This paper describes an experiment and results from a related meta-analysis that investigated the efficacy of worked examples for improving learning and transfer performance. Worked examples are designed to demonstrate the correct steps to take in a problem-solving process. Preventing errors during training can reduce the difficulty of the task and the associated demands on the learner. Meta-analysis results suggest that worked examples may benefit transfer effectiveness for lower complexity problems and near transfer tasks. We present results from an experiment that used worked examples to train unmanned vehicle route planning. This is an example of a complex decision making task coupled with procedural data entry on a digital system. Transfer tasks were used to evaluate the effectiveness of the worked examples in supporting 1) transfer from structured training tasks to problems requiring more inferential reasoning and 2) transfer from paper-based training to performance in a simulation environment. Results suggest that worked examples did not provide a transfer benefit, either in terms of plan content or plan sophistication. The implications of worked examples and training media are discussed as well as the use of the Tower of Hanoi puzzle as a measure of problem-solving aptitude.