Innovative technologies in psychiatry education: a systematic review

Abstract

Introduction

The rapid integration of innovative technologies into healthcare education has reshaped how medical training is delivered. While simulation-based learning, virtual reality (VR), and gamified platforms are well established in other specialties, their use in psychiatry education remains limited. Given the increasing demand for interactive methods that suit digitally oriented learners, this review evaluates the effectiveness of these technologies in psychiatry training.

Method

A systematic review was conducted in accordance with the PRISMA guidelines. We searched PubMed, Scopus, ScienceDirect, Cochrane, and PsycINFO to identify eligible randomized and non-randomized trials up to May 2025. Inclusion criteria required studies involving medical students, residents, or physicians using technology-based interventions such as VR, virtual patients (VPs), serious games, high-fidelity simulation (HFS), or mobile applications. We included only studies that reported quantitative educational outcomes (e.g., knowledge, skills, diagnostic accuracy, confidence). We assessed risk of bias using the RoB2, ROBINS-I, and JBI tools, depending on the study design.

Results

Out of 2,005 initial records, nine studies met the inclusion criteria. Interventions included VPs (n = 4), HFSs (n = 2), comprehensive clinic simulators (n = 1), VR simulations (n = 1), and gamified mobile microlearning (n = 1). Most studies demonstrated positive short-term effects on knowledge, diagnostic accuracy, skills, confidence, and empathy. Virtual reality simulations reduced stigma and increased empathy toward patients with psychosis, while high-fidelity simulations significantly improved procedural skills such as electroconvulsive therapy (ECT). However, evidence of sustained learning and long-term outcomes was inconsistent. Risk of bias varied from low to moderate across studies, with methodological heterogeneity limiting comparability.

Conclusion

Innovative technologies show promising potential in psychiatry education by enhancing engagement, clinical reasoning, and skill acquisition. Virtual patients support diagnostic training, high-fidelity simulations strengthen procedural skills, and VR interventions can positively influence attitudes such as empathy and stigma reduction. Gamified mobile applications provide scalable, accessible tools for reinforcing psychiatric knowledge. Nevertheless, findings remain mixed due to variability in study design, small sample sizes, and limited follow-up. Future research should focus on large-scale, well-controlled trials that assess long-term effectiveness, cost-efficiency, and integration into curricula. Overall, innovative technologies represent a valuable complement to traditional psychiatry education and warrant broader adoption and evaluation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12909-026-09573-1.

Introduction

As technological advancements continue to make significant changes in the healthcare landscape, the integration of innovative technologies in the healthcare training system has become increasingly essential [1–3]. Many residents and healthcare trainees have been raised in a digitally oriented environment, leading to a strong preference for technology-based learning methods. Therefore, it is crucial to utilize these innovative technologies to improve the efficacy and relevance of healthcare education [4, 5].

The application of new technologies within the medical field is not a novel concept. Various medical specialists, including surgeons, obstetricians, and critical care professionals, have used simulation technology as a standard tool to assess clinical competency in their residency training programs. The range of technologies employed has varied from sophisticated, “high-fidelity” simulations with human-sized computerized mannequins or virtual reality (VR) environments [6, 7] to simpler, “low-fidelity” simulations, such as using porcine eyes for the training of specific procedures to medical students [8, 9]. Despite recent advancements, the adoption of new technologies in the training of medical students, residents, and postgraduate psychiatry trainees remains rare [10, 11].

Innovative technologies, including VR, VPs, serious games, mobile applications, and web-based simulation platforms, have shown significant efficacy in medical training. Employing these technologies enhances experiential learning, clinical decision-making, and skill acquisition. Such approaches often provide a level of interactivity and immersion that proves to be better than traditional educational methods, thereby providing a more engaging and effective learning environment for medical professionals [12–15].

Research shows that virtual reality (VR) can significantly enhance knowledge and skills compared to traditional educational methods and various forms of digital education, whether online or offline [12]. Virtual reality can increase the likelihood of users experiencing a heightened sense of presence compared to non-immersive or less interactive educational modalities, such as traditional text-based instruction, video-based learning, or standard computer-based training [16]. Research indicates that immersive experiences can enhance visual memory, provide realistic environments, and increase engagement, along with eliciting feelings of excitement and joy [17, 18].

