Data from: Predicting and measuring decision rules for social recognition in a Neotropical frog

Many animals use signals to recognize familiar individuals but risk
mistakes because the signal properties of different individuals often
overlap. Further, outcomes of correct and incorrect decisions yield
different fitness payoffs, and animals incur these payoffs at different
frequencies depending on interaction rates. To understand how signal
variation, payoffs, and interaction rates shape recognition decision
rules, we studied male golden rocket frogs, which recognize the calls of
territory neighbors and are less aggressive to neighbors than to
strangers. We first quantified patterns of individual variation in call
properties and predicted optimal discrimination thresholds using signal
variation. We then measured thresholds for discriminating between
neighbors and strangers using a habituation-discrimination field playback
experiment. Territorial males discriminated between calls differing by 9%
to 12% in temporal properties, slightly higher than the predicted
thresholds (5-10%). Finally, we used a signal detection theory model to
explore payoff and interaction rate parameters and found that the
empirical threshold matched those predicted under ecologically realistic
assumptions of infrequent encounters with strangers and relatively costly
missed detections of strangers. We demonstrate that receivers group
continuous variation in vocalizations into discrete social categories and
that signal detection theory can be applied to understand evolved decision
rules.