J. psychiar. Rex, Vol. 24, NO. 2, pp. 155-163. Printed in Great Britain.
AUGMENTATION
IW
GU22-3956/90 $3.00 + .OO Pergamon Press plc
OF AUDITORY SOMATIZATION
LEIGH JAMES,* EVIAN GORDON,
EVOKED
POTENTIALS
IN
DISORDER
CLAUDIA
KRAIUHIN, ALAN HOWSON?
and RUSSELL MEARES Department of Psychiatry, Westmead Hospital, Westmead, NSW 2145, Australia; Severalls Hospital, Colchester, Essex, CO4 5HG and TDepartment of Economics, NSW 2113, Australia (Received
15 July 1988; revised
12 March
*Department of Psychology, Macquarie University, Ryde,
1990)
Summary-Sensory input regulation was examined in terms of augmenting/reducing of auditory evoked potentials in 10 patients with somatization disorder (8 males and 2 females) and 10 ageand sex-matched normal controls. The slope of PI-N1 amplitude change as a function of stimulus intensity was greater in patients compared with controls, suggesting an enhanced central nervous system response to sensory input. Taken together with previous findings of a failure to habituate to incoming stimuli in a similar group of patients, and evidence obtained in somatizers of both over-responding to background stimuli in a simple tone-discrimination task and enhanced parietal activation during selective attention, this finding suggests disturbances in the processes of attention and in the regulation of afferent stimuli in somatization disorder, and may help explain the multiple and chronic complaints characteristic of patients with the disorder.
INTRODUCTION PATIENTS
recurring
with somatization disorder (SMD) attend many clinics for the investigation complaints which are often minor or for which no organic cause is evident.
of In
many cases, the description of these symptoms corresponds quite closely to known disorders. Pain is a common feature of these symptoms which often result in extensive but ultimately fruitless investigations. According to DSM-III-R (AMERICAN PSYCHIATRIC ASSOCIATION, 1987) criteria, at least 13 symptoms of several years’ duration and of unknown organic pathology are required to satisfy a diagnosis of SMD. Somatization disorder is related historically to hysteria; it partially replaced the diagnosis hysteria in DSM-III (AMERICAN PSYCHIATRIC ASSOCIATION, 1980; for review, see GORDON, KRAIUHIN, KELLY, & MEARES, 1984). Unlike hysteria, a diagnosis of somatization disorder does not require the presence of conversion symptoms-those which mimic symptoms of functional loss due to neurological damage. Nevertheless, in many cases of SMD such symptoms are evident. Given the relationship between somatization disorder and hysteria, it is not unreasonable to look to aetiological theories of hysteria for hypotheses concerning the bases of the disorder. Theories of acute and chronic hysteria have been dominated by psychodynamic (primarily sexual aetiology) and sociological (primarily secondary gain) perspectives. However, JANET (1901) in seeking the underlying causes of chronic hysteria, proposed a neurophysiological model which concerned a disturbance in the awareness of or attention 155
LEIGHJAMESet al.
156
paid to sensory data impinging upon the nervous system. WHITLOCK (1967) also regarded hysteria primarily as a disorder of attention and vigilance. The theory of a subtle dysfunction in chronic hysteria with respect to the neurophysiological processing of afferent stimuli has received some experimental support (SHAGASS & SCHWARTZ, 1963; MEARES & HORVATH, 1972; MUSHIN & LEVY, 1974; MOLDOFSKY & ENGLAND, 1975). MEARES and HORVATH, for example, reported that such patients failed to habituate to a repetitive and meaningless auditory stimulus; a finding which suggests a deficiency in the ability to ignore irrelevant sensory input. However, there is still a paucity of studies in which neurophysiological measures of responses to afferent stimuli have been recorded from patients diagnosed according to explicit criteria as having somatization disorder. The hypothesis examined in this study is that somatization disorder is associated with a basic disturbance in sensory input regulation, that is, returning to JANET’S hypothesis, it is caused by a dysfunction in the processing of afferent stimuli. Sensory input regulation was examined in terms of the augmenting/reducing phenomena. To evoke these physiological phenomena, a series of stimuli of varying intensity are presented to the subject (BUCHSBAUM & SILVERMAN, 1968); in some individuals the amplitudes of evoked potentials (EPs) increase with increasing stimulus intensity (i.e. ‘augment’), whereas in others, smaller amplitudes or even a reduction in amplitude with increasing stimulus intensity occur (i.e. ‘reduced’). The resultant slopes of augmentation and reduction are presumed to reflect ‘stimulus intensity control’ (BUCHSBAUM & SILVERMAN, 1968; SILVERMAN, BUCKSBAUM, & HENKIN, 1969). Our hypothesis grew out of previous findings that excessive augmentation is associated with pain that has been called psychogenic (MUSHIN & LEW, 1974; BUCHSBAUM, 1975; MOLDOFSKY & ENGLAND, 1975). Psychogenic pain is frequently associated with depression; so also is the evoked potential pattern of augmentation (BUCHSBAUM, GOODWIN, MURPHY, & BORGE, 1971; BUCHSBAUM, LANDAU, MURPHY, 8z G~~DwN, 1973). Furthermore, it appears that, during the depressed state, the degree of augmentation correlates with the severity of depression (FRIEDMAN & MEARES, 1979b). BUCHSBAUM and his group believe that the augmenting pattern is not a direct correlate of depression but a genetic marker of predisposition to the illness (GERSHON & BUCHSBAUM, 1977). BUCHSBAUM and his colleagues, however, did not study patients in remission. It has since been shown that, on recovery, the EP amplitudes of those who suffer depression decrease significantly (FRIEDMAN & MEARES, 1979a; FRIEDMAN, MCCALLUM, & MEARES, 1980). The possibility arises that those with somatization disorder have particular patterns of sensory input regulation which are similar to depressives, but which are immutable and unrelated to temporary mood states. Given the association between the augmenting pattern and psychogenic pain, it was anticipated that somatizers would show enhanced augmentation compared with normal individuals. This hypothesis has also been advanced by LUDWIG (1972). However, to our knowledge, the hypothesis concerning augmentation of EP responses to stimuli of varying intensity has not been tested previously in somatization disorder patients diagnosed according to explicit and standardized criteria. METHOD
Subjects The patient
group consisted
of 8 females
and 2 males aged between
25 and 67 years
AUGMENTATION OFAEP’s IN SMD
157