The integration of gaming and serious games -defined as digital games specifically designed for the educational purpose of enhancing medical knowledge of health professionals- has shown significant benefits for medical trainees. These interactive platforms provide active learning by immersing learners in clinical problem-solving scenarios, allowing them to acquire practical experience in a risk-free environment [19]. Engagement with serious games allows learners to develop valuable abilities, including analytical skills, strategic thinking, knowledge acquisition, multitasking abilities, decision-making proficiency, effective communication, and psychomotor skills. Additionally [20, 21], , when these games incorporate multiplayer functionalities, they increase collaborative learning experiences and enhance teamwork and interpersonal skills among participants [22].

Virtual patients (VPs) are interactive, computer-based clinical scenarios that provide clinical students with valuable opportunities for repeated, safe practice. These simulations allow learners to encounter a variety of clinical situations in a controlled environment, enhancing their diagnostic and decision-making skills without the risks associated with direct patient interactions [13]. Utilizing VP technology offers several significant advantages over traditional standardized patient methodologies. One of the primary benefits is the reduction of training costs. Additionally, VP technology enables remote training, giving students access to learning modules from anywhere. This flexibility is especially beneficial for institutions with diverse student populations or those located in rural areas, where access to standardized patient scenarios may be limited. Furthermore, the ease of organizing training sessions with VPs enhances the logistical efficiency of clinical education programs [23, 24].

Simulation technologies also offer significant opportunities for evaluating skills and enhancing clinical practice. When employed wisely, these technologies serve as a valuable complement to assist educators tasked with assessing physician competence [25].

While innovative technologies offer numerous benefits in medical education, psychiatry presents unique educational challenges compared to other medical fields. This includes variability in patient availability and limited exposure to specific conditions during routine clinical training. These factors may influence how emerging educational technologies are applied and evaluated in psychiatric education. As previously noted, unfortunately, the integration of innovative technologies into the training of psychiatry students has been limited. This study aims to shed light on potential applications of such technologies for psychiatry residents and medical students in the field of psychiatry. By investigating various technological advancements, we seek to identify effective strategies that can enhance educational outcomes, improve clinical skills, and offer a deeper understanding of psychiatric principles. Our goal is to contribute to the ongoing discourse regarding the modernization of psychiatric education and to support the adoption of these tools to better prepare future practitioners for the complexities of mental health care.

Methods

This study followed the guidelines established by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [26] and is registered in PROSPERO repository (registration ID: 1119120).

Eligibility

We included only studies published in English conducted exclusively with medical students, residents, and general physicians. We systematically excluded studies involving nursing, psychology, or other non-medical student populations from consideration in this review.

We considered interventions that involved technology-based or simulation-based innovations such as VR, mobile applications, VPs, gamification, and HFSs. Eligible studies included randomized controlled trials, non-randomized controlled trials, and uncontrolled before–after studies, provided that they reported at least one quantitative educational outcome such as knowledge, confidence levels, and procedural skills. These outcomes could be assessed using pre–post comparisons, between-group comparisons, or both. We excluded studies reporting only qualitative outcomes. This restriction was applied to ensure consistency in outcome measurement and to enable objective comparison of the effectiveness of different interventions. Qualitative and mixed-method studies were excluded because their findings, while valuable for understanding user experiences and implementation contexts, are less suitable for direct comparison of educational outcomes across heterogeneous interventions.

Search strategy and study selection

We conducted a search across various databases, including PubMed, Scopus, ScienceDirect, Cochrane, and PsycINFO, in collaboration with an expert research librarian. We aimed to identify randomized trials published from database inception through May 2025. Search terms combined controlled vocabulary (MeSH) and free-text keywords related to psychiatry, education, and innovative technologies (e.g., “psychiatry,” “residency,” “virtual reality,” “simulation,” “serious games,” “mobile applications”). We applied Boolean operators to construct search strings tailored to each database. The full search strategies for all databases are provided in Supplementary Appendix 1. The flowchart detailing the search process, including the steps taken to filter and select studies, is presented in Fig. 1, and the search process and the terms employed in our systematic review are shown in Table 1.

