Hit a glitch in your research? Some ‘night science’ thinking could move it forward