The Effect of High-Fidelity Simulation vs. Simulation with Standardized Patients on the Development of Reflective Practice Among Medical Students

Abstract

Background: This study evaluated the impact of high-fidelity simulation versus simulation with standardized patients on the development of reflective practice among medical students. Methods: A randomized controlled trial design with both pre- and post-simulation assessments was adopted. Thirty-two final-year medical students were randomly assigned to two groups (Group 1: high-fidelity simulation (n = 16); Group 2: simulation with standardized patients (n = 16)). Each group participated in six sessions over the course of two months, including six identical scenarios for both groups. The Groningen Reflection Ability Scale (GRAS) was used to assess the participants’ reflection skills before and after the simulation. Data were analyzed using descriptive statistics, paired t-tests for within-group changes, and independent t-tests for between-group comparisons. Results: Reflection scores improved significantly from pre- to post-simulation across the combined sample (p < 0.05). Within-group analyses demonstrated statistically significant improvements in self-reflection (31.3 ± 7.11 vs. 36.8 ± 5.34; p < 0.001), empathic reflection (19.1 ± 4.68 vs. 20.6 ± 4.51; p = 0.020), and reflective communication (23.1 ± 5.11 vs. 25.5 ± 4.35; p < 0.001). Additionally, between-group comparison revealed that the high-fidelity simulation group attained a significantly higher total reflection ability score compared with the standardized patient group (91.8 ± 7.70 vs. 74.0 ± 11.55; p <0.001). Conclusions: Simulation practice, whether high-fidelity or with standardized patients, helps to improve students’ reflection. However, high-fidelity simulation was proven to be more effective than simulation with standardized patients. This study reinforces the use of simulation as a tool for developing reflective practice skills in medical training.

1. Introduction

Simulation has come a long way, from low-fidelity simulation to the integration of innovative technologies and the development of modern high-fidelity tools [1].

Simulation has made enormous progress in the last few years, and its use has become an essential part of the training of healthcare professionals [2]. The advanced technology currently used in simulation provides a variety of learning experiences that allow learners to immerse themselves in a controlled, safe and realistic environment [3]. High-fidelity simulation plays a crucial role in medical education by supporting the development of technical and non-technical skills [4].

This simulation method involves the use of computerized mannequins, which are capable of reproducing realistic physiological parameters and responses, thereby providing students with a high level of realism and interactivity [5]. High-fidelity simulation emphasizes a global approach centered on the patient and potential events, and therefore enables students to be better prepared for real-life situations [6]. The crucial objective of this technique is to recreate realistic clinical environments in order to enhance the learner’s skills [7]. In fact, simulation plays an important role in the development of non-technical skills.

Reflective practice is the essential cognitive skill required to develop both a therapeutic relationship and professional expertise for future doctors [8]. Reflective practice can be defined as a dynamic and cyclical process, including reflection, adjusting actions based on that reflection, and then applying newly adapted practices. In this context, reflection is an essential component, enabling individuals to examine their experiences, question their decisions, and identify areas for improvement. However, reflective practice is not limited to individual reflection; it also includes critical reflection and social and systemic reflexivity [9]. Reflective practice is widely recognized as a key educational strategy for enhancing medical students’ learning. It plays a crucial role in the development of clinical reasoning, emotional intelligence, and professional identity formation [10]. Reflection is one of the processes that enables students to explore their learning activities and thoughts, discover their mistakes, and then correct them [11]. Furthermore reflection can reduce anxiety and negative emotions, build confidence, and positively influence mental health, resulting in a stronger dedication to providing quality patient care [12]. Due to the recent integration of simulation in the Arab world, assessing medical student reflection remains a challenge [13].

Although simulation is widely recognized for fostering experiential learning and reflection among health professions students, most studies have examined simulation modalities in isolation or emphasized outcomes such as performance or technical skills. Research comparing high-fidelity simulation and standardized patient simulation has primarily targeted nursing students and prioritized outcomes other than reflective practice [14,15]. Consequently, direct comparisons between these modalities’ effects on reflective practice as a primary outcome remain scarce.

