What approach should I use for remote testing?
There are really three big questions you need to answer when deciding how to set up an auditory experiment for remote data collection: what hardware will I use, what software will I use, and who are the subjects I want to test. In all three cases, the alternatives range from convenient and less controlled to more time consuming and well specified.
|
hardware: calibration & interfaces
|
- loose control of auditory stimuli with respect to calibration & frequency response
- flexible graphic and temporal specs
- no “special” data collected (e.g., ambient noise levels)
|
Any user hardware
e.g., PC & headphones
or more controlled solutions (below)
|
- stimulus level & frequency response defined within ~5 dB
- controlled graphic and temporal specs
- some specialized data collection capabilities (e.g., touch screen input)
|
specified hardware
e.g., iPad
or more controlled solution (below)
|
- strict control of level and freq. response
- other specialized measurement or interface required (e.g., calibrated ambient noise recordings, software permitting)
|
lab hardware
e.g., deliver tablet & sound level monitor
|
|
software: data handling and experimental control
|
- standard procedures w/out modifications required
|
Preconfigured hearing research packages
e.g., PART
or more controlled solutions (below)
|
- custom stimuli
- standard interface and response sufficient
|
build-your-own systems
e.g., Gorilla
or more controlled solution (below)
|
- custom procedures or real-time processing
- non-standard interface or data collection (e.g., voice recordings, hardware permitting)
|
fully custom scripts
e.g., MATLAB or Python
|
|
subjects: demographics & instruction
|
- anyone can participate
- no one-on-one instruction required
|
anonymous & unsupervised
e.g., Mturk
or more controlled solutions (below)
|
- targeted population
- specialized instruction (e.g,. via zoom)
|
“by invitation” access
most hardware/software will work
|
- populations requiring real-time supervision (e.g., children)
- protocols requiring rigorous training
|
supervision by proxy (e.g., parent)
experimenter-administered protocol hybrid model (e.g., training in person, data collection at home)
most hardware/software will work
|
