Journal of Psycholinguistic Research, Vol. 4, No. 3, 1975
P e r s o n a l i t y Organization and Language Behavior: The Imprint of P s y c h o l o g i c a l D i f f e r e n t i a t i o n on Language Behavior in Varying C o m m u n i c a t i o n Conditions Irving Steingart, 1 Norbert Freedman, 1 Stanley Grand, l and Charles Buchwald 1
Received July 25, 1974 The language behavior of field-independent (F-l} and field-dependent (F-D) clinically normal, verbally resourceful female college students was examined in three different communication conditions: Dialogue, Warm (visually supportive) monologue, and CoM (visually nonsupportive and stressful} Monologue. F-I and F-D Ss produced similar amounts of the different types of language behavior evaluated in each of the three communicative conditions. However, they differed with respect to verbal output and length of sentence "packaging" unit in Monologue conditions. F-D Ss talked considerably less but at the same time produced different types o f grammatically more elaborate language behavior in Warm and Cold Monologue compared to their Dialogue language behavior. F-I Ss talked considerably more but also showed a type of language autonomy. The pattern of language behavior which characterized F-I speech in Dialogue remained the same in both Monologue conditions.
INTRODUCTION Leaving aside the important but sometimes quite ethereal preoccupation of contemporary psycholinguistic literature with the question of the psychological reality of generative-transformational rules, there is at least some solid This research was supported in part by Grants MH-14383, MH-1983, and MH-5090 awarded by the National Institute of Mental Health, United States Public Health Service. 1State University of New York, Downstate Medical Center, Brooklyn~ New York. 241 Q 1 9 7 5 Plenum Publishing Corporation, 227 West 17th Street, New York, N.Y. 10011. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permission of the publisher.
242
Steingart, Freedman, Grand, and Buchwald
observational footing about how language behavior actually does change with age. Developmental studies of changes in language behavior demonstrate an increase in words per sentence but also add important qualifications about such growth (Hunt, 1970; Loban, 1963; O'Donnell et al., 1967). Sentence elaboration via coordination devices, which can be referred roughly to deep structures effected by Chomsky's "conjoining transformations" (Lyons, 1968) and which shows in surface structure as compound sentences, is something that "seems to be learned relatively early" (Hunt, 1970, p. 35). Sixth graders can accomplish such sentence elaboration via coordination (together with deletion of redundant material) with a skill equal to that of a superior adult. Another finding (Hunt, 1970) is that even superior adults do not produce more full dependent clause formations per main clause than do eighth graders, which can be referred roughly to deep structures affected by Chomsky's "embedding transformations" and shows in surface structure as complex sentences (Lyons, 1968). What does distinguish the superior adult's grammatical skill-and what really accounts for significant increases in the amount of information "packaged" into longer surface sentences-is the skillful way in which the superior adult can delete verbal material into a variety of less than full clause formations (Hunt, 1970). Such clause reduction activity, according to a generative-transformational grammar, results from the application of deletion rules (Lyons, 1968) upon kernel strings. Recent developmental research by Hunt (1970) indicates that the ability to use such deletion rules especially to produce less than clause formation embeddings is a particularly important yardstick of syntactic maturity. Hunt's work (1970) and especially work by O'Donnell et al. (1967), which used a more spontaneous type of speech sample such as we employ, indicate that a simple and satisfactory measure exists for evaluating the extent of deletional, clausereducing activity which has gone into the production of the surface structure sentence-and this is the average number of words per sentence (AWS) which characterizes the speaker's language behavior. Along with these increases in syntactic skill per se, another distinction between adult language behavior and that of the young child has to do with the functional uses made of language skills for the structuring of experience. The development of language behavior shows distinct changes in this regard: At first, language functions o n l y to describe rather isolated, perceptually impelling, concrete frames of experience (Piaget, 1960). This changes into more complexly patterned but still descriptive portrayals of experiences. And, finally, language can function so as to structure some type of explanatory, deductive, purposive (i.e., conditional or contingency) matrix which is applied to immediate experience. As Hunt puts it, with development the language user
Personality Organization and Language Behavior
243
