There have been growing interests in clarifying perceptual mechanisms for different material features such as glossiness and transparency in recent years. In this study, we examined whether perceptual learning effects can be observed for material perceptions, and whether active conduct of material judgment tasks is important for the learning effects. The observers were divided into two groups; the glossiness group and the transparency group. The former observers repeated a glossiness discrimination task, while the latter observers repeated a transparency discrimination task for 10 days. We measured changes in correct response ratios and reaction times for discriminations of learned and unlearned material features between before and after the learning. In the results, in both learning groups, the reaction times for the learned material features were significantly shortened after the learning. These results suggest that the human visual system can increase glossiness and transparency discrimination performances by actively repeating material discrimination tasks.