The Future of Touch: Enhancing Artificial Senses for BCI Users

Rapid technological advancements are increasingly narrowing the line between humans and machines. At the forefront of this progress is a brain-computer interface (BCI), which creates a direct communication link between the brain’s electrical activity and an external output.

By facilitating communication between the brain and computer, BCI helps restore the capabilities of physically challenged people. It converts activity in our brains into signals that can replace or improve body functions, such as muscle movement, typically controlled by the brain. This way, BCI showcases huge potential in improving people’s quality of life. 

In development for about half a century now, BCI has seen massive progress over the years with researchers now demonstrating the technology’s ability to efficiently restore capabilities of people with disabilities, such as paralysis, motor impairments, speech difficulties, and those with schizophrenia symptoms.

Studies have also shown that using BCI helps disabled people experience the lost sensation of touch. Such tactile sensations, however, remain imperfect and are similar between objects with different textures or temperatures. Now, scientists are looking to create an intuitive sense of touch.

Touch is an integral part of our lives, helping us not only connect with others but also pick up objects and walk. According to Charles Greenspon, a neuroscientist at the University of Chicago:

“Most people don’t realize how often they rely on touch instead of vision. If you can’t feel, you have to constantly watch your hand while doing anything, and you still risk spilling, crushing or dropping objects.”

To restore sensation in prosthetic limbs, researchers use tiny electrode arrays placed in the brain areas responsible for that specific function. 

This allows participants to move their limbs using a robotic arm by simply thinking about movement while sensors on it trigger pulses of electrical activity in the regions of the brain dedicated to touch. While scientists could invoke feelings of touch, these were rather weak and difficult to localise as to just where the contact actually occurred.

But now, brand new research has brain-computer interface users designing unique tactile experiences for different objects shown on a screen and then, using sensation alone, guessing the object with some accuracy.

The Challenge with Integrating Sensory Feedback in Prosthetics

Scientists from the University of Pittsburgh School of Medicine have achieved a breakthrough that takes them all that closer to developing a BCI that allows people with tetraplegia¹, also known as quadriplegia, to restore their lost sense of touch.

Tetraplegia is when someone loses movement in both their arms and legs, and often the torso too, usually because of an injury to the cervical spinal cord, strokes, or other neurological damage. 

The damage disrupts the signals that allow the brain to receive and process sensory information like touch, resulting in the patient’s loss of feeling in the affected limbs.

While prosthetics provide an artificial limb to replace the lost function, this can only be achieved if they also provide a sense of touch, much like a real limb. Traditional artificial limbs focused mainly on restoring movement, but technological advancements have made it possible to have a sense of touch with the use of sensors and electrical stimulation.

To completely restore the lost function of limbs, the device has to be seamlessly integrated with a person’s existing sensorimotor system, which connects human perception and action.

To achieve this, tactile feedback is of key importance. This form of sensory interaction with devices is all about physical touch experiences.

A promising way to provide this kind of feedback is through intracortical microstimulation (ICMS) of the somatosensory cortex, which can evoke localized sensations on a person’s paralyzed limb. By delivering tactile information directly to the brain, ICMS makes for an attractive option for people with high-level amputation or spinal cord injury.

However, achieving this isn’t simple but rather very difficult because of our limited understanding of the neural processing of touch. Hardware restrictions also limit the ability to replicate neural responses naturally. Also, the stimulation parameter space is complex, and reports of measuring just how real the artificial experience with sensation was are prone to bias and difficult to interpret.

While the position and strength of the touch evoked through microstimulation of the somatosensory cortex can be conveyed reliably, there are issues with developing more intricate natural sensations, which involve insufficient techniques to effectively scan a wide stimulus space and problems with analyzing the perceptual quality. 

Most studies that have explored the psychophysics of ICMS have manipulated one stimulation parameter at a time. Most of the studies have been conducted in non-human primates that can’t verbalize their experienced sensations. 

There is simply a need for more efficient methods to explore the quality of sensations evoked through ICMS.

So, in the latest study, a collaboration between Pitt and the University of Chicago, scientists have presented an interface that addresses the challenges of creating complex naturalistic sensations. 

The study noted that, with people ultimately expected to use closed-loop BCIs in daily life, it is important to investigate the functional use and experience of ICMS-evoked sensations. 

However, the sample of participants available in such research is limited. Only about seven people with bidirectional intracortical implants in their somatosensory and motor cortex are available, and the study included three of them.

Empowering BCI Users to Define Their Sensory Experience

The interface created by the scientists has been used by three male individuals with tetraplegia to design their sensations for different virtual objects.

Artificial tactile sensations were designed to represent interactions with a cat, an apple, a key, a towel, and a piece of toast. These objects were chosen for their range of tactile dimensions, including familiarity, pleasantness, temperature, micro and macro texture, moisture, friction, and compliance.

Participants used the residual function in their left hand to interact with the tablet interface that generated “touch” sensations on their palm surface.

While exploring the objects through their artificial touch, participants described the cool roundness of an apple, the smooth, rigid surface of a door key, and the warm fur of a cat. This is completely different from earlier experiments, where artificial touch often felt like tingling or buzzing, which didn’t even vary from one object to another.

What set this experiment apart from previous research was that participants were in control of their own stimulation and could actively explore an object presented visually. 

In passive stimulation, where there is no vision and exploration, people usually report skin-level sensations such as pressure or vibration. This is because participants’ attention is focused on their body. 

In contrast, participants in active exploration are likely focused on the external world. Hence, they interpret the same percept as object-oriented sensations like roughness. This could be why the latest study participants spontaneously reported more object-oriented sensation descriptors.

This ability of participants to control stimulation presentation by “touching” an object displayed on the tablet helped create a more realistic experimental context. Here, the experienced sensations weren’t the result of experimenter-driven stimulation without a meaningful context, but rather targeted explorative movements.

Scientists basically gave BCI users control over the details of the electrical stimulation that creates tactile sensations rather than making those decisions themselves, allowing them to recreate a sense of touch that felt intuitive to them. According to lead author Ceci Verbaarschot, a former postdoc fellow at Pitt Rehab Neural Engineering Labs and currently the assistant professor of neurological surgery and biomedical engineering at the University of Texas-Southwestern:

“Touch is an important part of non-verbal social communication; it is a sensation that is personal and that carries a lot of meaning.”