Predicting a subject's ability to use the interface with good accuracy is one of the major issues in the motor Brain-Computer interface (BCI) domain. A few recent studies show that subjective questionnaires could be used to predict the performance of motor imagery (MI) based BCI. Indeed, the Kinesthetic and Visual Imagery Questionnaire (KVIQ), could allow a better predictability of BCI-illiterate cases [1]. Another more recent questionnaire called the Motor Imagery Questionnaire Revised-Second Edition (MIQ-RS) is a suitable option for examining MI ability [2]. In 2016, Marchesotti et al. found that the representation of subjective behaviour, calculated using the MIQ-RS questionnaire, and the control of the BCI were intimately linked [3]. However, in these studies [1, 3], the performance of the classifier was calculated for a right-hand MI versus a left-hand MI task. In this abstract, we classify between resting state and imagined movement, which is a relevant classification task in BCI research [4]. The aim of this study is to answer the following question for a resting state versus MI classification task: can the MIQ-RS be used to estimate the performance of an MI-based BCI?