The main aim of this study was to evaluate the impact of high-fidelity simulation versus simulation with standardized patients on the development of reflective practice among medical students.

2. Methods

2.1. Study Design and Settings

This was an experimental study, with a randomized controlled trial design conducted in the simulation center of the Faculty of Medicine and Pharmacy of Agadir (FMPA), from October to December 2025.

The FMPA is a public institution of higher education, affiliated with Ibn Zohr University in Agadir, established in 2016. It has a capacity to serve 4500 students, with 4 amphitheaters, 6 lecture halls, 16 tutorial rooms, 16 practical laboratories, and a medical simulation center.

The simulation center “Agadir Sim” is equipped with high-fidelity mannequins and simulation labs tailored to a wide range of clinical scenarios. It supports both team-based interprofessional learning and individual technical skill training, within controlled and realistic environments.

2.2. Participants

Thirty-two final-year medical students were recruited through announcements at the simulation center to participate in a simulation training program. Participation was voluntary, and students who had previously completed training on the same scenarios were excluded. The sample size was determined by the availability of eligible students and scheduling constraints, which is common in educational research involving naturally limited populations. The eligible students were randomly allocated to one of two simulation methods, high-fidelity simulation training (HFS, n = 16) or standardized patient (SP, n = 16).

2.3. Randomization

After providing informed consent, participants were randomly assigned to two groups: Group 1, trained with a high-fidelity mannequin, and Group 2, trained through a conventional role-play method with a standardized patient (Figure 1). Each participant was assigned a unique number between 1 and 32. A random number generator (RAND) function in Microsoft Excel was then used to assign a random number to each participant. The list was sorted according to these random values, then the first 16 participants were assigned to Group 1 and the next 16 to Group 2. After that, using the same random number generator, each main group was subsequently subdivided into 6 sub-groups of 2 to 3 students. A researcher not involved in training conducted the randomization.

2.4. Instruments

The socio-demographic questionnaire, the Groningen Reflection Ability Scale “GRAS”, was the instrument used for this study.

The socio-demographic questionnaire prepared by researchers included age, gender, previous experience in simulation. The GRAS is a tool designed to measure reflection that includes 23 items, with three subscales including self-reflection (10 questions), empathic reflection (6 questions), and reflective communication (7 questions). It uses a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5) with a total score between 23 and 115. The items cover three aspects of personal reflection: self-reflection, empathic reflection and reflective communication. The GRAS scale includes five negative formulated items (3, 4, 8, 12, 17 and 21), which should be reversed when scoring [16].

2.5. Intervention

Each group participated in 6 simulation sessions with two themed medical scenarios, each focusing on a distinct medical subject, and each of them was addressed in three sub-themes (

Table 1. Simulation scenarios.

Scenarios	Themes
Shock scenarios
Scenario 1	Hypovolemic shock
Scenario 2	Septic shock due to acute pyelonephritis
Scenario 3	Hemorrhagic shock secondary to a stab wound
Acute respiratory distress scenarios
Scenario 4	Rapidly progressing pneumonia with respiratory decompensation
Scenario 5	Pulmonary embolism
Scenario 6	Severe acute asthma with respiratory distress

). All scenarios were developed from critical care situations, in collaboration with an ICU professor and a simulation instructor, representing cases frequently encountered in clinical practice. To ensure comparability between groups, the scenarios were designed with parallel learning objectives and the same clinical complexity to both modalities: HFS and SP.

During each session, three key stages were systematically followed: a briefing, a training session (either by HFS or SP), and then a debriefing (Figure 2).

Briefing: Before each simulation session, a structured briefing lasting 5 min was carried out with the students in each group. It included an explanation of the objectives of the session, a description of how the simulation activity would unfold, familiarization of the participants with the equipment and the roles of the facilitators, as well as the duration of the scenario.