(mean = 47 years) who had been referred by staff psychiatrists at a number of teaching hospitals. Diagnosis of somatization disorder was made using the PERLEY and GUZE (1962) checklist in which 25 symptoms with no apparent organic explanations are required. (DSMIII-R criteria are a summary of the PERLEY and GUZE checklist). An unexplained symptom was accepted as such if it satisfied the following criteria: (a) the symptoms caused the patient to take medicine (other than aspirin), alter his or her life pattern, or see a physician (b) the symptom was not adequately explained by physical disorder or injury (c) the symptom was not the side effect of medication, drugs or alcohol. According to these criteria the number of symptoms in the patient group ranged from 25 to 42 (mean = 32). Due to the polysymptomatic nature of their disorder, most patients (n = 7) were taking medication, primarily analgesics, minor tranquilizers (benzodiazepines) or both. Three patients (and all controls) were on no regular medication. The control group consisted of 10 normal subjects aged between 23 and 67 years (mean = 42 years) who were matched for sex and age (within 5 years) to the patients. Controls were recruited on a voluntary basis from visitors to and staff of a large general hospital. The number of ‘unexplained’ symptoms in this group ranged from 0 to 8 (mean = 2). Neither patients nor normals had a history of hearing or neurological disorders. No normal subject had a psychiatric history. PROCEDURE
Each subject was presented with a total of 250 auditory stimuli in the form of tones which varied in intensity (number of intensities = 5). The amplified EEG response to each tone was averaged separately for each intensity. In this way, 5 separate EPs were obtained, each representing the averaged EEG response to 50 tones of a particular intensity. The 5 intensities of tones were presented in pseudo-randomized ‘blocks’, where each block consisted of 5 tones of the same intensity. The procedure was such that every block was preceded by every other intensity block, including itself, an equal number of times. The use of pseudo-randomized blocks presumably minimizes the development of arbitrary sets that could occur during a long series of stimuli of the same intensity, and also reduces the possible effect on the amplitude of response of a strong stimulus on a weak one, and vice versa. It is also a way of coping with problems of habituation, fatigue, and attentional drift that could occur during the course of a recording session. Following a description of the experimental procedure the informed consent of each subject was obtained. Stimuli The auditory stimuli were produced by a software controlled sine-wave generator incorporated in the EP recording computer, and were presented with an inter-stimulus interval of 1 set through stereo headphones. Each tone had a frequency of 1000 Hz and a duration of 16 msec. The levels of stimulus intensity used in this study were 45, 65, 85, 95 and 102 dB above threshold. (The maximum intensity was selected in accordance with the finding of PRESCOTT, CONNOLLY, & GNUZELIER, 1984, that the classification of
LEIGHJAMESet al.
158
augmenting be reliable).
and reducing
patterns
using maximum
intensities
lower that 100 dB may not
Event-related potential recording An Ag/AgCl electrode was applied at Cz (of the lo-20 system; JASPER, 1958) which was referenced to linked earlobes and grounded with an electrode on the forehead. Additional electrodes were placed above and below the right eye and referenced to each other to record eye-related potentials which were averaged in a separate channel. Electrode resistances were below 5 kfi. The signals were amplified to a total gain of 50,000 and filtered with a bandpass of .05-100 Hz. They were sampled by an A-to-D converter at 868 Hz for 600 msec (100 msec prestimulus and 500 msec poststimulus). Trials in which 250 points of the potential exceeded +75% of the input voltage range of the analog-to-digital converter (f 5 V) were automatically rejected. The averaged waveforms were then additionally filtered with a nonrecursive digital low-pass filter at 40 Hz. Data analysis Subjects’ EEG records were averaged separately for each stimulus intensity. Individual peaks were scored by a computer cursor programme. Amplitude was measured as the peakto-peak amplitude of the Pl and Nl components. Mean (*s.d.) values for Pl and Nl latency across both groups were 41( f 15) msec and 89 (f 17) msec, respectively. An amplitude-intensity slope value was obtained for the PI-N1 component recorded in each subject. This was calculated by the least squares technique, which fitted a straight line (y = mx + c) to event-related potential amplitude measurements at the five intensities. In these computations, arbitrary values of 1, 2, 3, 4 and 5 were assigned to the auditory stimulus intensities (5 = loudest). This provided a ‘slope’ which reflected change in amplitude as a function of intensity. A high slope is indicative of a relatively great increase in EP amplitude for an increase in stimulus intensity, and a low or negative slope is indicative of a small increase or decrease in EP amplitude for increasing intensity. SCHECHTER and BUCHSBAUM(1973) designated augmenting subjects by positive slopes and reducing subjects by zero or negative slopes. The average slope of the two groups was compared using Student’s t-test. RESULTS Grand average EP waveforms for the normal and patient groups (averaged over 50 responses to each stimulus) are presented in Fig. 1. Mean values of PI-N1 amplitude at each intensity for each group are plotted in Fig. 2 (mean and s.d. values are presented in Table 1). These data show that PI-N1 amplitude was enhanced (most markedly at the higher stimulus intensities) for the somatizers compared to the normals. For individual subjects, the slope values were significant at the .05 level or less for 4 of the normals and 6 of the patients. For the normals as a group, the equation representing the least squares fit was as follows: Pl-Nl amplitude = 1.88 x intensity + 5.91 (standard error = 4.92). The equation for the somatizers was as follows: Pl-Nl amplitude = 2.75 x intensity + 5.41 (standard error = 4.9). The difference between the slopes was significant (t,,,, (19) = 2.13).
AUGMENTATION OF AEP’s
159
IN SMD
il
i2
i3
..... i4
i5
+ I
SO
I
200
(in ms)
Latency
-
I
100
NORMALS SMD i -stimulus
FIG. 1. Group
intensity
average EP waveforms; averaged over 50 responses to each stimulus intensity. augmentation of responses in somatizers, especially at the higher intensities.
TABLE
1.
MEANS
(S.D.)
OF
Pl-Nl
AMPLITUDE (in pV) AT EACH STIMULUSINTENSITY FOR NORMALSAND PATIENTS Stimulus
Normals
(n = 10)
SMD (n = 10)
Mean s.d. Mean s.d.
