Why would anyone immobilise a caterpillar with thin strips of cling film and drip caffeine into its head? To help it to pass an exam? To enable it to make that gallery opening after work? No: in the name of science, of course. John Glendinning and colleagues of Columbia University, New York have been doling out the insect-equivalent of double espressos in an effort to shed light on the complex issue of taste sensitisation. Caterpillars, you see, are a bit of an open book when it comes to recording neural responses to taste, and caffeine is a useful bitter-tasting compound.

Just as people who eat a lot of chillies or sugary foods seem to become almost immune to them, "chronic exposure to specific compounds can [also] profoundly alter subsequent behavioural responsiveness to chemical stimuli in both vertebrates and invertebrates," as Glendinning's group explain in Journal of Experimental Biology [July 1999]. But what is not yet clear is where exactly these exposure-induced sensitivity changes occur in the 'taste pathway' which joins an animal's sensory organs with its brain.

So the team fed one group of Manduca sexta caterpillars a caffeine-rich diet for 48 hours. To a second group they applied the caffeine directly to the insects' taste organs for 48 hours without allowing them to ingest any of it (by siphoning off the drips with a bizarre miniature vacuum device). Other groups were fed or drip-exposed to a variety of other solutions. After all these experiments, the researchers measured the responsiveness of the bitter-sensitive taste cell to several bitter and pleasant-tasting substances, including caffeine.

This revealed that, although dietary exposure of the caterpillars to caffeine – as Glendinning's group puts it – "significantly (and specifically) decreased the responsiveness of [their] bitter-sensitive taste cell to all caffeine concentrations by about 70%", gustatory stimulation alone had more or less the same effect. Interestingly, both types of exposure also dulled the cell's sensitivity to salicin (another bitter substance that uses the same taste pathway as caffeine) but had no effect on its response to aristolochic acid, a third bitter compound, which acts on the cell in a different manner. The caterpillars' other taste cells (for salt and sugar, for example) within the same organ, however, were unaffected.

Furthermore, the bitter-sensitive taste cell in those insects that had eaten an agreeable control diet or who had had it dripped onto the organ for 48 hours was more responsive to caffeine (and salicin).

The finding that this numbing phenomenon can be localised to a single biochemical route within a single taste cell reveals that exposure-induced desensitisation does not necessarily render a taste cell 'blind' to all the substances to which it would usually respond. Which, for creatures needing to avoid ingestion of dangerous doses of toxic substances, makes sense.