Compensatory effect of general cognitive skills on non-literal language processing in schizophrenia: A preliminary study

Studies show that patients with schizophrenia have problems with understanding non-literal language, presumably due to theory of mind (TOM) dysfunction, while literal language processing remains intact in these patients. Furthermore, studies about the influence of general intelligence on non-literal language processing show incongruent results in schizophrenia. In the present study, our aim was to identify the influence of good intellectual skills (that is, good neurocognitive skills in general) on non-literal language comprehension on a preliminary sample of schizophrenic patients. 19 patients with schizophrenia and 19 healthy controls (carefully matched for all subtests of WAIS) completed different language tasks in the form of an interview, including conventional and unconventional metaphors, irony, Gricean implicatures, control implicatures and semantics. In the case of the Gricean implicatures, the comprehension of the linguistic incongruity of the utterance and also the comprehension of the intended meaning of the speaker were examined separately. To explore the effect of IQ on non-literal language comprehension, we divided the schizophrenia group into a lower-IQ (IQ <= 106) and a higher-IQ (IQ > 106) subgroup by making a median split (median IQ = 106). Our results showed that schizophrenics with lower-IQ were able to understand conventional metaphors presumably relying on their good semantic processing, while they were impaired in the comprehension of unconventional metaphors and irony. However, the higher-IQ schizophrenia subgroup was able to comprehend not only conventional metaphors, but also unconventional metaphors and irony, supposedly using IQ-dependent compensatory mechanisms. Interestingly, both the lower- and the higher-IQ schizophrenia subgroups showed significant impairment in the comprehension of the ToM question of the Gricean implicatures and the patients' IQ had no significant impact on the comprehension of such ToM questions. In conclusion, our findings assume a possible compensatory effect of general cognitive skills on non-literal language processing in schizophrenia. We also argue for the importance of using more complex TOM tasks in order to detect the existing ToM deficit in schizophrenic patients with good cognitive skills. (C) 2014 Elsevier Ltd. All rights reserved.