Figure 2
From: Sharing an environment with sick conspecifics alters odors of healthy animals
Biosensor mouse performance in Y-maze odor discrimination assays. Biosensor mice (N = 9) initially underwent reward-based training to identify the arm of the Y-maze containing urine odors from sick/LPS-injected versus healthy/PBS-injected mice, and were reinforced to (i.e., received a reward for choosing) odors of sick conspecifics. Here, we display performance of biosensors in extinction/unrewarded training trials, novel donor validation trials, and experimental generalization trials (for full descriptions of trial types see the “Biosensor mouse care and testing in the Y-maze” section of our Methods). For extinction and validation trials, bar height corresponds to the mean percentage of trials in which biosensor mice correctly identified the arm of the Y-maze containing odors of sick conspecifics. For generalization trials, bar height corresponds to the mean percentage of trials in which the same biosensor mice chose the arm of the Y-maze containing odors of exposed, versus healthy, conspecifics. Error bars represent +/−1 SEM and individual biosensor performance is denoted with open circles. The two biosensor mice below the 50% line were not the same individuals in validation and generalization trials. All nine circles corresponding to individual biosensor responses may not be visible because of overlapping symbols/biosensor performance. On average, the responses of biosensor mice within each trial type differed significantly from 50/50, or random choice of the odors/arms of the Y-maze (independent exact binomial tests for each trial type were all P < 0.01).
