Attention is a factor that affects the performance of various intelligent activities in humans. Up until now, the methods for measuring the level of attention have been mostly based on subjective reports or employing large and costly devices. In this paper, a new method of estimating the level of attention is proposed, based on posture and EEG measurements. These data can be recorded using easily available and less burdensome devices. From the obtained data, the time evolution of attention was explored. Experiments showed that there is negative correlation between posture variance and attention, and also between EEG and attention.