
There is not yet a scientific consensus on the nature of addiction. My layman’s view is that a synthesis needs to reflect a thorough understanding of choice theory.
The literature in psychology and behavioural economics on the role of choice in addiction started with Nobel Laureate Gary Becker’s 1988 theory of rational addiction. With a colleague from the Chicago School of Economics, Becker argued that addicts seek to maximise their future utility from their drug use.
In 1937, economist Paul Samuelson first brought precision to the measurement of utility in economics, and Becker applied rational choice theory to the analysis of addiction. That addicts are rational utility maximisers is the idea. According to this theory humans discount the value of a future reward according to an exponential curve over time. The exponential rate of discounting has been central to microeconomics ever since Samuelson. The meaning of exponential discounting is as follows: imagine a reward that we can obtain in the future. But we value any reward higher if we get it sooner. We therefore discount the value of the reward by a constant factor per unit time.
However George Ainslie, an American research psychiatrist, published experimental evidence in the 1970s that showed the rate of discounting of future rewards followed a hyperbolic rather than exponential curve. Since then an accumulation of research involving a range of human and animal subjects confirms the hyperbolic discount rather than the exponential rate assumed in standard economics. Hyperbolic discounting means that we don’t discount the value of a reward by a constant factor per unit time of delay before we can get it. The further into the future a reward lies, the less sensitive we are to delays.
The important consequence of hyperbolic discounting is as follows. There is a time when a larger but later future reward is more attractive than a smaller, sooner reward; there is a point in time where both rewards are equally attractive; and there is a time - just before you can get the smaller reward - when the smaller, sooner reward is more attractive. If we draw two curves that show the valuations of the two rewards over time, the point of equal valuation is the point where these two curves intersect.
Which mathematics do we intuitively follow in our everyday decisions? Exponential or hyperbolic discounting?
It appears that exponential discounting of a future reward is limited to circumstances where humans are at their most rational: calculating the value of money over time. Perhaps it is when we engage in mathematical calculation that our choices are the coolest. In other circumstances we probably make much hotter and less rational calculations.
Hyperbolic discounting represents a challenge to conventional economics. Research evidence and theories emanating in psychology and neuroscience have messed up the clean lines of rational economic theory. Microeconomics does not presently know how to contend with what Ainslie dubbed the picoeconomics of human motivation and brain functioning.
One of the strengths of Ainslie’s hyperbolic discounting is that it predicts an important feature of animal and human behaviour: preference reversal. Standard economics provides no explanation for preference reversal.
In standard economic theory, if we place a high valuation on our long-term health (the larger, later reward) then we will discount that value exponentially according to a curve with the same shape as we discount the low valuation we place on an indulgence in fried chicken.
If our valuation of future indulgence is lower than our valuation of future health at a certain point in time, it should be lower at every point in time, including the 30 seconds before we drive past KFC. This would be true if our valuation curves were exponential.
Yet while we might hold to the larger, later reward for a long period of time, there comes a point when the sooner reward is imminent, and we flip our preference!
Ainslie’s hyperbolic discount curves explain why we change our preference driving down the main street of Mareeba and we see the KFC Colonel chuckling at our impotence. When indulgence and health are both perceived to be distant rewards, health is valued higher. Close to the opportunity for gluttony the curves intersect. And in the final seconds the temptation is irresistible.
We often anticipate preference reversal, and we try to devise strategies to deal with anticipated weakness of will.
We can, earlier in the journey, when our preferences for health v fried chicken have not yet been reversed, choose a strategy of detouring through the back streets of Mareeba, thereby avoiding temptation. This is the strategy of pre-commitment.
It is one of the measures proposed by the Productivity Commission to curb problem gambling. Andrew Wilkie and Nick Xenophon are now pushing the Gillard government to introduce pre-commitment devices to poker machines, against opposition from clubs and state governments. Wilkie and Xenophon are dead right: without pre-commitment devices, no amount of “responsible gambling” education and awareness campaigns will match the power of addiction and preference reversal.
Which is why I am sceptical of the national marketing campaign aimed at tackling the appalling rates of tobacco smoking among indigenous Australians. If it helps to get the conversation going in communities, then fine, but the water-off-a-duck’s-back effect of education and awareness campaigns is well known. Much more decisive interventions will need to accompany the campaign.
If we are susceptible to subverting larger, later rewards in favour of smaller, sooner rewards, then one strategy that appears to avoid compromising our long-term wellbeing is to construct what I will call valuable rewards in the near-term.
This happens when employees who value their jobs are able to arrest their substance abuse problems when subjected to random drug testing as part of workplace health and safety requirements. This was evident in a trial of young indigenous workers from Cape York Peninsula who relocated to Victoria to work in abattoirs.
While many were habitual marijuana users, most readily adjusted to the random drug testing regime thanks to an appropriate amnesty by the meat processing company. In this case the calculation facing the young men was whether to give in to the smaller, sooner reward of indulging in marijuana, or to preserve their jobs and high incomes that were visible and valuable when they made their calculations.
The use of random drug testing in workplaces for health and safety reasons represents an effective intervention for thousands of people who would otherwise be struggling with drug addictions.
For similar reasons, programs enabling the reinstatement of practising certificates to doctors provided they comply with drug testing over an extended period, have proven highly effective in the rehabilitation of drug-addicted medical practitioners.
The discourse between visceral motivations involved in addiction and choice remains to be resolved. I am conscious that laymen such as me follow such interesting intellectual developments as if following a horse race. I expect Ainslie will ultimately perform the required synthesis and receive a Nobel for illuminating the picoeconomics of human motivation and choice.