Do Patients With Depression Prefer Literal or Metaphorical Expressions for Internal States? Evidence From Sentence Completion and Elicited Production.

Abstract

In everyday communication metaphoric expressions are frequently used to refer to abstract concepts, such as feelings or mental states. Patients with depression are said to prefer literal over figurative language, i.e. they may show a concreteness bias. Given that both emotional functioning and the processing of figurative language may be altered in this clinical population, our study aims at investigating whether and how these dysfunctions are reflected in the understanding and production of metaphorical expressions for internal states. We used two behavioral approaches: a sentence completion task and elicited speech production. In the first experiment, patients with ICD 10 depression (n = 26) and healthy controls (n = 32) were asked to complete sentences by selecting an appropriate word out of four alternatives (metaphorical expression, literal expression, concrete distractor, abstract distractor). All participants–irrespective of the presence of depression–chose more literal (60%) than metaphorical (40%) expressions. In the second experiment, patients with depression (n = 44) and healthy controls (n = 36) described pictures showing emotive events. The descriptions were transcribed and coded for type of expression (non-figurative words for internal states vs. metaphorical expressions, valence, type of metaphor, source and target domain of metaphor). In addition, the Thought and Language Index was applied to assess formal thought disorder. When talking about internal states, both groups used more literal than metaphorical expressions. The groups did not differ with respect to the composition of internal state language, but patients with depression tended to verbalize positive content to a lesser extent. Correlation analyses within the patients' group revealed that signs of disorganization in their speech were related to a higher use of internal state expressions, whereas a negative correlation was found with dysregulation phenomena. Taken together, results indicate that people with and without depression prefer literal means in order to verbalize internal states, but they additionally make use of figurative language. Since patients with depression were able to understand and produce metaphors for internal states similar to controls, the concreteness bias cannot be confirmed by the present study. The results contribute to existing research by demonstrating associations between symptoms of formal thought disorder and internal state language.