Journal of Mental Science (1944) 90: 595-602. doi: 10.1192/bjp.90.379.595
© 1944 The Royal College of Psychiatrists
Parotid Gland Secretion in Affective Mental Disorders
H. J. Eysenck, Ph.D.* and
P.-M. Yap, B.A.
Psychological Laboratory, Mill Hill Emergency Hospital
* With the support of the Rockefeller Foundation.
ABSTRACT
Summary and Conclusions: The salivary secretion of altogether 100 neurotic and psychotic patients was measured in c.c. by means of the Lashley disc, under a variety of different conditions, such as olfactory stimulation, rest, reading, mental work, etc. Under these experimental conditions the following results were found:
- Salivary secretion in neurotic patients suffering from affective disorders (anxiety, depression) is significantly less than secretion in neurotic patients suffering from hysterical and other non-affective disorders. The control neurotic group secreted 1.41 times as much saliva as the affective group.
- Salivary secretion in psychotic patients suffering from affective disorders (melancholia, manic-depressive psychosis) is significantly less than secretion in psychotics suffering from schizophrenia and other non-affective disorders. The control psychotic group secreted 1.36 times as much saliva as the affective group.
- Men serected 1.2 times as much saliva as women, a proportion similar to that which the body-size of one sex bears to that of the other.
- Salivary secretion is decreased during concentrated mental work. There is about ten times as much secretion during silent reading as there is during the most concentrated mental work.
- Food imagery and olfactory stimulation produce a reduction rather than an increase in the amount of salivary flow when the measurements are taken over a long enough period to allow "unconditioning" to take place.
- The curve of distribution of the "scores" (i.e. of the total amount of saliva secreted by each patient during the period of the experiment) is not normal, but resembles a J-curve.
- The reliability of the test, as measured by retesting a sample of patients after three weeks, is not very high, the correlations between test and re-test being .54 ± .15 S.E.