Lexical Differences in Autobiographical Narratives from Schizophrenic Patients and Healthy Controls

Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 37–47, Jeju Island, Korea, 12–14 July 2012. c⃝2012 Association for Computational Linguistics Lexical Differences in Autobiographical Narratives from Schizophrenic Patients and Healthy Controls Kai Hong1, Christian G. Kohler2, Mary E. March2, Amber A. Parker3, Ani Nenkova1 University of Pennsylvania Philadelphia, PA, 19104, USA 1{hongkai1,nenkova}@seas.upenn.edu 2{kohler,memarch}@mail.med.upenn.edu 3{parker}@sas.upenn.edu Abstract We present a system for automatic identification of schizophrenic patients and healthy controls based on narratives the subjects recounted about emotional experiences in their own life. The focus of the study is to identify the lexical features that distinguish the two populations. We report the results of feature selection experiments that demonstrate that the classifier can achieve accuracy on patient level prediction as high as 76.9% with only a small set of features. We provide an in-depth discussion of the lexical features that distinguish the two groups and the unexpected relationship between emotion types of the narratives and the accuracy of patient status prediction. 1 Introduction Recent studies have shown that automatic language analysis can be successfully applied to detect cognitive impairment and language disorders. Our work further extends this line of investigation with analysis of the lexical differences between patients suffering from schizophrenia and healthy controls. Prior work has reported on characteristic language peculiarities exhibited by schizophrenia patients. There are more repetitions in speech of patients compared to controls (Manschreck et al., 1985). Patients also tend to repeatedly refer back to themselves (Andreasen., 1986). Deviations from normal language use in patients on different levels, including phonetics and syntax, have been documented (Covington et al., 2005), however lexical differences have not been investigated in detail. In this paper we introduce a dataset of autobiographical narratives told by schizophrenic patients and by healthy controls. The narratives are related to emotional personal experiences of the subjects for five basic emotions: ANGER, SAD, HAPPY, DISGUST, FEAR. We train an SVM classifier to predict subject status. Our good results on the relatively small dataset indicate the potential of the approach. An automatic system for predicting patient status from autobiographical narratives can aid psychiatrists in tracking patients over time and can serve as an easy way to administer large scale screening. The detailed feature analysis we performed also pinpoints key differences between the two populations. We study a range of lexical features including individual words, repetitions as well as classes of words defined in specialized compiled by psychologists (Section 4).dictionaries We use several approaches for feature analysis to identify statistically significant differences in the two populations. There are 169 significant features among all of the 6057 features we examined. Through feature selection we are able to obtain a small set of 25 highly predictive features which lead to status classification accuracy significantly better than chance (Section 6.3). We also show that differences between patients and controls are revealed best in stories related to SAD and ANGRY narratives, they are decent in HAPPY stories, and that distinctions are poor for …

Content truncated to 500 words