Understanding how pre-service teachers learn remains a central objective of research-based academic teacher education (Swennen, 2025). Reflection and metacognition are important aspects of pre-service teachers’ learning (Korthagen, 2016; Matsumoto-Royo et al., 2022), yet many pre-service teachers graduate without sufficiently developed capacities of reflection and metacognition, which form part of the skills needed to succeed in increasingly diverse and complex educational contexts (Michalsky, 2024; Suphasri & Chinokul, 2021). The variation in pre-service teachers’ initial reflective and metacognitive capacities highlights the need for teacher education programs that support higher-order thinking while addressing individual differences, an enduring challenge in European and global educational contexts.

Well-developed reflection and metacognition are associated with multiple positive outcomes in pre-service teacher education. Reflection supports learning and professional development (Korthagen, 2016; Li, 2025), the ability to take multiple perspectives (Toom et al., 2015), and deeper understanding of the teaching profession (Li, 2025). Metacognition enhances pre-service teachers’ own learning (Bedel, 2012), academic performance (Hong et al., 2020) and teaching practices (Matsumoto-Royo et al., 2022). These findings highlight the importance of examining how reflection and metacognition contribute to pre-service teachers’ learning characteristics, informing both the design of teacher education programs and research on individual differences in higher-order thinking.

Both reflection and metacognition are broadly regarded as forms of higher-order thinking (Ritchhart & Perkins, 2005) with theoretical intersections (Tarricone, 2011). Dewey (1933) describes reflection as the “active, persistent, and careful consideration of any belief or supposed form of knowledge,” a process further elaborated in transformative learning theory (Mezirow, 1991). Metacognition refers to thinking about one’s own cognitive processes (Flavell, 1979), while metacognitive awareness specifically denotes individuals’ knowledge of their cognition and ability to regulate it through planning, monitoring, and evaluation (Schraw & Dennison, 1994). The present study focuses on pre-service teachers perceived reflective thinking (RT) and metacognitive awareness (MA), which capture measurable dimensions of these constructs, allowing investigation of higher-order thinking.

Pre-service teachers differ in RT and MA due to variation in readiness (Kitchener, King & DeLuca, 2006), developmental trajectories (Kleka et al., 2019) and prior cognitive resources (Larmar & Lodge, 2014). These capacities are influenced by internal factors, including prior knowledge, motivation, and cognitive resources, as well as by contextual conditions that can facilitate or constrain their learning (Sargent 2015; Abrouq, 2024). Research shows variability in university students’ MA (Hong, 2020), and in pre-service teachers’ development of MA (Mendoza & Elepano, 2023; Radulović et al., 2024). Similarly, RT develops individually, with varying contextual dependency and differences across programmes (Allas, Leijen & Toom, 2020; Gahlsdorf Terrell & Sherman, 2022; Tiainen et al., 2018). 

Prior research is largely qualitative or cross-sectional, with few studies examining associations between RT and MA (e.g. Adadan & Oner, 2018; Alt & Raichel, 2020; Ho & Lau, 2025; Pousi et al., 2025). To our knowledge, no prior study has investigated these capacities together using a person-centred approach in pre-service teacher education. The present study therefore empirically examines the joint structural patterns of RT and MA to understand pre-service teachers’ learning characteristics. Using latent profile analysis on repeated-measurement data, the study seeks to identify distinct subgroups and describes homogeneous profiles, providing insight into the variation and structure of higher-order thinking. Although the data are longitudinal, the present analysis is cross-sectional, emphasizing profile structures at three time points.

Accordingly, this study addresses the following research questions:

1. What kind of distinct profiles of reflective thinking and metacognitive awareness can be identified among pre-service teachers at different stages of teacher education?
2. How do the identified profiles differ in their overall patterns of reflective thinking and metacognitive awareness?

This study utilizes a longitudinal dataset with three measurement points. At autumn 2022 (T1), 262 Finnish first-year pre-service teachers from one university participated in the study at the beginning of their studies (response rate 72 %). At the second data collection in autumn 2023 (T2) 172 participants remained (65.6% retention from T1). By the third data collection in spring 2025 (T3), 105 participants completed the survey, corresponding to 40.1% of the original T1 sample and 61.0% of the T2 sample. As the present analysis is cross-sectional, participant dropout is not a primary concern. The participants’ voluntary involvement, informed consent, and anonymity were ensured during the research process. The participants were early childhood education (n=87), primary school (n=92), and subject pre-service teachers (n=83) aged 19 to 59 years (M = 24.2, SD = 7.1) (T1 sample). The sample consisted primarily of female participants (89.8%). Data were collected using the Finnish-adapted Reflective Thinking Questionnaire (RTQ) (Kember et al., 2000; Pousi et al., 2025) and a compressed version of the Metacognitive Awareness Inventory (MAI) (Kallio et al., 2018). The RTQ, measuring non-RT and RT, included four components: habitual action (non-RT), content reflection, process reflection, and critical reflection. The MAI included three components: knowledge of cognition, regulation of cognition and self-evaluation. Factor structure and measurement invariance were confirmed for each timepoint using confirmatory factory analysis, demonstrating acceptable fit. As for measurement invariance testing, metric invariance was reached for both questionnaires. Separate LPAs were conducted for each time point (T1=262, T2=172, T3=105), testing models with 1–5 profiles. Analysis was done with R using tidyLPA-package. Factor scores from CFAs were used as latent profile indicators. Separate LPAs were estimated for each time point rather than assuming profile stability across time. This approach allows for potential changes in the number, shape, and size of profiles as teacher education progresses (Edelsbrunner, 2023). Consistent with recommendations (Johnson, 2021; Rahden et al., 2025), the selection process proceeded in two steps where within-type comparison proceeds across-type comparison. First each variance-covariance model specification was evaluated separately to determine the optimal number of profiles. Second the most suitable models from types 1–3 were compared using a combination of statistical fit indices with preference for AIC and SABIC as Edelsbrunner (2023) suggests. To examine how the identified profiles differed in terms of the higher-order thinking variables used in the LPA, between-profile tests at each timepoint were carried out with Welch’s ANOVA.

