http://exhibits.library.stonybrook.edu/mfp/files/original/0a04fd8af8c99e319cfc9e14d05f3ffd.pdf d38bf0d44fbbfe5439d4cca003bd77b2 PDF Text Text CHANGES IN VERBAL TRANSACTIONS WITH INDUCED ALTERED BRAIN FUNCTION JOSEPH JAFFE, M.D., MAX FINK, MD. Reprinted from Tm: JOURNAL or AND ROBERT L. KAHN, PHJ). NERVOUS AND MENTAL DISEASE 130, No. 3, March 1960 Volume Printed in U.S.A. �Reprinted from THE JOURNAL OF NERVOUS AND MENTAL DISEASE Volume 130, No. 3, March 1960 Printed in U.S.A. CHANGES IN VERBAL TRANSACTIONS WITH INDUCED ALTERED BRAIN FUNCTION JOSEPH JAFFE, M.D.,1 MAX FINK, MD. Repeated interviews with patients under— going convulsive therapy reveal progressive changes in the interpersonal relationship, which are referable to verbal and non-verbal transactions. While non-verbal aspects of communication are difficult to quantify, techniques are available for the measure— ment of verbal behavior. Using such lin— guistic methods, we have observed systematic alterations in language patterns during convulsive therapy, which were related to independent evaluations of behavioral change and to improvement. The description of these language patterns has provided a useful quantitative method for understand— ing interpersonal changes which occur during therapy. In a syntactic—content analysis of recorded interviews during convulsive therapy (6), such changes as denial (negation), qualification (subjunctive and .adverbial modifiers), displacement (person and tense), and cryptic and clichéd remarks were scored. An increased incidence of these changes in the patients’ speech was related both to the degree of induced altered brain function and to the evaluation of therapeutic response. It has been clinically observed that when the language of the patient is affected by neurologic dysfunction, modification of the interviewer’s speech patterns occur. In the syntactic-content analyses, in which a structured interview was used, the examiner’s participation was restricted to statements in the questionnaire. This two-person inter— 1Department of Experimental Psychiatry, Hillside Hospital, Glen Oaks, Long Island, New York. This study was aided by grants 56-151, Foundations’ Fund for Research in Psychiatry, and M-927 of the National Institute of Mental Health, U. S. Public Health Service. The technical assistance of Mrs. Jean Kolodny and Mrs. Ann Horowitz is gratefully acknowledged. 235 AND ROBERT L. KAHN, PHD. view group (or dyad) therefore assumed special characteristics. Interactive effects were minimized. The constraint of the questionnaire interrupted the reciprocal modification of the examiner’s speech. In such a dyad, scoring of the patient’s responses alone constituted an adequate description of changes in the verbal transactions of the two-person communication system. Judgements as to mental status are usually arrived at in less structured conversa— tions between doctor and patient. It was suggested that measurable changes in language patterns would occur even in con~ ventional clinical interviews, and furthermore that such changes would be related to those observed in structured interviews. The methods of dyadic analysis were developed (3—5) to permit a quantitative description of verbal transactions in unstructured interviews (diagnostic, psycho— therapeutic). In this analysis, the speech of doctor and patient is considered as one continuous behavior, and the measurements are performed on consecutive units, irrespective of speaker. The verbal flow of the interview is thus considered as a single out— put, irrespective of speaker. The justification for this maneuver rests ultimately on the correlation of dyadic speech patterns with those obtained by other methods, and is an aim of the present study. These scoring measures of unstructured interviews have been applied to weekly interviews with patients undergoing convul— sive therapy. It is the purpose of this report to 1) determine the pattern of change in dyadic language measures with convulsive therapy; 2) study the relation of these measures to the degree of induced altered cerebral function; and 3) relate the dyadic scores to syntactic language measures ob- �236 JAFFE, FINK AND KAHN tained concurrently in structured inter- repetitive the interaction. For these studies, both the mean and standard deviation of views. the TTR distribution of each interview sam— SUBJECTS AND METHOD ple was obtained. For the syntactic speech analyses the paTwenty-seven consecutive referrals for convulsive therapy in a voluntary mental tient was seen by