Toothbrush
Prison

Toothbrush Prison

This project explores a novel, experimental approach to improving adherence to home-based physical therapy.

Developed as my graduation project for the MSc in Interaction Technology, it investigates the potential of coercive behavior strategies that aim to guide behavior through structured limitations rather than persuasion. Research indicates that low adherence to home-based physical therapy is a substantial problem. Only 30 to 50 percent of patients perform their exercises as prescribed (Holden, 2014; Peek, 2025). Which has implications for treatment cost, related to extended and/or inefficient treatment. While the root of the adherence problem is complex, it was found that patients often want to adhere to their physical therapy regime, but they have difficulty implementing the required behaviors and then integrating them into their lives so they become routine (Bassett, 2012). This is known as the difference between intention and actual performance of behavior, the so-called intention-behavior gap. Existing tools offer limited support in helping patients bridge the gap between intention and behavior. While mobile apps can offer reminders and instructional guidance, they often fail to sustain behavior, especially when life becomes unpredictable or motivation dips occur.
Developed as my graduation project for the MSc in Interaction Technology, it investigates the potential of coercive behavior strategies that aim to guide behavior through structured limitations rather than persuasion. Research indicates that low adherence to home-based physical therapy is a substantial problem. Only 30 to 50 percent of patients perform their exercises as prescribed (Holden, 2014; Peek, 2025). Which has implications for treatment cost, related to extended and/or inefficient treatment. While the root of the adherence problem is complex, it was found that patients often want to adhere to their physical therapy regime, but they have difficulty implementing the required behaviors and then integrating them into their lives so they become routine (Bassett, 2012). This is known as the difference between intention and actual performance of behavior, the so-called intention-behavior gap. Existing tools offer limited support in helping patients bridge the gap between intention and behavior. While mobile apps can offer reminders and instructional guidance, they often fail to sustain behavior, especially when life becomes unpredictable or motivation dips occur.
Developed as my graduation project for the MSc in Interaction Technology, it investigates the potential of coercive behavior strategies that aim to guide behavior through structured limitations rather than persuasion. Research indicates that low adherence to home-based physical therapy is a substantial problem. Only 30 to 50 percent of patients perform their exercises as prescribed (Holden, 2014; Peek, 2025). Which has implications for treatment cost, related to extended and/or inefficient treatment. While the root of the adherence problem is complex, it was found that patients often want to adhere to their physical therapy regime, but they have difficulty implementing the required behaviors and then integrating them into their lives so they become routine (Bassett, 2012). This is known as the difference between intention and actual performance of behavior, the so-called intention-behavior gap. Existing tools offer limited support in helping patients bridge the gap between intention and behavior. While mobile apps can offer reminders and instructional guidance, they often fail to sustain behavior, especially when life becomes unpredictable or motivation dips occur.

This project explores the use of a commitment-based interactive system to sustain behavior during moments of low motivation or competing life priorities.

This project explores the use of a commitment-based interactive system to sustain behavior during moments of low motivation or competing life priorities.

This project explores the use of a commitment-based interactive system to sustain behavior during moments of low motivation or competing life priorities.

In the context of this project, a commitment-based system deliberately restricting future choices. An example would be that if you want to avoid using your phone in bed, placing it in another room before going to sleep creates a physical barrier that reinforces your intention. By adding friction to undesired behavior, such devices can make the undesired behavior less likely to occur. The act of placement is a commitment-based method o achieve the behavior. This project focused specifically on rehabilitation following a rotator cuff tear, a common shoulder injury for which long term home-based physical therapy is viewed as a crucial part of recovery, however adherence is often low (Burns et al., 2021). Some examples of physical therapy exercises for this condition can be viewed below.
In the context of this project, a commitment-based system deliberately restricting future choices. An example would be that if you want to avoid using your phone in bed, placing it in another room before going to sleep creates a physical barrier that reinforces your intention. By adding friction to undesired behavior, such devices can make the undesired behavior less likely to occur. The act of placement is a commitment-based method o achieve the behavior. This project focused specifically on rehabilitation following a rotator cuff tear, a common shoulder injury for which long term home-based physical therapy is viewed as a crucial part of recovery, however adherence is often low (Burns et al., 2021). Some examples of physical therapy exercises for this condition can be viewed below.
In the context of this project, a commitment-based system deliberately restricting future choices. An example would be that if you want to avoid using your phone in bed, placing it in another room before going to sleep creates a physical barrier that reinforces your intention. By adding friction to undesired behavior, such devices can make the undesired behavior less likely to occur. The act of placement is a commitment-based method o achieve the behavior. This project focused specifically on rehabilitation following a rotator cuff tear, a common shoulder injury for which long term home-based physical therapy is viewed as a crucial part of recovery, however adherence is often low (Burns et al., 2021). Some examples of physical therapy exercises for this condition can be viewed below.

The Concept

A toothbrush holder that restricts access to a user’s toothbrush until prescribed exercises have been completed.

