Change detection is biased to listeners' known languages in complex auditory scenes

When we cross a busy street, we may hear traffic sounds, strangers' conversations, birds chirping, and music all at once. Despite the presence of many possible sounds, we are biased to detect changes to speech over other real-world sounds in our environments. Yet, it is unclear what factors are responsible for listeners' biases toward speech: Is there a processing bias toward all speech or is this bias sensitive to the learned acoustic patterns of listeners' known languages? Here, we examined whether listeners' language backgrounds influenced change detection of speech in auditory scenes. Adult listeners from three language backgrounds (English monolingual, English-Mandarin bilingual, and English-other language/non-Mandarin bilingual) completed a change detection task in which they determined whether pairs of auditory scenes consisting of speech (nonsemantic English and Mandarin sentences), music, environmental, and animal sounds were the same or different from each other. We replicated the attentional bias toward speech and showed that only English-Mandarin bilinguals detected Mandarin speech changes as well as English speech changes. Beyond a general bias toward all human speech, listeners are biased toward speech in their known language(s). Our work suggests that linguistic experiences contribute to encoding and remembering communicative sounds in busy sound environments. (PsycInfo Database Record (c) 2026 APA, all rights reserved).