Thursday, July 2, 2009

Decision making under uncertainty: A behavioural model

Recent research in behavioral economics has revealed the important role cognitive biases play in influencing decision making. Certain interesting truths that economists find surfacing from the prosaic behavior of apparently rational minded consumers reveal that you and me are oriented towards front-loaded benefits and back-ended costs, that we are ignorant of the outcomes of most options which would have helped us choose best, nor can we accurately evaluate complex choices. We are more averse to risks and losses than attracted to similar sized gains, and have an overwhelming motivation to be liked and regarded, just as we do not always know or care what is in our best interests, and see specific conditions as more probable than a single general one (conjunction fallacy).

Nicola Gennaioli and Andrei Shleifer present a model of judgment under such uncertainty, which can account for some of the evidence on the aforementioned commonly observed heuristics and congitive biases. They use the model to explain how cognitive biases like availability and anchoring heuristics come to mind of or are recalled by individuals facing a decision making moment.

In their model, "an agent combines data received from the external world with information retrieved from memory to evaluate a hypothesis. (They) focus on what comes to mind immediately, as the agent makes quick, intuitive evaluations. Because the automatic retrieval of data from memory is both limited and selected, the agent's evaluations may be severely biased... Some, but not all, of the missing scenarios come to mind of the decision maker. Moreover, the hypothesis in question primes the selective retrieval of scenarios from memory, with those most predictive of the hypothesis itself relative to the other hypotheses – the representative scenarios - being retrieved first. In many situations, such intuitive judgment works well, and does not lead to large biases in probability assessments. But in situations where the representativeness and likelihood of scenarios diverge, intuitive judgment becomes faulty."

In standard economic models of agents making decisions under uncertainty, they are assumed to combine the limited information they receive from the outside world with everything they already know, to make evaluations and decisions. In the behavioural model proposed, the authors claim that people do not bring everything they know to bear on their decisions but recall information (most often not even those most useful) selectively from passive memory. The authors make the distinction between intuition (of the behavioural economists) and reasoning (of the conventional economists), not as two different modes of thought, but in what is retrieved from memory to make an evaluation. In the former, the retrieval is not only quick, but also partial and selective, whereas in the latter the retrieval is complete.

Update 1
John Tierney points to studies by behavioural economists that find that humans often waste money (on expensive things) because of the unconscious — and mistaken — belief that our costly stuff will signal our intelligence and sterling personality traits to potential mates and allies. See also here on more such research.

Update 2
In the light of the worst economic crisis since the Great Depression, Gary Stix (also here) makes a reassessment of how financial markets work and how people make decisions about money.

No comments: