📄 medicine

Chatbot versus Actor-Based Simulation for Sexuality Communication Training in Occupational Therapy Education

This study found that while actor-based simulations offer greater authenticity, chatbot-based simulations provide a feasible, low-cost, and less stressful alternative for occupational therapy students to practice sensitive sexuality communication, with both modalities effectively enhancing professional self-efficacy.

Original authors: Hagit Meishar-Tal, Osnat Burak, Gili Hoter Ishay

Published 2026-07-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Hagit Meishar-Tal, Osnat Burak, Gili Hoter Ishay

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are training to be a superhero, but your superpower isn't flying or shooting lasers; it's the ability to talk to people about the most awkward, embarrassing, and personal topics imaginable. In the world of healthcare, specifically for Occupational Therapists (the pros who help people get back to doing the things they love, like dressing, working, or playing), there is a huge gap in their training. They learn how to fix broken bones and help with daily tasks, but they often feel totally lost when it comes to discussing sexuality. It's a normal part of being human and staying healthy, yet many students feel their faces turn red just thinking about asking a patient, "How is your sex life?"

To fix this, schools use simulations. Think of these like flight simulators for pilots. Instead of crashing a real plane, you practice in a safe, fake environment. Usually, this involves hiring actors to play the patients. You talk to them face-to-face, and they react with real emotions, awkward silences, and body language. It's the "gold standard" because it feels real, but it's also expensive, hard to schedule, and for shy students, it can be terrifyingly stressful. Recently, a new tool has entered the arena: the Chatbot. This is an AI program you type to, like a super-smart text message friend. It can pretend to be a patient, answer your questions, and even tell you how it "feels" (e.g., "I am looking down at my hands"). The big question researchers wanted to answer was: Can a robot text conversation teach you how to handle these awkward human moments as well as a real actor can?

This study, conducted by researchers at the Holon Institute of Technology and Ono Academic College, decided to put this to the test with 24 occupational therapy students. They split the group into two teams: one practiced their "awkward talk" with a trained actor, and the other practiced with a text-based AI chatbot. Both groups had to handle a scenario involving a patient with bipolar disorder who was struggling with sexual health issues. Afterward, the researchers interviewed the students to see how they felt, what they learned, and which method they preferred.

Here is what the students discovered. The Chatbot turned out to be a surprisingly powerful training tool, especially for the "beginner" phase. The students loved it because it felt like a safe zone. Since they were typing to a computer, not a human, the fear of making a mistake or looking silly vanished. One student noted that because it wasn't a "real person," they felt more willing to take risks and try different ways of asking questions. It was like practicing a speech in front of a mirror instead of a crowd; you can stumble, pause, and think for a long time without anyone judging you. The chatbot also allowed for repeated practice. Students could go back and try the conversation again and again, refining their words until they felt confident. It was also great for active learning, forcing them to think and write their own responses rather than just watching a video.

However, the Actor had its own superpower: Authenticity. While the chatbot was great for building confidence, the actor felt real. The students could see the actor's body language, their facial expressions, and the awkward silence in the room. The chatbot tried to help by typing out descriptions like "the patient looks down," but the students admitted it wasn't the same as actually seeing it with their own eyes. The actor simulation was the "final boss" level of training—it was more intense and stressful, but it felt like the real deal.

The study found that both methods helped the students feel more ready to handle these conversations in the real world. The chatbot was a fantastic, low-cost, and accessible way to get students over their initial fear and embarrassment. It acted as a gentle ramp-up, letting them practice the basics without the pressure of a human staring at them. But it wasn't a perfect replacement. The students saw the chatbot as a stepping stone, a necessary first step before they could face the real, messy, emotional complexity of a human-to-human conversation.

In the end, the researchers suggest that schools shouldn't choose one or the other. Instead, they should use the chatbot as a training wheel. Let students type out their awkward questions and get feedback from the AI first. Once they feel brave and know what to say, then send them to the actor for the real-life simulation. This combo could help future therapists become confident, compassionate, and ready to talk about the things that matter most, without the fear of embarrassment holding them back. The study didn't prove that robots are better than humans, but it did show that a little bit of artificial intelligence can go a long way in helping humans learn to be more human.

Technical Summary: Chatbot versus Actor-Based Simulation for Sexuality Communication Training in Occupational Therapy Education

Problem Statement
Sexuality is a critical dimension of health, well-being, and occupational functioning, yet it remains under-addressed in occupational therapy (OT) education and practice. Professionals and students frequently report discomfort, embarrassment, and a lack of confidence when initiating sexuality-related conversations with clients. While professional frameworks like the PLISSIT model exist, educational challenges persist in translating theoretical knowledge into practical communication skills. Traditional actor-based simulation (standardized patient simulation) offers a realistic environment for practice but is resource-intensive (requiring recruitment, training, and scheduling of actors) and may intensify emotional demands for students already anxious about sensitive topics. Conversely, conversational artificial intelligence (chatbots) has emerged as a potential alternative, yet empirical knowledge regarding their specific utility for training in sexuality communication—and how they compare to actor-based methods—is limited.

