Investigation of bone conduction sound propagation by osseous and non-osseous pathways and their interactions based upon the stimulation site and coupling method of the actuator from a bone conduction hearing aid (BCHA).Experiments were conducted on five Thiel embalmed whole head cadaver specimens. The electromagnetic actuator from a commercial bone conduction hearing aid (BCHA) (Baha® Cordelle II) was used to provide a stepped sine stimulus in the range of 0.1-10 kHz. Osseous pathways (direct bone stimulation or transcutaneous stimulation) were sequentially activated by stimulation at the mastoid or the BAHA side using several methods including a percutaneously implanted screw, Baha® Attract transcutaneous magnet and a 5-N (5-N) steel headband. Non-osseous pathways (only soft tissue or intra-cranial contents) were activated by actuator stimulation on the eye or neck via attachment to a 5-N steel headband, and were compared with stimulation via equivalent attachment on the mastoid and forehead. The response of the skull was measured as motions of the ipsi- and contralateral promontory and intracranial pressure (ICP) in the central, anterior, posterior, ipsilateral and contralateral temporal regions of the cranial space. Promontory motion was monitored using a 3-dimensional Laser Doppler vibrometer (3D LDV) system.The promontory undergoes spatially complex motion with similar contributions from all motion components, regardless of stimulation mode. Combined 3D promontory motion provided lower inter-sample variability than did any individual component. Transcranial transmission showed gain for the low frequencies and attenuation above 1 kHz, independent of stimulation mode This effect was not only for the magnitude but also its spatial composition such that contralateral promontory motion did not follow the direction of ipsilateral stimulation above 0.5 kHz. Non-osseous stimulation on the neck and eye induced comparable ICP relative to percutaneous (via screw) mastoid stimulation. Corresponding phase data indicated lower phase delays for ICP when stimulation was via non-osseous means (i.e., to the eye) versus osseous means (i.e., to the mastoid or forehead). Sound propagation due to skull stimulation passes through the thicker bony sections first before activating the CSF.Utilization of 3D promontory motion measurements provides more precise (lower inter-sample variability) information about bone vibrations than does any individual component. It also provides a more detailed description of transcranial attenuation. A comprehensive combination of motion and pressures measurements across the head, combined with a variation of the stimulation condition, could reveal details about sound transmission within the skull.