Asymmetric effects of sudden changes in timbre on auditory stream segregation

These datasets comprise the stimuli and listeners’ judgments of stream segregation for Experiments 1 and 2 of the article of the same title (Roberts and Haywood, 2023; Journal of the Acoustical Society of America).

There are two spreadsheets for each experiment, one comprising the results when responses to the test sequence are divided into 1-s time bins (segregation scores) and the other comprising the keypress timing data for every trial.

Each spreadsheet for the segregation scores comprises four worksheets showing the extent of stream segregation (proportion of time the stimulus was reported as segregated) per time bin for each condition. The "Summary" worksheet shows the means and inter-subject standard errors when the results are averaged across listeners and is organized into 1-s time bins. The "DF2", "DF4", and "DF6" worksheets show the results separately for each listener for frequency separations between low and high subsets of sounds of 2, 4, and 6 semitones, respectively. There are two tables in each of the "DF2", "DF4", and "DF6" worksheets. The upper table contains cells coded 'Nan' (missing values) indicating cases for which no mean could be computed (see article for full details). In the lower table, these missing values have been replaced using mean imputation. All figures and statistics presented in the article used the data with mean imputation.

Each spreadsheet for the raw data comprises a single worksheet showing, for each listener on every trial, the timing of each keypress and the nature of the response (one stream or two). For each trial, the trial number, condition number, frequency separation, and stimulus file name are identified. Note that all responses were recorded but instances where a given response was the same as the previous one were ignored when computing the time-bin data.

Also available are two Zip files, one for each experiment, containing the stimuli used (WAV files). All have been resynthesized at the standard CD sampling rate of 44.1 kHz (48 kHz was used during the experiments) and reduced from full scale for comfortable listening. The stimuli have also been padded with 100-ms silences at onset and offset. These silences were not present during the experiments but have been inserted here to remove any onset/offset artefacts that might otherwise be heard with some playback software. Details as follows:

Experiment 1: 
Transitions (if present) occur 6 s after the onset of the first sound. Filenames are of the form:
COND{N}_{Stimulus type <6 s}_{Stimulus type >6 s}_{Δf}.wav 
•	N is the condition number (1-9)
•	Stimulus types are pure tones (PT), slower-modulation dyads (SD), or faster-modulation dyads (FD) 
•	Δf is the frequency separation between the H and L subsets of sounds (2, 4, or 6 semitones)

Experiment 2: 
Transitions (if present) occur 6 s after the onset of the first sound. Filenames are of the form:
COND{N}_{Stimulus type <6 s}_{Stimulus type >6 s}_{Δf}.wav 
•	N is the condition number (1-11)
•	Stimulus types are pure tones (PT), slow-attack tones (SA), fast-attack tones (FA), narrow-band noises (NB), and faster-modulation dyads (FD) 
•	Δf is the frequency separation between the H and L subsets of sounds (2, 4, or 6 semitones)
•	NOTE: Five examples have been provided (suffixes _1 to _5) for each sequence involving NB stimuli (conditions 4, 8, and 9)