The Editor of the British Journal of Psychiatry has been kind enough to ask me to write a summary and a critical survey of the objective data bearing on the causation of schizophrenia by environmental factors, especially those of a psychogenic nature. I was also asked to combine with this a general assessment of our knowledge in this field. At first sight the task as defined seems a simple and easy one, namely to collect together all those facts which point to the probability of schizophrenia being caused by the psychogenic influences of the patient's environment. But almost at once, a number of serious doubts arise. What are we to understand by the concept of “Schizophrenia“—should we take it in its widest sense as including all the “schizophrenic reaction types”, as do many American authors, or in the far narrower sense favoured by Scandinavian psychiatrists, who recognize only a central or “nuclear group” as true schizophrenia, and describe all related conditions as “schizophreniform psychoses“? Further, what do we mean here by “causation”, and what do we call “objective data”? We shall see later that if we were to collect as data only such as a “non participant observer” (H. Stierlin) thinks he can perceive in the schizophrenic's environment, we should miss the significance of those very factors which are most relevant to the development of the psychosis. It is indeed a complex question—and one which has been the subject of much philosophical reflection— whether it is at all legitimate to assume the existence of “causes” for psychological processes, and whether a search for “causes” is a meaningful undertaking. This, however, lies outside the scope of the present survey; all we need to do is to bear in mind that the use of a one-sided conception of causality might from the very start greatly restrict our field of enquiry, so that essential processes by which schizophrenia arises might not even come to our notice.

This enquiry was prompted by a discussion on training in psychotherapy which took place at the Troon Conference on the Teaching of Psychiatry (Ingram, 1963). We felt that more precise information was required about what is meant by the term psychotherapy, which methods are in use and what opinions are held by psychotherapists in training. This information could be of value in assessing the workload of psychiatrists and in deciding what training might be given to psychiatrists and medical students. It was therefore decided to seek the opinions of Scottish psychiatrists on some aspects of their current practice in psychotherapy.

In a number of previous communications (10, 11, 12) I have advanced the thesis that in successful psychotherapy what goes on is a unique kind of learning process in which the therapist employs his particular theoretical frame of reference as a rational basis for explaining the patient's past and present difficulties, and by non-verbal as well as verbal cues uses the instrument of a warm empathic human relationship as a means of helping the patient to persist in the difficult task of combating his anxieties, overcoming his resistance to change, and learning more mature patterns of adaptation. This underlying modus operandi is essentially the same in different “schools” of dynamic psychiatry, despite their ideological variations. In the present communication I propose to outline some of the significant areas in which the findings from learning theory experiments are applicable to what goes on in psychotherapy.

Behaviour therapy is conducted in this department under combined general psychiatric and behaviour therapist supervision. Patients considered suitable for behaviour therapy are first assessed clinically at a phenomenological and psychodynamic level. On the basis of this assessment an appropriate symptom or number of symptoms are selected for treatment by behaviour therapy. The patient is thereafter observed by the original clinician and assessed at regular intervals. Meanwhile the behaviour therapist treats the specific symptom or symptoms.

The significance of observations made in clinical work can really only be found by reference to a hypothetical position. It is necessary therefore first to indicate very briefly the theoretical constructs relating to therapeutic community work.

In view of recommendations for the establishment of Units for the treatment of addiction to Heroin and Cocaine (H & C) (1) and (2), it may be of interest to relate our experiences of treating addicts to H & C and to alcohol in an Addiction Unit.

Sleep is essential for physical and mental health. In the last 15 years there has grown up the concept of the brain stem reticular activating system. Electroencephalographic studies have shown two qualitatively different and alternating kinds of sleep, the orthodox (“slow wave”, or “forebrain“) and the paradoxical (”hind-brain“, “rapid eye movement”, “activated“, or “dreaming”) phases (Akert et al., 1965). It may be predicted that in the next decade attention will turn increasingly to the chemical basis of sleep. If a man is deprived of sleep for 100 hours, it is extremely difficult to keep him awake and one may suppose that an abnormal biochemical state exists within his central nervous system.

Patients with idiopathic narcolepsy are seen by both neurologists and psychiatrists. The diagnosis has traditionally been made on the basis of a history of frequently falling asleep against the patient's wishes, attacks of cataplexy or loss of muscle tone upon sudden emotion, episodes of sleep paralysis, and an account of vivid, dream-like experiences during the periods of involuntary sleep.

The validity of judgments based upon clinical methods such as interviews or projective tests has been persistently questioned over the past 15 years. The disappointing results of clinicians' efforts to predict the professional performance of Veterans' Administration clinical psychology trainees or Menninger psychiatric residents are well-known examples (Holt and Luborsky, 1958; Kelly and Fiske, 1950). Other investigators (Glass, et al., 1956; Holtzman and Sells, 1954; Plag, 1964; Thorne, 1960) also have reached highly critical conclusions with regard to the validity of clinical procedures, based upon their studies in military settings. Eysenck (1952), in particular, has strongly rejected claims of validity for the clinical approach. Hunt (1959), after many years devoted to the problem, reports progress in achieving stable validities by application of quasi-experimental approaches to the analysis of clinical judgments. Meehl (1954), in summarizing research on the accuracy of clinical judgments more than a decade ago, suggested that such predictions often err because the clinician typically does not know what weights to give the available information nor how to combine the various cues efficiently.

Foulds (1961) has drawn what appears to be a useful distinction between personality traits and neurotic and psychotic symptoms and signs. This distinction is helpful both in diagnosis and in the assessment of treatment. Within the framework of this personality-trait and symptom-sign differentiation, Caine and Hawkins (1963) introduced a measure of what they describe as the hysteroid-obsessoid dimension of personality; and reasons in defence of this admittedly rather ugly terminology have been stated recently by Caine and Hope (1964). The self assessment hysteroid-obsessoid questionnaire (HOQ) was developed as a refinement of the hysteroid-obsessoid rating scale previously devised by Foulds and Caine (1958), which they had not found to be adequate in differentiating between groups of obsessoid and groups of hysteroid persons. The questions in the HOQ are framed in simple language and require only a true-false response. Caine and Hawkins have reported the results they obtained from administering this questionnaire to neurotic in-patients, some of whom also completed the Maudsley Personality Inventory (MPI), albeit at different stages of treatment. Although the MPI has been used extensively with samples drawn from non-psychiatric populations, both in the United States and the United Kingdom, Caine's questionnaire, a relatively new instrument, has not been as widely tested.

Since the introduction of the WAIS many short forms have been proposed, but specific problems associated with the assessment of the aged have not been factors influencing their development. The measurement of cognitive functioning in persons over 65 presents the clinician with many difficulties, not least of which is a consideration of time.