Article Text

  1. W. S. Wang
  1. University of Colorado Health Sciences Center, Denver


This study examines the use of a nonlinear measure to quantitate and differentiate cortical activity during performance of three mental tasks. Specifically, an electroencephalogram (EEG) was recorded during task performance and analyzed using the correlation dimension (D2). The correlation dimension quantifies the nonlinearity and complexity of a signal, and when applied to an EEG may serve as an index of cortical activation. Larger absolute values of D2 suggest greater signal complexity. EEG signals were recorded while subjects performed three mental tasks: graphical reasoning, mental arithmetic, and vocabulary association. For each task, twenty to thirty seconds of EEG signal were chosen for analysis. The correlation dimension increased from resting values during performance of all mental tasks. When compared between the three tasks, D2 values were greatest during graphical reasoning, suggesting greater EEG signal complexity and stronger cortical activation during its performance. Mental arithmetic exhibited D2 increases primarily in the left temporal regions, while vocabulary association displayed increases in both hemispheres of the brain. Application of D2 values to EEG electrode location generated a map of cortical activation specific for each mental task, and illustrated the differences in activation associated with task performance.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.