Methodology
The study employed a qualitative comparative design with an experimental component involving 24 undergraduate occupational therapy students (3rd and 4th year) from the 2025–2026 academic year.

Procedure: Following a six-hour workshop on sexuality in OT, students were randomly assigned to one of two simulation conditions based on the same clinical scenario (a patient with bipolar disorder and sexual functioning difficulties).
- Chatbot Condition ( $n=7$ ): Students engaged in a text-based interaction with an AI chatbot (CESURA.AI platform) portraying the client. The system provided written responses and descriptions of nonverbal cues (e.g., gaze, posture). Post-simulation, students received AI-guided reflective feedback.
- Actor Condition ( $n=8$ ): Students interacted face-to-face with a trained actor in a simulation center, experiencing embodied presence and spontaneous dialogue. Post-simulation, they received feedback from the actor and a facilitator.
Data Collection: Semi-structured interviews were conducted within 48 hours of the simulation with 15 participants (7 from the chatbot group, 8 from the actor group). Two participants initially assigned to the actor group requested reassignment to the chatbot condition due to intimidation and were excluded from the interviews.
Analysis: Transcripts were analyzed using thematic analysis (inductive and deductive) to identify recurring themes regarding learning experience, challenges, and effectiveness. A descriptive frequency count of themes across groups was also performed.

Key Results
Thematic analysis identified six primary themes: perceived educational value, active learning, emotional safety, repeated practice/feedback, authenticity, and professional competence.

Chatbot Simulation:
- Strengths: Participants perceived the chatbot as highly valuable for promoting active learning (71.4% of chatbot group vs. 37.5% of actor group) and enabling repeated practice (42.8% vs. 0%). It was notably effective in reducing stress and anxiety (57.1% vs. 0%) and providing time to think (71.4% vs. 0%). The text-based nature allowed students to refine wording and practice in a psychologically safe, non-judgmental space.
- Limitations: While some found it authentic, others noted a lack of visual anchors (e.g., facial expressions, physical appearance) and a sense that it was a "preliminary stage" rather than a full substitute for face-to-face interaction.
- Feedback: Structured feedback from the chatbot was highly valued (71.4%) for self-exploration and identifying strengths/weaknesses.
Actor-Based Simulation:
- Strengths: The actor condition was perceived as significantly more authentic (75% vs. 57.1%) due to embodied presence and spontaneous dialogue. It was universally associated with gains in professional competence (100% vs. 85.7%).
- Limitations: Some participants found the face-to-face encounter intimidating, leading to higher stress levels and, in two cases, a refusal to participate in the study.
Comparative Outcomes: Both modalities successfully increased professional self-efficacy. However, the chatbot excelled in lowering the barrier to entry for sensitive topics through reduced anxiety and the ability to pause and reflect, whereas the actor simulation excelled in replicating the complexity of human interaction.

Key Contributions
The study contributes to the literature by:

Empirical Comparison: Providing one of the first direct comparisons of chatbot-based versus actor-based simulation specifically for sexuality communication training in health professions.
Pedagogical Distinction: Demonstrating that chatbot simulation is not merely a low-cost substitute but a distinct pedagogical environment with unique affordances: specifically, the capacity to create a "psychologically safe entry point" that reduces the fear of embarrassment, thereby facilitating active learning and repeated practice.
Feedback Mechanism: Highlighting the value of immediate, structured AI-generated feedback in supporting reflective learning and competence building in sensitive clinical areas.

Significance and Claims
The authors claim that chatbot-based simulation is a feasible, accessible, and low-cost tool for the initial training of sensitive clinical communication skills. They posit that it should be viewed as a complementary tool to live actor-based simulation rather than a complete replacement. The significance lies in its potential to address the "pedagogical challenge" of sexuality training by lowering emotional barriers (stress, embarrassment) while maintaining educational rigor through active learning and feedback loops. The study suggests that integrating chatbots into curricula could expand opportunities for practice, particularly for students who may otherwise avoid such training due to discomfort, ultimately contributing to improved professional readiness in addressing client sexuality.

Limitations Noted
The authors modestly acknowledge that the small sample size ( $n=15$ for interviews) limits generalizability. The study relied on subjective self-reports immediately following the intervention and did not measure long-term retention or actual performance in real clinical settings. Future research is needed to validate these findings across larger, diverse populations and to assess the transfer of skills to real-world practice.

Technical Summary: Chatbot versus Actor-Based Simulation for Sexuality Communication Training in Occupational Therapy Education

More like this