Fractions remain one of the most persistently difficult topics in mathematics education, and Ghanaian learners are no exception. National assessment reports have repeatedly identified fractions as an area of weakness among Basic Education Certificate Examination candidates, yet limited classroom-based evidence exists on instructional strategies for addressing fraction-related errors and misconceptions in Ghanaian basic schools. This study investigated the use of mind mapping to address errors and misconceptions in fractions among Basic 7 pupils in a rural junior high school in Ghana. Using an action research design, the study involved 48 pupils (22 boys and 26 girls) from one Basic 7 class over a three-week intervention. Data were collected through a researcher-developed achievement test, classroom observations, and semi-structured interviews administered before and after the intervention. The intervention focused primarily on the addition and subtraction of unlike fractions and involved guiding pupils to visually organize concepts, procedures, examples, and relationships using mind maps. Pre-intervention findings identified several common difficulties, including adding fractions without finding a common denominator (75.00%), difficulty identifying equivalent fractions (68.75%), confusion between numerators and denominators (62.50%), and incorrect simplification (56.25%). The mean pre-test score was 3.03 out of 10, compared with 7.13 in the post-test, representing a mean improvement of 4.10 points. Post-intervention observations and interviews also indicated increased participation and improved ability to explain fraction-solving procedures. The findings suggest that mind mapping can serve as a promising, low-cost, learner-centred strategy for supporting pupils’ conceptual understanding of fractions and addressing common errors and misconceptions. However, because the study used a single-group action research design without a control group, the findings do not establish a causal effect of mind mapping. Further research using larger samples, comparison groups, and inferential statistical methods is recommended.