In contemporary neuroscience, a traditional representation approach is complemented by growing interest in neuronal dynamics. We used tetrode technique to study in hippocampal CA1 activity how spatial representations and neuronal dynamics are affected by environmental and task complexity and by rats’ prior learning experience. Each neuronal ensemble was recorded in four recording sessions: a session where a rat was exposed to a stationary rat-sized robot and a session with the robot slowly moving, were flanked by two sessions without the robot. Some rats were trained to avoid a location defined relative to the robot; other rats were not trained for avoidance behavior. Spatial organization of putative pyramidal cell activity was weaker and less stable across sessions in trained compared to untrained rats. Correlations in firing of pairs of cells were higher in trained rats and were relatively stable across recording sessions in both groups. Dynamics of simultaneously recorded ensembles of neurons was further assessed by correlations of ensemble activity vectors calculated for one-second intervals and visualized by dimensionality-reduction methods. So far, no differences in stability and dynamics of ensemble vector activity between trained and untrained rats and no effects of type of recording session were detected. In summary, the main characteristics of neuronal dynamics were similar regardless of training experience or conditions of recording session. Cells’ firing rate correlations were more stable than spatial responses. Training for a more complex avoidance behavior increased variability in spatial organization of firing. Work was supported by Czech Science Foundation grant 22-16717S.