Note enhanced
Intensity 1
2
3
4
5
8.4 2.8 8.7 2.9
8.9 3.9 11.1 3.2
11.6 4.0 12.8 3.7
13.2 6.0 15.5 5.3
15.6 7.0 20.2 7.9
It is worth noting, however, that the significance of the results appears to have been contributed to by a very large Pl-Nl response at the higher intensities in one of the patients (i.e. 24.3 PV and 41.3 PV at stimulus intensities 4 and 5, respectively). The data from this subject were included in the analysis due to the unquestionable validity of the subject’s recordings. (The second largest responses from an SMD patient were 22.2 PV and 21.66 PV at intensities 4 and 5, respectively.) When this subject’s data were removed from the analysis the between-groups difference was not significant. Nevertheless, even without this subject’s data the somatizers still had larger responses, indicating a tendency for the somatizers to over-respond to the stimuli compared to the normals. DISCUSSION
Our finding that augmentation of auditory evoked potentials tends to be greater in patients with somatization disorder compared to normals suggests a disturbance in sensory regulation
LEIGH JAMESef al.
160 22r
20-
M
NORMALS
O-cl
SMD
16-
16-
8-
Intensity
of stimulus
FIG. 2. Means of Pl-Nl amplitudefor each stimulusintensityand for each group (n = 10 per group). based on recordings from the vertex (Cz).
Figures
in these individuals and requires further investigation. Specifically, the pattern of results indicates that somatizers over-respond to auditory stimuli of higher intensity. These results are consistent with the prediction of LUDWIG (1972) that the amplitudes of the P l-N1 and Nl-Pl waves in the averaged evoked responses of patients with hysteria would be minimally greater over moderate intensities with an exaggerated response at the highest stimulus intensities. The effect obtained in our study was observed in recordings from the vertex. In the light of findings, by CONNOLLY and GRUZELIER (1982) for example, the augmenting-reducing may vary at different scalp sites in the same individual to the same stimuli. Future research will explore our finding topographically. Although most patients in our study were not drug-free it seems unlikely that medication accounts for the experimental effect since other research (e.g. MILLIGAN, HOWARD, & DUNDEE, 1987; KOCHS, SCHULTE, & ESCH, 1988) have shown that benzodiazepines suppress the middle to late components of the evoked potential. Studies of sensory evoked potentials
AUGMENTATION OF AEP’s IN SMD
161
in patients with psychogenic pain (MUSHIN dz LEVY, 1972) and hysterical anaesthesia (MOLDOFSKY & ENGLAND, 1975) have revealed excessive augmentation in EPs of patients compared to controls. In the visual modality, VON KNORRING, ALMAY, JOHANSSON and TERENIUS, (1979) have demonstrated an augmenting response in the majority of patients with chronic pain. Our finding of increased augmentation in EPs of patients with somatization disorder further substantiates, in the auditory modality, the theory of neurophysiological disturbance in patients with psychogenic pain. Referring to the results of other research relating changes in average EPs to pain, BUCHSBAUM (1975) found that a reducing pattern of evoked responses is associated with a high pain tolerance. It might therefore be expected that a low tolerance of pain would be found in those whom augmentation is exaggerated. Our finding that patients with multiple symptoms, most of which involve pain, show an exaggerated augmenting response to auditory EPs is consistent with this hypothesis. Also consistent with this hypothesis is the finding of VON KNORRING et al. (1979) that the majority of patients with chronic pain have an augmenting pattern of visual evoked responses. (Their augmenting patients also had significantly lower CSF endorphin levels than those with reducing response.) In a previous publication (GORDON, KRAIUHIN, KELLY, & MEARES, 1986a) we reported an abnormality in the amplitude of the Nl component of auditory EPs to task-irrelevant (background) stimuli in somatizers. It is well known that the amplitude of the Nl components varies with the amount of attention directed toward stimuli (HILLYARD, HINK, SCHWENT, & PICTON, 1973; PICTON & HILLYARD, 1974; SCHWENT & HILLYARD, 1975; HINK VAN HOORIS, HILLYARD, & SMITH, 1977; HINK, FENTON, PFEFFERBAUM, TINKLENBERG, & KOPELL, 1978). The finding of an abnormal increase in Nl amplitude in somatizers, together with the subsequent finding that this abnormality is unlikely to be due to abnormally high levels of arousal engendered by anxiety, (GORDON, KRAIUHIN, MEARES, & HOWSON, 1986b) was interpreted as suggesting that such patients pay an undue amount of attention to stimulus input that others are better able to ignore. The current finding of increased PlNl amplitude in somatizers provides additional support for our hypothesis that somatizers over-respond to afferent input. Further support for this hypothesis comes from studies of other physiological indices of CNS activation. JAMES, SINGER, ZURYNSKI, GORDON, KRA~JHIN, and HARRIS, (1987) found evidence of hyperactivity in the right parietal lobe in a study of regional cerebral blood flow in somatizers. This region of the brain is implicated in the cortical network for the control of attention to external stimuli (e.g. GUR, GUR, & ROSEN, 1983; REIVICH, GUR, & ALAVI, 1983). In conclusion, the pattern of results from our laboratory reveals disturbances in the processes of attention and in the regulation of sensory input in SMD. These disturbances may be associated with the unusual incidence of recurrent pains and discomfort in patients with the disorder. However, given that other psychiatric disorders (e.g. depression) are associated with augmentation of evoked potentials, the specificity of our interpretations to somatization disorder needs to be investigated. It may be, for example, that an abnormally large degree of EP augmentation is associated with the psychogenic pains and the depressive symptoms experienced by patients with SMD, but is not the determinant of the entire syndrome of symptoms which represents somatization disorder.
162
LEIGH JAMES et al.
Acknowledgements-This research was supported by grant no. 860242 from the Australian National Health and Medical Research Council. We wish to thank Chris Rennie, BSc., M. Biomed. Eng., of the department of Medical Physics at Westmead Hospital, for his valuable contribution, in terms of soft-ware programming, to this study.
REFERENCES AMERICAN PSYCHIATRIC ASSOCLUION (1980). Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: Author. AMERICAN PSYCHIATRIC ASSOCIATION (1987). Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: Author. BUCHSBAUM, M. (1975). Average evoked response augmenting/reducing in schizophrenia and affective disorders. In D. X. Freedman (Ed.), Biology of the major psychoses: a comparative analysis. New York: Raven Press. BUCHSBAUM, M., GOODWIN, F. K., MURPHY, D. L., & BORGE, D. F. (1971). AER in-affective
Journal of Psychiatry
disorders.
American
128, 19-25.
BUCHSBAUM, M., LANDAU, S., MURPHY, D., &GOODWIN, F. K. (1973). Average unipolar affective disorders: relationship to sex, age of onset, and monoamine
evoked response in bipolar and oxidase. Biological Psychiatry
7, 199-212. BUCHSBAUM, M.,
Psychosomatic
& SILVERMAN, J. (1968).
