Cognitive network science as a framework for detecting structural patterns and emotions in suicide letters

Communicating one's mindset means transmitting complex relationships between concepts and emotions. Using cognitive network science, we reconstruct the mindset around suicide as communicated in 139 genuine suicide notes. Despite their negative context, suicide notes are surprisingly positively valenced and their ending statements are markedly more emotional, i.e. elicit deeper fear/sadness but also stronger joy/trust and anticipation, than their main body. By using emotional states from the Emotional Recall Task, we "open the lid" of suicidal narratives and compare their emotional backbone against emotion recall in mentally healthy individuals. Supported by psychological literature, we introduce emotional complexity as an affective analogue of structural balance theory, measuring how elementary cycles (closed triads) of emotion co-occurrences mix positive, negative and neutral states in narratives and recollections. Both authors of suicide notes and healthy individuals exhibit less complexity and more emotional coherence than expected by chance. However, suicide narratives display higher complexity, i.e. a lower level of coherently valenced triads, than healthy individuals recalling the same states. Entropy measures identified a similar tendency for suicide letters to shift more frequently between contrasting emotional states. Our results demonstrate that suicide notes possess highly contrastive narratives of emotions, more complex than expected by null models and healthy populations.