another examiner during hospital were studied. On a random basis, the same intervals. This interview consisted ten patients were assigned to a control group of a standardized questionnaire composed and the remaining seventeen constituted the of specific items concerning the major comexperimental group. Both groups were com— plaint, reasons for coming to the hospital, parable for age and education; the convul- and temporal and spatial orientation (9). sive group had a mean age of 47 and 11.7 The verbatim responses were analyzed for mean years of education, while the mean the presence of syntactic language changes age of the subconvulsive group was 47.8, previously described as occurring with inwith a mean of 10.5 years of education. The duced cerebral dysfunction (6). Prior to, and at weekly intervals during investigators had no part in the treatment process, and did not know which cases treatment, an electroencephalogram was reserved as controls until data collection was corded in each patient. These records were measured for the per cent time of induced completed. The experimental (convulsive) group was slow wave activity (1). In the dyadic TTR analyses, experimengiven grand mal electro-convulsive therapy three times weekly, under pentothal pre- tal (convulsive) and control (subconvulmedication for a minimum of 12 treatments. sive) groups were compared. In relating The control subjects were treated in identi— dyadic TTR changes to induced slow wave cal fashion except that they received sub- activity and to syntactic language analyses, convulsive electrostimulation while under the mode of treatment was disregarded, all pentothal premedication. patients being considered as a single group. All patients were interviewed prior to RESULTS treatment, and in the week of the 12th 1. Dyadic TTR: A consistent change was treatment. An unstructured clinical diagnostic interview technique was used centered observed in the TTR patterns of the experiabout the patient’s symptoms and life prob- mental group. Scores for consecutive 25 lems. The patient was encouraged to talk word units of interaction in 500 word samfreely, with occasional guiding interventions ples were plotted before and during the by the interviewer. Long silences resulted treatment course (end of the fourth week). in increased interviewer activity. All inter- Figure 1 shows a graph of the TTR patviews were tape—recorded. terns for one patient. Next to each graph is For the formal dyadic analyses (3, 5), the frequency distribution of the 20 consecu— the first 500 words of each interview were tive scores shown. The change in the distranscribed in temporal sequence without tribution for this case was a decrease in the regard to the speaker. This sample of dyadic mean and an increase in the standard deviaspeech was divided into consecutive 25 word tion of the dyadic TTR. This pattern of units. The type-token ratio (TTR) was cal- change was characteristic of the expericulated for each unit. The type—token ratio mental group. is the number of different words (types) Table 1 shows changes in the group mean divided by the total number of words (to- T TR score during treatment. Although there kens). Thus, the lower the TTR the more was a decrease in both groups, the change �237 LANGUAGE CHANGES WITH BRAIN DYSFUNCTION was significant (p < .01) only in the convulsive group and not in the control (sub— convulsive) group. Table 2 shows the changes in standard deviation of the group TTR scores during treatment. There was a significant increase in standard deviation (p < .01) in the convulsive group. In the control (subcon— vulsivc) group the standard deviation was decreased during treatment. The change, however, did not reach statistical signifi— cance. 2. Relation of Dyadic TTR to EEG changes: The changes in dyadic TTR scores were related to changes in brain function as reflected in measurements of the amount of slow wave activity on the electroencephalogram. For this purpose, the per cent time delta activity in the EEG record obtained in the same week as the interview was used. While almost all members of the experimental group developed prominent amounts of EEG delta activity during treatment, none of the control group demonstrated such changes. Using the method of rank order correlation, the change in standard, deviation of the dyadic TTR correlated +65 with the per cent time of delta activity TABLE 1 Change in Mean TTR with Electroshock Subconvulsive Convulsive .92 j .xv-rp/w. p. l— .76 . /. - - with Electroshoclc Subconvulsive