Medicine
Stimulus
intensity
control
and
the cortical
evoked
response.
30, 600-611.
CONNOLLY, J. F., & GRUZELIER, J. H. (1982). Amplitute and latency changes in the visual evoked potential to different stimulus intensities. Psychophysiology 19, 599-608. FRIEDMAN, J., MCCALLUM, P., & MEARES, R. (1980). Stimulus intensity control in depression: a study of the comparative effect of doxepin and amitryptaline on cortical evoked potentials. Australian and New Zealand Journal of Psychiatry 14, 115-l 19. FRIEDMAN, J., & MEARES, R. (1979a). The effect of placebo and tricyclic antidepressants on cortical evoked potentials in depressed patients. Biological Psychology 8, 291-302. FRIEDMAN, J., & MEARES, R. (1979b). Cortical evoked potentials and severity of depression. American Journal
of Psychiatry
136, 1218-1220.
GERSHON, E. S., & BUCHSBAUM, M. (1977). A genetic study of averaged response augmentation/reduction in affective disorders. In Shagass, C., Gershon, S. & Friedhoff, A. (Eds.), Psychopathology and brain dysfunction. New York: Raven Press. GORDON, E., KRAIUHIN, C., KELLY, P., & MEARES, R. (1984). The development of hysteria as a psychiatric concept.
Comparative
Psychiatry
25, 532-537.
GORDON, E., KRAIUHIN C., KELLY, P., MEARES, R., & HOWSON, A (1986a). somatization disorder. Comparative Psychiatry 27, 295-301. GORDON, E., KRAIUHIN, C., MEARES, R., & HOWSON, A. (1986b). Auditory
somatization
disorder.
Journal of Psychiatric
Research
A neurophysiological evoked
response
study of potentials
in
20, 237-248.
GUR, R. C., GUR, R. E., & ROSEN, R. D. (1983). A cognitive-motor network demonstrated by positron emission tomography. Neurophsychofogia 21, 601-606. HILLYARD, S. A., HINK, R. F., SCHWENT, V. L., & PICTON, T. W. (1973). Electrical signs of selective attention in the human brain. Science 91, 177-180. HINK, R. F., FENTON, W. H., PFEFFERBAIJM,A., TINKLENBURG, J. R., & KOPELL, B. S. (1978). The distribution of attention across auditory input channels: an assessment using the human evoked potential. Psychophysiology
15, 466-473. HINK, R. G., VAN HOORIS, S. T., HILLYARD, S. A., &SMITH, T. (1977). The division of attention and the human auditory evoked potential. Neuropsychologia 15, 597-605. JAMES, L., SINGER, A., ZURYNSKI, Y., GORDON, E., KRAIUHIN, C., HARRIS, A., HOWSON, A., & MEARES, R. (1987). Evoked response potentials and regional cerebral blood flow in somatization disorder. Psychotherapy
and Psychosomatics 47, 190-196. The mental state of hystericals
JANET, P. (1901). JASPER, H. (1958).
(Translated by CORSON, C. R.). New York: Putnam Press. The ten-twenty electrode system of the International Federation. Electroencephalography and Clinical Neurophysiology 10, 371-375. VON KNORRING, L., ALMAY, B. G. L., JOHANSSON, F., & TERENIUS, L. (1979). Endorphins in CSF of chronic pain patients, in relation to augmenting-reducing response in visual averaged evoked response.
Neurophyschobiology
5, 322-326.
KOCHS, E., SCHULTE, A. M., & ESCH, J. (1988).
potential)
and the effects
of benzodiazepines.
Neurophysiological
Anasthesie,
Monitoring
(electroencephalogram,
evoked
Intensivtherapie, Notfallmedizin 27, 145-152. LUDWIG, A. M. (1972). Hysteria. A neurobiological theory. Archives of Genera/ Psychiatry 27, 771-777. MEARES, R., & HORVATH, T. (1972). Acute and chronic hysteria. British Journal of Psychiatry 121, 653-657.
AUGMENTATION OF AEP’s
IN
SMD
163
MELZAC, R., & WALL, P. D. (1965). Pain mechanisms; a new theory. Science 150, 971-979. MILLIGAN, K. R., HOWARD, R. C., & DUNDEE, .I. W. (1987). The effect on benzodiazepines on evoked potentials. Anuesthesiu 42, 1237-1238. MOLDOFSKY, H., &ENGLAND, R. S. (1975). Facilitation of somatosensory average-evoked potentials in hysterical anaesthesia and pain. Archives of General Psychiatry 32, 193-197. MUSHIN, J., & LEVY, R. (1974). Averaged evoked responses in patients with psychogenic pain. Psychologicol Medicine 4, 19-27. PERLEY, M. J., & GUZE, S. B. (1962). Hysteria-the stability and usefulness of clinical criteria. New England Journal of Medicine 266, 421-426. PICTON, T. W., & HILLYARD, S. A. (1974). Human auditory evoked potentials II: effects of attention. Electroencephalography and Clinical Neurophysiology 36, 191-199. PRESCOTT, J., CONNOLLY,J. F., & GRUZELIER, J. H. (1984). The augmenting/reducing phenomenon in the auditory evoked potential. Biological Psychology 19, 31-44. REIVICH, M., GUR, R., & ALAVI, A. (1983). Positron emission tomographic studies of sensory stimuli, cognitive processes and anxiety. Human Neurobiology 2, 25-33. SCHECHTER, G., & BUCHSBAUM,M. (1973). The effects of attention, stimulus intensity and individual differences on the average evoked response. Psychophysiology 10,392-400. SCHWENT, V. L., & HILLYARD, S. A. (1975). Evoked potential correlates of selective attention with multi-channel auditory inputs. Electroencephalography and Clinical Neurophysiology 38, 131-131. SHAGASS, C., & SCHWARTZ, M. (1963). Psychiatric disorder and deviant cerebral responsiveness to sensory stimulation. In J. Wortis (Ed.), Recent advances in biologicalpsychiatry, Vol. 5 (pp. 321-330). New York: Plenum Press. SILVERMAN,J., BUCHSBAUM,J., & HENKIN, R. (1969). Stimulus intensity and stimulus intensity control. Perceptive Motor Skills 28, 71-78. WHITLOCK, F. A. (1967). The aetiology of hysteria. Acta Psychiatrica Scandinavica 43, 144-162.
