The first documented use of a functional approach using electrical stimulation was carried out by Fritsch and Hitzig (17). They stimulated the cortical surface of dog brains with short DC discharges from a battery and reported five areas that gave rise to different types of movements. A few years later, Ferrier reported his experiments with dogs, cats, rabbits, guinea pigs and later with monkeys where he used seconds of continuous alternating current to induce movements (18, 19). Ferrier is credited as describing “purposive movements” and the first somatically arranged motor map. These two sets of researchers argued over the “correct” method of stimulation. Ferrier took the position that shorter trains of stimulation did not allow complex movements whereas Hitzig countered that longer stimulation trains evoked seizures (20). During the later part of the nineteenth century, Beevor and Horsley (21) working with monkeys used both short trains to elicit a muscle twitch and long trains to create a map of purposeful movements. They argued that both approaches were valuable ways to view cortical organization. While their later work was primarily based on short trains they also demonstrated that careful removal of the gray matter and stimulation of the remaining descending white matter tracts produced the same map of the body as did short train stimulation of the cortex itself. These influential researchers appear to have set the research field on a course typified by the view that the function of the motor cortex was defined by the simple activation of the descending connections to the spinal cord.

Sherrington made many important insights into the function of motor cortex by stimulating the surface of the motor cortex with short bursts of electrical stimulation at the lowest intensity possible that allowed the observation of a movement. He examined the relation between cortical points by stimulating one site and observing the movement evoked by stimulation at another site. He noted that prior stimulation of one site could alter the movement at another site. Even though he was using short bursts of stimulation he reached the important conclusion that the map was not anatomically fixed but plastic and dependent on its activation history. That is to say the motor cortex is the repository of a large variety of sequences of movements (22). Several researchers still using cortical surface stimulation with short trains in primates, including humans, detailed multiple cortical maps (23–27). However, the “problem” with surface stimulation is that it requires relatively high intensity stimulation currents to elicit movements. Thus according to the view that larger currents lead to more current spread and deriving a map should not involve overlapping areas of excitation, the technique of surface stimulation leads to a relatively low resolution motor map. In order to stimulate with smaller current intensities and derive the highest resolution motor map possible, Asanuma pioneered the intracortical microstimulation technique for deriving movement representations.

In the late 1960s and early 1970s, Hiroshi Asanuma and ­colleagues worked in the motor cortex of cats (28, 29) and capuchin monkeys (30). They inserted fine insulated microelectrodes with an exposed tip into different layers of the motor neocortex and passed a minimal amount of cathodal (negative charge) current with short pulse-widths (0.2 ms) and short trains (40 ms trains of 13 pulses at 300 Hz) to elicit a muscle or joint movement. On average their current intensities were 1/100th the intensity of those used to evoke movement on the cortical surface. These “standard” short train methodologies have been successfully used for decades and have generated a substantial literature on the functioning of motor cortex. Hall and Lindholm (31) were the first to adopt the short train ICMS technique to work in rats. In the 2000s, Michael Graziano reinvigorated the use of long-train intracortical microstimulation methodologies. Using stimulation trains an order of magnitude longer than traditional ICMS (500 ms versus <50 ms), he evoked complex, coordinated movements involving multiple joints in monkeys (32, 33). The train duration used was similar to both the timescale of forelimb reaches and motor cortex neuron activity during movement in monkeys (34). It appears that train duration is a critical factor mediating evoked movements by ICMS. At behaviourally relevant time scales, complex movements similar to spontaneous behaviour are evoked. As train duration is reduced, the movements become truncated, eventually leading to individual joint flexions. Long-train ICMS has more recently been successfully applied in the rat (35).

ICMS provides a valuable tool in the study of both the functional organization and sensitivity of cortical movement representations in addition to the nature and property of cortical-evoked movements (36). Below we provide a detailed description of the use of short-train stimulation techniques in both rats and mice and long-train stimulation techniques in rats.