Abstract
In the study of the brain and how it adapts to changes or injury, researchers sometimes come across situations where apparently similar types of tests or training do not achieve similar outcome results. This is true, in particular, within the field of computer-based rehabilitation where paper-and-pencil tests and training is converted to computer. This paper raises the attention to the fact that supposedly similar settings may not, in fact, elicit similar results and caution therapists and researchers who work with rehabilitation of brain injury. The paper suggests that the underlying mechanisms behind this may be illuminated by using the REF (Reorganization of Elementary Functions) model and suggests that further research into the use of advanced technologies such as computer-generated virtual reality is required.