Co-variation between subjectively estimated mood/activation and EEG characteristics, based on spectral power parameters, was investigated. Subjective estimation of mood was made by using Thayer's Activation-Deactivation Adjective Checklist, which yielded two dimensions: Energy-Tiredness (with Energy pole having positive valence connotation) and Tension-Calmness (negative connotation for Tension). A within-subject design with two sessions of EEG recording immediately followed by mood assessment was applied. These were separated by a cognitive task, introduced in order to modify the subjects' mood. The correlations between changes in mood estimation and changes in EEG spectral power parameters were calculated. Both ADACL dimensions co-varied with EEG in a specific way according to frequency and localization. Subjective estimation of Energy correlated negatively with alpha1 and, surprisingly, positively with delta, theta1 as well as theta2 relative power. Estimation of Tension correlated positively with theta1 and beta1, and negatively with alpha2 relative power. Presented results suggest that the adjective description of mood has objectively-measurable brain correlates in the EEG.