The first objective of this study was to investigate if the experimental students' post-test knowledge of nutrition and plant reproduction would be improved more significantly than that of their control group counterparts based on their treatment, attitudes to science, self-esteem, gender and socio-economic background. Treatment involved teaching the experimental students under three learning modes--pure cooperative, cooperative-competitive and individualistic whole class interpersonal competitive condition--using concept and vee mappings and the lecture method. The control groups received the same treatment but were not exposed to concept and vee mappings. This study's second objective was to determine which of the three learning modes would produce the highest post-test mean gain in the subjects' knowledge of the two biology concepts. The study's sample comprised 932 eighth graders (12-13-year-olds) in 14 co-educational comprehensive high schools randomly selected from two Jamaican parishes. An integrated science performance test, an attitudes to science questionnaire and a self-esteem questionnaire were used to collect data. The results indicated that the experimental students (a) under the three learning modes, (b) with high, moderate, and low attitudes to science, and (c) with high, moderate, and low self-esteem, performed significantly better than their control group counterparts. The individualist whole class learning mode engendered the highest mean gain on the experimental students' knowledge, while the cooperative-competitive learning mode generated the highest mean gain for the control group students.