Examples using fully online / browser-based platforms
Online Sound Level Setting Procedure
Questions of Interest:
One issue with running online auditory experiments is the lack of control over auditory presentation levels. Here we investigate how feasible it is to manage the range of experimental stimulus amplitude levels in online testing conducted in uncontrolled environments. Our method entails having online participants set the volume on their own computer to serve as a ‘self-calibrated audiometer’ in the context of their specific testing equipment and acoustic surroundings
Platform / Infrastructure:
The experiment was run and hosted on the Pavlovia.org online experimental platform using Google Chrome browser on a MacBook Pro and Beyerdynamic headphones; we have also verified the approach on an older Windows laptop as well as with less expensive over-the-ear headphones.
Participant Recruiting:
Validation of the online volume setting procedure required testing in-person participants on a common consumer laptop with consumer headphones. For the in-person experiments, participants were Carnegie Mellon or University of Pittsburgh students or staff; the same individuals were tested in the outdoor environment as well as in an anechoic sound booth under well-controlled laboratory conditions.
Activities:
The demo of the procedure is available here: https://run.pavlovia.org/sijiazhao/amplitudechecking_demo. The implementation is available in JavaScript [https://gitlab.pavlovia.org/sijiazhao/amplitudechecking_demo] and via the Gorilla experimental platform [https://gorilla.sc/openmaterials/261557].
A 1-second Gaussian white noise was generated, band-pass filtered between 80-8000 Hz, and adjusted to achieve a nominal RMS level of 0.000399. A stereo sound file in the lossless FLAC format was used. 100-ms cosine ramps were added to both ends of the noise, so when it played on a continuous loop without gaps, it would sound like a sequence of gentle noise bursts.
Written instructions presented on the laptop screen asked participants to set the computer’s volume to about 50% and then to click a button labeled ‘play’ to hear the pulsed noise played on a continuous loop until the participant pressed pause or proceeded to the next page. (The continuous loop was achieved by in-house JavaScript code incorporated into the Pavlovia routine). Next, participants were instructed to use the computer’s volume setting buttons to adjust the volume to a level at which the noise was barely audible. Specifically, they were asked to slowly lower the volume until they could no longer hear the noise, and then to increase the volume a very small amount until they could again just barely hear the noise. For online experiments, the participant will be re-directed to start other online experiments. For the in-person validation, after the participant was satisfied with their setting, the experimenter manually entered the final volume for further analysis.
Conclusions / Impressions:
Initial results indicate comparable performance between cochlear implant users who tested in the lab with an audiologist and those who self-tested at home following online instructions. The results are reported in Shafiro, V., Hebb, M., Walker, C., Hsiao, Y., Brown, K., Sheft, S., Li, Y., Vasil, K., Moberly, A., (2020). Development of the Basic Auditory Skills Evaluation (BASE) battery for online testing of cochlear implant listeners. American Journal of Audiology. https://pubs.asha.org/doi/10.1044/2020_AJA-19-00083
Contact:
Sijia Zhao, sijiazhao@psy.ox.ac.uk ; Fred Dick, f.dick@ucl.ac.uk