"brings to the whole discourse his mature knowledge of the socio-physical world. It is that knowledge which tells him of the purposive relations, and his syntax allows him to express it" (Hunt, 1970, p. 43). For Piaget (1960), this "mature knowledge of the socio-physical world" is the objectification of thought conceptualized as a change from preoperational to operational thought organization. What is important to stress is that this change to operational thought organization has a significant consequence for the organization of not only physical reality but social reality as well (Feffer, 1970). This objectification of social experience means, first, that now the child differentiates his own, immediate subjective experience as subject from the experience of the other as object, and, second, that language communication now is structured in accordance with a simultaneous coordination of subject and object viewpoints (whether the other as object be someone who possesses small or great emotional significance). The above considerations about the development of language behavior are incorporated in a system for the analysis of language behavior developed and employed in this laboratory for the past several years. Three categories of grammatically coherent, surface sentence structures are identified according to the kinds of structural information about experience made possible by an individual's choice of language behavior: Narrative language behavior consists of surface, simple sentence structures, i.e., single independent clauses, certain types of appositive formations, or a more elaborate "chaining" of independent clauses which can be accomplished by coordination devices (and these constructions can involve deletions of identical subject terms, verbs, etc.). Such language behavior, no matter how elaborate, at best only describes connections between fairly statically conceived experiences. Complex Portrayal language behavior consists of surface complex sentences which uses grammatical devices to describe some patterned interrelation among experiences (and can of course also show a good deal of deletional activity). Conditional language behavior consists of surface, complex sentence structures which also articulate a patterned interrelation among experiences but now within a purposive, causal, deductive, or explanatory framework via the selection of appropriate grammatical devices. There is another kind of language behavior through which the expression of such a conditional framework can be accomplished. This involves a type of contiguous placement of two sentences (any combination of simple and complex). The syntax of the second sentence, then, can establish a conditional framework by the introductory use of some kind of causal or illative (inference-making) coordinating conjunction. The second sentence can also accomplish this conditional reference in other ways, e.g., the introductory deployment of a demonstrative pronoun (or a demon-
244
Steingart, Freedman, Grand, and Buchwald
strative adjective, etc., etc.). We already have obtained evidence that such categories of language behavior discriminate important differences in psychopathology, such as changes in clinical state (Steingart and Freedman, 1972), and social adaptation in chronic schizophrenia (Grand et al., 1973), and we now want to consider the relevance of these language behaviors for individual differences in a group of clinically normal Ss. The average number of words per sentence (AWS) for each of these three types of surface language behavior is calculated. In addition, the percentage clause output for each of these types of language behavior is calculated, as well as the percentage of grammatically incoherent Fragmented language (false starts, etc.). Table I contains examples of both short and long AWSs for each of these three types of grammatically coherent, surface language behaviors actually uttered by the Ss of this study 2 We now introduce an operational distinction between monologue communication and dialogue. Monologue communication, while it presupposes a communication task, is nevertheless quite different from dialogue. In monologue communication, a speaker utters soliloquy directed at another who functions only as listener. Whether such monologue communication is instigated by special situational design (e.g., in Freudian analysis), or by endogenous personality determinants, we consider that important individual differences probably exist among adults with respect to the kinds of language behavior which will be evident in their monologue communication. It was precisely the purpose of this study to examine such relationships.
METHOD
The data used in this research were collected from audio-video tapes originally obtained from another study. This original study examined the effects of psychological differentiation on kinetic behavior in varying communication conditions, and it describes the method employed in greater detail (Freedman et aL, 1972).
2A more detailed description of these language construction categories as well as data which support the reliability of the scoring system is available elsewhere (Steingart and Freedman, 1972). Our procedure for the identification of a sentence unit is similar to the "minimal terminal unit" (T-unit) described by Hunt (1970). There are, however, two differences between Hunt's system and ours: first, we permit compounding of full independent clauses, and, second, we identify a type of Fragmented (grammatically incomplete) language word string.
Pezsonality Organization and Language Behavior
245
0
0 ~.~
~
E 0
Z
9~ ~
~
.~. ,.~
>, 0
<~r..) o
e~
~ = 9~
.
.~ ~
9
0 o
r..)
