Edges are ecologically important environmental features that have been well researched in agricultural and urban landscapes. However, little work has been conducted in flammable ecosystems where spatially and temporally dynamic fire edges are expected to influence animal activity patterns, particularly for animals reliant on vegetation for cover, foraging or nesting. We quantified the response of ground-dwelling mammals to fire edges and sought to determine whether animal activity mirrored temporal changes in regenerating understorey vegetation. We used a space-for-time substitution sampling design and selected a series of 26 treatment sites burnt by prescribed fire, where time since fire ranged from 0 to 7 years. Ten long-unburnt sites acted as controls. At each treatment site we identified a burnt/unburnt edge and used camera traps and Elliott traps to survey ground-dwelling mammals on either side. Habitat structure was measured at all 36 sites. We used general and generalised linear mixed models to determine the response of both habitat and animals to time since fire on both burnt and unburnt sides of edges. In addition, we used a resource selection index to assess the congruence between changes in understorey complexity and animal activity identified in the first set of analyses. On the unburnt side of the edge understorey complexity remained constant over time. On the burnt side understorey complexity followed a hump-shaped trend, peaking at 3 years post-fire where it exceeded the level of complexity on the unburnt edge. Larger animals with general resource requirements were more active at burnt compared to unburnt edges immediately after fire, but similarly active on both sides of fire edges from three years post-fire. Despite some activity on the burnt side of edges immediately after fire, small mammals were generally less active on burnt edges compared to unburnt edges for up to three years. Native species’ activity did not follow patterns of temporal change in structurally complex understorey vegetation. For all species, selection was strongest at recently burnt edges with little vegetation and substantially lower at 3-7 year old regenerating edges where understorey complexity was higher. In general, patterns of selection on the unburnt side of edges were similar over time. Our findings suggest that vegetation change on the burnt side of fire edges may not be a good predictor of native mammal use. Foxes and cats, exotic predators in our system, were also using the burnt edge more than expected immediately after fire. Immediate post-fire predation may be higher at fire edges than elsewhere, and recently burnt edge zones could be suitable locations for integrated predator and fire management.