Fig. 3.

Sensory perception is required for the induction of aversive cues following exposure to dead flies. a When flies were exposed to dead flies in the dark, they failed to evoke avoidance behavior in naive choosing females (N = 19 for light exposure and N = 20 for dark exposure). b Blind, norpA mutant flies exposed under lighted conditions also failed to induce aversive cues following exposure to dead flies (N = 8 for each treatment). c Orco² mutant flies, which have impaired olfaction, evoked a small, but significant, avoidance behavior in choosing females following death exposure (N = 9 for Orco² and N = 8 for control). d Flies carrying the Poxn^{ΔM22-B5-ΔXB} mutation, which have impaired taste function, exhibited a similar induction of aversive cues in response to death exposure as did control flies (N = 9 for each treatment, P = 0.002 for control and P < 0.001 for poxn). e The sight of starvation-killed flies was sufficient to induce aversive cues to the same extent as direct exposure, while the sight of flies killed by immersion in liquid nitrogen had no effect (N = 19 for direct exposure, N = 20 for vision only). f The smell of starvation-killed flies, which was provided by isolating dead animals behind a fine mesh screen, failed to induce aversive cues (N = 10 for each treatment). g The sight of flies killed by freezing in liquid nitrogen failed to induce aversive cues (frozen; vision only), and this was not affected by simultaneous smell of starvation-killed animals (N = 8 for frozen only, N = 13 for starved and starved + frozen). h Homogenized dead flies failed to evoke avoidance behavior in naive flies (N = 10 for ground up and control treatments). For binary choice assays, all exposed flies and naive choosing flies were from the Canton-S strain. Each T-maze sample tests 20 flies. Error bars represent standard error of the mean (SEM). P values for binary choice were determined by non-parametric randomization