Third and final installment of Nicolelis day. Here, the Nicolelis lab develops a cortical sensory prosthesis which allows rats to detect and respond to infrarad light (23403583). A IR detector is mounted on the head, and the detected intensity determines the level of stimulation over the a whisker-barrel of the somatosensory cortex. With training, the rats become quite proficient at using the prosthesis–they sweep their heads back and forth to scan the IR field and can reliably choose a lever cued by an IR signal. Importantly, the input to S1 is not actually deleted prior to the prosthesis being turned on, and neurons end up tuning to *both* IR and touch. This is early stages, but one limitation is the use of a simple point-source as the sensation (no spatial dimension to the new sense). Very clever, and they give an appropriate shout-out to the pioneering work of Paul Bach-y-Rita on sensory substitution.