Journal of Neuroscience Methods 182 (2009) 34–42 Contents lists available at ScienceDirect Journal of Neuroscience Methods journal homepage: www.elsevier.com/locate/jneumeth Video imaging system for automated shaping and analysis of complex locomotory behavior Nelson G. Publicover a , Linda J. Hayes b , L. Fernando Guerrero c , Kenneth W. Hunter Jr. d,∗ a Department of Electrical and Biomedical Engineering, University of Nevada, Reno, NV 89557, USA b Department of Psychology, University of Nevada, Reno, NV 89557, USA c Autism Spectrum Therapies, Los Angeles, CA 90230, USA d Department of Microbiology and Immunology, University of Nevada, Reno, NV 89557, USA article info Article history: Received 3 March 2009 Received in revised form 20 May 2009 Accepted 25 May 2009 Keywords: Operant conditioning Shaping Behavior Locomotory Video imaging Arthritis abstract Although many observational technologies have been developed for the study of behavior, most of these technologies have suffered from the inability to engender highly reproducible behaviors that can be observed and modified. We have developed ACROBAT (Automated Control in Real-Time of Operant Behavior and Training), a video imaging system and associated computer algorithms that allow the fully automated shaping and analysis of complex locomotory behaviors. While this operant conditioning sys- tem is particularly useful for measuring the acquisition and maintenance of complex topographies, it also provides a more general and user friendly platform on which to develop novel paradigms for the study of learning and memory in animals. In this paper we describe the instrumentation and software developed, demonstrate the use of ACROBAT to shape a specific topography, and show how the system can be used to facilitate the study of arthritic pain in mice. © 2009 Elsevier B.V. All rights reserved. 1. Introduction While human behavior can often be analyzed through verbal interactions, animal behavior must be directly observed. Animal movement (locomotory or topographical behavior) is easily observ- able and has been widely used in behavioral studies (Ossenkopp and Kavaliers, 1996). In addition to the obvious studies of natu- ral or induced disease states that present with motor dysfunction (Sakic et al., 1996; Costa et al., 1999; Simon et al., 2000; Min et al., 2001), locomotory behavior has also been used as a metric in stud- ies of learning, memory, and other cognitive functions (Reeves et al., 1995; Bach et al., 1995; Rogers et al., 1997; Markowska et al., 1998; Smith et al., 1998; Prusky et al., 2000; Morris et al., 2001). A wide variety of methods have been developed to observe and record animal locomotory behavior (Ossenkopp and Kavaliers, 1996; Young et al., 2000; Fowler et al., 2001). However, the most robust systems involve video imaging, and many such systems have been developed (Livesey and Leppard, 1981; Godden and Graham, 1983; Sanberg et al., 1984; Vorhees et al., 1992; Schwarting et al., 1993; Park et al., 1995; Sams-Dodd, 1995; Hoy et al., 1996; Rousseau et al., 2000; Noldus et al., 2001; Spink et al., 2001; Lind et al., 2005; Tort et al., 2006). While the development of sophisticated tech- ∗ Corresponding author. Tel.: +1 775 327 5255. E-mail address: khunter@medicine.nevada.edu (K.W. Hunter Jr.). nologies for the observation and measurement of movement is a very valuable contribution to the study of animal behavior, in some cases the movements of interest are those illustrating changes in behavior brought about by deliberate means. That is to say, for some purposes the effects of biological or environmental interven- tions are best observed as changes in the speed with which new topographical behaviors are learned, remembered, become elab- orated into more complex forms, or show deviations from stable, established patterns. In such cases technologies designed simply to observe behavior, even in highly precise ways, are not sufficient to answer the questions being addressed. Therefore, the utility of auto- mated systems for the measurement of behavior would be greatly enhanced if systems were designed in such a way as to also control the occurrence of this behavior. Although simple technologies capable of both observation and control of behavior have been reported (Pear, 1985; Pear and Legris, 1987; Hori and Watanabe, 1987; Pear et al., 1989), no fully automated method for the robust shaping of specific topographic behavior presently exists. In this paper we describe ACROBAT (Automated Control in Real-Time of Operant Behavior and Train- ing), a fully automated system that utilizes video imaging and computer-controlled operant conditioning to train complex and highly reproducible locomotory behaviors in rodents. We provide brief demonstrations of the use of ACROBAT in the shaping of com- plex locomotory behavior in mice, and provide an example of its use in the study of pain in a mouse arthritis model. 0165-0270/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jneumeth.2009.05.016