Listeners are sensitive to numerous fine-grained acoustic cues in speech. However, there has been little workexamining how listeners encode these cues at early stages of perception. The event-related potential (ERP) technique providesa tool to help us address this. Previous work shows that the amplitude of the auditory N1 ERP component varies with differencesalong VOT continua, but it is not clear which other cues show similar effects. We present data examining a large set of minimalpair stimuli spanning 18 consonants. Results reveal widespread differences in N1 amplitude for stops, fricatives, and nasals,including distinctions primarily caused by temporal cues (stop voicing; /b,d,g/ vs. /p,t,k/) and spectral cues (place of articulation;/b,p/ vs. /d,t/ vs. /g,k/). Our results suggest that early speech processing is based on fine-grained acoustic cues, rather thanarticulatory differences, and that the ERP technique provides a useful tool for measuring speech sound encoding.

In order to understand speech, listeners must weight and combine multiple acoustic cues. For example, voiceonset time (VOT) is a reliable cue to stop consonant voicing, while onset F0 provides information, but is much less reliable.Consequently, we would expect listeners to weight VOT higher than F0. This is the pattern observed for most listeners.However, these cue weights also change over time, and older adults tend to rely less on VOT than young adults, even inlisteners without hearing loss. One hypothesized mechanism for this change is a decreased ability to detect temporal differencesin sounds, which renders temporal cues (e.g., VOT) less reliable and leads to a greater reliance on spectral information (F0). Wesimulate this using a weighted Gaussian mixture model and find evidence in support of this mechanism: decreased temporalcue reliability leads to the same pattern of differences observed between younger and older listeners.