Abstract Damage to axonal tracts of the central nervous system results in costly and permanent disability. The observations of aborted neurite outgrowth and disorganized scarring in injured central nervous system tissue have motivated the hypothesis that engineered bridging devices might facilitate regeneration. It is thought that both the shape and surface chemistry are important design parameters, however, their relative importance is poorly understood. Previously, we utilized smooth cylindrical surfaces to demonstrate that surfaces designed with directionally varying curvature bias in a stereotyped way postnatal dorsal root ganglion axonal regeneration in the direction of minimum curvature independent of surface chemistry. In the present study, we extend this analysis to include adult dorsal root ganglion neurons and cerebellar granule cells, cell types more representative of the challenge faced clinically. We found that axonal outgrowth of both the adult neuron and the central neuron was less sensitive to substrate curvature than the outgrowth of the postnatal neurons. These differences were quantified by constructing distributions describing the probability of outgrowth for a defined range of surface curvatures. Both the adult neuron and the central neuron exhibited a higher probability of extension in high-curvature directions compared to the postnatal neuron implying that surface geometry may not be as potent a cue in directing the regeneration of these neurons. A microtubule-stabilizing agent enhanced the sensitivity to curvature of the adult neuron, partially reversing the increased probability of growing in a high-curvature direction. The results suggest novel methods to enhance directed neuron regeneration using bridging substrates.