Though emotional processing deficits are often conceptualized as a core feature of psychopathy, the common assessment of these deficits using the percentage correct (or hit rate) on affect recognition tasks may not provide a full or accurate picture of facial affect recognition in psychopathic individuals. Signal detection theory (SDT) provides a more informative statistical approach by providing independent measures of perceptual sensitivity (d’) and willingness to report perceiving a signal or response criterion (c). The current study employed signal detection methods to test the predictions of the integrated emotion systems and hostile attribution bias perspectives, two theoretical perspectives that make specific predictions regarding facial affect recognition. These perspectives were tested in a sample of 280 adult male incarcerated offenders who were assessed for psychopathy using the Psychopathy Checklist-Revised and who completed a novel test of facial affect recognition presenting 324 digital morphs of faces reflecting systematic combinations of pixels from neutral and affective face images (displaying six different types of emotion) as expressed by four different actors. The findings were generally not consistent with either of these perspectives. Psychopathy was negatively associated with d` for anger. Results also indicated an effect of psychopathy on response criterion for fear and effects of psychopathy on response criterion for anger and surprise that were evident only for some actors. The implications of these findings are considered through the lenses of several theoretical perspectives, and theoretical and methodological limitations of the current study are considered.