Effective learning strategies are crucial for lifelong learning, particularly during the transition to more autonomous learning environments in lower secondary school. This transition is characterized by increased academic demands and a shift towards independent learning. Despite significant investment in STEM education, student achievement in science, especially Physics, remains a concern globally, including in Uganda. Gender disparities in Physics achievement also persist, with boys often outperforming girls, which impacts career choices. This study investigates the interplay of cognitive and metacognitive learning strategies, motivation, and autonomy support in shaping learner profiles and understanding gender differences in Physics learning. The study focuses on the predictive effects of autonomy support on learner profiles characterized by their use of cognitive and metacognitive learning strategies. Prior research suggests that learner performance in science is influenced by affective factors, such as motivation, interest, attitudes, and learning strategy use. Autonomous motivation is particularly beneficial for achievement. Teachers often integrate autonomy-supportive aspects into instruction; however, students’ perception of this support is crucial, as perceived teacher autonomy support directly and indirectly impacts motivational beliefs, attitudes, achievement emotions, skill choices, and achievement. This study aims to understand how perceived teacher autonomy support affects the likelihood of students belonging to different learner profiles based on their learning strategy use, specifically in the context of Physics learning in lower secondary school in Uganda. This educational level marks a significant transition toward independent learning and increased academic complexities, and it is a critical period for developing effective learning strategy repertoires. This study uses a person-centered approach to explore learner profiles, considering the variability in learning strategies and the limitations of variable-centered methods. A particular focus is given to gender differences in cognitive learning strategies to inform gender-based pedagogical interventions aimed at addressing the gender gap in Physics achievement and subsequent career choices.

Existing research highlights individual differences in learning strategy use. Autonomously motivated students tend to employ adaptive strategies like critical thinking, while less motivated students may resort to maladaptive strategies such as rote learning. A lack of well-adapted strategies can negatively affect academic progress. While research exists on learning strategies in general science and mathematics, less is known about students' cognitive strategy use in Physics during lower secondary school, a period of transition to more independent learning and a potential decline in motivation. The self-regulated learning theory posits that learning is governed by cognitive, metacognitive, and motivational components. Cognitive skills encompass encoding, memorizing, and recalling information, while metacognition involves understanding, monitoring, and regulating the cognitive process. Motivation comprises beliefs that drive the use and development of cognitive and metacognitive skills. Intrinsic motivation—the inherent enjoyment of a task—is particularly advantageous for achievement and is fostered by autonomy support (a perception of choice and control over one's learning). Cognitive learning strategies, such as rehearsal, elaboration, organization, and critical thinking, vary in depth of information processing. Deep-level strategies emphasize understanding and application, while surface-level strategies focus on memorization and basic comprehension. Metacognitive skills include planning, monitoring, and evaluating learning processes. Existing research shows variability in cognitive strategy use across cultures, subjects, and grade levels. Person-centered approaches (like latent profile analysis) complement variable-centered studies by revealing latent groups or profiles of individuals with similar characteristics regarding various variables. Previous research using person-centered analyses have identified diverse learner profiles based on learning strategies in different educational levels and contexts. These studies reveal the importance of a person-centered approach to unveil distinct learner profiles and their characteristics. However, studies on learner profiles in Physics learning during lower secondary school, especially in developing countries, are limited. Furthermore, research on gender differences in learning strategy use shows inconsistencies, suggesting the influence of cultural contexts and task nature. Therefore, examining gender differences in cognitive strategies during Physics learning in a developing country is crucial.

This quantitative study used a self-report questionnaire administered to 573 ninth-grade students (56% female) in six schools in central Uganda. The questionnaire assessed socio-demographic characteristics, cognitive and metacognitive learning strategies (adapted from the Motivated Strategies for Learning Questionnaire, MSLQ), perceived teacher autonomy support (adapted from the Learning Climate Questionnaire), and intrinsic motivation for Physics (adapted from the Science Motivation Questionnaire II). Confirmatory factor analysis (CFA) was conducted to validate the instrument's subscales. Measurement invariance of the learning strategy scale was assessed across genders. Latent profile analysis (LPA) was employed using a three-step approach: (1) determining the optimal number of profiles using various model fit indices (AIC, BIC, ABIC, LMR, aLMR, BLRT, entropy), (2) characterizing each profile based on the patterns of cognitive and metacognitive learning strategies, and (3) examining the predictive effects of gender, perceived autonomy support, and intrinsic motivation on profile membership using multinomial logistic regression. Paired t-tests were conducted to compare gender differences in cognitive strategy use.

CFA revealed good model fit for the questionnaire sections. Measurement invariance across genders supported comparisons of mean scores between genders. LPA identified four distinct learner profiles: 1. **Competent Strategy Users (26.5%):** High scores on all cognitive and metacognitive strategies (elaboration, organization, critical thinking, metacognition). 2. **Struggling Strategy Users (10.9%):** Lowest scores on all strategies. 3. **Surface-Level Learners (35.6%):** High scores on rehearsal (memorization), low scores on other strategies. 4. **Deep-Level Learners (27%):** High scores on critical thinking and metacognition, moderate scores on others. Paired t-tests revealed girls used elaboration and organization strategies more than boys. Multinomial logistic regression indicated that girls were 2.4-2.7 times more likely to be competent strategy users or surface-level learners than boys. Higher perceived teacher autonomy support increased the likelihood of being a competent strategy user, while higher intrinsic motivation predicted membership in the deep-level learner profile. Lower intrinsic motivation and autonomy support predicted membership in the struggling user profile.

The findings highlight the prevalence of surface-level learning strategies and the significant proportion of students struggling with cognitive strategy use in Physics. The gender differences suggest that girls, while exhibiting higher use of some strategies, may still be disproportionately represented in the surface-level learning profile. The relationship between autonomy support and intrinsic motivation with profile membership reinforces the importance of creating supportive learning environments. The unexpected finding that deep-level learners didn't have the highest scores on autonomy support and intrinsic motivation suggests a need for future exploration. The results align with findings from other studies highlighting the importance of critical thinking and metacognition for academic success, while also noting that the relative importance of frequency versus quality of strategy use remains unclear and requires further investigation. The study's findings could help guide teachers to create better learning environments and provide appropriate instruction for different learner types, particularly in Physics and potentially other science domains. The large number of students in the struggling user profile underscores the need for targeted interventions to improve their cognitive strategy use.

This study revealed four distinct learner profiles based on their cognitive and metacognitive strategy use in Physics, with differences in gender, motivation, and perceived autonomy support. The findings highlight the need for pedagogical approaches that foster deep learning, address gender disparities, and promote self-regulated learning. Future research should investigate the longitudinal stability of these profiles, explore other science domains, and examine the relationship between profiles and academic achievement in Physics.

The study relied on self-reported data, which may be subject to biases. The cross-sectional design limits causal inferences, and the stability of profiles over time is unknown. The study focused solely on Physics and lacked achievement data to assess the relationship between profiles and performance. Future research should address these limitations by incorporating multiple data sources, longitudinal designs, and measures of academic achievement.