By Dr. Lise Johnson (CNT Education Manager)
At the end of my last post I left you with the question of whether or not electrical stimulation could be used to provide somatosensory feedback for a brain-computer interface. If you read my last post, then you already know that it’s possible to artificially evoke sensations by electrically stimulating a particular part of the brain such as the primary somatosensory cortex. It may seem like this answers the question already, but like so many things in life, it’s more complicated than it seems. Before we can use electrical stimulation as a substitute for natural sensory feedback, we need to consider how electrical stimulation works.
Sensory receptors are neurons, and like all neurons they represent information using an electrical code. As sensory signals come in, they are sent up to the cortex where other neurons are electrically activated, and then others, and then others. This eventually leads to the conscious perception of a sensory event, and it is at that point that you realize you have just stepped on a rock (or whatever). Electrical stimulation bypasses the sensory receptors and jumps straight to the cortex, activating the neurons that usually get input from the sensory receptors. The chain of neuron activation after that point is (theoretically) the same, and so electrical stimulation also leads to the conscious perception of a sensory event. So far so good. But, the type of electrical stimulation that you get from an electrode is really different from the electrical stimulation that the nervous system uses internally. Really, really different. In truth, the stimulus you provide with an electrode is nowhere close to natural. A standard-sized electrocorticography electrode is 2.3mm in diameter, which is enormous compared to the size of a single cell. Electrically stimulating through one of these electrodes is like trying to fill a thimble with a fire hose. As a result, the sensations that are elicited by electrical stimulation don’t feel like natural sensations. If you ask people who have received this kind of stimulation what it feels like, most often the first thing they’ll say is “it feels weird.” The sensation that is evoked may not be enough like natural sensory input for a person to understand.
Another thing that the sensory receptors do very well is communicate small changes in the inputs. For example, you can tell when something is touching you very lightly, when something is poking you very sharply, and many different shades in-between. This kind of graded feedback is really important – it keeps you from dropping things, crushing things, hurting yourself and generally being a nuisance wherever you go. This kind of information is also particularly difficult to get from visual feedback, so, it is especially important to get it from electrical stimulation. In order for this to work researchers need to be able to change something about the stimulus in such a way that it feels like a change in intensity. In other words, you don’t want the type of feeling to change when you change the stimulus, you just want the strength of the feeling to change. There are basically two things you can change about the electrical stimulus, the magnitude and the frequency (these are called the stimulation parameters). But, electrical stimulation is such a weird thing, it isn’t clear that changing the stimulation will feel different at all, or, if it does feel different, it might feel like something else entirely.
To answer these questions, Jeff Ojemann designed a simple study for his volunteer epilepsy patients. In this experiment, an electrode on the primary sensory cortex was stimulated with different stimulation parameters (different combination of magnitude and frequency) and the person was asked to report how it felt. The experiment worked very much like an eye exam – the doctor provided one stimulus, and then another and simply asked, “Was the first one stronger, the second one stronger, or were they the same?” Importantly, the subject didn’t know what the stimulation parameters were, and the pairs were mixed up in a random order. In addition to asking which stimulus was stronger, the researchers asked the subjects what the stimulus felt like. This is called a qualitative report, and even though it is impossible to measure with numbers, this kind of information is important to know whether the different kinds of stimulation were perceived in a similar way, or if changing the parameters changed the feeling entirely.
The results of this experiment were very encouraging. Subjects reported that while the stimulation felt weird, the weirdness was constant. That is to say, increases or decreases in amplitude and frequency changed only the strength of the feeling, it didn’t change the essence of the feeling, or the perceptual qualities of the sensation (if you want to be technical). The other good news was that, almost all the time, subjects could identify the correct relationship between different kinds of stimuli – they thought the lower magnitude stimulation felt weaker and the higher magnitude stimulation felt stronger, and the same from lower and higher frequencies.
Of course, there is still more that we need to know (in science there is almost always still more to know!). For example, what would happen if two electrodes were stimulated at the same time? And what happens if the same electrode is stimulated for a long period of time? And just how small can we make the changes in stimulus parameters before subjects can’t tell them apart? All of these questions need to be answered before we can say for sure that electrical stimulation will be useful for feedback on BCIs, but given the results of Ojemann’s most recent experiment, we have reason to hope that it can.