Spatial cognition is crucial for navigation and survival and has been extensively studied in mammals, thereby revealing specialized cell types such as place cells, grid cells, and head direction cells. Avian species like titmice, quails, barn owls, streaked shearwaters, and black-capped chickadees also exhibit these cell types. Despite pigeons demonstrating remarkable spatial cognition and navigation abilities, the neural basis remains poorly understood. Previous research suggests that spatial coding in neurons, such as place cells, can be influenced by factors like distances from walls or environmental shape.This study addresses the often-overlooked aspect of encoding geometry and topology in hippocampal coding, offering significant computational advantages. To explore this in pigeons, we designed a resizable experimental arena. Experiments investigated the neurons associated with spatial cognition in pigeon hippocampus. The enclosed 5m x 2m x 2m arena allowed free pigeon movement, with 18 food feeders on the ceiling for random food drops. The arena's long axis direction was resized to half (2.5 m x 2 m x 2 m) and one-third (1.67 m x 2 m x 2 m) for experimentation. Neural responses were recorded using an 8-channel, 4-shank linear silicon probe, and chronic recordings over two months or longer were achieved by altering electrode positions. Pigeons freely moved in the arena during wireless neural response recordings for 30 minutes to 1 hour. A machine-learning approach facilitated the extraction of behavioral parameters like moving trajectory, head direction, and speed. This approach successfully obtained multiple spatial cell types relevant for encoding spatial properties.