First Computational Psychiatry Day in Ferrara

On December 2, 2023, Ferrara hosted the First Italian Day of Computational Psychiatry, marking a milestone in the formal launch of the Italian Network for Computational and Precision Psychiatry. The event brought together researchers from psychiatry, neuroscience, psychology, and computational sciences to share perspectives on how mathematical models and data-driven approaches can reshape our understanding and treatment of mental illness.

The day opened with greetings from Martino Belvederi Murri (University of Ferrara), who underlined the ambition of creating a national community devoted to computational and precision approaches in psychiatry. Paolo Ossola (University of Parma) followed by framing the discipline: computational psychiatry as a bridge between clinical phenomena and quantitative models, and as a crucial step toward personalized care.

The first part of the symposium focused on theory-driven approaches.

Rodolfo Rossi (University of Rome Tor Vergata) illustrated Bayesian belief updating as a framework for psychosis, showing how probabilistic models can formalize alterations in inference.

Paolo Ossola then presented applications of the Drift Diffusion Model to mood and temperament, highlighting how decision-making dynamics can illuminate affective processes.

After the break, Chiara Montemitro (Linköping University, Sweden) introduced reinforcement learning paradigms in the study of addictive behaviors.

Closing the morning, Martino Belvederi Murri outlined “hybrid” models that integrate theory-driven and data-driven approaches in the study of affective disorders.

After lunch, the stage was given to the “Young Computationals”, a new generation of researchers presenting short communications and open calls for collaboration:

Federica Folesani (University of Ferrara) on computational models of cognition in psychiatric and general hospital settings.

Tommaso Jannini (University of Rome Tor Vergata) on network approaches to complex PTSD and adolescent addictions.

Andrea Costumati (University of Parma) on data-driven classification of bipolar disorder trajectories.

The second part of the symposium turned to data-driven methods.

Linda Antonucci (University of Bari) introduced the fundamentals of machine learning and its applications to psychosis.

Alessandro Pigoni (Policlinico of Milan) compared supervised and unsupervised learning approaches in psychotic disorders.

Filippo Corponi (University of Edinburgh, UK) discussed neural networks and self-supervised learning applied to affective disorders.

The meeting concluded with an open discussion, emphasizing the complementarity of theoretical and empirical approaches, and the need to build sustainable collaborations across disciplines and institutions.

The success of this first gathering confirmed both the urgency and enthusiasm for computational psychiatry in Italy. By establishing common ground between clinicians, data scientists, and theoreticians, the Ferrara meeting laid the foundation for a national and international network. The Italian Network for Computational and Precision Psychiatry aspires not only to foster research and training, but also to translate computational insights into tangible advances in clinical practice, while maintaining a strong humanistic and patient-centered orientation.