[The understanding of metaphors in schizophrenia and depression. An experimental approach]

Since the first clinical descriptions of schizophrenia, clinical practitioners have been interested in the difficulties experienced by patients with schizophrenia in interpreting the meaning of metaphors. A long tradition of proverb tests intended either for clinical or experimental use, has revealed that the difficulties in metaphor interpretation experienced by patients with schizophrenia refer to a large number of different types of interpretational errors (eg literality, concretism, or autistic, idiosyncratic and bizarre responses). A number of studies have adopted an experimental approach to investigating the dynamics of the cognitive processes that underlie the difficulties in accessing figurative meaning experienced by patients with schizophrenia. However, limited in the way they operationalise these phenomena and/or by their psycho-linguistic reference models, these studies have focused on only one aspect of metaphor: for example, multiple-choice tasks have tested the tendency of patients with schizophrenia to interpret metaphors literally ("literality bias"), whereas lexical decision tasks have investigated the importance that such patients attach to a single word in the expression ("concretism"). The first aim of our study was to investigate, in parallel and without confound, the respective contributions of the literality and concretism biases in the interpretation of metaphor in patients with schizophrenia. The second aim was to examine the question of the specificity of difficulties in accessing figurative meaning in patients with schizophrenia by comparing their performance profiles with those of patients with depression. The third aim was to examine the influence of the patients' clinical symptomatology on their result profiles.

The participants consisted of 25 patients with schizophrenia (DSM IV, 1994), 18 patients with major depression (DSM IV) and 22 healthy controls. All the participants were matched on socio-demographic variables (age, vocabulary level and level of education). The participants had to complete a task consisting of 10 metaphorical sentences (eg "Ce milieu est un panier de crabes"=lit. "This place is a basket of crabs", fig. "What a bunch of sharks"). The participants had to choose only one word from a set of 4 responses: the figurative meaning (eg "magouille"=dishonest person), the literal meaning (eg "vivier"="pond"), the concrete meaning of the final word (eg "crustacé"="crustacean") and one unassociated word (eg "journal"=newspaper). The inter-group comparison and the symptomatic assessment of the patients (PANSS, Kay et al., 1987, TLC, Andreasen et al. 1979 for the schizo-phrenic and HAMD, Hamilton, 1960 and ERD, Widlöcher, 1983 for the depressive patients) made it possible to investigate the specificity of these difficulties in metaphor understanding amongst patients with schizophrenia, together with the effect of the severity of the symptomatology on the response profiles.

The results reveal that the literality bias and concreteness bias influenced the interpretation of metaphors in both groups of patients when compared with the control subjects. Furthermore, the results reveal a common bias towards literal responses (11%) and to concrete responses (4%) among both the patients with schizophrenia and those with depression. An important finding of our study is the heterogeneity of the performance observed in the schizo-phrenic and the depressive patients. Amongst the patients with schizophrenia, erroneous metaphor interpretation was influenced by the severity of the formal thought disorders (Andreasen et al., 1979), whereas in the patients with depression, it depended on the severity of the depressive symptomatology (Hamilton, 1960) and the psychomotor-retardation (Widlöcher, 1983).

This study represents a preliminary stage in studying metaphor understanding among patients with schizophrenia and major depression, and addresses new questions for further research, which may enhance exploration of the cognitive bases of these disorders.