Convulsive ,5 3 83 ° .5660 ,0 .92 El e4 ...76 .68 .60 . .68 D—URING TREATMENT / f .76 T” text). .92 Lo 55 5“ 83 m Ea /\T I .56/6068 .76 TTR FIG. 1. .84 fl—I 6 0 10 17 During Difference treatment treatment 8.6 8.1 76 9.2 —0.5 +1.6* *Significant at 0.01 level, using Wilcoxon’s method of paired replicates. TABLE 3 Relation of Syntactic and Dyadic TTR Language ZWeasnres Dyadic Analysis: Syntactic Analysis N Change in Mean Fewer than two changes Two or more changes Change in Standard Dev1at10n 15 —0.8 —0.1 12 —3.0** +1.4* < .01). The greater increase in variability in the language measure was thus associated With the greater degrees of altered brain function. Changes in the mean T TR, however, were not significantly related to the changes in brain function (r = +19). 3. Relation of Dyadic TTR to Syntactic Language Changes: A comparison of dyadic TTR scores with syntactic aspects of the patient’s speech obtained in independent structured interviews was made. The patients were divided into those who showed two or more syntactic language changes, and those who showed fewer than two such changes, regardless of the type of treatment (Table 3). For the patients showing (19 0 N Group Pre— Significant at 0.02 level. ** Significant at 0.01 level. 0’) . —0.8 —2.4* TABLE 2 Change in Standard Deviation of TTR 25 WORD UNITS) E2 80.3 79.2 * ————“‘7 /' 17 81.1 81.6 Significant at 0.02 level, using Wilcoxon’s method of paired replicates. PRE-TREATMENT 1.0 10 treaItirient tiggffiiltegnt Difference * DYADIC TTR PATTERN (CONSECUTIVE N Group .84 .92 LIO Plot of TTR patterns for one patient (see �238 JAFFE, FINK AND KAHN two or more syntactic changes, both dyadic indices showed a significant change during treatment. There were no significant altera— tions in TTR indices for the group showing fewer than two syntactic changes. DISCUSSION These observations indicate a significant difference in dyadic transactions in the experimental (convulsive) and control (subconvulsive) groups—a difference which is referable to a consistent change in the subjects receiving convulsive therapy. The find: ings are consistent with those reported by Weinstein and Kahn (9) in their studies of patients with altered brain function. They observed increased use of the second and third person, non-aphasic misnaming, cli— chés, stereotyped expressions, condensations and neologisms. These language patterns were termed the “language of denial” and were regarded as symbolic adaptations. Kahn and Fink (6) noted similar language changes in patients with brain function al— tered by convulsive therapy. The present observations indicate that verbal transactions during altered brain function are not only more stereotyped qualitatively, as in the use of clichés, but are also more stereotyped quantitatively as in the increased repetition of words. Thus, analyses of the more formal aspects of speech parallel analyses of content. Alteration in brain function was also related to changes in the dyadic indices. The low, but significant correlation between the dyadic scores and EEG delta activity suggests that the two-person communication system as a whole may reflect neurophysio— logical alteration in one of its participants. The low correlation is consistent with previous observations that the dyadic TTR pattern is sensitive to factors other than al— tered brain function (5). Other studies of the dyadic TTR and syntactic language measures during drug administration (2) are consistent with the present findings. Administration of agents which produce EEG hypersynchrony similar to that of convulsive therapy was associated with changes in both language measures in the direction of increased stereotypy and repetitiveness. Agents which produced EEG desynchronization, however, were associated with decreased repetitiveness and a decreased number of syntactic alterations. These observations, though limited to acute drug interviews, indicate that similar changes in language patterns can be antici— pated in subjects following the chronic administration of psychotropic compounds. We anticipate that alteration in patterns of lan— guage may provide cues for the evaluation of behavioral change in drug therapies as well as in convulsive therapy. It is of significance that changes in dyadic speech are measurable when the neurophysiological status of only one of the participants is altered. This observation is consistent with concepts of verbal behavior as a two-person phenomenon, inseparable from its interpersonal context (7, 8, 10). The method and