Educational animation, recognized for its potential accessibility and engaging qualities, has become increasingly prevalent in classroom instruction. However, not all educational animations exhibit high quality or significantly enhance learning outcomes. This study addresses the need for optimizing educational animation design to enhance student learning outcomes and experiences, employing the construction-integration model. We developed three types of educational animations: subtitled textual cue (STC), keyword textual cue (KTC), and structured textual cue (CTC). Through a quasi-experimental research design, 257 fifth-grade students were assigned to three groups, each exposed to one type of textual cue. The results indicate that CTC leads to superior achievement, knowledge retention, higher self-efficacy, and the lowest cognitive load. In comparison, KTC demonstrates moderate results, while STC yields the poorest outcomes. Furthermore, there is a significant negative correlation between achievement and cognitive load, and a significant positive correlation between achievement and self-efficacy. Additionally, there is a significant positive correlation between the "faded effect" of knowledge retention and self-efficacy. These findings highlight the superior learning outcomes and experiences associated with CTC. Based on these findings, recommendations are provided for future educational animation design and instructional practices.