AUGMENTATION
IW
GU22-3956/90 $3.00 + .OO Pergamon Press plc
OF AUDITORY SOMATIZATION
LEIGH JAMES,* EVIAN GORDON,
EVOKED
POTENTIALS
IN
DISORDER
CLAUDIA
KRAIUHIN, ALAN HOWSON?
and RUSSELL MEARES Department of Psychiatry, Westmead Hospital, Westmead, NSW 2145, Australia; Severalls Hospital, Colchester, Essex, CO4 5HG and TDepartment of Economics, NSW 2113, Australia (Received
15 July 1988; revised
12 March
*Department of Psychology, Macquarie University, Ryde,
1990)
Summary-Sensory input regulation was examined in terms of augmenting/reducing of auditory evoked potentials in 10 patients with somatization disorder (8 males and 2 females) and 10 ageand sex-matched normal controls. The slope of PI-N1 amplitude change as a function of stimulus intensity was greater in patients compared with controls, suggesting an enhanced central nervous system response to sensory input. Taken together with previous findings of a failure to habituate to incoming stimuli in a similar group of patients, and evidence obtained in somatizers of both over-responding to background stimuli in a simple tone-discrimination task and enhanced parietal activation during selective attention, this finding suggests disturbances in the processes of attention and in the regulation of afferent stimuli in somatization disorder, and may help explain the multiple and chronic complaints characteristic of patients with the disorder.
INTRODUCTION PATIENTS
recurring
with somatization disorder (SMD) attend many clinics for the investigation complaints which are often minor or for which no organic cause is evident.
of In
many cases, the description of these symptoms corresponds quite closely to known disorders. Pain is a common feature of these symptoms which often result in extensive but ultimately fruitless investigations. According to DSM-III-R (AMERICAN PSYCHIATRIC ASSOCIATION, 1987) criteria, at least 13 symptoms of several years’ duration and of unknown organic pathology are required to satisfy a diagnosis of SMD. Somatization disorder is related historically to hysteria; it partially replaced the diagnosis hysteria in DSM-III (AMERICAN PSYCHIATRIC ASSOCIATION, 1980; for review, see GORDON, KRAIUHIN, KELLY, & MEARES, 1984). Unlike hysteria, a diagnosis of somatization disorder does not require the presence of conversion symptoms-those which mimic symptoms of functional loss due to neurological damage. Nevertheless, in many cases of SMD such symptoms are evident. Given the relationship between somatization disorder and hysteria, it is not unreasonable to look to aetiological theories of hysteria for hypotheses concerning the bases of the disorder. Theories of acute and chronic hysteria have been dominated by psychodynamic (primarily sexual aetiology) and sociological (primarily secondary gain) perspectives. However, JANET (1901) in seeking the underlying causes of chronic hysteria, proposed a neurophysiological model which concerned a disturbance in the awareness of or attention 155
LEIGHJAMESet al.
156
paid to sensory data impinging upon the nervous system. WHITLOCK (1967) also regarded hysteria primarily as a disorder of attention and vigilance. The theory of a subtle dysfunction in chronic hysteria with respect to the neurophysiological processing of afferent stimuli has received some experimental support (SHAGASS & SCHWARTZ, 1963; MEARES & HORVATH, 1972; MUSHIN & LEVY, 1974; MOLDOFSKY & ENGLAND, 1975). MEARES and HORVATH, for example, reported that such patients failed to habituate to a repetitive and meaningless auditory stimulus; a finding which suggests a deficiency in the ability to ignore irrelevant sensory input. However, there is still a paucity of studies in which neurophysiological measures of responses to afferent stimuli have been recorded from patients diagnosed according to explicit criteria as having somatization disorder. The hypothesis examined in this study is that somatization disorder is associated with a basic disturbance in sensory input regulation, that is, returning to JANET’S hypothesis, it is caused by a dysfunction in the processing of afferent stimuli. Sensory input regulation was examined in terms of the augmenting/reducing phenomena. To evoke these physiological phenomena, a series of stimuli of varying intensity are presented to the subject (BUCHSBAUM & SILVERMAN, 1968); in some individuals the amplitudes of evoked potentials (EPs) increase with increasing stimulus intensity (i.e. ‘augment’), whereas in others, smaller amplitudes or even a reduction in amplitude with increasing stimulus intensity occur (i.e. ‘reduced’). The resultant slopes of augmentation and reduction are presumed to reflect ‘stimulus intensity control’ (BUCHSBAUM & SILVERMAN, 1968; SILVERMAN, BUCKSBAUM, & HENKIN, 1969). Our hypothesis grew out of previous findings that excessive augmentation is associated with pain that has been called psychogenic (MUSHIN & LEW, 1974; BUCHSBAUM, 1975; MOLDOFSKY & ENGLAND, 1975). Psychogenic pain is frequently associated with depression; so also is the evoked potential pattern of augmentation (BUCHSBAUM, GOODWIN, MURPHY, & BORGE, 1971; BUCHSBAUM, LANDAU, MURPHY, 8z G~~DwN, 1973). Furthermore, it appears that, during the depressed state, the degree of augmentation correlates with the severity of depression (FRIEDMAN & MEARES, 1979b). BUCHSBAUM and his group believe that the augmenting pattern is not a direct correlate of depression but a genetic marker of predisposition to the illness (GERSHON & BUCHSBAUM, 1977). BUCHSBAUM and his colleagues, however, did not study patients in remission. It has since been shown that, on recovery, the EP amplitudes of those who suffer depression decrease significantly (FRIEDMAN & MEARES, 1979a; FRIEDMAN, MCCALLUM, & MEARES, 1980). The possibility arises that those with somatization disorder have particular patterns of sensory input regulation which are similar to depressives, but which are immutable and unrelated to temporary mood states. Given the association between the augmenting pattern and psychogenic pain, it was anticipated that somatizers would show enhanced augmentation compared with normal individuals. This hypothesis has also been advanced by LUDWIG (1972). However, to our knowledge, the hypothesis concerning augmentation of EP responses to stimuli of varying intensity has not been tested previously in somatization disorder patients diagnosed according to explicit and standardized criteria. METHOD
Subjects The patient
group consisted
of 8 females
and 2 males aged between
25 and 67 years
AUGMENTATION OFAEP’s IN SMD
157
