Table 1.
Characteristics of G. rubens calling songs (n = 25) on which the parameters of the test stimulus were based.
Figure 1.
Proportions of positive phonotaxis (A) and acoustic response (B) toward the test stimulus in the playback experiment.
At a given age, males were tested with four intensity levels (0 dB SPL = 20 µPa) over a two-day trial period (n = 49 males and 579 trials). See text for experimental methods.
Table 2.
Contingency table of positive phonotactic and acoustic responses across three age classes and four intensity levels in the playback experiments.
Table 3.
The best models for positive phonotactic (a; -2 Res Log Pseudo-likelihood = 3143.73, Generalized chi-square = 341.47, Generalized chi-square/df = 0.59) and acoustic (b; -2 Res Log Pseudo-likelihood = 2916.42, Generalized chi-square = 422.65, Generalized chi-square/df = 0.73) responses determined by a generalized linear mixed model with forward selection.
Figure 2.
Distributions of song characters produced in the playback experiment (n = 96).
In the top panel (A), pulse rate is plotted against pulse duration, and in the middle panel (B), trill rate is plotted against trill duration. Rectangular areas in both panels represent the ranges of respective characters in calling songs of field-caught males (see Table 1). The box plot in the bottom panel (C) shows the distribution of carrier frequency of songs produced in the playback experiment. Top, middle, and bottom lines of the boxes indicate quartiles, and the upper and lower whiskers indicate 90th and 10th percentiles, respectively. X denotes an outlier. The vertical bar represents the range of carrier frequency in calling songs of field-caught populations (see Table 1).
Table 4.
Result of the analysis of variance to test for variation in properties of acoustic responses over three age classes and four intensity levels.