Intonations in different languages serve the universal function of conveying communicative information and expressing affective meaning, while the prosodic encoding of the same intonational meaning is language-specific. For native speakers of Uyghur in China learning Mandarin as L2 and English as a third language (L3), few studies have been on their cross-linguistic acquisition of prosodic encoding of interrogative and declarative intonations in non-native languages. Applying the L2 Intonation Learning theory (LILt), this study investigated the prosodic patterns of declarative and interrogative intonations in Uyghur, Mandarin, and English by 20 Uyghur speakers and compared them with 20 L1-Mandarin speakers and 10 L1-English speakers. Findings showed that the prosodic encoding of Mandarin intonation by Uyghur speakers was assimilated to Uyghur. When the pitch competition between lexical tones and interrogative intonation occurred, Uyghur speakers prioritized the prosodic encoding of intonation rather than the lexical tones, resulting in the gradual rising pitch level and rising boundary tones in Mandarin questions with the dipping tones (T3) and falling tones (T4). Regarding their English intonation, the delayed divergence between the downward declarative and upward interrogative intonations suggests that Uyghur speakers relied more on the sentence-final prosody in L1-like norms to distinguish the two intonations.