Skip to main content
. 2020 Jun 5;11:2839. doi: 10.1038/s41467-020-16743-2

Fig. 5. Revised auditory feedback control loop of the DIVA model.

Fig. 5

The specified model includes two parallel auditory feedback control subsystems that preferentially monitor and correct temporal (blue) or spectral (orange) properties of the produced speech sound. Projections from the left premotor cortex to the bilateral auditory association cortices transmit the expected sensory consequences of the produced speech sound via the left TPJ, both for normal speaking and for speaking with altered auditory feedback4. The predicted auditory consequences of speaking are compared with the incoming auditory information taking the temporal delay of feedback information into account. The left STG monitors especially temporal and the right STG especially spectral speech features. Potential mismatches in either domain are transmitted via temporo-frontal projections and translated into corrective commands for the timing of articulator movements in left IFG and corrective commands that specify articulator target positions in right IFG. Corrective motor commands are transmitted to the feedforward system targeting M1 and the vocal tract and to the internal model for updating auditory-motor associations. Neither potential interhemispheric connections nor somatosensory feedback control loops are shown for the sake of clarity. PMV ventral premotor cortex, IFG inferior frontal gyrus, STG superior temporal gyrus, TPJ temporo-parietal junction.