(mean = 47 years) who had been referred by staff psychiatrists at a number of teaching hospitals. Diagnosis of somatization disorder was made using the PERLEY and GUZE (1962) checklist in which 25 symptoms with no apparent organic explanations are required. (DSMIII-R criteria are a summary of the PERLEY and GUZE checklist). An unexplained symptom was accepted as such if it satisfied the following criteria: (a) the symptoms caused the patient to take medicine (other than aspirin), alter his or her life pattern, or see a physician (b) the symptom was not adequately explained by physical disorder or injury (c) the symptom was not the side effect of medication, drugs or alcohol. According to these criteria the number of symptoms in the patient group ranged from 25 to 42 (mean = 32). Due to the polysymptomatic nature of their disorder, most patients (n = 7) were taking medication, primarily analgesics, minor tranquilizers (benzodiazepines) or both. Three patients (and all controls) were on no regular medication. The control group consisted of 10 normal subjects aged between 23 and 67 years (mean = 42 years) who were matched for sex and age (within 5 years) to the patients. Controls were recruited on a voluntary basis from visitors to and staff of a large general hospital. The number of ‘unexplained’ symptoms in this group ranged from 0 to 8 (mean = 2). Neither patients nor normals had a history of hearing or neurological disorders. No normal subject had a psychiatric history. PROCEDURE
Each subject was presented with a total of 250 auditory stimuli in the form of tones which varied in intensity (number of intensities = 5). The amplified EEG response to each tone was averaged separately for each intensity. In this way, 5 separate EPs were obtained, each representing the averaged EEG response to 50 tones of a particular intensity. The 5 intensities of tones were presented in pseudo-randomized ‘blocks’, where each block consisted of 5 tones of the same intensity. The procedure was such that every block was preceded by every other intensity block, including itself, an equal number of times. The use of pseudo-randomized blocks presumably minimizes the development of arbitrary sets that could occur during a long series of stimuli of the same intensity, and also reduces the possible effect on the amplitude of response of a strong stimulus on a weak one, and vice versa. It is also a way of coping with problems of habituation, fatigue, and attentional drift that could occur during the course of a recording session. Following a description of the experimental procedure the informed consent of each subject was obtained. Stimuli The auditory stimuli were produced by a software controlled sine-wave generator incorporated in the EP recording computer, and were presented with an inter-stimulus interval of 1 set through stereo headphones. Each tone had a frequency of 1000 Hz and a duration of 16 msec. The levels of stimulus intensity used in this study were 45, 65, 85, 95 and 102 dB above threshold. (The maximum intensity was selected in accordance with the finding of PRESCOTT, CONNOLLY, & GNUZELIER, 1984, that the classification of
LEIGHJAMESet al.
158
augmenting be reliable).
and reducing
patterns
using maximum
intensities
lower that 100 dB may not
Event-related potential recording An Ag/AgCl electrode was applied at Cz (of the lo-20 system; JASPER, 1958) which was referenced to linked earlobes and grounded with an electrode on the forehead. Additional electrodes were placed above and below the right eye and referenced to each other to record eye-related potentials which were averaged in a separate channel. Electrode resistances were below 5 kfi. The signals were amplified to a total gain of 50,000 and filtered with a bandpass of .05-100 Hz. They were sampled by an A-to-D converter at 868 Hz for 600 msec (100 msec prestimulus and 500 msec poststimulus). Trials in which 250 points of the potential exceeded +75% of the input voltage range of the analog-to-digital converter (f 5 V) were automatically rejected. The averaged waveforms were then additionally filtered with a nonrecursive digital low-pass filter at 40 Hz. Data analysis Subjects’ EEG records were averaged separately for each stimulus intensity. Individual peaks were scored by a computer cursor programme. Amplitude was measured as the peakto-peak amplitude of the Pl and Nl components. Mean (*s.d.) values for Pl and Nl latency across both groups were 41( f 15) msec and 89 (f 17) msec, respectively. An amplitude-intensity slope value was obtained for the PI-N1 component recorded in each subject. This was calculated by the least squares technique, which fitted a straight line (y = mx + c) to event-related potential amplitude measurements at the five intensities. In these computations, arbitrary values of 1, 2, 3, 4 and 5 were assigned to the auditory stimulus intensities (5 = loudest). This provided a ‘slope’ which reflected change in amplitude as a function of intensity. A high slope is indicative of a relatively great increase in EP amplitude for an increase in stimulus intensity, and a low or negative slope is indicative of a small increase or decrease in EP amplitude for increasing intensity. SCHECHTER and BUCHSBAUM(1973) designated augmenting subjects by positive slopes and reducing subjects by zero or negative slopes. The average slope of the two groups was compared using Student’s t-test. RESULTS Grand average EP waveforms for the normal and patient groups (averaged over 50 responses to each stimulus) are presented in Fig. 1. Mean values of PI-N1 amplitude at each intensity for each group are plotted in Fig. 2 (mean and s.d. values are presented in Table 1). These data show that PI-N1 amplitude was enhanced (most markedly at the higher stimulus intensities) for the somatizers compared to the normals. For individual subjects, the slope values were significant at the .05 level or less for 4 of the normals and 6 of the patients. For the normals as a group, the equation representing the least squares fit was as follows: Pl-Nl amplitude = 1.88 x intensity + 5.91 (standard error = 4.92). The equation for the somatizers was as follows: Pl-Nl amplitude = 2.75 x intensity + 5.41 (standard error = 4.9). The difference between the slopes was significant (t,,,, (19) = 2.13).
AUGMENTATION OF AEP’s
159
IN SMD
il
i2
i3
..... i4
i5
+ I
SO
I
200
(in ms)
Latency
-
I
100
NORMALS SMD i -stimulus
FIG. 1. Group
intensity
average EP waveforms; averaged over 50 responses to each stimulus intensity. augmentation of responses in somatizers, especially at the higher intensities.
TABLE
1.
MEANS
(S.D.)
OF
Pl-Nl
AMPLITUDE (in pV) AT EACH STIMULUSINTENSITY FOR NORMALSAND PATIENTS Stimulus
Normals
(n = 10)
SMD (n = 10)
Mean s.d. Mean s.d.
