本研究應用「生物認知偏好測驗」(Test of Biology Cognitive Preferences, TBCP)、「認知策略調查表」(Inventory of Cognitive Strategies, ICS)以及「生物科學思考測驗」(Test of Scientific Thinking in Biology, TSTB)三項工具，探討台北地區1567位國中一年級學生之認知偏好與認知策略運用情形間的關係，並探討此二者與生物科學業成就及科學思考表現的關係。上述三項工具均經嚴謹之發展過程，「認知策略調查表」(ICS)與「生物科學思考測驗」(TSTB)之內部均質性信度（以Cronbach a表示）分別為0.96與0.76，效度考驗的結果亦均稱滿意。「生物認知偏好測驗」(TBCP)所含R、P、Q、A四種偏好型式之信度值分別為：0.71，0.63，0.82，0.48，與原工具相較，信度值略為偏低，可能係因題數較少與部份學生未具有明顯之認知偏好所致。全體樣本之認知偏好取向為A（應用）> R（回憶），P（原理原則）> Q（批判質疑），顯示學生偏好講求生物學知識的應用性，而不喜好對知識批判質疑。男女學生間的認知偏好有明顯差異，男生之認知偏好取向為A, R, P > Q，女生為A > R > P > Q。全體樣本之認知策略運用情形尚佳，女生的認知策略運用情形顯著較男生為佳。全體樣本之生物科學思考能力表現並不理想，男女學生的生物科學思考能力並無差異存在。國中一年級學生的認知策略與P偏好型式間呈顯著正相關，與R、A偏好型式間呈顯著負相關，若僅以具明顯認知偏好風格的樣本言，認知偏好與認知策略間的相關情形較全體樣本明顯。認知策略運用愈佳的學生，認知偏好強度（Dc2值）亦愈大。認知偏好在性別與認知策略兩變項之間有交互作用存在，認知策略運用較差的女生，具有較強的A認知偏好，認知策略運用較差的男生，具有較強的R認知偏好。全體樣本的生物科學業成就僅與R偏好型式呈顯著負相關，而具明顯認知偏好風格的樣本，生物科學業成就與R偏好型式亦呈顯著負相關，與Q偏好型式呈顯著正相關，與P、A偏好型式則無顯著的相關。生物科學業成就與認知策略呈顯著正相關。另外，全體樣本和具明顯認知偏好風格的樣本之生物科學思考能力與R偏好型式呈顯著負相關，與Q偏好型式呈顯著正相關，生物科學思考能力較佳的學生，具有較強的Q認知偏好與較弱的R認知偏好。生物科學思考能力與認知策略呈低度正相關，生物科學思考能力愈佳的學生，認知策略運用情形亦愈佳。生物科學思考能力在認知策略與認知偏好兩變項之間有交互作用存在，認知策略運用較差者，生物科學思考能力亦較差；認知策略為中等以上者，認知偏好為「批判分析型」的學生之生物科學思考能力，顯著較「知識記憶型」與「中間型」的學生為佳。

In this study, three instruments, namely, Test of Biology Cognitive Preferences (TBCP), Inventory of Cognitive Strategies (ICS), and Test of Scientific Thinking in Biology (TSTB), were used to investigate the relationships among cognitive preferences, cognitive strategies, academic achievement and scientific thinking in biology of 1567 seventh grade students in Taipei area. The internal consistency reliabilities (Cronbach a coefficients) of the ICS and the TSTB were 0.96 and 0.76 respectively. The internal consistency reliabilities of the R, P, Q, A preference modes of the TBCP were 0.71, 0.63, 0.82, and 0.48 respectively. These reliability coefficients were somewhat low maybe due to the decrease of number of items and the lack of distinct cognitive preference patterns for some subjects. In general, the reliability and validity of the ICS ,the TSTB and the TBCP were satisfactory. The results of this study showed that the subjects exhibited a biology cognitive preference style of A (Application) > R (Recall), P (Principle) > Q (Questionig), which indicated a strong preference for application and a weak preference for critical questioning. There was significant difference in cognitive preference style between male and female subjects.The males exhibited a preference style of A, R, P > Q, and the females exhibited a preference style of A > R > P > Q. In general, the cognitive strategies of the entire subjects were fairly well. Further examination showed that the females performed better cognitive strategies than the males. The scientific thinking ability of the subjects was not proficientas expected. No significant differences between the male and female subjects on the TSTB scores were found. The cognitive strategies correlated positively with P mode and negativelywith R and A modes. The correlation coefficients for the subjects with distinct cognitive preference patterns were higher than those for the total subjects. Subjects with better cognitive strategies exhibited greater cognitive preference intensity (Dc2). Two-way ANOVA showed that there were interactions between gender and cognitive strategies for cognitive preferences. The female subjects with low ICS score exhibited stronger preference for A mode, and the male subjects with low ICS scoresexhibited stronger preference for R mode. Biology academic achievement correlated negatively with R mode for the entire subjects. For the subjects with distinct cognitive preference patterns, the biology academic achievement correlated negatively with R mode but positively with Q mode. No significant correlations were found between achievement and P and A modes. Biology academic achievement correlated positively with cognitive strategies. Scientific thinking ability in biology correlated positively with Q mode but negatively with R mode. Subjects with high scientific thinking ability exhibited stronger preference for Q mode and weaker preference for R mode. Slightly positive correlations were found between scientific thinking ability and cognitive strategies. There was an interaction between cognitive strategies and cognitive preferences for scientific thinking ability. Subjects with low ICS scores exhibited lowest TSTB score. For subjects with medium and high ICS scores, the TSTB scores obtained by Q-type (critical questioning) cognitive preference students were significantly better than those obtained by M-type (medium) and R-type (memorizing information) cognitive preference students.