Statistics in motion

Tuesday 4 June 2024, 4:30 pm
PhD candidate
T. Ghilardi MSc.
prof. dr. S. Hunnius
dr. M. Meyer

Upon entering the world, infants immediately face the difficult task of interpreting a highly complex environment, filled with a variety of signals like sounds, visuals, and actions. Surprisingly, most of the stimuli that surround infants are not as disorganised as they might first appear. On the contrary, they contain considerable amounts of information and structure, even if not immediately apparent. For example, a parent's daily routine can reveal hidden patterns. These patterns help infants understand the constant flow of information they experience, which in turn helps them learn and adapt quickly to their surroundings. This work investigates how infants and adults can use statistical learning - the ability to detect patterns in continuous streams of stimuli - as a key mechanism for acquiring new knowledge about the world. It demonstrates that both infants and adults can identify statistical regularities in action sequences, which is reflected in the predictive activity of their motor system. Furthermore, this work uncovers evidence that infants can extend their use of statistical learning beyond tangible stimuli, such as visual or motor inputs, by learning and predicting where information can be found. The experiments described in this thesis collectively contribute to a better understanding of how statistical learning may be leveraged to extract new knowledge from the world around us.

Tommaso Ghilardi pursued his Bachelor's degree in Psychological Sciences and Techniques at the University of Trento in Rovereto, Italy. During this time, he joined the Affiliative Behaviour and Physiology Lab (ABP lab) to study the relationships between genes, psychological traits, and behaviors under Prof. Gianluca Esposito. He then completed his Master’s degree in Psychology at the University of Trento. As part of his master’s program, he visited the Graduate School of Biomedical Sciences at Nagasaki University. Here, he conducted his master’s thesis project on the recognition of cross-cultural emotional faces under the supervision of Prof. Kazuyuki Shinohara and Prof. Gianluca Esposito.
For his PhD, Tommaso moved to the Donders Institute where he worked under the supervision of Prof. Sabine Hunnius and Prof. Marlene Meyer. During his PhD, he visited the Centre for Brain and Cognitive Development (CBCD) at Birkbeck University, working with Prof. Denis Mareschal on a project exploring the neural correlates of informativity prediction. Since July 2023, Tommaso has been working as a postdoctoral researcher at Birkbeck University, studying children’s planning abilities.