ESPE Abstracts (2014) 82 P-D-2-2-332

Game Interaction Between a Humanoid Robot and a Diabetic Teenager: Might This Improve Motivation to Fill in the Nutritional Diary?

Alberto Sannab, Ilaria Baronic, Elettra Olearib, Maria Iole Colombinia, Clara Pozzia, Gianni Russoa, Andrea Rigamontia, Giulio Frontinoa, Valeria Favallia, Roseila Battaglinoa, Clara Bonuraa, Riccardo Bonfantia, Giusy Ferroa & Giuseppe Chiumelloa


aPediatric Department, San Raffaele Hospital, Milan, Italy; bFondazione San Raffaele, Milano, Italy; cTelbios, Milano, Italy


Introduction: This study describes the experience of introducing Nao, a humanoid robot, into a Summer Camp for children with diabetes (August 2013, Misano Adriatico, Italy), with the aim to provide them a companion capable to support and motivate. Our goal was to investigate if, Nao’s interactions with children could positively affect the adherence to specific medical recommendations during their stay. Namely children were asked to fill in a specific nutritional diary. This activity was carried out in the context of the ALIZ-E EU co-funded project, which develops the theory and the practice behind the development of embodied cognitive robots capable of long-term interaction with children.

Methods: Among all the participants attending the Camp (age: 11–14), 58 were involved in the study: 20 volunteered to interact individually with Nao, the remaining 38 were the control group. During the interaction, the child and the robot played different activities related to nutrition and diabetes. Meanwhile Nao provided motivational hints about the diary, underlining its importance to fill it. Finally, adherence of diabetic children to this task was measured by checking whether the child, after the interaction, filled in the diary at least once during the following days.

Results: All the 58 children showed a good glicometabolic control (Hba1c M 7.2%, ds 0.93). Eight of 20 children who met Nao filled in the diary whereas four of 38 in control group did it. A two-tailed t-test comparing the two means confirmed statistical significance (t=2.39 with P=0.0103).

Conclusions: This study revealed a better adherence to fill the diary thanks to the individual child–robot interactions, compared to that of the control group. The result shows the efficacy of this useful and enjoyable way to motivate diabetic children.