Simulation running: The environment, theme, duration, instructions, flow, patient status and parameters, script, and patient responses were rigorously identical for both groups. Two different simulation methods were implemented:

-HFS (Group 1): The high-fidelity simulation sessions were carried out using the Gaumard HAL^® high-fidelity manikin. This simulator was handled by simulation-trained technicians along with an expert professor. The interaction between students and the simulation occurred through standardized responses delivered by the instructor, real-time voice communication via a voice-over system, and recognition of technical gestures using integrated sensors, all remotely supervised by the trainer.

-SP (Group 2): The role-play simulation session with a standardized patient (SP) involved three experienced residents who were trained consistently and realistically to play the role of a patient. These residents were trained by an experienced SP trainer.

Each simulation scenario lasted 20 min. Groups were subdivided into subgroups. Each subgroup participated in one scenario as physicians while the other students viewed a live video feed in an adjacent observation room. Each student participated in at least one case and observed the others. Each simulation session was led by an ICU/anesthesia professor and an ICU assistant professor.

Debriefing: A structured debriefing took place at the end of each simulation scenario in the dedicated room. Lasting 40 min, the debriefing encouraged reflective learning and was based on four phases:

• The emotional phase (reaction phase) gave students time to express their emotions and feelings, helping to offer a secure climate.
• The descriptive phase aimed to establish and understand the events that occurred during the simulation.
• The analysis phase was to explore the students’ actions, clinical decisions, and the reasons behind any errors made during the simulation, as well as any problems encountered.
• The synthesis phase aimed to summarize the key learning and teaching objectives, and to discuss the transfer of this learning to real clinical contexts.

A consistent protocol was followed by trained debriefers to minimize facilitator variability. To prevent contamination of information between the two groups during the simulation sessions, a strict protocol was implemented: each subtheme session was held one week apart. On the same day, the two groups took part in the session, with a 30 min interval to allow for a slight rearrangement of the simulation room. The second group waited in a separate area until the first group had left the room completely. A member of the research team supervised the entrances and exits to prevent any crossovers. Mobile phones were strictly forbidden throughout the sessions to prevent the dissemination or exchange of information between the groups.

2.6. Statistical Analysis

All data were analyzed using Jamovi software version 2.6.2.0, which is widely adopted in medical education research for its user-friendly interface and robust analytical tool. Descriptive statistics were used to summarize participant demographics and GRAS scores, presenting continuous variables as means and standard deviations, and categorical variables as frequencies. The internal consistency of the GRAS tool was confirmed with a high Cronbach’s alpha (α = 0.968), indicating excellent reliability in this context. Normality tests ensured appropriateness for parametric analyses. Paired t-tests were then used to assess within-group changes in GRAS scores from pre- to post-simulation, reflecting the impact of the educational intervention on reflective abilities. Comparisons between the two independent simulation groups were performed using independent-sample t-tests to assess the impact of simulation type on reflective capacity. Statistical significance was set at p < 0.05.

3. Results

3.1. Participant Characteristics

The study sample consisted of 32 participants equally divided into two groups: High-Fidelity Simulation (HFS) and Standardized Patient (SP). The mean age was 24.8 ± 2.00 years, with no significant difference observed between the HFS group (25.2 ± 2.20) and the SP group (24.5 ± 1.79; p = 0.34). The gender distribution was similarly balanced, with females representing 62.5% of the total sample and demonstrating no statistically significant difference between groups (HFS: 68.75%, 56.25%; p = 0.46). All participants had prior experience with simulation-based training, ensuring homogeneity across groups for this variable. These findings indicate comparable baseline demographic and experiential characteristics between groups, minimizing potential confounding factors related to age, sex, or previous experience with simulation. A summary of the findings is presented in

Table 2. The participants’ characteristics.

Characteristics		Total (n = 32)	HFS Group (n = 16)	SP Group (n = 16)	p Value
Age (mean ± SD)		24.8 ± 2.00	25.2 ± 2.20	24.5 ± 1.79	0.34
Sex—n(%)	Female	20 (62.5)	11 (68.75)	9 (56.25)	0.46
	Male	12 (37.5)	5 (31.25)	7 (43.75)
Previous simulation experience—n(%)	Yes	32 (100)	16 (100)	16 (100)

.