0
r
e~
. ~ Z '~
[..,
0
0
0
246
Steingart, Freedman,Grand, and Buchwald
Subject Selection The original research entailed audio-video tape-recorded interviews of female college students previously identified as either field independent (F-I) or field dependent (F-D). Each S was confronted with two interview conditions, one by a "Warm Interviewer" and one by a "Cold Interviewer." Ss who participated in the original study of kinetic behavior were all between the ages of 18 and 22 years; they were native-born Americans predominantly of Jewish descent. In the original study, two extreme groups, 12 F-I and 12 F-D, were selected from a group of 134 on the basis of their ability to disembed figure from background with the portable rod-and-frame test (Oltman, 1968). The scores of these groups fell into the F-I and F-D thirds of the population, respectively. In addition, various forms of group embedded-figures tests (Oltman e t al., 1972) were given the Ss, and mean Z scores of F-I and F-D groups in these tests were 1.21 and -1.02, respectively. In order to control in the present study for vocabulary level and verbal fluency, an effort was made to recall all of these original 24 Ss. We were able to recall 19 Ss (11 F-I and eight F-D). These 19 Ss were given the Vocabulary subtest of the WAIS (Wechsler, 1972). In addition, three tests of verbal fluency devised by Guilford (1967) were administered. These are speed tests which first require the utterance of words that begin with "P," then words that begin with "PRO," then words which end with "AY," and finally words that begin with "S" and end with "P." A sum verbal fluency score is used as a measure of best estimate for each S.
Experimental Procedures The Ss had previously been contacted by telephone by a member of the research staff, a person other than the interviewers. The Ss had previously been told that they were to participate in an experiment on speech patterns and that they were simply expected to talk. They were offered payment, which they received at the end of the experimental sessions for participation in the study. These 19 Ss in the original study (Freedman e t al., 1972) had been confronted with both a "Warm" or "Cold" communication condition which they encountered in one of two sequences. However, no sequence effects were obtained for any of the language behavior data which will be described here. 3 3This fact, that no sequence effects were obtained for language construction behavior but were obtained for kinetic behavior (Freedman et al., 1972), is worthy of consideration in itself but beyond the purview of this paper.
Personality Organization and Language Behavior
247
Communication Conditions The Ss were met on 4heir arrival by either Dr. R. (a Warm interviewer) or Dr. Q. (a Cold interviewer), whom for convenience' sake we shall henceforth call Dr. Warm and Dr. Cold. Drs. Warm and Cold were research psychiatrists working at a psychiatric research center. Even though in this study they assume the role of interviewer, they were introduced as "Doctor." Drs. Warm and Cold deliberately assumed their respective roles and thus did not follow a predetermined set of instructions. They also did not "switch" roles, each one staying with his procedure throughout the experiment.
The W-C Sequence In the W-C sequence, Dr. Warm greeted a S with a handshake, asked her to take a seat, and explained some of the functions of the equipment in the recording room. He then inquired into several areas of the S's life, her major interest, hopes, social activities, and dating patterns. In each case, he avoided probing into conflict areas and his attitude might best be described as reflecting a sustained, shared-focused attention on whatever the S was telling him. This attitude was reflected also in his nonverbal behavior, for example, nodding in agreement and leaning toward the S. This Dialogue "warmup" lasted for about 15 min. Dr. Warm then introduced the task using a modification of Gottschalk's instructions (Gottschalk et al., 1969): "I would like you to talk for 5 minutes. I would like you to talk about something that is personally meaningful and significant to you. I shall not interrupt you. Just talk until the time is up." This constituted the Warm Monologue communication condition. At the end of this 5-min period, Dr. Warm was called out of the room on the pretext that there was some equipment difficulty which required his special competence. After a few minutes, Dr. Cold appeared and introduced himself as Dr. Warm's assistant. He said that the 5-min talk period needed to be repeated because of technical difficulties and that Dr. Warm was needed to adjust the recording equipment. Dr. Cold was impatient and indicated that he wanted to "get this over quickly" and told the S "let's get on with it." He yawned, fidgeted, and looked away while the S was talking. This constituted the Cold Monologue communication condition.
