The episodic encoding of talker voice attributes across diverse voices

Talker-specificity and token-specificity in recognition memory

“…While such studies have been illuminating for the field and helped highlight the high-level group dynamics at play in memory for speech, it has long been known that variation is a hallmark of language production (e.g., Bell, 1984;Labov, 1966), and there is no one-to-one correspondence between the use of a particular variant and membership in a particular macro-demographic category (Eckert, 2008;. Under a theory of speech perception that treats exemplars as asymmetrically weighted based on social characteristics (e.g., Clapp et al, 2023;Sumner, Kim, King, & McGowan, 2014), and given that the use of sociolinguistic variables ranges widely among individuals, it follows that the strength of lexical encoding and recall may vary between talkers even when they are matched for macro-demographic category membership. The present study is only a small step in squaring the psycholinguistic study of speech perception with recent developments in sociolinguistics, but takes seriously the need for theories of episodic representations to draw on these findings.…”

“…This contrast may help elucidate the distinction between memory for physical acoustic events and memory for higher-level properties of talker attributes. Further, preliminary evidence suggests that talker-specificity effects may be found asymmetrically across individual talkers (Clapp, Vaughn, & Sumner, 2023), but to date, this hypothesis has not been tested directly. If talkers are encoded with different levels of specificity, it also follows that the type of token heard at a word's repetition (novel vs. identical) would influence recognition performance asymmetrically across talkers.…”