A.R. Cools and D. Heeren

Department of Psychoneuropharmacology, University of Nijmegen, Nijmegen, The Netherlands

Today, there are several image-analyzing systems that allow one to analyze automatically spatio-temporal transitions between stable configurations of objects or subjects. In animal studies on behaviour, these systems usually reduce the subject (rodent) to a single spot and provide its x and y coordinates: then, the distance, duration and velocity of the moving spot in space is calculated. Although these systems provide quick, objective and reliable measures of the animal's progression in space, the available systems do not provide insight into the organization of the animal's movement in space. We will present three systems that enable us to study the rules of order in the organization of movements of rodents. The rules themselves are discovered and described in detail by Golani and his coworkers (review: Behav. Brain Sci. 12:249, 1992). Analysis of these rules has been found to be a powerful tool in brain and behaviour studies, including studies on the effects of centrally acting agents.

We have focussed our attention on two sets of rules: one set directing the organization of home base behaviour in a novel environment and one set of rules directing the organization of postural transitions during ontogeny, aging, drug-induced changes and recovery from brain damage.

For studying spatio-temporal transitions (open field behaviour, home base behaviour, behaviour in elevated plus-maze or Morris maze), we have developed two video tracking systems consisting of: a CCD camera (Sony DXC-107), a frame grabber (Matrox PIP 1024 or PC-Vision Plus) and a MS-DOS 286 system. For studying postural transitions in time, we have developed a new system consisting of: a CCD camera, a frame grabber (Fast FPS60) and a Pentium (100 Mhz, Windows 95). Although the three systems allow a sampling rate of 20 frames per second, practice has shown that a sampling rate of 10 frames per second is sufficient for analyzing spatial and postural transitions between stable configurations in time.

A. Home base behaviour has been found to be an invariant feature of locomotor behaviour of rats in a novel environment. All features are described in detail by Eilam (University of Tel Aviv thesis, 1987) and Golani et al. (Behav. Brain Res. 53:21-33, 1993). They describe the main features as follows: "When rats are placed in a novel environment, they alternate between progression and stopping: in the course of a session they stop briefly in many places, but in one or two places they also stop for very long periods. The place in which they stay for the longest cumulative time is defined as the rat's home base. In this place the incidences of grooming and rearing are high and often the highest. In addition, the number of visits to the home base is typically the highest. Some rats establish a secondary base with similar properties to those of the main home base. The location of the base influences the mode of progression throughout the environment: progression away from base is slower and includes more stops than progression back." These authors have suggested that this home base paradigm may be used for the analysis of the spatial organization of locomotor behaviour in neuroscience research. Indeed, this approach turned out to be an excellent paradigm for the analysis of behavioural changes elicited by psychostimulants such as dexamphetamine (Eilam et al., Behav. Brain Res. 36:161-170, 1990) as well as for the analysis of the role of striatal structures in the organization of this behaviour (Cools, Behav. Brain Sci. 15:271, 1992).

The above-mentioned fully computerized and automated video tracking systems for spatio-temporal transitions allow us to study home base behaviour in detail. It automatically delineates (1) the location and number of home bases and (2) length, duration, number and velocity of distinct types of excursions. An excursion is defined as the route starting immediately after leaving the home base and ending just before stopping again at the home base; such an excursion could be either a "round-trip" being an excursion that starts and ends at the same base or a "home trip" being an excursion between two different home bases. A "stop" is defined as the interruption of an excursion, during which the rat ceases to progress forward and freezes in place, or ceases to progress forward and performs lateral and/or vertical scanning movements with any or all parts of its trunk while staying in place. Finally, the excursions are classified according to the number of stops. The validity of the computer program is tested in two manners: (a) the open field behaviour of 12 naive rats (recording time/rat: 60 min) was videotaped and visually analyzed in terms of number of home bases: both the visual analysis and the computerized analysis had the same outcome: 9 rats had 1 home base, and 3 rats had 2 home bases; (b) the open field behaviour of 2 naive rats (recording time/rat: 60 min) was videotaped and visually analyzed in terms of number of stops; the visual analysis revealed the presence of 200 stops, whereas the computerized program revealed the presence of 209 stops: accordingly, the mismatch was less than 5%. Finally, the systems provide the full tracking pattern of each distinct type of excursion.

The usefulness of these systems will be illustrated by presenting the outcome of a study, in which differences in the vulnerability to dexamphetamine in Nijmegen High Responders to novelty and Nijmegen Low Responders to novelty are analyzed.

The systems under discussion have a far broader application. Two additional applications will be illustrated: analysis of behaviour in the elevated plus-maze, namely a validated animal model of anxiety, and analysis of behaviour in the Morris water maze, namely a validated animal model of learning and memory.

Plus-maze. The image-analyzing system allows the computerized and automated measurement of: total arm entries, open arm entries, closed arm entries, % open arm entries, % open arm time, entry latency, % central platform time, % closed arm time, closed arm returns, etc.

Morris water maze. The image-analyzing system allows the computerized and automated measurement of: platform latency, returns to startbase, left or right turns and % time spent in small segments (for that purpose, the maze is divided into 3 imaginary circles as well as in 4 quadrants). Again, these systems provide an image of the full tracking pattern.

B. Postural transitions. After pronounced immobility, rat movement is comprised of phases of continuous movement interrupted by intervals of complete arrest. Sequence analysis has revealed a process of motorial expansion constrained by rules such as (a) every part of the trunk must first move laterally, then forward, then up, and (b) a part can never move along a dimension unless the part anterior to it has already moved along that dimension (Eilam et al., Develop. Psychobiol. 21:679, 1988). Following complete motorial expansion, the rules are applied in reverse. These rules apply to postural transitions displayed during ontogeny, aging, drug-induced changes and recovery from brain damage.

The third image-analyzing system that we have developed allows the computerized and automated measurements of lateral movements of the head, lateral movements of head and torso, and lateral movements of head, torso and torso. The usefulness of this system will be illustrated by presenting the outcome of a study, in which we compared the role of the olfactory tubercle, nucleus accumbens and the neostriatum in directing distinct subsets of these rules.

Paper presented at Measuring Behavior '96, International Workshop on Methods and Techniques in Behavioral Research, 16-18 October 1996, Utrecht, The Netherlands