The C-W Sequence In the C-W sequence, Dr. Cold brusquely greeted the S, introduced himself as Dr. Warm's assistant, and indicated that he had to take over because of Dr. Warm's absence. As above, he was impatient, almost corn-
248
Steingart, Freedman, Grand, and Buchwald
manded the S to sit down, and asked her name. If she replied, he turned away. He then gave the Gottschalk instructions as above and maintained his bored, impatient demeanor throughout the 5-min Cold Monologue sample. At the end of the 5-min verbal sample, Dr. Warm knocked at the door. He looked surprised and apologized for being late, and for Dr. Cold's conduct. He then conducted his 15-rain Dialogue "warmup" and the 5-min Warm Monologue sample as described above. Thus, aside from the Warm and Cold Monologue conditions, in which both of the interviewers obtained a 5-min talking sample, we also had available for analysis the "warmup" Dialogue condition with Dr. Warm. The last 5 min of this "warmup" Dialogue is used as the Dialogue language sample. For technical reasons, seven of the original 24 Ss did not have their Dialogue "warmup" condition videotaped (Freedman et al., 1972). Unfortunately, five of these seven Ss were among the 19 Ss we were able to recall for our language behavior study, so that we have Dialogue language data for only 14 Ss (eight F-I and six F-D).
RESULTS Monologue Conditions
Table II describes correlations among (sum) verbal fluency, Vocabulary, field independence, and AWS for our different types of language behavior in the Warm and Cold Monologue conditions. F-I is significantly and negatively correlated with AWS-Narrative language in Cold Monologue, and all other correlations are not significant. We did obtain a significant, positive correlation for our 19 Monologue Ss between F-I and (sum) verbal fluency (0.464, p < 0.05). Table III describes correlations in the Monologue conditions between F-I and AWSs for our different types of language behavior with (sum) verbal fluency controlled by the application of a part correlation between (sum) verbal fluency and F-I. The introduction of this control for verbal fluency by this part correlation leaves unaffected the initial significant, negative correlation between F-I and AWS-Narrative language in Cold Monologue. In addition, controlling for verbal fluency within F-I by part correlation produces a second, negative significant correlation between F-I and AWS-Portrayal language behavior in the Warm Monologue condition. We did not obtain, nor expect (Goodenough and Karp, 1961), a significant correlation between Vocabulary and F4 for our 19 Monologue Ss.
Pexsonality Organization and Language Behavior .~ ~ .o
249 .~ ~ o
,.~0~ 0 0 o
I
I
I
t'q
Z~,,a
e~
0
~:~ ~.~ 0
}1
m.~.
tt~
0
Z_ 0
~
>-. 0
0 ~ . ~
~ ~~
r
O
0 0 0
o ~ 0
.~u'~ ~
0
.~ ~.o
O
~. m. m. I
I
c..)
[
O
l
0
~
"6 >
~
g.r
~
0.~
O
tm o
r O
]
0 0
0 r~
O
>o 0 9
<
O
~o 0
r..- o o~ ~ t--I o
O
0
~:m ~_
o
o
o
~.~
I
O
0
o V
V
~'~ ~,
o
~
t~ 0 ~
~8
=
0~
r162
o t~
.~
II V
I~ V 0)
7-
250
Steingaxt, Freedman, Grand, and Buchwald
Dialogue Condition
No significant correlations were obtained in the Dialogue condition between AWSs for our different types of language behavior and F-I, (sum) verbal fluency, and Vocabulary. A correlation of 0.441 was present between F-I and (sum) verbal fluency for our 14 Dialogue Ss. The introduction of a control for (sum) verbal fluency within F-I, by the application of a part correlation between (sum) verbal fluency and F-I, left unaffected our initial insignificant correlations between F-I and AWSs for our various types of Dialogue language behavior. The effect of our analysis up to this point is to identify certain relationships which occur between cognitive style and language behavior in a situation of Monologue communication, but not in Dialogue: First, field dependence is associated with AWS-Narrative language in Cold Monologue. Second, but only upon the application of a control for differences in (sum) verbal fluency within F-I, field dependence is associated with AWS-Portrayal language in Warm Monologue. This latter finding indicates that differences in (sum) verbal fluency operated to mask an interaction effect of cognitive style, plus the situation of Warm Monologue communication, AWS-Portrayal language behavior.4 Three further, more detailed analyses of our data were performed in order to facilitate our interpretation of these findings which relate language behavior to psychological differentiation in monologue communication. First, one may ask whether the obtained correlations have been spuriously affected because sizable and significant differences exist between F-I and F-D Ss with respect to how much of each type of language behavior actually is uttered by Ss in Dialogue compared to the Warm and Cold Monologue conditions. We performed analyses of covariance controlling for any effects of verbal fluency about this issue for those 14 Ss (eight F-I and six F-D) for whom we had baseline Dialogue data. Neither for Dialogue nor for either of the two Monologue conditions are there any appreciable or significant differences between F-I and F-D Ss with respect to percentage output for Narrative, Portrayal, or Conditional language behavior. Moreover, both F-I and F-D Ss show a similar and significant pattern of change from Dialogue to Warm or Cold Monologue conditions. Narrative and Portrayal language behavior both noticeably increase in Monologue compared to Dialogue. Neces4To put this another way, differences in covert skill with verbal thought processes
contribute to the solution of the overtly "nonverbal" rod-and-frame and embeddedfigures tasks which define our extreme F-I and F-D Ss; and, for our group of successful female college Ss, such covert verbal skill must be high as well as relatively restricted in range compared to the general population.
