Dataset: "Auditory Localization of Multiple Stationary Electric Vehicles"

This repository contains data accompanying the publication "Auditory Localization of Multiple Stationary Electric Vehicles", published in the Journal of the Acoustical Society of America (JASA))

The following abbreviations are used in the filenames and data

c
Combustion Noise

n
Noise AVAS

t
Two-Tone AVAS

m
Multi-Tone AVAS

le
Localization Error in °

lt
Localization Time in s

fl
Failed Localizations in %

The dataset contains:

stimuli.zip: Sound pressure of all evaluated stimuli as calibrated 32-bit float .wav files

trials.zip: Binaural sound pressure recordings of all 72 trials as calibrated 32-bit float .wav files, including parking lot background noise. These recordings were obtained by placing a HeadAcoustics HMS-V artificial head at the listening position. The files are named as TrialNr_Stim1Abbreviation_Stim1PositionInDegree_Stim2Abbreviation_Stim2PositionInDegree_Stim3Abbreviation_Stim3PositionInDegree.wav

experimentProcedure.mp4: Video showcasing the experiment procedure.

rawData.xlsx: The unprocessed experiment data. Each row represents one individual trial.

processedData.xlsx: The pre-processed experiment data for each participant. Each column represents the mean localization error (le), mean localization time (lt), or percentage of failed detections (fd) for a single stimulus. E.g., le_mn_m is the mean localization error of the Multi-Tone AVAS, averaged for all 4 trials where Multi-Tone AVAS and Noise AVAS were played simultaneously (TrialNr. 61-64, see Table in Paper). fl_ccc is the percentage of failed localizations for the Combustion Noise in the 4 trials where three combustion sounds were played simultaneously. For the statistical evaluations, participants "HZPS", "TGCI" and "XKOU" have been removed as outliers.

All stimuli are compliant with both UNECE R138 and US FMVSS 141 as illustrated in Figure 3 of the paper.