Summer school - Quantifying natural behavior –
The road toward computational neuroethology

Short abstract

Over recent years, there has been a proliferation of methods aimed at automating the quantification of animal behavior at an unprecedented level of detail. This surge has given rise to the burgeoning field of computational ethology, promising to revolutionize our ability to comprehensively understand animal actions within their environments. The imperative to refine our behavior characterization techniques is underscored by the growing realization that unraveling behavior is not only crucial but often a prerequisite for tackling questions in neuroscience. The utility of these methodologies extends beyond studying behavior solely in natural or strictly ethological contexts. Modern tools for behavioral quantification can enrich a broad spectrum of approaches used to correlate brain function with behavior, spanning from psychophysics to cognitive tasks. These methods offer the potential to augment traditional measurements with nuanced insights into how animals navigate various challenges. In this 5-day course, we will delve into computational strategies for quantifying animal behaviors. Participants will explore recent advancements in behavior quantification, focusing particularly on tracking animal motion, characterizing dynamics using both supervised and unsupervised methods, and assessing group or collective behaviors. The course will feature a blend of introductory lectures, guest talks highlighting scientific applications of behavioral quantification, and hands-on practical sessions. The workshop is open to 30 advanced (MSc and PhD) students, postdocs and research scientists from all universities. Prerequisite courses include Python programming and basic statistics. The course is given in English.