Providing pre-training on new material can simplify complex content for learners who may need guidance to understand basic facts and organize their efforts. However, the effect of pre-training on learning outcomes is controversial because it tends to vary by context. Our aim was to investigate the effectiveness of pre-training in reducing intrinsic cognitive load and improving learning in a computer-based learning environment. Predictors of learning were also investigated. A total of 384 undergraduate students were randomly assigned to experimental conditions within a Solomon four-group design (n = 293). An additional manipulation control group (n = 91) was also included to determine the effectiveness of the current pre-training activity in improving achievement scores. Results showed significant relationships between cognitive load and achievement. Intrinsic and extraneous cognitive load were negatively correlated with performance, while germane cognitive load was positively correlated. Cognitive load explained 20% of the variance in achievement. In addition, topic interest and achievement motivation were associated with cognitive load. Interestingly, despite evidence of the usefulness of pre-training from the manipulation control group, it did not significantly affect learning gains. That is, all groups exposed to the video instruction had significantly better post-test scores, but pre-training did not make a difference between the groups. However, pre-training did appear to be somewhat effective in reducing intrinsic cognitive load. Findings underscore the importance of considering cognitive load dynamics and motivational factors in instructional design strategies aimed at optimizing learning experiences and outcomes for undergraduate students.