Abstract
The analysis of epistemological beliefs underlying psychotherapeutic interventions has been largely neglected by research in psychotherapy, training in psychotherapy and psychology, and often by theorists of different clinical orientations. The main risks of this neglect are the unexamined adoption of the epistemology that is taken for granted by the culture and is largely inconsistent with the specifics of the object of study of psychological science; the reduction of intervention effectiveness due to the inconsistency between epistemology, theory, and practice; and the maintenance of the gap between research and practice due to the different epistemologies used by clinicians and researchers. This article discusses the scientific status of the computerized linguistic measures of the Referential Process when used for clinical and research purposes. Our claim is that these measures developed to test Wilma Bucci's multiple code theory don't represent an objective examination of the psychotherapeutic process, but rather a methodological option to guide the researcher and clinician in identifying the most plausible scientific hypotheses about the complex phenomenon of emotional communication between speakers. Comparison of data from different points of view (therapist, patient, external observer, computerized linguistic analysis, etc.) and in different contexts (therapies, psychological tests, everyday conversations, experimental situations, etc.) will be presented as promising and viable ways to examine the validity of the hypotheses based on the Referential Process theory.