We search for principles of animal behavior across species.

The behavior of animals is not the behavior of their brains, but the processes emerging from the interaction between neural activity, body biomechanics and environmental constraints. Recent advances in neuroscience comprise a wide range of “big tools” enabling the collection of “big data”, both being promissory notes for understanding the brain and explaining behavior. This has lead to much emphasis on techniques and causal accounts of explanation in the flavour of the latest interventionist techniques and reductionist views, thus giving the impression that detailed studies of behavior and its algorithmic composition are less important. However, dissecting “necessary and sufficient” neural circuits for behavior is no shortcut to the proper study of behavior itself. After all, to ask how the brain works is different than (and requires) to ask what it is for — neurons indeed compute information yet nervous systems evolved to produce adaptive behavior. Thus, in the lab we try to avoid missing the forest for the trees.

We advocate for a more pluralistic notion of neuroscience where the dissection of neural processors (“hardware explanations”) are best investigated after a careful decomposition of behavioral processes (“software explanations”). This has lead us to pursue a theoretical/computational approach to animal behavior, and across species. From worms and flies to mice and humans, we study shared principles of animal movement from which the fundamental properties of these complex systems should be derivable, interpretable and explainable. Our current efforts target three fronts: (i) seeking the perceptual origins of the speed-curvature power-law in human drawing and maggot locomotion, (ii) exploring the organization of posture sequences in foraging worms and fish, and (iii) establishing behavioral homologies in the unfolding of locomotor degrees of freedom in flies and rodents.

We are hopeful that searching for principles of animal behavior across species will offer general insights into the neurobiology, ecology and evolution of animal behavior. Seeking to fulfill the promise of nowadays “big science”, our more abstract complementary approach moves towards a grounded integrative grasp of animal behavior. Quoting Woese, “without the proper technological advances the road ahead is blocked, without a guiding vision there is no road ahead”. Or, as Gallistel put it: “No Mendel, no Watson & Crick”.




“The important advance from this level of explanation that is made by turning to the nervous system as a controlling entity has unfortunately had a similar effect in discouraging a direct descriptive attack upon behavior. The change is an advance because the new entity beyond behavior to which appeal is made has a definite physical status of its own and is susceptible to scientific investigation. Its chief function with regard to a science of behavior, however, is again to divert attention away from behavior as a subject matter. The use of the nervous system as a fictional explanation of behavior was a common practice even before Descartes, and it is now much more widely current than is generally realized. At a popular level a man is said to be capable (a fact about his behavior) because he has brains (a fact about his nervous system). Whether or not such a statement has any meaning for the person who makes it is scarcely important; in either case it exemplifies the practice of explaining an obvious (if unorganized) fact by appeal to something about which little is known. The more sophisticated neurological views generally agree with the popular view in contending that behavior is in itself incomprehensible but may be reduced to law if it can be shown to be controlled by an internal system susceptible to scientific treatment. Facts about behavior are not treated in their own right, but are regarded as something to be explained or even explained away by the prior facts of the nervous system. (I am not attempting to discount the importance of a science of neurology but am referring simply to the primitive use of the nervous system as an explanatory principle in avoiding a direct description of behavior.)” — B. F. Skinner (1938) The Behavior of Organisms


“If we look far into the future of our science, what will it mean to say we ‘understand’ the mechanism of behaviour? The obvious answer is what may be called the neurophysiologist’s nirvana: the complete wiring diagram of the nervous system of a species, every synapse labelled as excitatory or inhibitory; presumably, also a graph, for each axon, of nerve impulses as a function of time during the course of each behaviour pattern. This ideal is the logical end point of much contemporary neuroanatomical and neurophysiological endeavour, and because we are still in the early stages, the ultimate conclusion does not worry us. But it would not constitute understanding of how behaviour works in any real sense at all. No man could hold such a mass of detail in his head. Real understanding will only come from distillation of general principles at a higher level, to parallel for example the great principles of genetics— particulate inheritance, continuity of germ-line and non-inheritance of acquired characteristics, dominance, linkage, mutation, and so on. Of course neurophysiology has been discovering principles for a long time, the all-or-none nerve impulse, temporal and spatial summation and other synaptic properties, y-efferent servo-control and so on. But it seems possible that at higher levels some important principles may be anticipated from behavioural evidence alone. The major principles of genetics were all inferred from external evidence long before the internal molecular structure of the gene was even seriously thought about. Three computers with the same programming instruction set are in an important sense isomorphic in principle, even though their wiring diagrams may be utterly different, one employing valves, another transistors and the third integrated circuits; how all three work is best explained without reference to particular hardware at all” — Richard Dawkins (1976) The Need for General Principles


The principal function of the nervous system is to produce behavior. Thus, the ultimate goal of most behavioral work with laboratory animals in neuroscience is to understand how molecular events in the nervous system come to produce behavior and, as a corollary, how changes in molecular events produce differences in behavior. Understanding these issues offers hope for understanding the nature of the human mind, which some may argue is the fundamental question in neuroscience. But perhaps even more important is that understanding brain-behavior relationships offers a way to find treatments for dysfunctions of behavior, whether they are in the province of neurology or psychiatry. Advances in molecular and cellular neuroscience have been dramatic over the past two decades, but most of these advances have been independent of an understanding of how they relate to behavior. This is changing. Neuroscientists oriented toward molecular research are increasingly looking to the ultimate function of the phenomenon that they have been studying— behavior. For the majority of behavioral studies, this means studying the behavior of the laboratory rat.” — I. Q. Whishaw (1939) The Behavior of the Laboratory Rat


“In writing the target article I had two purposes that I believe are now fulfilled: The first was to highlight a blind spot in the behavioral sciences with regard to the need for a universal symbolic language for the description of whole-animal movement. The state of this art is now documented in the commentaries. The second purpose was to demonstrate how, by using an appropriate language, a simple set of common generative rules can be shown to produce the transition from stereotyped to what appears to be free behavior. As I see it, the main value of this proposal is not in the particulars of the model, but in providing an integrative method for establishing behavioral continuities. In using this integrative method, attention will be attracted to new meaningful details that will either fit the model or modify it, or even suggest a better one.” — Ilan Golani (1992) The mobility gradient as an integrating model in the organization of vertebrate movement


“And if we can formulate an adequate theory of language in psychological terms, there are bound to be important implications for neurophysiology. In so far as the neurophysiologist is concerned to understand ‘how the brain works’, he must equip himself with a non-physiological account of the tasks which the brain and its peripheral organs are able to perform; only then can he form mature hypothesis as to how these tasks are carried out by the available ‘hardware’ — to borrow a phrase from computing science.” — H. C. Longuet-Higgins (1972) The algorithmic description of natural language


“Variability in behavior, even if the root causes of it remain unknown, does not render the behavior of organisms completely, or even largely, unintelligible – but it may be the case that precise, moment-to-moment behavioral prediction is not possible. By appealing to a closed-loop conception of organisms, the understanding of behavior can, instead, be approached on a different level. Instead of asking how a particular experimental manipulation alters the subsequent behavior of an organism, one might instead ask how an experimental manipulation alters the parameters of the system. This is a subtly different question, but the difference is important, and requires that the parameters of the system be understood to begin with. Understanding what variables organisms may be controlling necessitates that organisms be understood on their own terms before they are used as model systems to answer larger questions.” — H. C. Bell (2014) Behavioral Variability in the Service of Constancy