Note enhanced
Intensity 1
2
3
4
5
8.4 2.8 8.7 2.9
8.9 3.9 11.1 3.2
11.6 4.0 12.8 3.7
13.2 6.0 15.5 5.3
15.6 7.0 20.2 7.9
It is worth noting, however, that the significance of the results appears to have been contributed to by a very large Pl-Nl response at the higher intensities in one of the patients (i.e. 24.3 PV and 41.3 PV at stimulus intensities 4 and 5, respectively). The data from this subject were included in the analysis due to the unquestionable validity of the subject’s recordings. (The second largest responses from an SMD patient were 22.2 PV and 21.66 PV at intensities 4 and 5, respectively.) When this subject’s data were removed from the analysis the between-groups difference was not significant. Nevertheless, even without this subject’s data the somatizers still had larger responses, indicating a tendency for the somatizers to over-respond to the stimuli compared to the normals. DISCUSSION
Our finding that augmentation of auditory evoked potentials tends to be greater in patients with somatization disorder compared to normals suggests a disturbance in sensory regulation
LEIGH JAMESef al.
160 22r
20-
M
NORMALS
O-cl
SMD
16-
16-
8-
Intensity
of stimulus
FIG. 2. Means of Pl-Nl amplitudefor each stimulusintensityand for each group (n = 10 per group). based on recordings from the vertex (Cz).
Figures
in these individuals and requires further investigation. Specifically, the pattern of results indicates that somatizers over-respond to auditory stimuli of higher intensity. These results are consistent with the prediction of LUDWIG (1972) that the amplitudes of the P l-N1 and Nl-Pl waves in the averaged evoked responses of patients with hysteria would be minimally greater over moderate intensities with an exaggerated response at the highest stimulus intensities. The effect obtained in our study was observed in recordings from the vertex. In the light of findings, by CONNOLLY and GRUZELIER (1982) for example, the augmenting-reducing may vary at different scalp sites in the same individual to the same stimuli. Future research will explore our finding topographically. Although most patients in our study were not drug-free it seems unlikely that medication accounts for the experimental effect since other research (e.g. MILLIGAN, HOWARD, & DUNDEE, 1987; KOCHS, SCHULTE, & ESCH, 1988) have shown that benzodiazepines suppress the middle to late components of the evoked potential. Studies of sensory evoked potentials
AUGMENTATION OF AEP’s IN SMD
161
in patients with psychogenic pain (MUSHIN dz LEVY, 1972) and hysterical anaesthesia (MOLDOFSKY & ENGLAND, 1975) have revealed excessive augmentation in EPs of patients compared to controls. In the visual modality, VON KNORRING, ALMAY, JOHANSSON and TERENIUS, (1979) have demonstrated an augmenting response in the majority of patients with chronic pain. Our finding of increased augmentation in EPs of patients with somatization disorder further substantiates, in the auditory modality, the theory of neurophysiological disturbance in patients with psychogenic pain. Referring to the results of other research relating changes in average EPs to pain, BUCHSBAUM (1975) found that a reducing pattern of evoked responses is associated with a high pain tolerance. It might therefore be expected that a low tolerance of pain would be found in those whom augmentation is exaggerated. Our finding that patients with multiple symptoms, most of which involve pain, show an exaggerated augmenting response to auditory EPs is consistent with this hypothesis. Also consistent with this hypothesis is the finding of VON KNORRING et al. (1979) that the majority of patients with chronic pain have an augmenting pattern of visual evoked responses. (Their augmenting patients also had significantly lower CSF endorphin levels than those with reducing response.) In a previous publication (GORDON, KRAIUHIN, KELLY, & MEARES, 1986a) we reported an abnormality in the amplitude of the Nl component of auditory EPs to task-irrelevant (background) stimuli in somatizers. It is well known that the amplitude of the Nl components varies with the amount of attention directed toward stimuli (HILLYARD, HINK, SCHWENT, & PICTON, 1973; PICTON & HILLYARD, 1974; SCHWENT & HILLYARD, 1975; HINK VAN HOORIS, HILLYARD, & SMITH, 1977; HINK, FENTON, PFEFFERBAUM, TINKLENBERG, & KOPELL, 1978). The finding of an abnormal increase in Nl amplitude in somatizers, together with the subsequent finding that this abnormality is unlikely to be due to abnormally high levels of arousal engendered by anxiety, (GORDON, KRAIUHIN, MEARES, & HOWSON, 1986b) was interpreted as suggesting that such patients pay an undue amount of attention to stimulus input that others are better able to ignore. The current finding of increased PlNl amplitude in somatizers provides additional support for our hypothesis that somatizers over-respond to afferent input. Further support for this hypothesis comes from studies of other physiological indices of CNS activation. JAMES, SINGER, ZURYNSKI, GORDON, KRA~JHIN, and HARRIS, (1987) found evidence of hyperactivity in the right parietal lobe in a study of regional cerebral blood flow in somatizers. This region of the brain is implicated in the cortical network for the control of attention to external stimuli (e.g. GUR, GUR, & ROSEN, 1983; REIVICH, GUR, & ALAVI, 1983). In conclusion, the pattern of results from our laboratory reveals disturbances in the processes of attention and in the regulation of sensory input in SMD. These disturbances may be associated with the unusual incidence of recurrent pains and discomfort in patients with the disorder. However, given that other psychiatric disorders (e.g. depression) are associated with augmentation of evoked potentials, the specificity of our interpretations to somatization disorder needs to be investigated. It may be, for example, that an abnormally large degree of EP augmentation is associated with the psychogenic pains and the depressive symptoms experienced by patients with SMD, but is not the determinant of the entire syndrome of symptoms which represents somatization disorder.
162
LEIGH JAMES et al.
Acknowledgements-This research was supported by grant no. 860242 from the Australian National Health and Medical Research Council. We wish to thank Chris Rennie, BSc., M. Biomed. Eng., of the department of Medical Physics at Westmead Hospital, for his valuable contribution, in terms of soft-ware programming, to this study.
REFERENCES AMERICAN PSYCHIATRIC ASSOCLUION (1980). Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: Author. AMERICAN PSYCHIATRIC ASSOCIATION (1987). Diagnostic and Statistical Manual of Mental Disorders (3rd ed.). Washington, DC: Author. BUCHSBAUM, M. (1975). Average evoked response augmenting/reducing in schizophrenia and affective disorders. In D. X. Freedman (Ed.), Biology of the major psychoses: a comparative analysis. New York: Raven Press. BUCHSBAUM, M., GOODWIN, F. K., MURPHY, D. L., & BORGE, D. F. (1971). AER in-affective
Journal of Psychiatry
disorders.
American
128, 19-25.
BUCHSBAUM, M., LANDAU, S., MURPHY, D., &GOODWIN, F. K. (1973). Average unipolar affective disorders: relationship to sex, age of onset, and monoamine
evoked response in bipolar and oxidase. Biological Psychiatry
7, 199-212. BUCHSBAUM, M.,
Psychosomatic
& SILVERMAN, J. (1968).
