Personal Views Personal views

Railway signals passed at danger: psychology matters

BMJ 2000; 320 doi: (Published 03 June 2000) Cite this as: BMJ 2000;320:1549
  1. Glin Bennet, formerly senior lecturer and consultant psychiatrist
  1. Bristol

    Seventy five per cent, eighty per cent. Over the years this is the seemingly unvarying proportion of transport accidents ascribed to human factors. Investigations are meticulous, the engineers take care to see that every nut and bolt is studied, yet so often the conclusion is the same: “human failure.”

    This seems to be accepted as an explanation, and with it the matter can be closed. For any psychiatrist, reports of odd behaviour, such as driving a car or a train through a red light, are merely the beginning of an investigation, a warning that there may be problems below the surface.

    In the aftermath of the Paddington rail disaster in October 1999, attention has been focused on signalling systems, and rightly so. The assumption seems to be that if the signals and points can all be made reliable then accidents should cease. But that does not explain the 600 or so signals passed at danger each year in the United Kingdom.

    We need to understand more about the drivers

    There is nothing new in all this, as Professor R A Cocks has pointed out (BMJ 1999;319:1018-9). In the 1940s and 1950s the late Professor Derek Russell Davis published research on the human factors which lead to accidents. His researches into rail incidents, usually signals passed at danger, were made with the full cooperation of British Rail, and in private interviews drivers spoke openly to him about their behaviour in the accidents, their anxieties, and distracting worries.

    In my own studies of small boat sailors, airline pilots, train drivers, and motorists, numerous examples emerged of psychological processes which can lead to accidents. For example, seeing the signal that you want to see, in the colour that you want, not what is actually there (Lewisham train disaster, 1957–90 dead; Norton Fitzwarren, 1940–27 dead; perhaps also Harrow and Wealdstone, 1952–112 dead). Sailors can misread lights in the same way.

    Preoccupation with matters remote from the scene may account for some of the 600. At the Hayward's Heath accident in 1973 the driver misread the signals and drove his loopline train on to the main line and it was struck from behind by an express. This driver was reported as going through a divorce at the time, had financial worries, and a stomach ulcer. In November 1981 the driver of a commuter train was involved in a collision from misreading the signals. He was evidently dosing himself with pain killing drugs for an aching wisdom tooth, his wife had just given birth to their first child, and his mother was suffering from cancer.

    The Moorgate underground train disaster in 1975, 43 killed, was never satisfactorily explained, or possibly was never satisfactorily investigated. The medical experts involved gave speculative diagnoses that the driver (who was killed) was suffering from “akinesis with mutism” or “transient global amnesia.” I know nothing of these as acute conditions, but I would have wanted to inquire more into the unfortunate driver's circumstances.

    Embedded Image

    The reports of inquiries that I have read tended to be chaired by military men, with a robust approach to mechanical detail and not much interest in psychological processes. There is also the subtle British resistance to looking into the psychological process of others, because of the unspoken acknowledgment from so doing that you might have a vulnerable psyche yourself, and that would not do. Fortunately, the drivers of most rail incidents survive, unlike their colleagues in the air, and so there is a real opportunity to learn about the causes of the 75% or 80% of transport accidents. In fact the railways are excellent laboratories for such studies, and they have a relevance to all modes of transport.

    Improvements in signalling and track are important, of course, but the drivers are also important. These people can confound safety devices when they fall victim to adverse psychological processes, and if we are not going to opt for driverless trains we need to understand more about the drivers.

    View Abstract

    Sign in

    Log in through your institution

    Free trial

    Register for a free trial to to receive unlimited access to all content on for 14 days.
    Sign up for a free trial