Data for Reichert & Ronacher, "Noise affects the shape of female preference functions for acoustic signals", Evolution, Raw data Compiled by: Michael Reichert Institute for Biology Abteilung Verhaltensphysiologie Humboldt-Universität zu Berlin Invalidenstr. 43 10115 Berlin Germany reichems@hu-berlin.de Files are saved as tab-delimited .txt files. Each file contains the responses of females to a given Experiment. All of the trial types within an experiment are included in a single file (see Methods). Each row describes the response of a given female to the specified combination of stimulus characteristic and noise level. The female ID variable ("Indiv", column 1) identifies individual females within a trial type; however these do not correspond with females in a different trial type. Within each Experiment (except for Experiment 1, where this variable was not applicable because there was only a single trial type), a separate column ("CombinedIndiv") is a female ID variable that is valid across trial types. Some females were tested in more than one trial type and this is a global identification variable that denotes which female participated in which trial types. Note that this variable is only valid within an Experiment. Comparisons were not made across Experiments, so the "CombinedIndiv" columns do not correspond to one another across experiments. The name of the file specifies that song characteristic that was varied (song variant) to generate the preference function and denotes the corresponding Experiment and Figure in the text. The column headings are defined as follows: "Indiv": (as described above) identifies individual females within a trial type; however these do not correspond with females in a different trial type. "CombinedIndiv": (as described above) a female ID variable that is valid across trial types. Some females were tested in more than one trial type and this is a global identification variable that denotes which female participated in which trial types. Note that this variable is only valid within an Experiment. Comparisons were not made across Experiments, so the "CombinedIndiv" columns do not correspond to one another across experiments. This variable is not applicable to Experiment 1. "Captive": is a term for whether the individual originated from a wild population, where it was caught as a nymph (Captive=0) or from eggs that were reared in the lab (Captive=1). "Noise": gives the Noise level in dB SPL. Note that the "no noise" treatment here is coded (arbitrarily) as 20. This does not affect the analyses because Noise was modeled as a factor. The next column gives the value of the song characteristic that was varied to generate the preference function. These were for Experiments 1-4, respectively: "OnsetAccent" The onset accentuation in dB SPL relative to the stimulus plateau; "Pause" the pause duration in ms; "Offset" the offset accentuation defined as the amount in dB SPL by which the amplitude of the pause was decreased relative to that of the stimulus plateau (which always had an amplitude of 70 dB), "GapDur" the duration in ms of gaps within syllables. "Resp": gives the female responses as a proportion of the total number of repetitions of the specified signal/noise combination (normally 20 repetitions, see "Weight" column) "Weight": gives the number of times the female was exposed to the signal and noise combination specified in the preceding columns. This was normally 20 but occasionally a number less than 20 when the equipment malfunctions before a test was completed. This value was entered as a "weight" variable in the GEE model "NumResp": gives the number of times the female responded to the stimulus "NumNonResp": gives the number of times the female did not respond to the stimulus "TrialType": denotes the trial type, for those Experiments in which a preference function for one characteristic was measured at multiple values of a second characteristic (not applicable for Experiment 1). Values correspond to the characteristic that differentiated the different trial types (for Experiment 2 & 3, values correspond to the onset accentuation of the stimulus in dB SPL; for expt. 4, values correspond to the pause duration of the stimulus in ms). See methods and figures for further details.