The total energy the brain consumes and the intensities of information flows across different brain regions in an intellectual activity may help to explain an individual's intelligence level.

To verify this assumption, 43 students aged 18–25 were recruited as the research subjects. Their intelligence quotients (IQ) were scored by using Wechsler Adult Intelligence Scale (WAIS), while their electroencephalogram (EEG) signals were recorded simultaneously by using the Neuroscan system.

The total energy and distribution patterns of EEG signals were acquired in Curry 8.0. The intensities of information flow across different brain regions were measured by Phase Slope Index (PSI). 20 channels and 190 channel combinations were selected for data analysis.

It was found that the energy of EEG signals in an intellectual activity was negatively correlated with the IQ score. A possible reason could be that a person of high intelligence level may characterize a mental task as “light" while a person of relatively low intelligence level characterize it as “heavy"; the former therefore spend less energy to fulfill the mental task.

In this study, we found that the information °ow in long distance channel combinations were positively correlated to IQ scores on speci¯c frequencies and a®ected the subject's intelligence level more than the information °ow in short distance channel combinations did, which is inconsistent with previous studies.