We investigate whether hypothesis testing can be improved by a simple prompt to ‘think in opposites’, a strategy suggested by a growing body of literature as being beneficial in various reasoning and problem-solving contexts.