In this paper we argue that for brain-computer interfaces (BCIs) to be used reliably for extended periods of time, they must be able to adapt to the user’s evolving needs. This adaptation should not only be a function of the environmental (external) context, but should also consider the internal context, such as cognitive states and brain signal reliability. We demonstrate two successful approaches to modulating the level of assistance: by using online task performance metrics; and by monitoring the reliability of the BCI decoders. We then describe how these approaches could be fused together, resulting in a more user-centred solution.