text-content

Strong Inference: A Year 10 Athenaeum Lecture

During Science Week, the Year 10 Athenaeum explored a version of standard Baconian inductive inference that emphasises the importance of formulating alternative hypotheses. The method, known as “strong inference” was set out by University of Chicago biophysicist John R. Platt in the journal Science in 1964.

The key difference to other forms of inductive inference is in the a prior formulation of multiple alternative hypotheses, an open-minded approach that protects against blinkered vision and conformation bias. It was this version of the scientific method, Platt argued, that explained the rapid progress made in molecular biology and high-energy physics in the post-war period.

Biology teacher Dr Henry Nicholls set the Year 10 scholars the task of formulating as many alternative hypotheses as possible to explain the spectacularly coloured feet of the blue-footed booby. In about five minutes, the students had brainstormed about ten different possibilities and then began to think about “crucial experiments” that would falsify each hypothesis. They were able to dismiss, for instance, a role for the feet in camouflage against both prey and predators or as a response to temperature variation in the birds’ environment. By the end of the session, there were only two alternative hypotheses still standing: perhaps the birds’ diet could supply a pigment that might account for the colour and maybe the intensity of the blueness could play a role in mate choice.

In fact, researchers have explored both these hypotheses, working on a colony of blue-footed boobies off Mexico around 20 years ago. In one experiment, birds with a reduced diet quickly lost the blueness of their feet, but recovered it when given their normal food. In other experiments that involved manipulating the colour of feet, the duller the colour the lower a bird’s reproductive success and the brighter the blue the higher its reproductive success. Rather than disproving the diet and mating hypotheses, these crucial experiments lend them support. Until dismissed, they must remain on the board.

As it stands then, both female and male boobies appear to use the intensity of colour as a reliable cue of foraging skill and, therefore, as an accurate indication of parental potential.