Emotions are believed to be socially adaptive tools (Keltner & Kring, 1998); yet, emotion deficits are common in many individuals with mental illness (e.g., Kring, 2001). The implications of these deficits for mentally ill individuals may therefore include day-to-day social functioning as well as clinical outcomes. In an inpatient setting, much of the treatment is formally or informally based upon a therapeutic milieu (Peplau, 1989), which may affect its residents positively or negatively (e.g., Buehler, Patterson, & Furniss, 1966; Moos, Shelton, & Petty, 1973). Thus, both individual and environmental factors are likely to influence patient outcomes. The current study sought to examine the relation of individual factors (emotion expressivity), contextual factors (treatment milieu), and the person by environment interaction on patient functioning in a forensic inpatient population. Patient emotion expressivity (or the outward display of emotion) was a focus of investigation based on its adaptive function in social interactions and its relation to mental health outcomes. Patient-staff discrepancies on perceptions of the treatment milieu were of interest as a contextual factor, as smaller discrepancies have been associated with greater program "success." Further, treatment milieu discrepancies were explored as a moderating variable of the relation between patient emotion expressivity and functioning. Participants included 53 patients and 36 staff from a secure forensic hospital. Overall, results suggested that patient ratings of emotion expressivity predicted behavioral functioning, whereas staff ratings of patient emotion expressivity predicted clinical functioning. In both cases, greater emotion expressivity was generally related to superior functioning. Discrepancies on treatment milieu only moderated the relation between patient-rated emotion expressivity and behavioral functioning. Additional descriptive findings based on patient legal status, differences between the two programs studied, and differences between patient and staff perceptions were explored. Implications for educational and clinical interventions are discussed. Future directions include generalizing current findings to other measures of functioning and to longitudinal outcomes.