Cueing and prior knowledge affect learners’ bottom-up and top-down processing in multimedia learning respectively. Researchers of previous studies paid much more attention to some questions, namely “whether”, “what kind of” or “where” cues should be presented, and a strong cueing effect was confirmed as expected (for a review, see De Koning, Tabbers, Rikers, & Paas, 2009). However, when to present cues, that is, the temporal synchronization between the presentation of cues and words, which might significantly affect the integration of texts and diagrams (Scheiter & Eitel, 2015; Twyford & Craig, 2013), was neglected by previous researchers. In this study, an animation related to the process of chemical synaptic transmission (Ozcelik, Karakus, Kursun, & Cagiltay, 2009) were used as multimedia learning materials in which the words were presented in an audio format, rather than a visual format, to adhere to the modality principle (see Mayer, 2009). One hundred and twenty-five high (HPK) or low prior knowledge (LPK) college students (96 women) aged between 17 and 27 were recruited as participants and randomly assigned to three cueing conditions, namely advanced cues, synchronized cues, and delayed cues. A total of 13 keywords of commentaries (e.g. action potential) and corresponding cues in the picture were reasonably selected in each condition. Cues were respectively presented 2.6s before or after the corresponding commentaries in the condition of advanced or delayed cues, and each cue lasted for 1.4s. SMI RED 250 Eye tracker was used to record eye movement data. The aim of present study was to explore the effects of synchronized or unsynchronized cues as well as learners’ prior knowledge on the efficiency of multimedia learning and their online distribution of attention. The results showed that learners in the condition of synchronized cues got higher scores and lower perceived difficulty of the material in all of the three tests (retention, transfer, and matching tests) than that of unsynchronized ones (advanced and delayed cues). During 3 seconds after the keywords of commentaries were presented, especially compared with delayed cues, both of the fixation time and first fixation duration of AOIs (areas of interest) in the condition of synchronized cues were longer. Besides, more glance count and fixation count of them were found when cues and words were synchronized. In comparison to LPK learners, HPK learners performed better and perceived lower difficulty in all of the tests, thus showing a great superiority. However, there was almost no significant difference between them on fixation processing. Furthermore, the cueing by prior knowledge interaction effect was not significant either on learning outcomes or on eye movement indicators. In conclusion, the synchronization of cueing and audio commentary affects participants’ learning outcomes and fixation processing as expected. Synchronized cues contribute to text-diagram integration and help building connections between word-based and image-based representations, then reduce the perceived subjective difficulty, promote fixation processing and thus enhance learners’ performance. Besides, learners’ prior knowledge influences the effectiveness of multimedia learning. HPK learners show a significant superiority effect. However, learners’ prior knowledge is not served as a moderator between cues and performance in this study.