Stable neural representations are essential for consistent behavior and learning, but recent findings on representational drift challenge our understanding of how individual neurons contribute to stable brain computations over time. While representational drift has been observed across multiple brain regions, its relationship to structural connectivity and homeostatic statistics remains unclear [1]. To address this, we explored how drift interacts with structural recurrency and noise in a recurrent neural network (RNN) model. By simulating synaptic-level noise through random connectivity manipulation, we tested the model under different levels of homeostasis noise and recurrency. Our simulations revealed a strong dependence of drift on connectivity and the statistical properties of noise. Influenced by these theoretical predictions, we studied the stability of place field representations in distal and proximal subregions of the CA3 hippocampus of mice while the animal performed the same task. The targeted subregions offer the advantage of differing in recurrency while sharing similar functional roles, allowing us to isolate the effects of connectivity on drift. Using longitudinal imaging across multiple days, we found greater stability in proximal CA3 place cells, where recurrency is sparser, supporting our exponential noise model for drift. Our findings support the hypothesis that representational drift varies across brain regions depending on structural connectivity, challenging the view that drift is uniform across biological networks. By linking drift statistics to underlying neuronal connectivity, our results offer a mechanistic explanation for the variability of drift dynamics across different brain regions, suggesting that this variability may extend to other sensory, motor, and cognitive areas, where diverse structural connectivity could lead to different degrees of representational stability.