BibTeX download is not available in the pre-proceeding
Despite extensive research on human sentence processing, little is known about whether large audio language models (LALMs) exhibit similar sensitivity to prosodic and semantic cues during syntactic ambiguity resolution. We evaluated the ability of five state-of-the-art LALMs to resolve relative clause (RC) attachment ambiguities across three typologically diverse languages: English, Chinese, and Arabic. We constructed audio stimuli with two types of attachment cues: (1) prosodic cues, implemented through manipulated boundary placement and prosodic prominence to favor either high or low attachment, and (2) semantic cues, in which world knowledge favored one attachment site. Each stimulus was probed using both yes/no and forced-choice prompts. Our findings revealed limited and inconsistent sensitivity to prosodic cues, with low attachment remaining the dominant interpretation. In contrast, models demonstrated greater sensitivity to semantic and world knowledge cues. Forced-choice prompts yielded more stable attachment preferences and fewer uncertain responses than yes/no prompts. These results raise questions about the cognitive capacities of current LALMs, suggesting that their syntactic processing is driven primarily by semantic information, with prosodic cues playing a limited and inconsistent role in ambiguity resolution.