Fig. 1.

Flowchart of study selection

Table 1.

Search strategy and key words

Question	Keywords
Participants: Medical students, residents, and general physicians	education [Title/abstract] residents [Title/abstract] residency [Title/abstract] Psychiatry* [Title/abstract]
Intervention: Innovative technologies	Web-based [Title/abstract] blended learning [Title/abstract] Educational Technology [Mesh Terms] Computer-Assisted Instruction [Mesh Terms] Virtual Reality [Title/abstract] VR [Title/abstract] Smartphone [Title/abstract] Application Software [Title/abstract]
Comparison: Same group before the intervention	trial [Title/abstract] clinical trial [pt]

The initial search and review were conducted by a librarian. After duplicate removal, two independent reviewers (Author 1 and Author 2) conducted study screening, eligibility assessment, and data extraction. Disagreements at all stages were resolved through discussion and, when necessary, adjudication by a third senior reviewer (corresponding author).

Risk of bias assessment

The methodological quality of the included studies was assessed using appropriate, study-design–specific tools. Randomized controlled trials were evaluated using the Cochrane Risk of Bias 2.0 (RoB 2) [27] tool across five domains: randomization process, deviations from intended interventions, missing outcome data, measurement of outcomes, and selection of reported results. Non-randomized controlled trials were appraised using the Risk of Bias in Non-randomized Studies – of Interventions (ROBINS-I) tool [27], which classifies risk as low, moderate, severe, or critical. Uncontrolled before–and–after studies were assessed with the Joanna Briggs Institute (JBI) [28] critical appraisal checklist for quasi-experimental studies, which evaluates internal validity across eight items (e.g., cause–and–effect clarity, multiple measurements, complete follow-up, and statistical appropriateness).

Detailed domain-level judgments are provided in the Supplementary Appendix 2.

Results

Study characteristics

The database search identified 2,005 records. After removing 93 duplicates, 1,912 studies remain for screening. From the remaining records, 1,882 studies were removed after evaluating titles and abstracts. From the 30 full-text articles, nine studies met the inclusion criteria and were eligible for the study (Fig. 1).

The final set of nine eligible studies examines the use of innovative technologies in psychiatry education. Seven studies were controlled trials with pre- and post-measurements, and two studies lacked a control group. Five studies assessed the psychiatry training of medical students and physicians, while four studies focused on the education of psychiatry residents. Studies reported a variety of outcome variables, including knowledge, diagnostic performance, procedural proficiency, skill, communication, and confidence.

Technologies included HFS (n = 2), VPs and branched-narrative simulations (n = 4), avatar-based simulations (n = 1), comprehensive clinic simulators with integrated electronic medical record (EMR) (n = 1), VR simulation (n = 1), and mobile-based gamified microlearning (n = 1). The details of the studies are described in Table 2.

Table 2.

