Transient and sustained responses have been shown to play distinct functional roles in auditory processing: Transient responses may subserve rapid stimulus detection, sustained responses contribute to a more detailed sound characterization. While numerous fMRI studies have reported audiovisual interactions at multiple levels of the cortical hierarchy, they were not able to dissociate transient and sustained responses. This fMRI study optimized the design to disentangle the contributions of sustained, onset and offset responses to superadditive and subadditive interactions and localize the effects within the visual and auditory processing hierarchies.
Seventeen subjects participated in this fMRI study (Siemens TimTrio 3T scanner, GE-EPI, TE = 40 ms, 38 axial slices, TR = 3.08 s). While engaged in a target detection task, they were presented with 1 s, 10 s, 20 s, 30 s blocks of (i) video clips of an expanding radial star-field, (ii) auditory pink noise or (iii) both. The velocity of the star-field and the sound amplitude were jointly modulated according to 0.1 Hz sine-wave function. The regressors of the general linear model were formed by convolving (i) delta functions encoding the onset and offset of each block and (ii) box car functions adjusted for block length with the hemodynamic response functions. Blocks of 1 s duration were modeled only as onsets. In addition, the model included targets and parametric modulators encoding the amplitude / velocity modulation. To allow for a random-effects analysis (SPM5), contrast images for each subject were entered into second level one sample t-tests. We tested for superadditive and subadditive interactions separately for onset, offset and sustained block responses. Results are reported at p<0.05 whole brain corrected.
Significant audiovisual interactions were observed only for the transients: For the onsets, the interactions were superadditive in the fusiform gyrus (FFG), anterior calcarine sulcus (aCaS) and the cuneus (Cun) and subadditive in the posterior superior temporal gyrus/sulcus (pSTS/STG) and the precuneus (PrCun). For the offsets, the interactions were subadditive in the pSTS/STG region and the anterior intraparietal sulcus (aIPS). The regional response profiles were further characterized by their general responsiveness to visual, auditory and audiovisual onsets, offsets and sustained stimulation. This dissociated three activation profiles: (i) In FFG, only the onsets elicited a strong positive response with moderate responses to offsets and sustained stimulation. Further, the onset responses were positive for visual and audiovisual stimuli and negative for auditory stimuli. (ii) In aCaS, only the offsets elicited a positive response for all sensory modalities. (iii) In the remaining regions, both onsets and offsets elicited a positive response for all sensory modalities.
In conclusion, audiovisual interactions are observed primarily for transient rather than sustained stimulation. Furthermore, these AV interactions are located in regions that respond primarily to transients. In contrast, no significant interactions were observed in regions that exhibited sustained responses to extended blocks of audiovisual stimulation.