Poster C10, Friday, August 17, 10:30 am – 12:15 pm, Room 2000AB

Talker-anchoring deficit in lexical tone processing in Cantonese-speaking congenital amusics: Evidence from event-related potentials

Jing Shao1,2, Rebecca Yick Man Lau1, Caicai Zhang1,2;1The Hong Kong Polytechnic University, 2Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences

Congenital amusia (amusia hereafter) is a neurodevelopmental disorder that impacts the perception of music negatively. Previous studies have found that amusics are also impaired in lexical tone perception; importantly, amusics are more impaired in taking advantage of low acoustic variation (i.e., syllable and talker variation) in lexical tone perception, suggesting a deficit in anchoring to acoustic constancy (i.e., a constant syllable or talker) in perception. However, a fuller understanding of the neural mechanism underlying such anchoring impairment in tone perception remains to be achieved. To pinpoint the deficiency neural mechanism, we examined the event-related potentials (ERPs) in 24 Cantonese-speaking amusics and 24 controls in an active oddball paradigm. ERPs were recorded during tone change detection in Cantonese tone pairs with small pitch differences (mid level-low level, or T3-T6) vs. large pitch differences (high level-low level, or T1-T6) presented in blocked-talker vs. mixed-talker conditions. In the blocked-talker condition, tone stimuli from each of four talkers were presented within a separated block (i.e., low talker variation); in the mixed-talker condition, tone stimuli from the four talkers were mixed within one block (i.e., high talker variation). All stimuli were presented in an oddball paradigm. Higher talker variability in the mixed-talker condition requires greater attentional resources to process the talker voice idiosyncrasies and form a stable tonal “template”, thus making the detection of the deviant much more demanding in this condition. It is predicted that in general the P300 activities will be reduced in the mixed-talker condition. Given that amusics have deficits in anchoring to the acoustic constancy, amusics were expected to show reduced P300, especially in the blocked-talker condition. The results demonstrated that in the time-window of P3a (350-450 ms after auditory onset), the P3a amplitude was enhanced in the blocked-talker condition relative to the mixed-talker condition, but the effect size was smaller in the amusic group; moreover, the amusic group exhibited significantly reduced P3a amplitude compared to the control group in the blocked-talker condition, but the group difference was not significant in the mixed-talker condition. In the time window of P3b (500-800 ms after auditory onset), the amusic group showed overall reduced P3b amplitude than the control group, irrespective of blocked- and mixed-talker presentation manners. The P3a results suggested that the amusic brain was impaired in consciously processing the novelty of lexical tone changes, with reduced attentional switch to such changes when the talker variation was low. This implies that the attentional resources to attend to a constant talker may be reduced in the amusic brain, which presumably underlies the anchoring deficit observed in amusics. As the P3b is associated with stimulus categorization, the reduced P3b in amusics may imply that the amusic brain was overall impoverished in the ability to categorize lexical tone changes, irrespective of low and high variation conditions. This is consistent with the previous finding that amusics are impaired in consciously detecting and categorizing lexical tone differences. Altogether, these findings shed some light on the neural underpinnings of amusia in lexical tone perception.

Topic Area: Perception: Speech Perception and Audiovisual Integration