findings also demonstrate that neurophysiologic effects can be investigated in unstructured interviews and that the results may be related directly to those obtained under more structured experimental conditions. Thus, the measurement of formal aspects of language in clinical interviews may be viewed as another tool of neurophysiologic investigation. CONCLUSION Formal language measures in unstruc— tured clinical interviews were undertaken in the course of a study of convulsive and subconvulsive therapies in a hospitalized psychiatric population. Dyadic TTR (Type—Token-Ratio) measures showed a significant decrease in the mean and an increase in the standard deviation in the subjects receiving convulsive therapy, but no differences in those receiv— ing subconvulsive therapy. The degree of change in dyadic indices was related both to the degree of induced delta activity on �LANGUAGE CHANGES WITH BRAIN DYSFUNCTION the electroencephalogram, and to changes in syntactic language patterns obtained in independent structured interviews. Theoretic implications for the understanding of language changes during altered brain function were discussed. 1. REFERENCES FINK, M. AND KAHN, R. L. Relation of EEG delta activity to behavioral response in electroshock. AMA. Arch. Neurol. & Psychiat., 78: 516—525, 1957. J. AND KAI-IN, R. L. Drug induced changes in interview patterns: Linguistic and neurophysiologic indices. In The Dynamics of Psychiatric Drug Therapy, SarwerFoner, G., ed. C. C Thomas, Springfield, Ill. 2. FINK, M., JAFFE, 3. In press. JAFFE, J. An objective study of communication in psychiatric interviews. J. Hillside Hosp., 6: 207—215,1957. 239 J. Dyadic analysis of two psychoanalytic interviews. Presented in Symposium on Psycholinguistic Analysis of the Psychiatric Interview, Divisional Meeting of A.P.A., New York City, November 28, 1959. JAFFE, J. Language of the dyad. Psychiatry, 21: 4. JAFFE, . 249—258, 1958. . . KAHN, R. L. AND FINK, M. Changes in language during electroshock. In Psychopathology of Communication, Hoch, P. and Zubin, J ., eds., Grune & Stratton, New York, 1958. SAPIR, E. Language: An Introduction to the Study of Speech. Harcourt-Brace, New York, 1949. .SKINNER, B. F. Verbal Behavior. Appleton— Century-Crofts, New York, 1957. . WEINSTEIN, E. A. AND KAHN, R. L. Denial of 10. Illness: Symbolic and Physiological Aspects. C. C Thomas, Springfield, Ill., 1955. ZIPF, G. K. Human Behavior and the Principle of Least Efiort. Addison-Wesley, Cambridge, Mass, 1949. � Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Published Works Text A resource consisting primarily of words for reading. Examples include books, letters, dissertations, poems, newspapers, articles, archives of mailing lists. Note that facsimiles or images of texts are still of the genre Text. Dublin Core The Dublin Core metadata element set is common to all Omeka records, including items, files, and collections. For more information see, http://dublincore.org/documents/dces/. Title A name given to the resource Changes in verbal transactions with induced altered brain function. J Nerv Ment Dis. 1960 Mar; 130:235-9. Type The nature or genre of the resource Text Identifier An unambiguous reference to the resource within a given context mfp-02-01-003-44-010 Date A point or period of time associated with an event in the lifecycle of the resource 1960 Creator An entity primarily responsible for making the resource Jaffe, Joseph; <a title="Fink, Max, 1923-" href="http://id.loc.gov/authorities/names/n79039548" target="_blank">Fink, Max, 1923-</a>; Kahn, Robert L. Subject The topic of the resource Published Works -- Articles and Reviews Relation A related resource The Max Fink Collection Description An account of the resource Reprinted from The JOURNAL of NERVOUS AND MENTAL DISEASE Volume 130, No. 3, March 1960 Rights Information about rights held in and over the resource <a title="IN COPYRIGHT - EDUCATIONAL USE PERMITTED" href="http://rightsstatements.org/vocab/InC-EDU/1.0/" target="_blank">IN COPYRIGHT - EDUCATIONAL USE PERMITTED</a> Source A related resource from which the described resource is derived Special Collections and University Archives, University Libraries. Stony Brook University Libraries (State University of New York). Language A language of the resource en-US Format The file format, physical medium, or dimensions of the resource application/pdf Publisher An entity responsible for making the resource available Contributor An entity responsible for making contributions to the resource Published