3.2. Comparison of Pre- and Post-Simulation GRAS Scores

The comparison of GRAS scores between pre- and post-simulation demonstrated a statistically significant improvement in the participants’ reflective abilities. Self-reflection scores increased from a mean ± SD of 31.3 ± 7.11 pre-test to 36.8 ± 5.34 post-test (p < 0.001). Similarly, Empathic Reflection showed a significant rise from 19.1 ± 4.68 to 20.6 ± 4.51 (p = 0.020), and Reflective Communication increased from 23.1 ± 5.11 to 25.5 ± 4.35 (p = 0.001). The total GRAS score also reflected a significant overall enhancement from 73.5 ± 15.93 pre-test to 82.9 ± 13.23 post-test (p < 0.001). A summary of the findings is presented in

Table 3. Scores comparison between pre- and post-simulation of all participants.

Items	Pre-Test (Mean ± SD)	Post Test (Mean ± SD)	p Value
Self-reflection	31.3 ± 7.11	36.8 ± 5.34	<0.001
Empathic reflection	19.1 ± 4.68	20.6 ± 4.51	0.020
Reflective Communication	23.1 ± 5.11	25.5 ± 4.35	0.001
Total score GRAS	73.5 ± 15.93	82.9 ± 13.23	<0.001

.

3.3. Comparison of GRAS Scores Across Simulation Types

The comparison of GRAS scores between the two simulation groups: the High-Fidelity Simulation group and the Standardized Patient group was assessed across three subdomains: Self-Reflection, Empathic Reflection, and Reflective Communication, as well as the total GRAS score. Scores showed significant increases in the HFS group compared to the PS group for Self-Reflection (40.6 ± 2.73 vs. 33.0 ± 4.52; p <0.001, Cohen’s d = 2.04, 95% CI [1.17–2.89]), Empathic Reflection (22.8 ± 3.56 vs. 18.3 ± 4.32; p = 0.003, Cohen’s d = 1.14, 95% CI [0.38–1.88]), Reflective Communication (28.4 ± 2.75 vs. 22.7 ± 3.77; p < 0.001, Cohen’s d = 1.72, 95% CI [0.89–2.53]), and total GRAS score (91.8 ± 7.70 vs. 74.0 ± 11.55; p < 0.001, Cohen’s d = 1.82, 95% CI [0.97–2.63]). A summary of the findings is presented in

Table 4. Comparison of post-test scores by type of simulation.

Items	Post-Test Group 1: HFS (Mean ± SD)	Post-Test Group 2: PS (Mean ± SD)	p Value	Cohen’s d	95% CI
Self-reflection	40.6 ± 2.73	33.0 ± 4.52	<0.001	2.04	[1.17–2.89]
Empathic reflection	22.8 ± 3.56	18.3 ± 4.32	0.003	1.14	[0.38–1.88]
Reflective Communication	28.4 ± 2.75	22.7 ± 3.77	<0.001	1.72	[0.89–2.53]
Total score GRAS	91.8 ± 7.70	74.0 ± 11.55	<0.001	1.82	[0.97–2.63]

.

4. Discussion

This study demonstrated a significant improvement in reflection scores post-simulation in both the High-Fidelity Simulation (HFS) and Standardized Patient (SP) groups. Overall, GRAS scores improved significantly from pre- to post-simulation among all participants, with notable increases in Self-Reflection, Empathic Reflection, Reflective Communication, and total scores. This finding is consistent with previous research, which reported similar improvements in overall reflective thinking score after video-assisted debriefing in simulation sessions [17].

Importantly, between group comparisons revealed significantly higher outcomes in the HFS group compared to the SP group across all GRAS subdomains and total scores, indicating that HFS is associated with higher reflective practice scores compared to SP in this study.

