Biological and bionic hands: natural neural coding and artificial perception

Abstract

The first decade and a half of the twenty-first century brought about two major innovations in neuroprosthetics: the development of anthropomorphic robotic limbs that replicate much of the function of a native human arm and the refinement of algorithms that decode intended movements from brain activity. However, skilled manipulation of objects requires somatosensory feedback, for which vision is a poor substitute. For upper-limb neuroprostheses to be clinically viable, they must therefore provide for the restoration of touch and proprioception. In this review, I discuss efforts to elicit meaningful tactile sensations through stimulation of neurons in somatosensory cortex. I focus on biomimetic approaches to sensory restoration, which leverage our current understanding about how information about grasped objects is encoded in the brain of intact individuals. I argue that not only can sensory neuroscience inform the development of sensory neuroprostheses, but also that the converse is true: stimulating the brain offers an exceptional opportunity to causally interrogate neural circuits and test hypotheses about natural neural coding.

Keywords: biomimicry; neuroprosthetics; proprioception; somatosensory cortex; touch.

Figure 1.

Diagram of a somatosensory neuroprosthesis. Signals from sensors on the prosthetic hand (a) are converted into trains of intracortical microstimulation (ICMS) delivered to S1 (b), designed to elicit meaningful sensations that convey information about objects grasped in the hand.

Figure 2.

(a) Somatotopic organization of primary somatosensory cortex with the hand region highlighted (adapted from [9]). (b) Combination of skin locations at which pokes were delivered as well as receptive fields of electrodes through which stimulation was delivered. The dotted lines link conditions that were paired in a trial. As can be seen, each electrode replaced a poke; that is, the receptive field of each stimulated electrode corresponded to one of the poke locations. (c) Performance on mechanical and hybrid trials. Each dot represents a condition, bars represent the mean performance (adapted from [10]).

Figure 3.

(a) Neuronal activation evoked over a 4 × 4 mm patch of cortex by indentations delivered to the tip of the little finger at four amplitudes. As the amplitude increases, the firing rate increases and the area of activated neurons also increases. (b) Psychometric equivalence functions that map electrical amplitude onto mechanical amplitude such that the ICMS and the corresponding poke are of equal perceptual magnitude (each curve corresponds to one electrode/skin location pair; different colours denote different animals; reproduced from [10]). (c) Animals perform identically on a pressure discrimination task whether pokes are delivered to their native finger (blue) or to a prosthetic one (red). The standard amplitude for these experiments was 150 µm (reproduced from [10]).