Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Revisiting Laurel vs. Yanny: insights from an online experiment and an FFR experiment
Poster Session D, Saturday, September 13, 5:00 - 6:30 pm, Field House
Ja Young Choi1, G. Nike Gnanateja2, Anoop Basavanahalli Jagadeesh3, Bharath Chandrasekaran1; 1Northwestern Univeristy, 2University of Wisconsin-Madison, 3University of Montana
In May 2018, a short audio clip went viral on social media, sparking widespread debate over whether it said “Laurel” or “Yanny.” The acoustic basis for this perceptual ambiguity was quickly identified as an unusually strong resonance that drops from 3 kHz to below 2 kHz, which can be interpreted either as the third formant of /ɔ/ in “Laurel” or as the second formant of /i/ in “Yanny,” depending on the listener. The basis of this perceptual variability between listeners and change within listeners remains unclear. In this study, we revisited the viral phenomenon to examine the factors contributing to these perceptual differences. A total of 598 participants took part in this online experiment. Following the approach of Pressnitzer et al. (2018), we used an 11-step audio continuum, created by mixing highpass- and lowpass-filtered versions of the original clip at varying amplitude ratios. The sixth step in the continuum corresponded to the unaltered original sound. During the main task, participants listened to each of the 11 stimuli three times in a random order and reported whether they heard “Laurel” or “Yanny” on each trial. They also reported their demographic information, whether they had previously heard the “Laurel/Yanny” clip, and, if they had, what they had heard at the time. We found a significant effect of both age and sex on the likelihood of perceiving the ambiguous sound as “Laurel.” Older participants and men were more likely to hear “Laurel” than younger participants and women. Audio equipment also played a role: participants using in-ear earbuds or over-ear headphones were more likely to perceive “Laurel” than those using computer speakers. Interestingly, among the 26 participants who had previously heard the original clip as “Yanny” only, 38% perceived it as “Laurel” in the current study. In contrast, of the 55 participants who had previously heard “Laurel” only, just one switched to hearing “Yanny.” These findings suggest that auditory perception is shaped by a combination of stable individual traits and changing factors such as age, listening environment, and experience. In a separate part of the study, we recorded frequency-following responses (FFR) from 32 young adults listening to four versions of the stimulus: the viral clip, the original “Laurel” recording from vocabulary.com, and its highpass- and lowpass-filtered versions. Both the envelope and the temporal fine structure of the neural responses were cross-correlated to each of the four stimuli. No significant differences in neural response patterns were observed between participants with different perceptual experiences of the clip. Overall, our findings highlight the complexity of auditory perception and demonstrate how it can be influenced by a variety of biological, environmental, and experiential factors.
Topic Areas: Speech Perception,