The role of segmental and tonal information in Thai spoken word recognition

Abstract

An issue that continues to be debated in psycholinguistic research concerns the role of phonological information in spoken word processing, particularly segmental information (e.g., consonants and vowels) in comparison with suprasegmental information (e.g., lexical tone). While segmental information can be critical for lexical access across languages, tonal languages like Thai and Mandarin also rely on variations in tonal pitch which are fundamental for distinguishing word meaning. Previous research on this topic has found mixed results: some suggest segmental cues are more important, while others emphasize the role of lexical tone in speech processing. This study investigated the relative importance of segmental and suprasegmental phonological information in Thai spoken word processing by examining the effects of vowel, consonant, and tone priming on word recognition using an auditory lexical decision task. Native Thai speakers heard prime-target word pairs that varied in the degree of phonological overlap in consonants, vowels, and lexical tone. Depending on the condition, the prime and target either shared all three phonological components or overlapped only partially in segmental or tonal information. The findings revealed faster response times when the prime and target shared phonological information across all three components—onset, rime, and tone—compared with a phonologically unrelated baseline condition. Conversely, no facilitation effects were observed in the conditions with only partial phonological overlap. One possible explanation is that the partial-overlap prime stimuli were nonwords, which may not have provided sufficient phonological information to effectively activate lexical representations during spoken word recognition. These results contribute to models of spoken word recognition by demonstrating that a single segmental or tonal cue may not be sufficient to facilitate spoken word recognition under nonword priming conditions, providing a valuable cross-linguistic comparison relevant to existing speech processing models in other tonal languages.