Much development occurs as children explore and react to their social environments. To understand the rules governing early social interaction, Dr. Messinger employs machine learning of audio, video, and ultrawideband signals deployed in naturalistic contexts. Guided by a dynamic systems perspective, he uses statistical tools to model early face-to-face interaction, predict attachment, and characterize autism spectrum disorder. Dr. Messinger will discuss ongoing research harnessing multimodal objective measures of vocalizations and location captured with wearable sensors in inclusion classrooms containing children with and without disabilities such as autism. The research incorporates both machine learning and tools from statistical physics to predict children’s speech and language development in everyday contexts. 

Short Bio:  Dr. Messinger is a professor in the departments of Psychology, Pediatrics, and Electrical and Computer Engineering. He is an interdisciplinary developmental psychologist, and the author of over 120 scientific publications appearing in journals such as Science Reports, Developmental Science, and Molecular Autism. Dr. Messinger employs computational approaches to big behavioral data to understand social, language and emotional development. His research has been continuously funded by the US federal government and private foundations for 25 years. 

Problem-solving is a fundamental cognitive process, that appears in the early stages of human development and is a prerequisite for human progress. However, this process involves complex faculties such as exploration, creativity, and curiosity, which are challenging to model, especially in young children. Understanding children’s problem-solving becomes even more challenging when it unfolds in collaborative settings consisting of humans or within hybrid teams where children interact with Artificial Intelligence systems, including social robots. Vicky Charisi will discuss recent and ongoing research that uses social robots and AI systems to study children’s problem-solving in structured experimental settings and unstructured play environments. She will tackle aspects such as the extraction of implicit knowledge and the identification of the Aha! moment in problem-solving. Her talk will conclude with design recommendations for AI systems to support children’s problem-solving, such as the importance of voluntary interaction, and a reference to the emerging opportunities and risks based on the recent policy guidance on AI and children’s rights published by UNICEF.

Short Bio: Vicky Charisi is a Research Scientist at the European Commission, Joint Research Centre. She studies the impact of Artificial Intelligence on Human Behaviour, and specializes in child-robot interaction. She is particularly interested in understanding how interactive and intelligent systems, including social robots, affect human cognitive development, such as the processes of structure emergence in early childhood. Her current work involves evidence-based policy support on the emerging AI opportunities and risks for child users, based on the United Nations Convention on the Rights of the Child. Vicky Charisi has published over 60 scientific publications and science-for-policy reports, and she serves as a Chair of the IEEE Computational Intelligence Society, Cognitive and Developmental Systems, TF of Human-Robot Interaction.