An online headphone screening test based on dichotic pitch

Abstract

Online experimental platforms can be used as an alternative, or complement, to lab-based research. However, when conducting auditory experiments via online methods, the researcher has limited control over the participants listening environment. We offer a new method to probe one aspect of that environment, headphone use. Headphones not only provide better control of sound presentation but can also "shield" the listener from background noise. Here we present a rapid (< 3 minute) headphone screening test based on Huggins Pitch (HP), a perceptual phenomenon that can only be detected when stimuli are presented dichotically. We validate this test using a cohort of "Trusted" online participants who completed the test using both headphones and loudspeakers. The same participants were also used to test an existing headphone test (AP test; Woods et al., 2017, Attention Perception Psychophysics). We demonstrate that compared to the AP test, the HP test has a higher selectivity for headphone users, rendering it as a compelling alternative to existing methods. Overall, the new HP test correctly detects 80% of headphone users and has a false positive rate of 20%. Moreover, we demonstrate that combining the HP test with an additional test - either the AP test or an alternative based on a beat test (BT) - can lower the false positive rate to [~]7%. This should be useful in situations where headphone use is particularly critical (e.g. dichotic or spatial manipulations). Code for implementing the new tests is publicly available in JavaScript and through Gorilla (gorilla.sc).