Several authors agreed that suicide behavior come from a need to escape from unbearable psychological pain. Since Shneidman definition of psychological pain several authors proposed different terms and definitions for this construct. Psychological pain is already widely evaluated in psychopathology due to its translational power. However, due to the subjectivity of each one about psychological pain feelings we are far from reaching a consensus. The efforts to define the construct lead to the development of different methods to evaluate the construct. But, assessing a construct that we do not yet understand has the risk of jumping to conclusions when we try to use it for predictive value. With the aim of clarifying what psychological pain is for a patient, we recruit 99 patients with current or past major depressive disorder at the emergency psychiatric department of the academic hospital of Montpellier (CHU Montpellier). We ask patients to answer the question “What do you think is the definition of psychological pain?”. Participants complete a battery of questionnaires composed by the Childhood trauma questionnaire (CTQ), the physical and psychological pain visual analogue scale (PPP-VAS) and the beck depression inventory (BDI). The open question was evaluated by two-different raters using 30 predefined dimensions of psychological pain. Discrepancies were solved by a third person. Cohen´s kappa showed fair (.20-.40) to almost perfect (>80) inter-rater consistency in all the dimensions rated in 5 or more patients. The definitions of psychological pain converge with our results in terms of unpleasant feelings (Sadness [40.4%], Anxiety [37.4%]), loss of self (Narcissistic wound [22.2%], Inability to make a project [19.2%]), concomitant physical pain [35.4%], suicidality [16.2%] and the characteristics of psychological pain as something permanent [22.2%], overwhelming [32.3%] and subjective [25.3%]. Pre-conditions of pain like trauma [5%] or loss [0%] appears less consistently or even not appear. Results showed not differences in pain categories between suicide attempters and not attempters. Using the dimensions of psychological pain, we performed cluster analysis that showed three categories: one of physical pain, one based mostly on sadness and the third one of suicidality and loss of self. The clusters showed differences only for childhood trauma showing lower trauma in the category of suicidality. This results give light to the definition of psychological pain according to patients. In this sense, pre-conditions of pain did not appear in the majority of patients. Moreover, characteristics of pain classify patients in consistent clusters with psychopathological meaning. This may guide future research in mental pain construct evaluation.