The majority of non-suicidal self-injury (NSSI) research has used self- or clinician-rated measures of behavior which (a) are subject to reporting biases, or (b) have limited use in experimental designs that could illuminate causal relationships. Laboratory-based behavioral tasks have therefore been developed to assess NSSI-related behaviors more directly. We reviewed the behavioral methods that have been developed to assess NSSI tendencies or behaviors over the past 30 years. Several categories of laboratory analogues were identified: NSSI-related stimuli (e.g., NSSI pictures, implicit association tasks, guided imagery), experimenter administered pain stimuli (e.g., cold, heat, pressure, shock, and blade), and self-selected pain stimuli (e.g., cold and shock). These behavioral methods assess various aspects of NSSI and all have distinct advantages and shortcomings. Overall, these approaches have made significant contributions to the field complementing self- and clinician-ratings.