The equipment used was a Swan Housewares Ltd automatic traditional style (i.e. squat rather than jug form) electric kettle, catalogue number 00162, circa 1989. This was connected to a standard British ring main operating at approximately 240v as per the kettle's instructions. On each test, the kettle was filled from cold and empty with 1 litre of tap water which had been allowed to reach the ambient temperature. Timing would commence at switch on, and the automatic switch off would be taken as the end point.

Stage one was to check that a watched kettle actually would boil. This was performed by giving the kettle full attention for the full time it was switched on. This took 3 minutes 34 seconds. (all times given are the mean result.)

Stage two was to boil the kettle whilst concentrating on doing something else in the vicinity. This resulted in a time of 3 minutes 50 seconds.

Stage three was to leave the kettle boiling whilst leaving the room, to suggest that we'd forgotten the kettle, but staying nearby. This took 3 minutes 05 seconds.

Stage four was to repeat stage three, but to go further away, and chance missing the actual end point. This took 3 minutes 01 seconds.

From these results, it is clear that a watched kettle does in fact boil, however it will take longer than if it believes that it is being completely ignored. It does not however get fooled by the mere act of doing something else nearby while it is boiling. On the contrary, this seems to initiate a malicious extension of the boil-time. Clearly the fastest way to get a kettle to boil is to switch on, then go somewhere completely different.

Environmentalists please note, the boiled water was subsequently used in the preparation of much needed caffeine rich beverages. Bottom Pineapple does not condone the excessive wasting of electrical energy or hot water.