UC Berkeley

Brain-controlled prostheses have the potential to improve the quality of life of a large number of paralyzed persons by allowing them to control prosthetic limbs simply by thought. An essential requirement for natural use of such neural prostheses is that the user should be provided with somatosensory feedback from the artificial limb. This can be achieved by electrically stimulating small populations of neurons in the cortex; a process known as cortical microstimulation. This dissertation describes the development of novel technologies for experimental neuroscience and their use to explore the neural and perceptual effects of cortical microstimulation in rodents.

The first part of this dissertation describes the various tools built to study cortical microstimulation in awake, behaving rodents. Circuits were developed to simultaneously record and stimulate neurons in the cortex; thus paving the way for future research into neural responses to stimulation. Further, electrode coatings based on conductive polymers were explored to allow chronic neural stimulation without causing long term damage to the implanted electrodes or neural tissue.

Two technologies were then developed to monitor different aspects of rodent behavior. Wireless accelerometers were built to monitor gross behavior and neural network based algorithms were developed to extract behavioral states from such acceleration data. Rats have poor visual acuity and actively scan their facial vibrissae or whiskers to feel the world around them. To study this whisking behavior, a video based whisker tracking system was developed which tracks the movement of a single whisker in real-time.

The second part of this dissertation describes advances in neuroscience enabled by these tools. The neural response to microstimulation was explored in awake, behaving rats and it was found that microstimulation in barrel cortex evokes 15-18 Hz oscillations that are strongly modulated by motor behavior. In freely whisking rats, the power of the microstimulation evoked oscillation in the local field potential was inversely correlated to the strength of whisking. This relationship was also present in rats performing a stimulus detection task suggesting that the effect was not due to sleep or drowsiness. Further, a computational model of the thalamocortical loop is presented which recreates the observed phenomenon and predicts some of its underlying causes. These findings demonstrate that stimulus-evoked oscillations are strongly influenced by motor modulation of afferent somatosensory circuits.

The perceptual effects of cortical microstimulation were then explored using behavioral studies. Tactile exploration of the environment involves the active movement of external mechanoreceptors and the integration of information across sensory and motor modalities. To explore the encoding of somatosensory feedback in such an active sensing system, a novel behavioral paradigm was introduced which precisely controls cortical microstimulation in real-time based on the movements of the animal. Using a real-time whisker tracking system, microstimulation was delivered in barrel cortex of actively whisking rats when their whisker crossed a software-defined target. Rats learned to rapidly integrate microstimulation cues with their knowledge of whisker position to compute target location along the rostro-caudal axis. This showed that rats can perform sensorimotor integration using electrically delivered stimuli. Moreover, it was discovered that rats trained to respond to cortical microstimulation responded similarly to physical whisker deflection suggesting that microstimulation in barrel cortex induces tactile percepts. This ability to encode tactile percepts in active sensing systems may be critical for providing the sense of touch to future users of motor neuroprostheses.