The aim of this study was to identify distinct profiles of RT and MA among pre-service teachers across three measurement points in teacher education. At each timepoint, solutions with three to four profiles and a Type 2 variance–covariance structure (varying variances, zero covariances) provided the most parsimonious and interpretable representation of the data. A four-profile solution best captured meaningful variation in pre-service teachers’ learning characteristics. Although mean levels of RT and MA varied somewhat across timepoints, the overall configuration of profiles remained structurally similar. At each timepoint, profiles could be ordered along a continuum of increasing RT and MA and, to some extent, decreasing non-RT, indicating a stable structural organization of higher-order thinking profiles across the stages of teacher education. However, because strict scalar measurement invariance was not established, the findings do not support longitudinal interpretations of change. The identified profiles reflected distinct configurations of RT and MA. Profile 1 was characterized by low RT and MA alongside higher non-RT, whereas Profile 2 represented generally below-average levels across both constructs. Profile 3 displayed moderately elevated RT and MA, while Profile 4 consisted of consistently high RT and MA and low non-RT. Overall, the findings suggest that RT and MA function as complementary, but distinct learning characteristics that cluster in meaningful ways. Furthermore, despite their conceptual relatedness and previously reported strong interconnections, the results indicate that RT and MA constitute separate, though closely associated, constructs. The presence of intermediate profiles indicates potential for developmental movement, even though such transitions cannot be examined within the present design. By adopting a person-centred approach, this study contributes empirical insight into how reflection and metacognition manifest jointly in pre-service teachers’ learning, offering a foundation for future longitudinal research and teacher education practices responsive to individual differences and that further enhance student teachers’ reflective capabilities.

Selected key references Adadan, E., & Oner, D. (2018). Examining preservice teachers’ reflective thinking skills in the context of web-based portfolios: The role of metacognitive awareness. Australian Journal of Teacher Education, 43(11), 26–50. Allas, R., Leijen, Ä., & Toom, A. (2020). Guided reflection procedure as a method to facilitate student teachers’ perception of their teaching to support the construction of practical knowledge. Teachers and Teaching: Theory and Practice, 26(2), 166–192. Alt, D., & Raichel, N. (2020). Reflective journaling and metacognitive awareness: Insights from a longitudinal study in higher education. Reflective Practice, 21(2), 145–158. Aquino Mendoza, J., & Elepaño, C. (2023). Metacognitive awareness levels of pre-service teachers. World Journal of Advanced Research and Reviews, 17(3), 365–375. Abrouq, N. (2024). Reflective thinking: From opacity to clarity. British Journal of Teacher Education and Pedagogy, 3(1), 87–96. Gahlsdorf Terrell, D., & Sherman, D. (2022). Mirror of mind: Eliciting critical reflections in preservice and novice teachers. Action in Teacher Education, 44(3), 230–251. Ho, W., & Lau, Y. (2025). Role of reflective practice and metacognitive awareness in the relationship between experiential learning and positive mirror effects: A serial mediation model. Teaching and Teacher Education, 157, 104947. Hong, W., Bernacki, M., & Perera, H. (2020). A latent profile analysis of undergraduates’ achievement motivations and metacognitive behaviors, and their relations to achievement in science. Journal of Educational Psychology, 112(7), 1409–1430. Larmar, S., & Lodge, J. (2014). Making sense of how I learn: Metacognitive capital and the first year university student. International Journal of the First Year in Higher Education, 5(1), 93–105. Matsumoto-Royo, K., Ramírez-Montoya, M., & Glasserman-Morales, L. (2022). Lifelong learning and metacognition in the assessment of pre-service teachers in practice-based teacher education. Frontiers in Education, 7. Pousi, I., Kallio, M., Hotulainen, R., & Toom, A. (2025). Reflective thinking among Finnish pre-service teachers and its relation to metacognitive awareness. European Journal of Teacher Education, 1–25. Radulović, B., Džinović, M., & Miščević, G. (2024). Using a longitudinal trajectory of pre-service elementary school teachers’ metacognition as a quality indicator of higher education. International Journal of Cognitive Research in Science, Engineering and Education, 12(2), 251–257. Suphasri, P., & Chinokul, S. (2021). Reflective practice in teacher education: Issues, challenges, and considerations. PASAA: A Journal of Language Teaching and Learning in Thailand, 62, 23A–2A4. Swennen, A. (2025). Good teachers need good teacher educators. European Journal of Teacher Education, 48(4), 671–680.