Personality Organization and Language Behavior
251
sarily, Conditional language behavior thus shows a steady decline in output from Dialogue, to Warm Monologue, to Cold Monologue. The change in the percentage output of Conditional language from Dialogue to Warm Monologue to Cold Monologue especially is striking (for both F-I and F-D Ss). In Dialogue, Conditional language behavior accounts for approximately 35% of language behavior, whereas in Cold Monologue it accounts for only approximately 15% of speech output (and Narrative and Portrayal language each increase in output approximately 10%). A second question about the findings in Tables II and III has to do with what actually is contributing to the correlations between field independence and AWS language behavior type. Have the AWSs of F-D Ss gotten longer, have the AWSs of F-I Ss gotten shorter, or is it both of these? Figure 1 compares mean AWSs in Narrative and Portrayal language behavior, for those 14 F-I and F-D Ss for whom we have baseline data, in the three communication conditions. A S by S analysis of these results makes clear how our Monologue correlations primarily are due to F-D Ss getting longer, whereas the AWSs of F-I Ss tend to remain about the same as Dialogue. The AWS of Portrayal language behavior in Warm Monologue for five out of eight of these F-I Ss was within a range of 0 to +4 words compared to the Dialogue; one F-I S had an AWS o f - 3 , and one F-I S an AWS of -2, compared to Dialogue. On the other hand, only one F-D S showed no change in A N A R R A T I V E LANGUAGE
B
PORTRAYAL L A N G U A G E
I0-
22-
8-
20-
18<[
4-
16 --
=
l
i
l
l
2-
14
O'
Dialogue
i Warm Cold \--~-~,M on o l o g ue < - J "
Dialogue
Worm Cold ~ - ~ - M a n a l o g ue ~-- j
Fig. 1. Mean AWS language behavior for F-I vs. F-D Ss in three communication conditions.
252
Steingart, Freedman, Grand, and Buchwald
AWS-Portrayal language behavior and one F-D S showed a change of+4 words from Dialogue to Warm Monologue; remaining F-D Ss all had much longer AWS-Portrayal language behavior compared to Dialogue, up to 10 and 12 words longer. A similar difference was obtained between F-I and F-D Ss with respect to AWS change in Narrative language behavior between Dialogue and Cold Monologue. The AWS-Narrative behavior length for six out of eight F-I Ss showed essentially no change (0 or 1), whereas AWS-Narrative language behavior for five out of the six F-D Ss showed larger increases, including doubling of sentence length from Dialogue to Cold Monologue. A third question has to do with verbal output. Further analyses of covariance were performed for clause output for our 14 Ss for whom we have baseline data, and it revealed both a communication condition effect and a cognitive style effect. 5 With respect to communication condition, inasmuch as Ss have more "floor" time in Monologue they tend to talk more in this condition compared to Dialogue. But besides this, F-I Ss generally talk more than F-D Ss, and this is especially so in Monologue in which they exceed F-D verbal output by some 45% (compared to only 15% in Dialogue). Thus F-I Ss especially talk more in Monologue compared to F-D Ss but this is not accompanied by any increase in percentage output for any of our types of language behavior. 6 Further, this increase in language behavior of our F-I Ss in Monologue compared to F-D Ss is not accompanied by any change in AWS for any of our types of language behavior. In sum: The greater amount of talking done by F-I Ss in Monologue serves to introduce more dialoguepatterned units o f language behavior into their Monologue condition. F-D Ss, on the other hand, not only talk less in Monologue but also show a definite change in the AWS patterning of their language behavior compared to their Dialogue performance. DISCUSSION These results can be interpreted in terms of what F-I and F-D Ss both can and cannot do with their language behavior when faced with a transition from Dialogue to Monologue. F-I Ss do not simply respond to Monologue 5We consider that clause output, i.e., clauses which participate in grammatically coherent surface sentence structures, is our best measure of verbal output. As would be expected, this clause output measure is highly correlated with word count (over 0.90 in all three communication conditions). 6However, percentage output for these three different types of language behavior does discriminate important individual differences which exist among a group of clinically abnormal Ss, i.e., individuals who are schizophrenic (Steingart and Freedman, 1972; Grand et al., 1973).
