Right parietal-prefrontal lateralization during somatosensory discrimination of shape

Introduction: Somatosensory discrimination (SSD) of shape by manual exploration engages extended fronto-parietal networks whose activity and hemispheric topography are task-dependend and reflects different stages of somatosensory processing. During SSD with the right hand (RH), the right dorsal intraparietal sulcus (dIPS) is active during explicit discrimination of object shape, whereas the left dIPS is involved in shape information maintenance. We used functional magnetic resonance imaging (fMRI) to test the hypothesis that explicit SSD with the left hand (LH) shows the same asymmetric involvement of parietal and possibly frontal areas.

Materials and Methods: We examined 15 right-handed adults (mean 67±3.5 years) with fMRI at 3 T using a two-alternative forced-choice SSD task. After exploration of the first object (P1), subjects were required to hold shape information in memory (R1) and compare it with the second object during the next exploration phase (P2). If the first object was perceived to be longer, subjects extended their thumb; if not, they opened the whole hand (end of P2). Two runs with 12 pairs were acquired for each subject. The task was videotaped for off-line analysis. MR Images were analyzed with BrainVoyager QX 1.10.4 using standard preprocessing, normalization into Talairach space, random effects general linear modelling and event-related averaging of BOLD responses in anatomically defined volumes of interest (VOI). Voxels were thresholded at p(Bonf)<0.05. Laterality indices (LI) using the number of active voxels in VOI were calculated as LI=100*(R – L)/(R + L), with +100 denoting maximal right lateralization. Homologue regions were used for calculation.

Results: Neither SSD performance, movement frequency of repetitive thumb movements nor contact time per object or pair showed hand differences (Table, mean±SD from both runs). In terms of the neural substrates activated by SSD with RH or LH, there were marked differences in activation topography in both hemispheres when contrasting P2 > R2. More bilaterally extended activations were seen in left hemispheric/RH SSD, notably in the primary somatosensory cortices. The pre-SMA and the dIPS were predominantly active in the right hemisphere, together with the ipsilateral dorsolateral prefrontal cortex (dlPFC), independent of hand use. Both dlPFC and dIPS showed maintenance (R1) activity, and all areas peaked higher during P2 than P1.

Conclusions: SSD of shape with either hand involves predominantly the right dIPS implying a right-hemispheric lateralization during explicit discrimination of haptic information. LI of VOI sizes increased with left hand use, supporting this view. Activation of the pre-SMA probably relates to correct response selection (go/no-go function). dIPS, together with right dlPFC, shows activity during the delay phase (R1) between objects, consistent with somatosensory working memory. Our data support an asymmetric organization of haptic information processing within fronto-parietal circuits.