The sample’s homogeneity in terms of age, sex, and prior simulation experience supports the reliability of these findings by minimizing confounding demographic effects. Several factors may have contributed to the relatively higher reflective scores observed with HFS. Fundamentally, HFS provides an immersive, highly realistic clinical environment that closely simulates authentic patient care scenarios. This fidelity engages learners in active clinical reasoning and decision-making processes, fostering deeper cognitive involvement and reflection on clinical judgments and professional behaviors [15,18]. The layered complexity of HFS scenarios further compels learners to critically appraise their real-time actions, detect errors promptly, and refine their approaches, aligning closely with established theories and reflective learning that emphasize iterative self-evaluation [19].

Beyond immersion, HFS enhances psychological safety by offering a controlled setting where learners can safely make and learn from mistakes without fear of real-world consequences or judgment [20,21]. This reduction in stress and performance anxiety contrasts with SP interactions, where direct human presence and perceived evaluation by live actors may increase anxiety levels, potentially inhibiting reflective engagement. The safer learning environment provided by HFS enables learners to devote cognitive resources more effectively to metacognitive processes required for reflection. A cornerstone of reflective practice in HFS is the structured debriefing session delivered immediately post-simulation. Debriefing has been repeatedly validated as essential in stimulating metacognition and facilitating critical reflection on clinical reasoning and decision-making [22]. This reflective discourse allows learners to reconstruct their thought processes, integrate experiential insights, and consolidate learning. This creates a feedback loop that enhances self-awareness and professional growth.

Moreover, reflective journaling, often integrated into simulation curricula, adds an introspective dimension by allowing learners to process emotions, uncertainties, and emerging insights beyond the immediate simulation debrief, consistent with Dewey’s conceptualization of reflection as an active, continuous learning process [23]. Mezirow’s transformative learning theory further supports the premise that critical reflection, scaffolded by experiential activities such as HFS and debriefing, promotes reassessment and transformation of learners’ mental models, facilitating autonomous and adaptive professional development [24]. Importantly, HFS sessions have demonstrated improvements in learners’ confidence and self-efficacy, engendering a positive cycle wherein greater confidence enhances engagement in reflective practice and skill acquisition [25,26]. This contrasts with SP settings, where less immersive experiences and higher anxiety may limit confidence gains and thus reflective depth.

Taken together, these elements create an enriched learning environment in HFS. These features likely underlie the significantly higher GRAS scores across all reflective domains in the HFS group compared to the SP group. Despite these encouraging results, limitations include the small sample size, which restricts generalizability, and the short-term post-test design that precludes an assessment of sustained transfer of reflective skills into clinical practice. Additionally, although the Groningen Reflection Ability Scale (GRAS) is a validated and widely used instrument, its reliance on self-reported data may introduce response bias, as participants may overestimate their reflective practices. These limitations should be considered when interpreting the results.

Future studies could complement self-reported measures with objective or observational assessments of reflective practice to reduce potential response bias. Future research should also incorporate larger cohorts and employ longitudinal designs to evaluate lasting impacts on clinical competence. Investigators are also warranted to dissect the interplay among stress reduction, confidence building, and immersive fidelity in shaping reflective outcomes across varying simulation modalities. Overall, this study supports the pedagogical power of integrating high-fidelity simulation, structured debriefing, and reflective practice to cultivate critical reflective competence.

5. Conclusions

This study highlights the superior effectiveness of High-Fidelity Simulation (HFS) in enhancing reflective practice among medical students compared to Standardized Patient (SP) methods. By providing a highly realistic and immersive learning environment, HFS promotes active clinical reasoning, critical self-assessment, and deeper engagement in reflective processes. The combination of psychological safety, structured debriefing, and opportunities for reflective journaling collectively supports learners in developing essential reflective competence and self-efficacy. These results underscore the critical role of integrating HFS into medical training programs to foster both technical proficiency and professional growth, ultimately preparing learners for the complex demands of clinical practice.