ok enough jokes. Computational neuroscience is neuroscience. But instead of doing experiments in real life, you do that with your computer.
Or sometimes, you don't even need a computer because you will develop some crazy theories like old-school mathematicians
My own research is mainly around neural representation and Recurrent Neural Networks (RNNs). I deal with both conceptual models and data-driven models. They are somewhat abstract and machine-learning-like, but now I think this could be the right path (copy-pasted from my old site).