The details of the studies

Study (Year)	Country	Design	Participants	Intervention	Outcome	Risk of Bias	Key Finding
Satter (2012) [28]	Canada	Non-randomized controlled	14 PCPs (7 VP, 7 control)	Avatar-based simulation vs. traditional text	Diagnostic accuracy (MDD/PTSD)	Moderate (ROBINS-I)	↑ Diagnostic accuracy with VP
Rabheru (2013) [29]	Canada	RCT	19 psychiatry residents (10 VP, 9 control)	HFS vs. traditional training	Skills (ECT-OSATS); Knowledge & confidence	Low (RoB 2)	↑ Procedural skills; no difference in knowledge, confidence
Foster (2015) [30]	USA	RCT	67 2nd-year med students (34 VP, 33 video)	Interactive VP vs. video module	Suicide risk & bipolar assessment, SP communication	Low (RoB2)	No significant difference
Pantziaras (2015) [25]	Sweden	Before–after (uncontrolled)	32 psychiatry residents	Refugee Trauma VP (RT-SIM)	Knowledge & confidence	Moderate (JBI)	Immediate gains in knowledge, confidence (not sustained)
Wilkening (2017) [31]	USA	Before–after (uncontrolled)	18 3rd-year psychiatry residents	Branched-narrative VP	Case knowledge	Moderate–High (JBI)	↑ Knowledge in 3 VP cases; plateau in review
Matsumura (2018) [32]	Japan	Non-randomized controlled	79 5th-year med students (36 VP, 43 control)	Comprehensive psychiatry clinic simulator	Knowledge (dementia care)	Low (ROBINS-I)	↑ Knowledge compared to traditional training
Raysin (2018) [33]	USA	Block-RCT	35 psychiatry residents (18 HFS, 17 control)	HFS vs. traditional training	Skills, confidence	Low (RoB 2)	↑ Knowledge, procedural skills, confidence
Zareh-Bidaki (2022) [34]	Iran	Cluster RCT	144 2nd–3rd year med students (72 VR, 72 control)	VR simulation of psychosis (VRSP)	Knowledge, empathy, stigma	Some concerns (RoB 2)	↑ Knowledge & empathy, ↓ stigma
Zolfaghari (2023) [20]	Iran	Quasi-experimental	72 med students (27 mobile, 35 control)	Gamified microlearning (Roozbehyar app)	Knowledge exam	Low (ROBINS-I)	↑ Written exam scores; no change in global assessment

PCP Primary Care Physician, VP Virtual Patient, MDD Major Depressive Disorder, PTSD Post-Traumatic Stress Disorder, RCT Randomized Clinical Trial, HFS High-Fidelity Simulation, ECT Electroconvulsive Therapy, OSATS Objective Structured Assessment of Technical Skills, SP Standardized Patient, RT-SIM Refugee Trauma Simulation, VR Virtual Reality, EMR Electronic Medical Record, JBI Joanna Briggs Institute, ROBINS-I Risk Of Bias In Non-randomized Studies - of Interventions, RoB 2 Risk of Bias 2 tool

Virtual patient simulation

VP simulations were the most commonly used innovative technology in psychiatry education, appearing in four out of nine studies, with mixed results. Satter [28] reported that avatar-based VP cases improved diagnostic accuracy for MDD and PTSD among PCPs. In contrast, Foster [29] found no significant advantage of VP modules over video teaching for suicide risk assessment and communication skills in medical students. Pantziaras [25] demonstrated immediate but unsustained gains in knowledge and confidence after a refugee trauma VP simulation for psychiatry residents. Similarly, Wilkening [30] observed consistent short-term improvements in knowledge across three branched-narrative VP cases, though effects plateaued during a review session. Taken together, these findings suggest that virtual patient simulations can enhance diagnostic reasoning and confidence, but evidence for sustained learning effects remains limited. We will review all the eligible studies in chronological order in detail:

Satter et al. [28] utilized an educational approach that involved digitally simulated patients (avatars) to evaluate the diagnostic and therapeutic reasoning of primary care physicians (PCPs). The study compared this method to traditional paper-based case scenarios. The researchers randomized 14 PCPs into two groups: one group of seven received traditional text-based case descriptions as a control, while the other group of seven interacted with VPs through avatar-based simulations. Each group worked through four standardized scenarios involving major depressive disorder (MDD) and post-traumatic stress disorder (PTSD).

The findings indicated that the avatar group demonstrated greater diagnostic accuracy for both MDD and PTSD compared to the control group. Furthermore, participants in the avatar group were more likely to initiate appropriate treatment.

Foster et al. [29] conducted a randomized controlled trial to evaluate the effectiveness of an interactive VP simulation compared to a video-based teaching module for second-year medical students. They assigned 67 students to either the VP group (n = 34) or the video group (n = 33). Both methods aimed to enhance students’ skills in assessing suicide risk and identifying symptoms of bipolar disorder.

