Lexical status influences response variability in older adults’ speech perception

Speech sound categorization is a building block of downstream language processes, however, speech categorization can itself be influenced by lexical knowledge: Speakers tend to resolve ambiguous speech sounds in favor of real-word rather than non-word interpretations (Ganong,1980). Some studies have shown that older adults exhibit an increased lexical bias (e.g., Mattys and Scharenborg, 2014). However, previous work establishing this link has relied on an experimental paradigm in which participants are forced to make a binary choice, a task that may be overly sensitive to higher-level strategies and can obscure the distinction between responses that are highly variable and highly gradient (Apfelbaum et al., 2022). We thus asked if the link between aging and lexical bias persists in a more sensitive paradigm, in which participants indicated the extent to which a stimulus matches one of two categories using a Visual Analogue Scale (VAS). We performed an online study (n = 60) in which older and younger adults responded to a nine-step continua between either two words, or a word and a non-word (à la Ganong, 1980). Preliminary results suggest that while older adults are not necessarily more lexically biased generally, those with increased response variability do show a stronger lexical bias.