In this talk Dawson's (2004) so-called 'synthetic psychological' method will be briefly introduced. It will then be applied on two problems. The first problem illustrates how connectionist systems can be used effectively as theory generators in domains where there is no prior theory. The second problem illustrates how this method can be used to reliably recover a solution to a problem that was generated by independant means. These examples show how the synthetic method can be usefully employed.