Fieldwork data of artificial prey survival of five different morphs with differences in their transparent element characteristics: O (opaque, 0 transparency), SW (small transparent windows), LW (large transparent windows), BE (large windows touching one prey edge) and B3E (large windows touching 3 prey edges). The experiment was performed in in two forests in southern France with evergreen and white oaks as predominant tree vegetation: La Rouvière forest (43.65°N, 3.64°E) and a large natural wooded area of the Montpellier Zoological Park (43.64°N, 3.87°E), for three 1-week sessions at each place. Artificial prey survival was monitored from predation by bird local communities once per day for four consecutive days. Morph order was randomized within blocks of 25 artificial prey, with 5 items per morph per block. Non-attacked prey were considered censoring data.