After the intervention, the researchers evaluated the students using standardized patient (SP) interviews. The students were scored based on their communication skills, rapport with the patient, and clinical accuracy. The results revealed no significant differences between the two groups across several outcome measures, including suicide risk assessment scores, accuracy on the bipolar symptom checklist, SP communication checklist, and rapport ratings. For example, the p-values were 0.16 for communication (Effect size measured with Cramer’s V: 0.03) and 0.40 for rapport (Effect size measured with Cramer’s V: 0.01).

The study concluded that while students appreciated the VP simulation, it did not demonstrate greater educational effectiveness compared to traditional video learning.

Pantziaras et al. [25] examined the effectiveness of the Refugee Trauma Simulation (RT-SIM) in training psychiatry residents to assess and manage trauma-related disorders in refugee groups. Thirty-two residents took a knowledge test before the intervention, used the VP system, and then completed a post-test. The study found a notable improvement in knowledge scores immediately after the simulation, but this improvement did not persist in the follow-up evaluation. Their confidence in handling culturally sensitive cases also increased significantly across various areas, particularly in trauma-related assessments.

Wilkening et al. [30] conducted a pilot trial with 18 participants, without a control group, on an interprofessional learning series that utilized branched-narrative VP simulations for third-year psychiatry residents. The VP cases were designed to simulate outpatient psychiatric encounters and focused on advanced psychopharmacological decision-making. Each case presented learners with clinical scenarios including patient history, symptoms, laboratory findings, and prior treatments, requiring them to make treatment decisions. The branched-narrative structure allowed different clinical outcomes based on learner choices, with suboptimal decisions leading to negative patient outcomes and expert feedback, and appropriate decisions resulting in positive outcomes with explanatory feedback. The sessions were integrated into a pharmacist-led educational program and followed by structured debriefing. Pre- and post-test assessments were given for each case to measure learning outcomes.

Results showed that knowledge scores increased significantly from pre- to post-assessment for the first three VP cases (p < 0.05). However, the fourth review module did not show further improvement, possibly due to a ceiling effect or a plateau from repetition. Importantly, participant satisfaction was consistently high, with all residents reporting a positive experience with the branched-narrative format and pharmacist review.

High-Fidelity Simulation (HFS)

High-Fidelity simulation in psychiatry education involves immersive, realistic training environments that imitate clinical settings. These environments use actors, detailed manikins, or simulation technology to recreate complex psychiatric scenarios. Unlike traditional teaching methods, HFS allows learners to practice hands-on, improve their procedural skills, and receive structured feedback in a safe, controlled setting. This review looks at two studies that used HFS to teach electroconvulsive therapy (ECT) techniques and psychiatric emergency procedures.

Rabheru et al. [31] conducted a randomized controlled trial to evaluate the effectiveness of HFS in teaching ECT to psychiatry residents. Nineteen residents were randomly assigned to either the HFS group (n = 10) or a traditional training group (n = 9). Both groups completed pre- and post-tests to assess their knowledge and confidence. Blinded observers independently evaluated their procedural competence using the ECT-OSATS (Objective Structured Assessment of Technical Skills) checklist.

The high-fidelity simulation intervention involved human patient simulation (HPS)-based training for electroconvulsive therapy (ECT), conducted in a simulated clinical environment. Residents practiced procedural steps such as scalp preparation, electrode placement, and interpretation of electroencephalographic activity using manikins, while also engaging in team-based interactions with actors representing anesthesia and nursing staff. The training emphasized both technical and non-technical skills, including communication and professionalism.

The residents in the simulation group scored significantly higher on the ECT-OSATS global rating, with all 10 trainees passing the assessment. In contrast, only one out of nine in the control group passed (p < 0.001, Cohen d = 2.8). However, no significant differences were detected between the groups in knowledge scores (p = 0.36) or self-reported confidence (p = 0.21). The results show that while HFS greatly improves hands-on technical skills in ECT, it does not significantly increase theoretical knowledge or confidence compared to traditional teaching methods.

Raysin et al. [32] evaluated the effectiveness of HFS for ECT training in psychiatry residents. Thirty-five psychiatry residents were randomly assigned to either simulation-based training (n = 18) or traditional didactic training (n = 17). High-fidelity simulation-based ECT training (SIM-ECT) was designed using a mannequin and ECT instrument by faculty educator. SIM-ECT covers the education of all of the steps involved in ECT including the pre-procedural discussion and education with the patient, the procedure itself, and the post procedure assessment of the patient.