To test a hard commitment device for the performance of shoulder exercises, this project introduces the concept of a toothbrush holder that restricts access to a user’s toothbrush until prescribed exercises have been completed at some point during the day. A small strap can be attached to either a manual or electric toothbrush, allowing it to be locked into a custom holder. On designated exercise days, the toothbrush is locked and a blue light signals that exercises need to be performed. These exercises can be completed using a resistance band equipped with a sensor. Once the required exercises are done, the toothbrush is unlocked.

This concept was selected based on a comprehensive set of requirements. It was primarily chosen because toothbrushing is a stable and consistent part of daily life for a large portion of the population. Additionally, a toothbrush is a highly personal item, making any commitment device based on it inherently individualized. Lastly, because oral hygiene is generally regarded as essential, users are unlikely to forgo brushing their teeth simply to avoid the exercises, thereby reinforcing adherence to the prescribed routine.

The Process

The project followed a structured and comprehensive process grounded in theoretical frameworks and literature analysis.

This began with a literature review, which was complemented by interviews with individuals recovering from shoulder injuries and a practicing physical therapist. This groundwork informed a design process guided by both the Persuasive System Design (PSD) model and B.J. Fogg’s 8-Step Design framework, ensuring that the resulting solution was both theoretically grounded and practically relevant. Based on the insights from the background research, a clear set of design requirements was established to help guide and converge the ideation process toward a single, focused concept. This ideation phase explored various concepts, but ultimately led to the development of one well-defined concept suitable for in-depth testing. The prototype was evaluated through a 7-day field study involving three experimental conditions and a total of nine participants selected to reflect the diversity of the target user group. The entire process and its outcomes were documented to meet academic standards, ensuring transparency, rigor, and reproducibility.

Real-World
Testing

Real-World Testing

Real-World Testing

The prototype was evaluated through a 7-day field study involving three experimental conditions.

The goal of this testing phase was to explore whether a commitment-based interactive system could offer real value in supporting the performance of home-based physical therapy exercises. To do so, I designed a study that combined both quantitative and qualitative methods to capture a well-rounded understanding of the system's impact. To measure its effectiveness, I focused on four main aspects: self-efficacy, perceived autonomy and competence (inspired by Self-Determination Theory), adherence to prescribed exercises, and participant reflections gathered through semi-structured interviews. Standardized questionnaires were used to assess both self-efficacy and the psychological needs of autonomy and competence, ensuring consistency and reliability in the collected data.
The study took place in and around Enschede. I personally visited each participant at home to introduce the system, explain its purpose, and set it up within their daily environment. Participants then used the system independently over the course of seven days. After this period, I returned to collect the device, administer follow-up measures, and conduct a final interview to gain deeper insight into their experiences. This setup allowed for a realistic understanding of how the system functioned in everyday life. Below are some of the environments in which the prototype was installed.

Outcome

The study found that commitment-based interactive systems can improve exercise adherence, particularly among users with low confidence in their ability to stick to the behavior over time.

However, the effectiveness of such systems is not universal. Their benefits appear to be most significant for a specific subset of patients, as adherence issues often arise from a variety of personal and contextual factors. Moreover, the findings suggest that long-term success with commitment-based systems depends heavily on sustained intrinsic motivation. Without it, users may find ways to bypass the commitment mechanism altogether.

The study found that commitment-based interactive systems can improve exercise adherence, particularly among users with initially low confidence in their ability to stick to a routine. However, the effectiveness of such systems is not universal. Their benefits appear to be most significant for a specific subset of patients, as adherence issues often arise from a variety of personal and contextual factors. Moreover, the findings suggest that long-term success with commitment-based systems depends heavily on sustained intrinsic motivation. Without it, users may find ways to bypass the commitment mechanism altogether.

My Role

This project was conducted as part of my master’s thesis, for which I was solely responsible. However, during the process I was guided by two supervisors: dr.ir. Juliet Haarman, dr. Armağan Karahanoğlu. Over the course of the project, I carried out every stage of the process, from initial research to final evaluation. This project gave me comprehensive, hands-on experience in independent research, product design, user-centered development, and academic evaluation.

Design &
User Research

Defined design requirements from background research and interviews with patients and a physical therapist. Developed and refined the prototype through iterative design cycles. Planned and executed a month long 7-day field study, using mixed-method data collection to assess user behavior and system impact.

Prototyping &
Engineering

Prototyping &
Technical Development

Built the physical prototype using SolidWorks and 3D printing. Integrated electronics for physical locking and sensing functionality. Developed supporting software for sensor data logging and feedback logic.

Academic Research & Documentation

Applied behavioral design theory to a real-world physical therapy challenge. Designed and conducted a field study with clear hypotheses, defined conditions, and mixed-method data collection. Documented the full process and findings to meet academic standards, suitable for evaluation as a master's thesis.

Home

-

Work

-

About

Back to top

Home

-

Work

-

About

Back to top

Home

Back to top