<![CDATA[This study profiled undergraduate students’ cognitive processes, mental models, and learning interest regarding moon phases using an integrated empirical approach. A convergent mixed-methods design was employed in which quantitative and qualitative data were collected concurrently and then merged. Cognitive processes were measured using an eight-item test aligned with the revised Bloom’s taxonomy (C3–C5), and learning interest was assessed using a Likert scale questionnaire; mental models were inferred from students’ written explanations and confirmed through semi-structured interviews. A total of 65 first-year students from Universitas Pendidikan Indonesia participated through purposive sampling. Quantitative data were analyzed using descriptive statistics and the Rasch model, whereas qualitative data were analyzed thematically. The Rasch analysis indicated acceptable item fit (MNSQ 0.65–1.27), high item reliability (0.96), and acceptable person reliability (0.73), although the two most difficult items (CE1 and T) exceeded most students’ ability level. Mean performance decreased systematically across cognitive levels, from applying (C3) to analyzing (C4) and evaluating (C5). Mental-model analysis showed that 40.0% of students held Initial models, 50.8% Synthetic models, and 9.2% Scientific models, while 61.5% reported high learning interest. These results reveal a consistent gap between students’ motivational readiness and their cognitive–conceptual readiness, indicating that high interest functions as an entry condition for engagement rather than an indicator of conceptual mastery. These findings indicate that students’ high learning interest should be treated as motivational readiness rather than evidence of conceptual mastery, thereby highlighting the need for scaffolded, diagnostic-based, and differentiated astronomy instruction. Keywords: cognitive processes, learning interest, mental models, moon phases, rasch analysis.]]>
