Based on computer programs that process digital video sequences, this multitracking video system used to observe fish designed by ULg research suggests many applications, such as the observation of desert locusts, patterns of human movement, or tracking cars on highways.

However things may stand with the living creatures observed, their collective behavior is a product of interactions between individuals in the group. Many important studies are conducted today in order to understand whether group behavior – the classic examples are schools of fish and swarms of insects – is organized in an automatic way, or is arrived at through some process of group decision. Johann Delcourt was initially interested in the behavior of predation, the subject of his doctoral research (1); he switched over to considering the behavior of the prey. This opened the way to the study of interesting concepts, such as the structure of a school of fish. Schools of fish are in fact an important social phenomenon, in which an association of individuals causes the emergence of certain properties and behavioral possibilities that are not reducible to the sum of characteristics of the individuals of the group. These structural or dynamic properties have attracted interest on the part of ethologists as well as researchers working in the field of statistical physics. The latter have observed significant analogies with phenomena in the world of physics. Delcourt is at present pursuing post-doctoral research for the Developmental biology unit at the ULg (Pr. Pascal Poncin), which works closely with GRASP (Group for Research and Application in Statistical Physics, ULg, Pr Nicolas Vandewalle), especially with respect to the work of Christophe Becco. “In general, most research is aimed at the ultimate origin of behaviour, the “why” in terms of function and evolution...what interests me is more knowing and understanding the “how” of behaviour, the mechanism and how it comes to be,” explains Johann Delcourt, co-author of a recent study involving an original multitracking video system (2).

His dissertation concerned the theme of group movement in animals, taking for a model schools of two kinds of African fish: Oreochromis niloticus (Nile tilapia) and Pelvicachromis pulcher, both from the Cichlidae family of freshwater fish. Why these particular species? “Because they exhibit different parental behaviours,” according to the young researcher. “The pulcher lays eggs on a substrate (in a natural cavity) and both parents help guard the young fish, while the Nile tilapia belongs to the family called “buccal incubators”, and the fry can take refuge in their mother’s mouth in case of danger.” The fry of both species exhibit group movements whose conditions of production require explaining, especially as regards parental protection. In addition, biologists do not always understand the processes that produce movements of schools: are there leaders? Is the group democratic? Do they move because of external stimuli such as water currents, light, the presence of predators? Etc. But in order to study and describe these group behaviours, there was no well-adapted tool, except for simulation programs often constructed on too simplistic a model. In the absence of any consensus concerning behavioural rules, the conclusions of studies employing the software simulation models rarely portrayed biological reality in an optimal manner. Still in recent years new systems for videotracking capable of following groups of animals had been developed, such as EthoVision Color-Pro, adapted for following fish in the late 1990s by Dr. Marc Ylieff, researcher at the University of Liège.

(1) “Structure and ontogenesis of school behaviour in two fish, members of the Cichlidae family: quantitative approach using automatic videotracking of Oreochromis niloticus and Pelvicachromis pulcher.” (May 2008).
(2) “A video multitracking system for quantification of individual behaviour in a large fish shoal: advantages and limits”, Johann Delcourt, Christophe Becco, Nicolas Vandewalle, Pascal Poncin, Behavior Research Methods, 2009.