The design of optimal training for complex systems can be a difficult problem. Generally, one of three types of training is used to train for a specific task – procedural, conceptual or perceptual training. In aviation, procedural training has been the primary form of training. However, procedural training does not always train all the skills needed for effective flying. This study evaluates the impact of perceptual and conceptual training on a particular aviation maneuver, the visual approach. A visual approach is an approach and landing technique that can be executed under visual flight rules (VFR). In this approach, the pilot does not rely on his or her instrumentation to execute the approach. Rather, the pilot uses a visual out-thecockpit reference to the runway to safely guide the aircraft to the ground. There are two keys to achieving a stabilized approach. First is maintenance of glideslope. Glideslope is the angle at which the aircraft descends onto the runway. For commercial aircraft, the recommended glideslope is 3°, although this can differ for a non-standard approach. Second is an understanding of the concept of energy management. During an approach there are speed restrictions below certain altitudes. Improper management of factors like speed and/or altitude can lead to the aircraft being too high and/or too fast to complete the landing safely. The researchers developed two different training methods, one that focuses on perceptual discrimination and one that focuses on conceptual knowledge. Most perceptual training focuses on repeatedly exposing individuals to task-relevant perceptual stimuli. One alternative to strict exposure training to achieve perceptual learning is discrimination training (see, e.g., Curtis, Maraj, Ritman & Jenstch, 2010; Curtis, Schuster, Jentsch, Harper-Sciarini & Swanson, 2008). Discrimination training is a method in which individuals are presented with two stimuli either simultaneously or in succession. The task is to compare the images based on relevant cues and determine whether they are the same or different. Because the training is composed of comparisons between two images, the individual receives twice as much exposure to the stimuli in a fixed period of time. Further, the focus on comparison reduces memory load and simultaneously helps individuals develop strategies to identify relevant cues and ignore irrelevant ones. The second approach focused on conceptual training, which has been shown to be effective in training pilots for crew resource management (for a description of these programs, see Helmreich, Merritt, & Wilhelm, 1999) and automation (e.g., Boehm-Davis, Smith & Prada, 2009). The effectiveness of conceptual training in these instances may be due to the fact that the use of rote procedures may not adequately support the development of a robust mental model of the system. In fact, Sarter and Woods (1994) have demonstrated that errors often occur when the pilot does not understand how the system operates. Therefore, providing training that supplements procedures with precise information on how the system works should help to reduce errors and improve performance. The experiment examined the performance of pilots who received the perceptual discrimination training, conceptual training, or both forms of training against a control condition, where pilots did not receive any training. The results indicate that pilots who received the perceptual discrimination training or both the perceptual and conceptual training performed better on a perceptual evaluation task than pilots who received just the conceptual training. There were no significant performance differences on the conceptual evaluation task between the various training groups. This may be due to the high experience levels of the pilots who participated in this study. More experienced pilots are more likely to already have knowledge of the visual approach concepts included in the training; a significant effect might be observed with less experienced pilots. Overall the results support the notion that different types of training are more effective for specific types of performance and that the type of training delivered should be matched to the demands of the task.