The simulation group showed significantly greater improvements in ECT knowledge, procedural skills, and confidence compared to the control group. Knowledge scores increased from 51% to 69% (p = 0.02), and procedural skills improved from 22% to 51% (p < 0.001). Confidence and comfort levels also increased in the simulation group, highlighting the benefits of using HFS in psychiatry education.

Taken together, these studies suggest that HFS may be particularly effective in enhancing practical skills, although its impact on knowledge and attitudinal outcomes requires further research.

Comprehensive clinical simulation

Matsumura et al. [33] developed a web-based psychiatry clinic simulator to improve medical students’ clinical reasoning and knowledge in dementia care. The system included multiple components to simulate a realistic outpatient experience. This involved VPs, a branching scenario engine, and a simulated EMR interface. Students used these elements to perform assessments, order tests, and document findings in a manner that closely mimics actual clinical workflows. In a controlled study involving 79 fifth-year medical students, those who utilized the simulator (n = 36) demonstrated significantly greater improvements in their knowledge scores compared to the control group (n = 43), who received traditional clinical training (p = 0.01).

Virtual Reality (VR) simulations

Zare-Bidaki et al. [34] examined the effectiveness of a virtual reality simulation of psychosis (VRSP) in enhancing knowledge, empathy, and reducing stigma among second- and third-year medical students. In this randomized study, 144 participants were divided into two groups: one that experienced the VR simulation and another that participated in standard supervised inpatient visits after a lecture on positive psychotic symptoms.

The virtual reality (VR) experience lasted about 4.5 min. It placed students in a culturally contextualized Iranian apartment setting. This simulation included auditory hallucinations and delusional experiences, such as hearing persecutory voices and having delusions of reference. Participants completed questionnaires before and after the sessions to assess their knowledge, empathy, and stigma levels, both immediately after the first session and after a second exposure.

The results showed significant reductions in stigma (p < 0.001) and increases in empathy (p < 0.001), along with improvements in knowledge related to psychosis (p < 0.001) for the VR group compared to the control group. Notably, reported side effects were minor and temporary, with no lasting negative symptoms.

Mobile gamification

Zolfaghari et al. [20] evaluated the effectiveness of a gamified mobile microlearning app, Roozbehyar, in educating psychiatry clerkship students. In this quasi-experimental study, 72 medical students in their fifth or sixth year of training were assigned to either the experimental group (n = 27), which used the app, or the control group (n = 35), which received traditional education during their four-week psychiatry rotation. The Roozbehyar app used short modules, quizzes, and gamified challenges to reinforce psychiatry concepts.

At the end of the rotation, the intervention group had significantly higher scores on the final written exam (p = 0.031). However, global clinical performance assessments did not differ significantly between the two groups. Participants’ feedback was mainly positive regarding content quality and educational value. However, usability and structural aspects of the app were rated less favorably.

Risk of bias

The overall methodological quality of the included studies was variable (see Supplementary Appendix 3). Among randomized controlled trials, one study (Raysin 2018) [32] used block randomization, demonstrated excellent inter-rater reliability, and was judged to be at low risk of bias. Two other RCTs (Rabheru 2013; Foster 2015) [29, 31] raised concerns because they didn’t report their randomization in detail. However, their overall risk of bias was assessed as low. The remaining RCT (Zare-Bidaki 2022) [34] were rated with some concerns, chiefly due to incomplete reporting of sequence generation/concealment and reliance on self-report outcomes.

The three non-randomized controlled studies showed low to moderate risk of bias on the ROBINS-I scale. Matsumura (2017/2018) [33] and Zolfaghari [20] were limited by potential confounding from non-random allocation and were judged as overall low risk, while Satter [28] was rated at moderate risk of bias owing to participant allocation and reliance on subjective outcome coding.

