Limb position sense: Role of limb posture, visual experience, and input from muscle spindle Ia afferents

Abstract

Limb position sense is the ability to determine location and orientation of limb segments with respect to each other and with respect to the external environment without vision. Limb position sense is critical for accurate control of posture and movement. The lack of position sense due to illness has devastating consequences for the performance of even simplest motor tasks performed under visual control. Position sense of limb segments with respect to each other originates from joint angle-related sensory information provided by the muscle spindles, Golgi tendon organs and cutaneous afferents in skin overlying joints. Limb endpoint position sense with respect to external space can be derived by the nervous system through a transformation of joint-related coordinates to endpoint external coordinates using estimated dimensions of limb segments. This coordinate transformation and factors that can potentially affect accuracy and precision of endpoint position sense in external space are not fully understood, and many important questions remain unanswered. For example, it is not clear why people perceive hand position more precisely in the radial than in azimuth direction and closer to the body than farther away. Moreover, since vision contributes to forming somatosensory representations of body segment dimensions and to integration of somatosensory information encoded in joint-based and external coordinates, would long-term blindness differentially affect the limb position sense in joint and external space coordinates? Furthermore, what does happen to precision of limb endpoint positioning in external space if input from muscle spindle Ia afferents from a major limb joint is compromised? My work addressed all these questions. Using a theoretical analysis of the transformation of random joint angle errors to random hand position errors for a two-joint kinematic arm model, I demonstrated that arm posture alone can explain the better precision of hand position sense in the radial than in azimuth direction and closer to the body than farther away. I confirmed the model predictions in experiments with healthy sighted individuals (n=11) who performed a hand position matching task. The fact that the predicted distributions of random hand position errors (precision ellipses) in the horizontal workspace were nearly orthogonal to the experimentally obtained arm stiffness ellipses reported in the literature, provides a mechanistic explanation for how the distribution of random hand position errors is shaped for any arm posture. To investigate the role of visual experience in arm position sense in joint and external space coordinates, I investigated long-term blind (n=7) and age-matched sighted individuals (n=7) who performed three arm position matching tasks: joint angle matching (JAM), hand distance and direction matching (DDM), and hand distance and mirror direction matching (MDDM). The latter hand position matching task was kinematically identical to the joint angle matching task JAM. The blind participants generally had lower accuracy and precision of arm position sense in joint angle and hand position matching than the sighted. In addition, the blind had the same precision of arm position sense in the joint angle matching and hand position matching tasks. Measuring the Contingent Negative Variation-inspired EEG potential (CNV), an indicator of neurophysiological functions related to task complexity during the performance of arm matching tasks, revealed that the blind group had a higher EEG negative potential than the sighted group. Both subject groups demonstrated a higher EEG negative potential in the hand position matching DDM task than in the joint angle JAM task or in another hand position matching MDDM. These results suggest that visual experience positively affects arm position sense possibly due to integration of visual and proprioceptive sensory information and the development of more accurate body schema. I suggest that similar precision of arm position sense in joint and external coordinates in the blind participants may result from a possible increase in perceptual acuity of other exteroceptors that could provide information about limb position in the peripersonal space. The lower accuracy and higher perceived task complexity in the DDM task, as judged from the EEG negative potential, may indicate more complex transformations from the joint coordinates to hand position coordinates required for this task. In the last series of experiments, I investigated precision of hindpaw position control just before and right after the stance phase during walking in 4 cats after their major knee muscles were self-reinnervated unilaterally. This procedure compromises sensory input from Ia afferents (Cope at al. 1994), the main source of joint-related position information. Contrary to my expectations, the precision of paw position control at the late and mid-swing phase during walking significantly increased bilaterally. The animals achieved this by adopting a more extended hindlimb posture bilaterally during mid-swing and prior to stance onset. This extended limb posture reduces the radial random error, as follows from my theoretical analysis of precision of limb position sense. These studies determined for the first time the contribution of limb posture, visual experience and proprioceptive input to accuracy and precision of limb position sense. This information will help design new diagnostic and therapeutic tools for people with visual and proprioceptive deficits. The gained understanding of random error distributions of hand position sense can be used to design control panels for blind and sighted operators that increase precision and accuracy of control.