ABSTRACTAnimals' assessments of predation risk are influenced by a variety of external and internal factors, including predator space use. However, it remains unclear what variables mediate prey species behavior within a landscape where predation risk is heterogeneous. To address this, we employed three assays to examine zebra (Equus quagga) responses to varying predation risk in a multiple‐use area of northern Tanzania: (1) quantifying head‐up posture as a proxy for vigilance through direct behavioral observation in areas of high and low likelihood of lion (Panthera leo) presence, (2) quantifying head‐up posture as a proxy for vigilance when exposed to a lion roar playback, and (3) measuring flight initiation distances (FIDs) when approached by a person. Using generalized linear (mixed) models, we tested how lion space use and habitat type (as proxies for predation risk), normalized difference vegetation index (NDVI, as proxy for primary productivity), time of the day, and zebra‐related variables (sex‐age category, zebra herd size, group size including heterospecifics, and location within the herd) influenced vigilance and flight responses. We found that (1) neither vigilance nor FID were markedly influenced by estimated lion space use, habitat type, and NDVI; (2) vigilance decreased with group size, was lower for zebras positioned centrally in the herd, and during midday; (3) FID increased with a greater number of associated heterospecifics; and (4) zebras increased vigilance when exposed to lion roar playbacks, irrespective of lion space use. These findings suggest that zebra vigilance and flight behavior are not necessarily mediated by spatial variation in apparent predation risk but instead reflect a strategy of maintaining a consistent monitoring of possible threats across the landscape. Rather than relying on spatial clues alone, zebras primarily mitigate predation risk by increasing group size and associating with other species.