Neuronal correlates of locomotion speed in the dorsal CA1 of the rat hippocampus: new insights on the speed cells

• Autor(a) principal: Góis, José Henrique Targino Dias
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFRN
• Texto Completo: https://repositorio.ufrn.br/jspui/handle/123456789/28788
• topic: CNPQ::OUTROS::CIENCIAS; Spatial navigation; Locomotion speed; Rats; Hippocampus; Neurons; Electrophysiology

Edward Tolman postulated the existence of cognitive maps in the brain to explain the animal capability of spatial navigation. Since then, neuroscience seeks to describe brain components underlying this capability. In the present work, we describe advancements in the characterization of a neural sub-population engaged in coding the scalar velocity of rat locomotion. Upon analyzing an open database, we re-discovered the existence of a velocity correlation present in the rate of emission of action potentials of neurons in the dorsal hippocampus. we found that this correlation is independent of theta oscillation frequency (4- 12 Hz) and stable over space and time; moreover, it also persists in different arenas. We then classified the neurons as excitatory and inhibitory by the action potential waveform shapes, the rate of emission of action potentials, and the temporal inter-dependency of action potential emission between pairs of neurons. This classification revealed that, by using Pearson’s r coefficient (speed score) as a correlation metric, in the square open-field arena, only inhibitory neurons high-correlated with locomotion speed (henceforth, speed), even though there was a modulation of the average of the excitatory neurons. Intriguingly, when checking speed correlates in the linear arena, speed score made indistinguishable the correlation among neuronal classes. We then formulated the hypothesis that the apparent locomotion speed correlation of pyramidal neurons in the linear arena is spurious, due to a by-product of their spatial code. To prove this, we simulated artificial neurons that emitted action potentials influenced by the actual animal behavior; the simulated neurons coded either speed or space. The simulation replicated real data Pearson’s r coefficient classifications – true positives in the square arena and false positives in the linear arena for those cells encoding speed. To solve this ethological interdependency, we adopted a new metric of speed correlation – the ratio of the difference of linear-non-linear models prediction accuracy based on speed and position over the prediction accuracy of the two-variable model. This analysis disentangled the ethological issue, satisfactorily classifying the simulated neurons and confirming the spurious hypothesis correlates, and the prevalence of speed correlates in interneurons. The results of the present work demonstrated that a genuine speed correlation is present in the dorsal CA1 of the hippocampus of rats in inhibitory neurons.