Today Matthijs Smakman will defend his doctoral thesis titled “Robots in Education: Implementing Robot Tutors in a Morally Justified Way”. Conducted under the supervision of Prof. Dr. Elly Konijn, his empirically grounded research represents a significant milestone in achieving a morally responsible integration of social robots as educational assistants in primary schools. Previous studies have already indicated that the utilization of social robots yields better outcomes compared to traditional educational technologies.
The guidelines proposed by Smakman provide a framework for the deployment of robots while adhering to the ethical requirements set by various stakeholders. Based on a palette of 17 ethical values, his comprehensive guideline consists of 22 key points. Smakman explains, “Preserving human interaction with teachers and classmates should not diminish due to the use of robots. The robot must genuinely add value, not detract from it. Privacy is crucial. How does the robot handle the data it collects through its cameras and microphones? Children readily confide secrets in robots, as they sometimes find it easier to express themselves to non-human entities. Therefore, trust is a paramount value that must never be compromised.”
Smakman describes a social robot as having a physical body, capable of movement and communication with humans through speech or other means such as lights. These robots possess a degree of autonomy and are not entirely controlled, but rather programmed, by humans. Due to their independent reactions and presence in physical spaces, people, especially children, perceive them as living beings and even consider them as buddies. Smakman’s research aimed to explore the social roles and associated ethical questions regarding robots in primary schools. Children, being receptive by nature, readily engage in conversations with robots, thus building relationships. This raised the question of how to design a robot that does not compromise the values cherished by parents and educators.
One of the primary concerns, particularly among parents, was the potential impact on a child’s socio-emotional development. Would children start imitating robotic behavior when interacting with their technological companions? Would their development be hindered? However, research indicates that such concerns may not be warranted, although children with specific conditions such as ADHD seem to form attachments to robots more quickly. The precise mechanisms underlying this phenomenon require further exploration. Therapeutic sessions with autistic children have revealed the importance of robots’ neutrality. These children appreciate the predictability of robotic responses. Furthermore, robots can be utilized to help these children understand and express certain emotions. This opens avenues for further investigation.
The acceptance of robots also varies based on socio-demographic factors. Individuals with lower incomes tend to be more skeptical than those with extensive technological experience. Elderly individuals exhibit more skepticism than younger generations, despite social robots proving beneficial in elderly care. Remarkably, teachers demonstrated a predominantly practical approach, focusing less on moral considerations and emphasizing the robot’s potential to alleviate their workload. Hence, profound philosophical questions were not at the forefront. Teachers often mentioned using hand puppets, which also create a fantasy world and are not perceived as deceitful.
Based on these insights, the research team had to determine which data the robot should or should not store, the role of technology, and how to handle the received data. It is crucial that a robot does not replace a teacher. The recommendation is to build a robot capable of performing one task effectively, such as storing only the data that contributes to meaningful lesson interactions. This may include information about a child’s progress in mathematics or language skills, along with personal details like hobbies. However, the robot should not foster deep social bonds that encourage the sharing of secrets, akin to a storytelling stuffed animal where children confide their feelings.
Smakman further explains, “We continually fine-tune the robots and enrich them with additional information to ensure they remain properly aligned with the child they interact with. This also demands effort from the teachers. The introduction of robots should not exacerbate their already high workload but rather alleviate it. When collecting data, it is essential to make use of that data effectively. Therefore, we transitioned from recording entire conversations to merely registering responses to a math problem or identifying specific words. This is the only information we store. Teachers can access the child’s scores, allowing them to evaluate whether it aligns with their understanding of that particular student.”
What lies ahead? “The next step is to develop robots that comply with the guidelines I have formulated. This will be the ultimate litmus test. Subsequently, we can explore the additional contributions robots can make to education and the necessary requirements for such advancements. Moreover, we need to identify the specific data we require from children to personalize interactions effectively. Companies involved in robot development have expressed great interest and are collaborating on future projects. Ultimately, an ethically responsible robot will foster quicker acceptance within our society.”
Parallel to this research, Smakman also had the opportunity to establish a specialization in social robotics within the ICT program at the HU, where he teaches three days a week. This combination of research and education has proven to be a remarkable endeavor. Robots are being utilized not only in education but also in healthcare, such as at the Wilhelmina Children’s Hospital and in elderly care. Gradually, Smakman finds himself crossing the line from education to the realm of healthcare, where there is significant demand for these technologies. This transition presents exciting challenges in exploring further ethical issues.
As the field of social robotics continues to evolve, Smakman’s pioneering research marks a critical step towards ensuring the responsible and ethical integration of robots as educational assistants. By addressing concerns related to privacy, trust, and human interaction, his guidelines provide valuable insights for stakeholders and pave the way for future advancements in this rapidly growing field.