Investigating the consistency of negative BOLD responses to combinations of visual, auditory, and somatosensory stimuli and their modulation by the level of task demand

Abstract

Negative BOLD fMRI responses (NBR) occur commonly in sensory cortex and default mode network regions but remain poorly utilized as a marker of brain function due to an incomplete understanding. To better understand how NBR manifest across the brain, compare between different sensory stimuli and how they are modulated by changes in task demand, we recorded fMRI during trials of visual, auditory, or somatosensory stimulation, delivered either alone or in concurrent pairs. Twenty young-adult participants were cued to attend to a single modality and detect targets in each trial. We found that NBR were consistently induced in all non-task-relevant primary sensory cortices and default mode regions during all stimuli. NBR were observed within the stimulated modality, in the cortex ipsilateral to the stimulus; as well as cross-modal responses bilaterally within the cortex of an unstimulated sensory modality. The NBR regions showed high spatial overlap with the primary sensory positive BOLD response (PBR) of the stimulated modality. The NBR occurred in spatially comparable regions across different modality stimuli such that the peak voxel location and spatial extent were comparable between within and cross-modal NBRs. Some specific differences were seen, such as stronger magnitude sensorimotor NBR to somatosensory stimuli than to visual or auditory. No significant relationships were found between subjects' PBR and NBR magnitude, but significant linear correlations were observed between NBRs indicating that subjects with high magnitude NBR within one sensory modality also displayed high magnitude cross-modal NBR in a different modality. These findings suggest that cortical NBR are largely consistent between different sensory stimuli but also contain stimulus-specific variability in magnitude and spatial extent. Finally, positive BOLD responses were stronger to dual stimuli in all contralateral primary sensory regions, whilst NBR were slightly increased in specific regions of ipsilateral visual and sensorimotor cortex. This finding suggests a strong contribution to NBR from bottom-up stimulus input that was further modulated by attention during dual conditions and that NBR is driven by a combination of bottom-up and top-down influences whereby contributions to its generation arise from both feed-forward signals from subcortical or activated sensory regions and feedback mechanisms such as higher-level attentional control.