Computational modelling and AI for brain and behaviour: integrating prediction with inference

Computational modelling, machine learning, and broader artificial (AI) intelligence approaches are now key approaches used to understanding and predicting cognitive function and behaviour. Although these approaches are closely related, they make distinct contributions. Computational modelling models provide theory driven accounts of cognitive and neural mechanisms. They do that by formalising hypotheses and testing them through algorithms development based on the findings and simulations. In contrast, machine learning and AI methods are successful in discovering structure in complicated and high dimensional data, enabling robust prediction, and supporting analysis across modalities. Together these tools can connect levels of explanation, strengthen inference, and improve reproducibility. Historically computational modelling has been used in computational neuroscience to explain brain processes and behaviour, including neural network and biologically inspired models. Examples of this work can be seen in modelling cognitive processes such as decision making, attention, language and memory retention processes. Machine learning approaches, use algorithms trained on experimental data to support prediction and classification, with some applications in EEG and fMRI data analysis. These methods are increasingly important in clinical oriented work, especially for prediction of outcomes and the development of precision- focused computational tools applied to precision medicine. Across the journey from understanding behaviour to applying that understanding in clinical approaches machine learning and computational modelling each contribute in distinct ways. Despite their widespread use, there is still relatively limited work that integrated these approaches to bridge the gap between understanding and prediction and to accelerate progress toward better clinical outcomes. The proposed research topic is building on this previous work and aims to present work that uses computational modelling and machine learning separately as well as any new work that combines these approaches to help progress further the research in the area. We particularly welcome submissions that link model-based inference with data driven prediction, and that use one or more of these methods to address key questions in brain processes, cognitive mechanisms, and neural dynamics. We welcome all article types. This Research Topic was initiated in collaboration with the Organization for Computational Neurosciences (OCNS) . A portion of the article processing charge revenue is shared with OCNS to support programs and initiatives that foster the growth and development of the computational neuroscience community.If you would like to host your own society-led Research Topic in collaboration with the Organization for Computational Neurosciences (OCNS) and Frontiers in Computational Neuroscience, please submit this form. Topic editors: Prof. Eirini Mavritsaki, Eleni Vasilaki, Zoe Kourtzi Source: https://www.frontiersin.org/research-topics/81008/computational-modelling-and-ai-for-brain-and-behaviour-integrating-prediction-with-inference