The two uncontrolled before–and–after studies were appraised using the JBI checklist: Pantziaras [25] was rated as moderate quality due to the use of a single pre/post-test and reliance on self-report, while Wilkening [30] was rated as moderate–high quality because of repeated measures, consistent gains, and complete follow-up.

Taken together, these evaluations indicate that while several studies demonstrate promising educational effects, the overall strength of evidence is limited by methodological heterogeneity and risk-of-bias concerns.

Discussion

This systematic review identified nine trial studies examining the use of innovative technologies in psychiatry education. A range of technologies was identified, including virtual patient simulations, high-fidelity simulators, comprehensive clinical simulations, VR, and gamified mobile applications. The included studies varied in methodological quality and reported outcomes, highlighting both the promise and heterogeneity of the evidence base in this emerging field.

Psychiatry presents unique challenges for the application of innovative technologies. In clinical training, access to patients with severe mental disorders may be limited, particularly when individuals are unable or unwilling to participate in teaching. In addition, certain conditions, such as phobias or obsessive-compulsive disorder, are more commonly encountered in outpatient settings and may be underrepresented in inpatient-based education. Virtual patient simulations may help address these gaps by providing standardized exposure to a broader range of clinical scenarios. Across studies, different technologies appeared to support different educational domains, with virtual patients enhancing diagnostic reasoning, high-fidelity simulations supporting procedural skill acquisition, and VR and gamified tools contributing to engagement and attitudinal outcomes.

These results align with broader studies in medical education, which indicate that learners’ engagement, realism, and interaction increase motivation and improve information retention [35, 36]. Virtual patients and immersive simulations are examples of innovative technologies that provide standardized case scenarios and education in a safe environment. They minimize the impact of chance and expose learners to complex or uncommon clinical situations that they may not encounter during standard rotations. Even when such situations do arise, it can be challenging to educate learners in those moments effectively [37, 38].

Although most of the studies we reported generally showed positive results, not all were statistically significant or showed sustained benefits after the intervention. Variability in outcomes across studies may be explained by methodological heterogeneity, including differences in study design, sample size, follow-up duration, and comparator conditions. When comparing findings across study designs, a clearer pattern emerges. Randomized controlled trials (RCTs) generally reported more conservative or mixed results, with some studies showing no significant differences between technology-based interventions and active comparators such as video-based learning. In contrast, non-randomized and uncontrolled before–after studies more consistently reported positive effects on knowledge, confidence, and skills. This discrepancy may reflect the greater susceptibility of non-randomized designs to bias, including confounding factors, lack of control groups, and reliance on self-reported outcomes.

Another explanation for this inconsistency could be the role of different comparator conditions. Reviewing all the studies reveals that most studies with passive control groups, such as lectures and readings, show a clear benefit for educational technologies. Conversely, comparing active learning modalities, such as video tutorials, shows less impact on the use of innovative technologies in psychiatry education. This highlights the need for further research on new tools compared with effective existing interventions, rather than outdated or baseline methods.

The same effect can be observed across different intervention groups. In addition to differences in the types of technologies used, not all of them have the same level of maturity or design quality. It is difficult to determine which technology is superior, as not all of them provide detailed and comprehensive methodologies for their design. However, based on the available information, it is clear that their designs exhibit varying levels of maturity. This aspect is crucial because a tool’s effectiveness relies on its design, alignment with curricular goals, and ease of use for learners [13, 39].

Learners’ prior knowledge, educational levels, and contextual differences can affect the effectiveness of interventions. We cannot determine the exact differences in participants’ needs across training levels because the studies are highly heterogeneous. However, it is evident that these differences can greatly affect the outcomes. This effect is well established in the broader educational literature, including traditional non-technology-based learning environments. Prior knowledge has been identified as a major determinant of learning outcomes, often accounting for a substantial proportion of post-intervention performance. It influences learning through multiple mechanisms, including attention, comprehension, and cognitive load [40–42].