Medicine
Stimulus
intensity
control
and
the cortical
evoked
response.
30, 600-611.
CONNOLLY, J. F., & GRUZELIER, J. H. (1982). Amplitute and latency changes in the visual evoked potential to different stimulus intensities. Psychophysiology 19, 599-608. FRIEDMAN, J., MCCALLUM, P., & MEARES, R. (1980). Stimulus intensity control in depression: a study of the comparative effect of doxepin and amitryptaline on cortical evoked potentials. Australian and New Zealand Journal of Psychiatry 14, 115-l 19. FRIEDMAN, J., & MEARES, R. (1979a). The effect of placebo and tricyclic antidepressants on cortical evoked potentials in depressed patients. Biological Psychology 8, 291-302. FRIEDMAN, J., & MEARES, R. (1979b). Cortical evoked potentials and severity of depression. American Journal
of Psychiatry
136, 1218-1220.
GERSHON, E. S., & BUCHSBAUM, M. (1977). A genetic study of averaged response augmentation/reduction in affective disorders. In Shagass, C., Gershon, S. & Friedhoff, A. (Eds.), Psychopathology and brain dysfunction. New York: Raven Press. GORDON, E., KRAIUHIN, C., KELLY, P., & MEARES, R. (1984). The development of hysteria as a psychiatric concept.
Comparative
Psychiatry
25, 532-537.
GORDON, E., KRAIUHIN C., KELLY, P., MEARES, R., & HOWSON, A (1986a). somatization disorder. Comparative Psychiatry 27, 295-301. GORDON, E., KRAIUHIN, C., MEARES, R., & HOWSON, A. (1986b). Auditory
somatization
disorder.
Journal of Psychiatric
Research
A neurophysiological evoked
response
study of potentials
in
20, 237-248.
GUR, R. C., GUR, R. E., & ROSEN, R. D. (1983). A cognitive-motor network demonstrated by positron emission tomography. Neurophsychofogia 21, 601-606. HILLYARD, S. A., HINK, R. F., SCHWENT, V. L., & PICTON, T. W. (1973). Electrical signs of selective attention in the human brain. Science 91, 177-180. HINK, R. F., FENTON, W. H., PFEFFERBAIJM,A., TINKLENBURG, J. R., & KOPELL, B. S. (1978). The distribution of attention across auditory input channels: an assessment using the human evoked potential. Psychophysiology
15, 466-473. HINK, R. G., VAN HOORIS, S. T., HILLYARD, S. A., &SMITH, T. (1977). The division of attention and the human auditory evoked potential. Neuropsychologia 15, 597-605. JAMES, L., SINGER, A., ZURYNSKI, Y., GORDON, E., KRAIUHIN, C., HARRIS, A., HOWSON, A., & MEARES, R. (1987). Evoked response potentials and regional cerebral blood flow in somatization disorder. Psychotherapy
and Psychosomatics 47, 190-196. The mental state of hystericals
JANET, P. (1901). JASPER, H. (1958).
(Translated by CORSON, C. R.). New York: Putnam Press. The ten-twenty electrode system of the International Federation. Electroencephalography and Clinical Neurophysiology 10, 371-375. VON KNORRING, L., ALMAY, B. G. L., JOHANSSON, F., & TERENIUS, L. (1979). Endorphins in CSF of chronic pain patients, in relation to augmenting-reducing response in visual averaged evoked response.
Neurophyschobiology
5, 322-326.
KOCHS, E., SCHULTE, A. M., & ESCH, J. (1988).
potential)
and the effects
of benzodiazepines.
Neurophysiological
Anasthesie,
Monitoring
(electroencephalogram,
evoked
Intensivtherapie, Notfallmedizin 27, 145-152. LUDWIG, A. M. (1972). Hysteria. A neurobiological theory. Archives of Genera/ Psychiatry 27, 771-777. MEARES, R., & HORVATH, T. (1972). Acute and chronic hysteria. British Journal of Psychiatry 121, 653-657.
AUGMENTATION OF AEP’s
IN
SMD
163
MELZAC, R., & WALL, P. D. (1965). Pain mechanisms; a new theory. Science 150, 971-979. MILLIGAN, K. R., HOWARD, R. C., & DUNDEE, .I. W. (1987). The effect on benzodiazepines on evoked potentials. Anuesthesiu 42, 1237-1238. MOLDOFSKY, H., &ENGLAND, R. S. (1975). Facilitation of somatosensory average-evoked potentials in hysterical anaesthesia and pain. Archives of General Psychiatry 32, 193-197. MUSHIN, J., & LEVY, R. (1974). Averaged evoked responses in patients with psychogenic pain. Psychologicol Medicine 4, 19-27. PERLEY, M. J., & GUZE, S. B. (1962). Hysteria-the stability and usefulness of clinical criteria. New England Journal of Medicine 266, 421-426. PICTON, T. W., & HILLYARD, S. A. (1974). Human auditory evoked potentials II: effects of attention. Electroencephalography and Clinical Neurophysiology 36, 191-199. PRESCOTT, J., CONNOLLY,J. F., & GRUZELIER, J. H. (1984). The augmenting/reducing phenomenon in the auditory evoked potential. Biological Psychology 19, 31-44. REIVICH, M., GUR, R., & ALAVI, A. (1983). Positron emission tomographic studies of sensory stimuli, cognitive processes and anxiety. Human Neurobiology 2, 25-33. SCHECHTER, G., & BUCHSBAUM,M. (1973). The effects of attention, stimulus intensity and individual differences on the average evoked response. Psychophysiology 10,392-400. SCHWENT, V. L., & HILLYARD, S. A. (1975). Evoked potential correlates of selective attention with multi-channel auditory inputs. Electroencephalography and Clinical Neurophysiology 38, 131-131. SHAGASS, C., & SCHWARTZ, M. (1963). Psychiatric disorder and deviant cerebral responsiveness to sensory stimulation. In J. Wortis (Ed.), Recent advances in biologicalpsychiatry, Vol. 5 (pp. 321-330). New York: Plenum Press. SILVERMAN,J., BUCHSBAUM,J., & HENKIN, R. (1969). Stimulus intensity and stimulus intensity control. Perceptive Motor Skills 28, 71-78. WHITLOCK, F. A. (1967). The aetiology of hysteria. Acta Psychiatrica Scandinavica 43, 144-162.
