Ferramentas de software para caixa de estudo comportamental de roedores em tarefas de discriminação tátil

• Autor(a) principal: Lima, Renan Araújo de
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFRN
• dARK ID: ark:/41046/001300001dzqq
• Texto Completo: https://repositorio.ufrn.br/handle/123456789/53026
• topic: Neurofisiologia do comportamento; Arquitetura de software; Software de supervisão; Visão computacional; CNPQ::ENGENHARIAS::ENGENHARIA ELETRICA

In the area of Neuroscience, the topic line of Neurophysiology of Social Behaviors in Rodents has as one of its objectives to investigate correlated electrophysiological models of social behaviors in rodents. Examples of studies in this line include the investigation of neural processing disorders and synaptic plasticity in animal models of autism and changes in synaptic plasticity in animal models of epilepsy. To carry out these and many other types of studies, sophisticated (and expensive) equipment is normally used. Most existing platforms are proprietary brands, being true black boxes in the sense of not allowing any change in their hardware and software. Still, many platforms are old, being obsolete, and with outdated technology, becoming prominent the need for improvement in this equipment to aid research, seeking alternatives with a cheaper cost, open construction, and using more current components. With that in mind, in this work we propose improvements in the software tools responsible for data acquisition and for controlling the necessary operations in the experiments, using as a baseline a new behavioral box model, also proposed by our group. Specifically, we contribute on three fronts. At first, we developed software aimed at automating the functions in the behavioral box developed for tactile discrimination tasks, which, in its new version, uses the Arduino microcontroller. Also, we developed the software part of the supervisory system, in Python, which is responsible for communicating with the Arduino, for viewing and defining the tasks to be performed in the behavioral box. Finally, we developed some basic computer vision routines for locating the rodent within the behavioral apparatus, indicating its position and orientation. The developments of this work collaborate so that the behavioral box can be automated, reducing the manual interaction with the experimenter, and thus contributing to improving the process of studying tactile discrimination in rodents.