Driver inattention is one of the leading causes of traffic crashes. In comparison to the extensive body of research examining the attention principles that govern distraction by external stimuli, the topic of attention and distraction by daydreaming has been relatively understudied. This work used a self-caught method to examine EEG and driving behavioral correlates of daydreaming during simulated driving tasks. Participants were asked to report any time they found themselves thinking something unrelated to the driving tasks. Measures of EEG and driving behaviors were compared prior to and after each daydreaming. Results showed that three EEG band ratios, beta/(alpha+theta), beta/alpha, and beta/theta were able to discriminate daydreaming from attentive driving, and the beta/alpha band ratio outperformed the other two indices. The power band activities of EEG observed from parietal lobe sites (Pz, P3, and P4) were more sensitive to reflect subtle changes in daydreaming. The current findings suggested that these EEG band ratios could be used for early detection of daydreaming in real driving.