Dj. Debowy et al., Comparison of neuronal firing rates in somatosensory and posterior parietal cortex during prehension, EXP BRAIN R, 137(3-4), 2001, pp. 269-291
To evaluate their functional roles during prehension, single-unit recording
s were made in the hand area of primary somatosensory areas 3b, 1 and 2 (S-
I) and posterior parietal areas 5 and 7 (PPC) of the same animal. Response
profiles of mean firing rate during performance of a multistage reach, gras
p, and lift task were analyzed to determine the period(s) of peak firing an
d to measure statistically significant rises or falls in rate compared with
baseline. We used the peak firing stage(s) to subdivide the population int
o classes tuned to single actions or two successive stages, or into multiac
tion groups that had sustained facilitation (BT) or inhibition (GI) during
hand-object interactions. Four times as many neurons fired at peak rates du
ring acquisition stages (approach, contact, grasp) than upon release, and t
heir firing rates were higher. Grasping evoked the strongest responses, as
grasp-tuned neurons had the highest peak rates in the population; BT, conta
ct-grasp, and grasp-lift cells also fired maximally in the grasp stage. Gra
sping also coincided with maximal inhibition of GI cells, as well as of neu
rons tuned to approach or relaxation of grasp. Holding evoked the lowest me
ant rates, and had the fewest tuned cells. S-I and PPC showed significant d
ifferences in behaviors evoking peak firing as well as facilitation and inh
ibition; these correlated with input modalities in each area. Hand contact
with the object and positioning of the fingers for grasp was the most stron
gly represented behavior in anterior S-I, where 61% received tactile inputs
from glabrous skin. Nearly 60% were facil itated at contact, 38% fired at
peak rates, and 10% were inhibited; release of grasp evoked peak firing in
only 5% of 3b-1 neurons. In posterior S-I, where proportions of tactile and
deep inputs were similar, positioning and grasping elicited peak responses
in 38% and 31%, respectively; 80% were facilitated or inhibited during gra
sping. During lift and hold, inhibition rose to 43%, while excitation decli
ned under 10%. PPC had the highest proportions firing at peak rates during
hand preshaping before contact (28%) and had the most facilitated responses
(38%) in this stage. Only 10% fired at peak rates during grasping. During
later manipulatory actions, proportions of facilitated and inhibited respon
ses in PPC were similar to those in posterior S-I. The data support models
in which PPC plans hand movements during prehension rather than guiding the
ir execution. Sensory monitoring of hand-object interaction occurs in S-I,
where cells sense specific hand behaviors, signal stage completion, enable
error correction, and may update grasp programs formulated in PPC. The resu
lts are discussed in relation to those obtained from lesion studies in huma
ns.