The purpose of this study was to investigate the relationship between cognitive load and cognitive style and explore the role cognitive load and cognitive style play in the achievement of learning outcomes, when using animation and static images as multimedia learning formats in an authentic learning environment. Two hundred and forty five 2nd year medical and dental students participated in the main study. The majority of the participants had a Analytic style on the Wholistic-Analytic dimension and an Imager style on the Verbaliser-Imager dimension. It is not clear from the literature whether this is a typical cognitive style profile for health education learners. Cognitive load was measured using a subjective rating technique. The cognitive loads of the respective research interventions were significantly different, yet neither version appeared to have an excessive cognitive load that negatively influenced learning. Significant learning took place for all the participants in this study. Surprisingly it was found that when the program was considered as a whole the version that used predominantly animation had the lower cognitive load. When the analysis drilled down to specific screens and compared animation and static images and text the results consistently showed that animation had a higher cognitive load than static images and text. This study established that there is empirical evidence that cognitive load influences learning performance. There are indicators that the Analytic cognitive style influences the subjective rating of cognitive load. Further empirical investigation of this relationship is necessary. The proposal is that the Analytic style influences the germane load experienced during learning. Since researchers are currently unable to measure the three different types of load separately this proposal remains an area for further investigation. The subjective cognitive load rating of the program was compared with the cognitive load rating measured using the direct measurement method. The direct measurement method found that the animation version had the higher cognitive load. The correlation between these two methods of measurement was very low and not significant, thereby confirming a suggestion in recent literature that each method might be measuring different aspects of cognitive load.