A good scientific model makes novel predictions (bonus points if unexpected), then we later confirm them through experiment. For example, Einstein's theory of relativity is continually tested in novel ways -- predictions are made using the mathematics then confirmed via experiment.
The challenge means that current computational neuroscience models by and large are mainly descriptions of the data we've gathered and seldom are able to extrapolate beyond that. Somewhat embarrassingly, perhaps the field's biggest success in achieving a deeper understanding is what bootstrapped it 60 years ago -- the Hodgkin-Huxley model. Admittedly, neuroscience is extremely challenging and our experimental techniques are still relatively primitive. However, having testable, predictive models is essential to moving the field forward.
The challenge means that current computational neuroscience models by and large are mainly descriptions of the data we've gathered and seldom are able to extrapolate beyond that. Somewhat embarrassingly, perhaps the field's biggest success in achieving a deeper understanding is what bootstrapped it 60 years ago -- the Hodgkin-Huxley model. Admittedly, neuroscience is extremely challenging and our experimental techniques are still relatively primitive. However, having testable, predictive models is essential to moving the field forward.