||Debriefing has been identified as an essential element of experiential learning in high-fidelity human patient simulation (HPS); however, identification of best practices in conducting debriefing sessions is lacking in the literature (Dreifuerst, 2009, 2012; Issenberg & Scalese, 2008; Jeffries, 2005). Nonetheless, it has been proposed that more significant or deeper learning occurs when using a facilitated, structured debriefing rather than an unstructured approach (Fanning & Gaba, 2007). Debriefing with Good Judgment is a structured facilitated model of debriefing designed for use in the simulated health care learning environment and has used extensively with medical trainees, but its effectiveness has not been investigated with graduate nursing students (Rudolph, Simon, Dufresne, & Raemer, 2006; Rudolph, Simon, Rivard, Dufresne, & Raemer, 2007). This structured model of debriefing has three phases: reactions, understanding and summary. The understanding phase is guided by the advocacy-inquiry approach that facilitates uncovering participants' internal mental frames. These frames or mental models drive the participants' decisions and actions in the clinical simulation. The purpose of this descriptive experimental repeated measures study was to explore the effectiveness of a structured model of debriefing, Debriefing with Good Judgment, on acute care nurse practitioner (ACNP) students' reflective ability and perspective transformation. Reflective ability was measured with the Groningen Reflective Ability Scale (GRAS) and perspective transformation with the Learning Activities Survey (LAS). A pilot study was conducted with 21 subjects to evaluate study protocols. Upon evaluation of the pilot study minor procedural changes were made in the research protocol. The full descriptive experimental study was conducted with a convenience sample of 22 ACNP students who were randomly assigned to either a control or experimental group. The style of debriefing was the independent variable with the control group assigned "usual" style debriefing while the experimental group received Debriefing with Good Judgment. Over one year, participants completed four simulation days, each with three cases for a total of 12 cases. The GRAS was administered as a baseline measure and then at each simulation day for a total of five assessments. The LAS was administered at the end of the study and 14 students participated in the semi-structured follow-up interviews. No significant differences were found between groups on the variables of gender ( p =.427), race ( p =.500), Marital status ( p = .115), age ( p = .095), number of years as a RN ( p = .635), number of years in critical care ( p = .592) and number of quarters enrolled in school ( p = .118). The experimental intervention fidelity was evaluated by four blinded external raters who reviewed 16 randomly assigned debriefing videos using the DASH-R © to ensure that the style of debriefing was consistent between the two debriefers, stable throughout the study and congruent with the model, Debriefing with Good Judgment. The mean scores on the DASH-R © for the experimental group ranged from 5.88 to 6.38 while the control group scores ranged from 2.63 to 3.75. There was a statistically significant difference between the experimental and control groups on the outcome variable of the overall DASH-R © scores [ F (5, 10) = 26.5, p = .000; Wilks λ = .070; η p 2 = .93]. There was not a statistically significant difference between groups on reflective ability; [ F (278, 18.23) = 1.282, p = .289]. The interaction between time and group assignment was statistically significant [ F (2.787, 18.23) = 2.851, p = .049, η p 2 = .125]. The one-way ANOVAs revealed that experimental group's change in mean GRAS scores over time was a significant quartic pattern. The experimental group reported a statistically significant higher incidence of perspective transformation than the control group; X 2 (2) = 6.443, p = .040. Therefore, although there was not a difference between groups on the reflective ability, a difference in the pattern of change in the mean scores on the GRAS was noted for the experimental group and there was a significantly higher incidence of perspective transformation in the experimental group supporting the premise that although the GRAS did not detect a difference, there was a qualitative difference in the experimental group's reflection process. The qualitative data confirmed the quantitative findings and identified the themes of safety and thinking deeply in the experimental group. In this context, the participants described safety as a safe environment to speak up and make mistakes. Thinking deeply and being willing to explore their reasons or internal "frames" that drove their behaviors in simulation would not have been possible without feeling safe to take the risk to engage in that behavior. Conversely, the control group themes included a more technical type of reflection.