Personality Organization and Language Behavior
253
with a significant increase in language behavior compared to F-D Ss. It is how the F-I Ss use this increased language behavior which is illuminating. Inasmuch as F-I Ss in both Warm and Cold Monologue continue to "package" their language as if they are Dialogue, they in effect respond to both Monologue conditions with an increase in their Dialogue language behavior. They talk, even more, as if they are in a Dialogue interchange in which they deliver their thoughts to another person unlike themselves, and whose verbal behavior must be taken into account with respect to their own ongoing language performance. That F-I Ss show what might be called an automony of Dialogue language pattern in the absence of any actual verbal interchange is congruent with the extensive literature about the psychological differentiation of such individuals compared to our F-D Ss (Witkin et al., 1962). F-I Ss talk in such a way so as to structure a linguistic environment which is relatively unarticulated (in the sense that the other only listens) into a condition of communicative interchange. More particularly, our results fit well with a finding from a recent study by Witkin et al. (1968) which contrasted the language behavior of F-I and F-D patients in Dialogue (i.e., ego-supportive psychotherapy). This study reports that significantly more verbal interactions take place in such therapy dialogue between F-D patients and therapists compared to a matched group of F-I patients treated by the same therapists. Thus, congruent with our findings, but now within the condition of Dialogue itself, F-I language behavior can be considered to show more dialogue-patterned autonomy in the sense that there is less necessity for any actual verbal feedback by the therapist. We can state these matters now from the vantage point of F-D Ss. The effect of Monologue on the language behavior of our F-D Ss clearly is different. But, again, it is not simply a matter of verbal output. F-D Ss do talk considerably less in Monologue than do F-I Ss but at the same time these F-D Ss are producing grammatically more elaborate (lengthier T-unit) surface sentence construction than F-I Ss. It is descriptive, Complex Portrayal language behavior which becomes grammatically more elaborate in Warm Monologue. The language behavior of our F-D Ss, in Cold Monologue, becomes grammatically more elaborate in a much less mature way. It is reasonable to suppose that the "lower" (developmentally much less mature) type of sentence elaboration represented by Narrative language, and shown by o u r F-D Ss in Cold Monologue, is due to the stress quality of this communication condition. One may say that the change in F-D language behavior in Monologue is responsive to the presence of the other only as listener. What occurs for our F-D Ss is the elaboration of language behavior which dwells upon descriptions of personal experience. Evidently, facial cues
254
Steingart, Freedman, Grand, and Buchwald
facilitate such kinds of language behavior for our F-D Ss. On the other hand, the presence of such a visual (but completely nonverbal human environment) has no such effect on the language behavior of our F-I Ss. Indeed, research already has established that F-D Ss selectively attend to and remember faces more than do F-I Ss (Crutchfield et al., 1958; Konstadt and Forman, 1965; Messick and Damarin, 1964). Thus one can speculate that even in a Dialogue situation verbal vs. visual inputs have entirely different sorts of significance for the organization of language behavior of F-I and F-D Ss. The foregoing considerations suggest several further lines of investigation. First, what has been suggested about the differential significance of verbal vs. visual cues for language behavior can be tested rather easily by experimental manipulation. One can, for example, simply have F-I and F-D Ss turn their backs to an interviewer, with the expectation that F-D language behavior ,will be much more adversely affected than that of matched F-I individuals, in both Dialogue and Monologue. Second, we consider that a more complex evaluation of individual differences is necessary in order to increase our understanding about this matter of a shift in language behavior from Dialogue to Monologue. For example, we already have implied that one out of eight F-I Ss in the Warm Monologue situation did show an elaboration of Complex-Portrayal language behavior very much like that which characterized the entire F-D group change from Dialogue to Monologue. We expect that some assessment of cognitive flexibility with language structures such as is measured by the Stroop Color-Word Test (Gardner et al., 1960), integrated with a field-independence evaluation, will reveal meaningful individual differences in this matter of a shift in language behavior from Dialogue to Monologue. Third, we consider that the overall deployment of these different types of language behavior may reflect important individual differences in the organization of consciousness. It has been our anecdotal impression that these different types of language behavior are associated with entirely different qualities of experience itself. For example, Narrative language behavior appears upon anecdotal examination to possess much more the qualities of concreteness and imagery compared to Conditional language behavior, and we intend to pursue this question in future research. Finally, we consider that we may be able to use these language behavior measures to identify not only relatively stable cognitive structure but process as well. It may be that we can use such vicissitudes of language behavior as a kind of "on-line" calculator to chart the dynamic ups and downs of psychological structuralization due to moments of outer stress, or states of inner psychopathology.
