The validation of personality questionnaires presents the psychologist with knotty problems. His reaction all too often is to plead that, as it is difficult to find criteria for validating measures of personality, he is justified in relying on face validity alone. Fortunately, this has not been true of the Maudsley Personality Inventory (MPI), nor even of its more recent successor, the Eysenck Personality Inventory (EPI). Serious attempts have been made to validate these questionnaires, and the problem has been tackled from a number of different angles. S. B. G. Eysenck (1962) and Eysenck and Eysenck (1963) found that subjects nominated by judges as extraverted, introverted, stable and neurotic scored significantly differently on scales measuring these dimensions. A further study (Eysenck and Eysenck, 1964), in which larger numbers of both judges and subjects were used, confirmed these findings. Vingoe (1966) found that subjects who rated themselves extraverted on a seven-point scale of extraversion–introversion differed significantly on the extraversion scale of the EPI from those who rated themselves introverted.