Klinische Neurophysiologie 2004; 35 - 118
DOI: 10.1055/s-2004-832030

Differential Interaction of BOLD Responses to Simultaneous Finger Stimulation in Subareas of Human Primary Somatosensory Cortex (SI)

R Jan 1, T Krause 2, B Taskin 3, A Villringer 4
  • 1Berlin
  • 2Berlin
  • 3Berlin
  • 4Berlin

Suppressive interaction of neural activity within the human somatosensory system following simultaneous stimulation of finger pairs has been described in electrophysiological studies (Hsieh et al 1995; Hoechstetter et al. 2001); SI and SII responses to simultaneous stimulation of two fingers appeared to be consistently smaller than the sum of the corresponding separate responses, indicating suppressive interaction between finger representations. In this fMRI study we investigated whether the BOLD responses to simultaneous stimulation of adjacent fingers exhibit a non-linear interference behavior and whether the extent of interference is dependent on subareas of SI. MR imaging was performed on a 1.5 T scanner using a surface coil. Functional images (9 slices, voxel size 2×2×3mm3) were acquired using a BOLD-sensitive T2*-GE-EPI sequence (TR=3s, TE=60 ms). Electrical stimuli (4 single 200-µs pulses, frequency 4Hz) of moderate intensities markedly below individual pain thresholds were delivered scan-triggered to the right hand (digit 2: 3.9±0.4 mA, digit 3: 3.6±0.3 mA). The experimental paradigm comprised three different stimulation conditions (separate stimulation of digit 2 and digit 3, simultaneous stimulation of digit pair 2 and 3) and null events (84 repetitions each, randomized order, mean interstimulus interval 6s). Twelve healthy subjects (mean 27 years) participated in the study. Ten out of the twelve subjects exhibited statistically significant activations (p<0.001) in SI due to the two separate stimulation conditions and were further analyzed. Group analysis demonstrated distinct activation foci associated with all three stimulation conditions presumably reflecting the activation of SI subareas 3b, 1 and 2. Within area 3b, the partially overlapping activations exhibited maxima which differed in location, pointing to a somatotopical arrangement of finger representations. In area 1/2 activation maxima shared identical coordinates. T-maps revealed significant interaction in projection to areas 1 and 2. However, as these results might be biased by the height of betas, we performed an independent calculation for all significantly activated voxels in SI. The resulting map revealed that interaction was also prevalent within area 3b. Interaction ratios showed a rostro-caudal increase in interaction from area 3b (26%) to area 1 (32.6%) and area 2 (42.2%). In parallel, the somatotopical arrangement was less clear cut within areas 1 and 2 as compared to area 3b. These results are in agreement with evidence for a rostro-caudal increase in convergence of afferent input and thus in overlap of receptive fields within SI, and concepts suggesting a hierarchical processing from 3b as the primary projection area of somatosensory information ('SI-proper') to areas 1 and 2 exhibiting integrative functions.