Another reason for the inconsistency across studies is the diversity of outcomes and measures. Outcomes varied widely across the included studies. They ranged from objective assessments, such as diagnostic accuracy and knowledge retention, to subjective measures, such as self-confidence and empathy. The tools used to assess these outcomes varied in validity and sensitivity. For instance, Zolfaghari et al. (2023) [20] reported improvements based on clinical examination scores and feedback forms. In contrast, Raysin et al. [32] used structured rating scales to evaluate behavioral performance during a simulated suicide risk interview. This diversity complicates direct comparison and may partly explain the inconsistent results.

To improve comparability and strengthen the evidence base, future studies may benefit from standardized assessment tools. Validated assessment methods such as the Objective Structured Clinical Examination (OSCE) [43] and Objective Structured Assessment of Technical Skills (OSATS) [44] offer reliable and reproducible ways to evaluate clinical competence and procedural performance. The use of such standardized tools could enhance consistency across studies and facilitate more meaningful comparisons of educational interventions in psychiatry.

Overall, the evidence suggests that innovative technologies may enhance psychiatric education when appropriately aligned with specific learning objectives. However, effects vary depending on the type of intervention, educational context, and outcome measure. Virtual patients appear most effective for supporting diagnostic reasoning, high-fidelity simulations for procedural training, and immersive technologies such as VR for attitudinal outcomes. Gamified and mobile-based approaches may be particularly useful for engagement and early-stage learning, although the overall strength of evidence remains limited by methodological variability.

Placing these findings within established educational frameworks helps to contextualize their significance. According to Kirkpatrick’s model [45], most studies demonstrated effects at Level 2 (learning), including improvements in knowledge, skills, empathy, or confidence. Only a few studies reported on Level 1 (learner satisfaction), while none extended to Levels 3 (behavior change in clinical practice) or 4 (patient outcomes). Similarly, when mapped to Miller’s pyramid of clinical competence [46], most interventions addressed the ‘knows’ and ‘knows how’ levels through knowledge tests and diagnostic reasoning tasks. HFSs and immersive VR achieved the ‘shows how’ level by enabling learners to demonstrate skills in realistic clinical scenarios. However, none of the included studies evaluated the ‘does’ level of actual clinical performance. These frameworks highlight both the promise of innovative technologies for intermediate educational outcomes and the critical gap in evidence for their long-term impact on professional practice and patient care.

While this review offers valuable insights into the current state of using innovative technologies in psychiatry education, several limitations should be noted. The number of eligible studies was small, and the methodological and design heterogeneity made meta-analysis impossible. Additionally, publication bias may have favored studies that reported positive results, while negative or null findings might be underrepresented.

Future research should focus on large-scale, well-controlled studies that compare innovative technologies to active learning interventions, assess long-term retention, and explore the cost-effectiveness and scalability of these interventions. Clear frameworks and detailed methodological design guidance are needed to enhance the maturity of these technologies and to instruct other researchers in their use.

Conclusion

This systematic review highlights the potential of innovative technologies in psychiatry education, including virtual patients, high-fidelity simulations, comprehensive clinical simulations, VR, and gamified mobile applications. These tools could improve knowledge, diagnostic reasoning, procedural skills, confidence, and empathy in medical students and psychiatry residents.

Despite these promising findings, the overall evidence is limited by methodological heterogeneity, small sample sizes, and variable follow-up periods. While several studies reported positive effects, these findings were more consistently observed in studies with moderate risk of bias or less rigorous designs, whereas studies with lower risk of bias demonstrated more variable outcomes, including both significant improvements and null effects.

The overall evidence remains limited by methodological heterogeneity and small sample sizes, with variability across study designs and outcome measures. Innovative technologies show promise as adjuncts to traditional psychiatry education, but stronger evidence from well-designed studies is needed to confirm their effectiveness and guide implementation.

Authors' contributions

Conceptualization: SM, SVS and MS; Data curation: SM, AM, and MS, Formal analysis: SM; Supervision: ANA, SVS, and MS; Writing – original draft: SM, and writing – review & editing: AM, SVS, ANA and MS. All authors read and approved the final manuscript.

Funding

The authors disclose receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by Iranian National Agency for Strategic Researches in Medical Sciences Education (NASR) to MS (Grant no: 4020282).

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Because of the type of article (Review), there was no need for obtaining ethical approval.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.