Personality Organization and Language Behavior
255
ACKNOWLEDGMENT We wish to express our thanks to Dr. Herman A. Witkin for his thoughtful comments and suggestions regarding this paper.
REFERENCES Crutchfietd, R. S., Woodworth, D. G,, and Albrecht, R. E. (1958). Perceptual performance and the effective person. Personnel Laboratory, Wright Air Development Center~TN-58-60, ASTIA Document No. AD 151039, Lackland Air Force Base, Texas. Feffer, M. (1970). Developmental analysis of interpersonal behavior. Psychol. Rev. 77(3):197-214. Freedman, N., O'Hanlon, J., Oltman, P., and Witkin, H. A. (1972). The imprint of psychological differentiation on kinetic behavior in varying communicative contexts. J. Abnorm. Psychol. 79(3):239-258. Gardner, F. W., Jackson, D. N., and Messick, S. J. (1960). Personality organization in cognitive controls and intellectual abilities. Psychol. Issues 4: Monograph No. 8. Goodenough, D. R., and Karp, S. A. (1961). Field dependence and intellectual functioning. J. Abnorm. Soc. Psychol. 93(2):241-246. Gottschalk, L. A., Winget, C. N., and Gleser, G. C. (1969). Manual of Instructions for Using the Gottschalk, Gleser Content Analysis Scales: Anxiety, Hostility, Social Alienation-Personal Disorganization, University of California Press, Berkeley. Grand, S., Freedman, N., and Steingart, I. (1973). A study of the representation of objects in schizophrenia. J. Am. Psychol. Assoc. 21:399-434. Guilford, J. P. (1967). The Nature o f Human Intelligence, McGraw-Hill, New York. Hunt, K. W. (1970). Syntactic maturity in school children and adults. Monogr. Soc. Res. Child Develop. 35: No. 1. Konstadt, N. and Forman, E. (1965). Field dependence and external directedness. J. Personal. Soc. PsyehoL 1:490-493. Loban, W. B. (1963). The langauge of elementary school children. National Council of Teachers of English Research Report No. 1, Champaign, IlL Lyons, S. (1968). Introduction of Theoretical Linguistics, Cambridge University Press, London. Messick, S., and Damarin, F. (1964). Cognitive styles and memory for faces. J. Abnorm. Soe. Psychol. 69:313-318. O'Donnell, R. C., Griffin, W. J., and Norris, R. C. (1967). Syntax of kindergarten and elementary school children: A transformational analysis. National Council of Teachers of English Research Report No. 8. Oltman, P. (1968). Portable rod-and-frame apparatus. Percept. Mot. Skills 26:503-506. Oltman, P., Raskin, E., and Witkin, H. A. (1972). The Group Embedded-Figures Test (G.E.F.T.], Consulting Psychologists Press, Palo Alto, Calif. Piaget, J. (1960). The Psychology o f Intelligence, Adams and Co., Littlefield, N.J. Steingart, I., and Freedman, N. (1972). A language construction approach for the examination of self-object representation in varying clinical states. In Psychoanalysis and Contemporary Science, Vol. 1, MacMillan, New York. Wechsler, D. (1972). The Measurement and Appraisal o f Adult Intelligence, Williams and Wilkins, Baltimore. Witkin, H. A., Dyk, R. B., Faterson, H. F., Goodenough, D. R., and Karp, S. A. (1962). Psychological differentiation: Studies o f Development, Wiley, New York. Witkin, H. A., Lewis, H. B., and Weil, E. (1968). Affeetive reactions and patient-therapist interactions among more differentiated and less differentiated patients early in therapy. J. Nerv. Ment. Dis. 146:193-208.