DREAM: A Toolbox to Decode Rhythms of the Brain System

Rhythms of the brain are generated by neural oscillations across multiple frequencies. Following the natural log linear law of frequency distribution, these oscillations can be decomposed into distinct frequency intervals associated with specific physiological processes. This perspective on neural oscillations has been increasingly applied to study human brain function and related behaviors. In practice, relevant signals are commonly measured as a discrete time series, and thus the sampling period and number of samples determine the number and ranges of decodable frequency intervals. However, these limits have been often ignored by researchers who instead decode measured oscillations into multiple frequency intervals using a fixed sample period and numbers of samples. One reason for such misuse is the lack of an easy-to-use toolbox to implement automatic decomposition of frequency intervals. We report on a toolbox with a graphical user interface for achieving local and remote decoding rhythms of the brain system (DREAM) which is accessible to the public via GITHUB. We provide worked examples of DREAM used to investigate frequency-specific performance of both neural (spontaneous brain activity) and neurobehavioral (in-scanner head motion) oscillations. Using the imaging data from the Human Connectome Project, DREAM mapped the amplitude of these neural oscillations into multiple frequency bands as well as their test-retest reliability. DREAM analyzed the head motion oscillation and found that younger children moved their heads more than older ones across all five frequency intervals, particularly in the higher frequency intervals. In the age interval from 7 to 9 years, boys moved more than girls across all frequency intervals. Such sex-related motion effects were not detectable for other ages. These findings demonstrate the applicability of DREAM for frequency-specific human brain mapping.