Department of Psychiatry, SUNY Stony Brook
Stony Brook, New York 11794
LDeLisi@notes.cc.sunysb.edu

The language abilities of modern humans are thought to be clearly distinct from non-human primates (Bickerton, 1990). Crow (1997, 1998a) has proposed that language disturbance is central to the genetic etiology of schizophrenia and may be the critical characteristic in the evolution of Homo sapiens. Oral soliloquys have been analyzed from schizophrenic patients at the onset of their illness, controls and families with a high density of illness. Anomalies in Bickerton’s 5 distinctly human language devices were examined (sequential order, null elements, arguments of verbs, mechanisms for expansion and grammatical correctness) using a structured scoring format on the transcribed passages. Analyses show reduced use of clausal embedding in chronic patients that is not present at the time of the first-episode. These data are preliminary and need to be replicated in a larger sample of subjects. However, they are consistent with previous literature on language in patients with schizophrenia and further show that there is a familial component to these measures. Thus, these findings suggest that deficits in specifically human aspects of language may be related to the genetics of schizophrenia.

Schizophrenia is a lifetime brain disorder expressed by thought, speech perception and organizational problems. It occurs in every human population (world-wide) and family, twin and adoption studies show a genetic predisposition for schizophrenia. Explanations for the thought and speech disorder of schizophrenia include that of (1) "dysfunctional executive control" (frontal lobe; e.g. Chaika, 1990; Morice,1995). (2) working/semantic memory deficits (e.g. Grove and Andreasen, 1985, Mortimer et al., 1995), and the extreme end of the variation in bilateral neural organization of the uniquely human components of language (Crow, 1998).

Bickerton (1990) proposes 5 unique characteristics of human language: (1) Differences in the superficial order of constituents, (2) use of null elements, (3) subcategorized argument structure of verbs, (4) mechanism for the expansion of utterances, and (5) grammatical items. It is hypothesized that the language disturbances of schizophrenia are related to defects in one or more of the uniquely human features of language and that the inheritance of the variation in these abilities is related to the inheritance of schizophrenia.

A relationship also exists between brain structural and functional asymmetries and both language abilities and schizophrenia (Corballis, 1991; Crow, 1998a; Geschwind and Galaburda, 1987). Furthermore, reduced functional asymmetry (hand skill) relate to reduced verbal ability (Crow et al., 1998). It has been proposed that the core clinical symptoms of schizophrenia relate to an underlying disorganization of language and its asymmetric perception and production (Crow, 1998b). Finally, some reduced posterior cerebral asymmetries present in schizophrenia are shown to be familial (DeLisi et al., 1997).

In previous studies (DeLisi et al., 1997; Shedlack et al., 1997) we found significantly more morphological errors and less clausal embedding in 1st episode patients compared with 52 controls. No associations were found with brain structural asymmetries. When ill and well members from 19 families with 2 or more siblings with schizophrenia were compared with 17 unrelated controls, soliloquys revealed that both ill and well siblings from these families had significantly less of one type of clausal embedding than controls, although the degree of embedding was not correlated within families and ill siblings did not have less embedding than their well siblings. The present study expands on our previous measures of language performance.

First-episode patients with schizophrenia, families with 2 or more siblings with schizophrenia and unrelated controls had their speech recorded and later transcribed (without punctuation) while an examiner asked each to use his/her imagination to describe what is happening in 7 pictures taken from the Thematic Aptitude Test (TAT). These pictures ranged from disturbing emotional scenes to tranquil scenery.

The following are examples of responses to two disturbing scenes, ones which tended to stipmulate more disorganized responses from patients.

In addition , each subject was asked to describe the steps taken to get up in the morning and get ready for an appointment, from waking-up to arrival at the destination (sequential ability).

Laterality was assessed by (1) a hand use questionaire (Annett 1967) and (2) by a relative hand-skill timed "dot" test (Tapley and Bryden, 1985). The following items were scored: total # of words, total # of sentences, SYNTAX: Order (# of awkward sentences, # with incorrect word order), # of null elements, # sentences missing arguments to verbs, mechanisms for expansion (# of conjoined clauses and # embedded, # grammatical mistakes, SEMANTICS: Misuse of Words (#), neologisms (#), MISCELLANEOUS: Appropriateness of content and logical order to sequential task and amount of detail.

Two separate 2-way ANOVA’s were performed (1) For the 1st epsode cases compared to chronic patients and controls covarying sex and age (2) For the 12 families only: with sex, diagnosis and family membership as factors, covarying for age.

Note: There were no significant sex or sex by diagnosis effects.

Table 1. Unrelated Patients vs. Controls

Table 1 summarizes the data on the 1st episode patients compared with controls, while table 2 summarizes the family data. Reduced percentages of conjoined and a trend for reduced embedded clauses were present in the schizophrenic patient groups compared with controls.

The Characteristic thought disorder present in patients with schizophrenia can be shown to be linguistically deviant. Particularly, chronic patients (ill for at least a few years) had reduced use of conjoined and embedded clauses in their speech, used fewer words overall, more inappropriate content and less connectivity between sentences, than controls. A lack of detectable language abnormalities at the time of the 1st episode suggests progressive development after the onset of illness. In addition, a familial basis for clausal mechanisms of expansion is suggested from analyses of the above families.

Thus, the uniquely human aspects of language, particularly mechanisms for expansion are shown to be reduced in patients with schizophrenia and also shown to be familial and related to illness within families.

 Conference site: http://www.infres.enst.fr/confs/evolang/