Abstract:
To understand whether a remark is intended ironic constitutes a challenge in everyday language comprehension. A selective deficit to indentify irony correctly has been demonstrated for several psychiatric disorders, with the best evidence for schizophrenia. The neuronal correlates behind this phenomenon remain unknown. Whereas some authors speculated about defective “theory of mind” function in medial prefrontal regions, recent fMRI evidence from schizotypy indicated a frontotemporal language comprehension deficit. This functional imaging study aimed to demonstrate brain regions involved in irony resolution in female patients with schizophrenia and to clarify the pathophysiology behind this phenomenon.
Methods: 15 female probands with schizophrenia and 15 female healthy control subjects, matched for age and verbal intelligence (MWT-B) participated in the fMRI experiment concerning prosody free irony comprehension (3-T-Scanner, whole brain, 32 slides, TR= 2s, TE= 40 ms). SPM5 software was used for Data analysis. Additionally 24 healthy controls took part in the offline Irony-Comprefension-Test, Irony-Rating Test, Proverb-Comprehension-Test and Schizotypal-Personality-Questionnaire (SPQ). During the fMRI experiment, the task was to read 44 short text vignettes silently and to understand their meaning. Each text vignette ended in a statement of the protagonist which is either ironic (n=22) or literal (n=22). After the magnetic resonance scanning, subjects performed an offline version, which contains identical stimuli and 10 additional meaningless sentences. In this off-line test, subjects indicated by button press whether the target sentence has an ironic, literal or meaningless content. In addition we asked the patients and the additionally healthy control group for the certainty of their decision and to evaluate the meanness and the jocularity of the statements using likert-scales.
Results: As expected, the patients showed significantly lower irony understanding. The Irony-Rating-Test showed, in contrast with our hypothesis, only limited variance. The distribution of errors both in the Irony-Comprehension-Test and Proverb-Comprehension-Test shows, the deficit cannot be explained exclusively by concretism. During fMRI both patients and controls activated a bilateral network during comprehension of ironic sentences. The comparision of the two groups showed hyperactivity of patients in the left parahippocampal gyrus and hypoactivity in the bihemispheric gyrus postcentralis, as well as in the left insula and anterior cingulum. In the analysis of correlation of image data an increasing schizotypal-score independently of the psychiatric diagnosis corresponded the nearing of brain activity in the group of patients.