Assessment of factors affecting the detectability of thermal radiations from the neural system of the brain.

The paper describes a technique by which it may be possible to locate and analyse neural activity within the brain by a non-invasive and non-contact method with (in principle) an unlimited bandwidth. The method detects the rapid temperature fluctuations which are believed to accompany individual action potentials and other neural phenomena, and which should be detectable as a noise-like fluctuation of thermal infra-red radiation, with components up to several hundred Hz. It differs from conventional surface thermography in that it ignores changes of mean temperature in blood vessels etc. and transfer of heat to the skull surface by conduction or blood flow. Energy is assumed to travel only as infra-red radiation directly from the neural material to the detector, and as such the signal is very heavily attenuated by the water content of the skull and other intervening material. Available data suggest that the viability of the method is considerably reduced by this factor. None the less, the scarcity of reliable data on the temperature fluctuations themselves and the immense advantages of the technique, should it be successfully developed, suggest that it warrants detailed theoretical and experimental study.