Changes in semantic uses of first person pronouns as possible linguistic markers of recovery in psychotherapy

Objective: To examine changes in linguistic markers in the course of psychotherapy, drawing on Frege's logic of relations to define semantic variables distinct from syntactic variables. Method: From a sample of 73 patients, 10 patients with the best and 10 patients with the worst outcomes were selected. Forty transcribed sessions of each outcome group were compared statistically for change between commencement and termination of psychotherapy in: (i) the syntactic usage of first person pronouns ('I', 'me', 'we', 'us', 'implied I', 'implied me'); (ii) semantic usage of first person pronouns (expressing alpha, omega, or unclear positions); and (iii) non-pronoun linguistic variables (passive voice, negative, copula, auxiliary verbs expressing a sense of obligation). Results: There were no significant differences between the best and worst outcome groups in the change of either syntactic usage, or of the non-pronoun linguistic variables. However, the outcome groups differed significantly in the change of their semantic usage (alpha: p=0.002; omega: p=0.028): The best outcome group showed an increase of alpha positions and a decrease of omega positions, whereas the worst outcome group showed the inverse (i.e. decrease of alpha and increase of omega positions). Conclusions: Results suggest only semantic, that is meaning-driven, usage of first person pronouns marks recovery in the course of psychotherapy. If replicated, this finding could be used to monitor treatment responses. Replication in other kinds of treatment could mean these semantic changes are markers of recovery more generally than in psychotherapy