Combining psychoanalytic concepts and computer science methodologies: an empirical study of the relationship between emotions and the Lacanian discourses

This research aimed to examine the interdisciplinary interaction between psychoanalysis and computer science, suggesting a mutually beneficial exchange. Indeed, psychoanalytic concepts can enrich technological applications that involve the human factor, such as social media and other interactive digital platforms. By providing deeper insights into the elusive, unconscious aspects of communication, psychoanalytic methods can enhance content-centric applications, including fake news detection and mental health diagnostics. Conversely, computer science, especially Artificial Intelligence (AI), can contribute quantitative concepts and methods to psychoanalysis, identifying patterns and emotional cues in human expression. In particular, this research aims to apply computer science methods to establish fundamental relationships between emotions and Lacanian discourses. These relationships are identified in our approach through empirical investigation and statistical analysis and ultimately validated through a theoretical (psychoanalytic) account. Notably, although emotions have been sporadically studied in Lacanian theory, a systematic, detailed investigation of their role is missing. Such a fine-grained understanding of the role of emotions can also make the identification of Lacanian discourses more effective and easier in practice. Our methods indicated the emotions with the highest differentiation power in the corresponding discourses; conversely, we identified for each discourse the most characteristic emotions it admits. We call this method Lacanian Discourse Discovery (LDD), and it simplifies (via systematizing) the identification of Lacanian discourses in texts. Although the main contribution of this study is inherently theoretical (psychoanalytic), it can also facilitate major practical applications in interactive digital systems. Indeed, our approach can be automated using Artificial Intelligence methods that effectively identify emotions (and corresponding discourses) in texts.