A new functional EIT (f-EIT) evaluation technique providing information on the local dynamic behaviour of the lung is presented. Out of a series of single EIT thoracic images local time courses of the impedance change are extracted. To detect regional differences in the dynamic behaviour of the lung tissue the local time courses at different locations are related to the average time course of the impedance change over the whole thoracic cross section. The time shifts between this reference signal and the signals from separate positions are calculated from the phase information of the complex cross spectra and evaluated in terms of the local phase angle. The computed phaseshifts are imaged over the cross section creating an 'f-EIT phase image' characterizing the local dynamic properties. To relate the observed differences to the proper lung location the resulting images are presented as a combination of the f-EIT ventilation images, which represent the local amplitude of ventilation and the f-EIT phase images. The new imaging technique was tested in spontaneously breathing humans. Alterations to pulmonary dynamics were induced by changing the body posture of the subjects. The f-EIT phase imaging procedure was shown to identify lung regions with different dynamics and it is expected that this technique will also distinguish pathologically determined alterations.