Visual discrimination tasks are an integral component to many professional domains, including radiology and fingerprint matching. However, there has been little research investigating the most effective and efficient ways to train novices to become experts, in visual discrimination tasks. To address this gap in the literature, 100 participants were tested on their accuracy to correctly categorise mammograms. The study was a conceptual replication of Searston and Tangen (2014), who investigated the effectiveness of three learning methods, (a) feedback, (b) elaboration, and (c) contrast, on novice accuracy in the identification of fingerprints. The aim of the study was to investigate the effectiveness of three learning methods on novice training, in categorising mammograms. The learning methods included in the study were (a) feedback, (b) elaboration, and (c) contrast. Participants were assigned to one of four possible training conditions (a) baseline, (b) feedback, (c) elaboration, and (d) contrast. They each viewed a short training video and completed a training phase where they were required to categorise mammograms as being the same or different, on a confidence rating scale. They then completed the test phase, where they were required to categorise mammograms as being normal or abnormal, on a confidence rating scale. The results revealed no significant differences in accuracy between conditions. The results were not consistent with Searston and Tangen’s (2014) study. This line of research has many practical implications in the training of experts in professional domains, such as the field of medicine.