The interaction and cooperation between human and machine is becoming even more important and, with that, also the question of what the differences between the two are. A small step in this direction is to examine the learning behaviour of both agents. A common approach in psychology and computer science is learning through reward and punishment. This is not only a natural way for humans to learn, but also an approach to mathematical algorithms in computer science. The talk will describe two major comparisons to investigate and compare the learning behaviour of humans and computers. On the one hand, a comparison between human and machine is conducted and on the other hand a comparison between the human participants learning by seeing and hearing. The following hypothesis is established: There is a difference in the learning behaviour of human and machine and at the same time a difference in learning across sense modalities. Specifically, the learning speed, accuracy of answers and confidence during learning will be examined more closely.