hide
Free keywords:
-
Abstract:
A “voice” region has recently been identified in the monkey auditory cortex with functional magnetic resonance imaging (fMRI) and electrophysiology, which shows a close functional correspondence to the known human-voice region. Both human and monkey voice regions lie anterior and superior on the temporal lobe and strongly prefer species-specific vocalizations over other categories of sounds and acoustical controls. The human and monkey voice regions are also sensitive to vocal differences among individuals and appear to be important centers for vocal sound processing within a network that is poorly understood. To clarify the in-vivo functional connectivity of the voice region in the rhesus monkey we used microstimulation in combination with high-resolution fMRI. First we functionally localized the voice region with blood-oxygen-level-dependent (BOLD) fMRI, as previously described. Then we microstimulated the voice region with glass-coated iridium microelectrodes, using biphasic, cathode leading, 250 to 500 μA pulses of 200 μs duration. We used the BOLD response to evaluate the anterograde targets of the microstimulation site. Microstimulation of the monkey voice region, which lies on the rostral superior-temporal plane (rSTP), elicited a BOLD response from hierarchically earlier auditory areas (feed-back), and the surrounding superior-temporal-plane (STP), -gyrus (STG) and -sulcus (STS) of the ipsilateral hemisphere. We observed no direct targets in the prefrontal cortex from voice region microstimulation, so we hypothesized that voice information might reach the frontal cortex indirectly. To test this idea we microstimulated a region in the upper bank of the STS that was one of the direct targets of the voice region, which resulted in medial and orbital prefrontal cortex activity, and neighboring regions on the STP, STG, STS and temporal pole. Our initial observations suggest that acoustical information from the voice region reaches the frontal cortex indirectly via other rostro-temporal regions such as the STS. Since the primate STS receives multisensory input and is known to contain face-recognition regions, we propose that voice information is paired with face information in the anterior temporal lobe before being transmitted to the prefrontal cortex.
