Hello all! While nearly every speech perception study I’ve ever done has eliminated any speech level differences among talkers and speaking conditions, I of course chose a global pandemic to do an experiment where some of the level differences
are being kept. My PhD student has been doing excellent work this summer running another experiment in Gorilla. She’s been trying very hard in Gorilla and jsPych to find a way to ensure that listeners don’t do anything to the presentation level while they’re
in the experiment. The test blocks will contain sentences spoken in 7 different conditions that at a recording dry run varied in level over a range of 15-20 dB.
My student has tried things like having the listener max out their
system volume and then adjust the browser volume to a certain target during the listening screening, but hasn’t found a way to prevent them from adjusting their level again (like they might be tempted to do if they got a couple of lower-level
stimuli in a row). If we can’t make this work online, we’re going to have to do it in my lab, and at my university’s current status, that will be challenging to do.
Any suggestions?
Sarah Hargus Ferguson, PhD, CCC-A
Associate Professor
Director of Undergraduate Studies
Department of Communication Sciences and Disorders
390 South 1530 East, Room 1201
Salt Lake City, UT 84112
Office: 801-585-6776
Speech Acoustics and Perception Laboratory: 801-585-0036
