top of page

The Empathy Machine: A Play between Electromyography and Electrical Muscle Stimulation

Updated: Mar 18

Aida and Alice showcasing the Empathy Machine during Demo Day
Aida and Alice showcasing the Empathy Machine during Demo Day

Overview & Purpose

Imagine you can embody someone so well, such that when they feel something, you can feel it too. People with mirror-touch synesthesia (MTS) report tactile sensations on their own bodies when seeing another person being touched. Mirror-touch synesthesia is found in approximately 1.6–2.5% of the general population.

With this background knowledge, we thought what if we can make other people link together through technology? How would it feel to be connected through something like Mirror-Touch Synesthesia? Would it feel like a connection or would it feel like a violation dynamic?

Is it Empathy or is it Power?

Project Goals

The project goal was mainly to develop a machine that read one person's Electromyography (EMG) values and then translated them to electronic muscle stimulation (EMS) signals to the other person. At least for this version of the project, the general motion will be more important - establishing a simple Sender-Receiver connection - than a more refined touch.

Technical Background


Generally speaking, synesthesia is your brain routing sensory information through multiple unrelated senses, causing you to experience more than one sense simultaneously. More common examples are if people link colors to certain numbers and names.

Generally, the brain detects what happens around you: your eyes for vision, your ears for sound, your skin for touch, and so on. When you sense something, your senses send a signal to your brain that describes what they are experiencing. Once the signal reaches the brain, it will do processing, understanding what you have just experienced.

This is a schematic that shows mirror-touch synesthesia
This is a schematic that shows mirror-touch synesthesia*

*This schematic is from this paper.

Electromyography (EMG)

Motor neurons - a specialized type of brain cell located within the spinal cord and the brain - cause muscles to contract by transmitting electrical signals. When you contract a muscle, the individual electrical signals within the fibers of the muscle cause them to contract. If there are more fibers contracted, then there is more electrical activity.

With the help of electrodes, EMG devices are able to measure muscle response or electrical activity in response to a nerve's stimulation of the muscle. The sensor basically converts the signal from the contraction into a voltage level: the level related to the amount of contraction of the muscle.

The EMG circuit requires three electrodes: positive input, negative input, and ground. The placement of these electrodes will obviously vary based on the muscle that you intend to measure.

Treyden and Aida trying out EMG electrode placement on the face
Treyden and Aida trying out EMG electrode placement on the face

For the final version of the project, we targeted the contraction through the bicep.

Picture of electrode placement to detect muscle contraction,
Electrode placement for bicep muscle contraction.*

*This picture is from this paper.

The red electrode - the positive - and the yellow electrode - the negative - are both placed on the biceps with a bit of distance. The green, ground, and reference electrode is placed on the elbow, though this position may differ depending on preference.

Electric Muscle Stimulation (EMS)

With electric muscle stimulation, you can send electrical impulses through the skin, helping to repair issues and strengthen muscles. These are an imitation of what occurs when someone contracts and releases muscles naturally. One of the most common forms of this EMS is transcutaneous electric nerve stimulation (TENS). A TENS unit delivers a small electrical impulse through electrodes. Usually, they are used to treat symptoms such as period, labor, neck, back, and many more types of pain. For our purposes of the empathy machine, we have used 3 TENS 7000 units to convey stimulation and contract muscles. Of course, you have to be aware of the safety risks when using these devices: generally, you should not put it in dangerous places, e.g. the heart, spinal cord, etc.), which is why the Empathy Machine used EMG-EMS communication mostly through either the arms (biceps) or the legs.

For the EMS device we experimented with a lot of muscles, the most interesting and most apparent effects could be felt on larger muscles groups though. This lead us to try a lot with the triceps, leading to straightening of the elbow,

and also, finally, the biceps, which bends the elbow.

Technical Implementation


  • 3x TENS 7000 Unit

  • 2x EMG sensors

  • Arduino UNO

  • 16-channel relay switch (this was out of convenience because we had the relay switch)


Each TENS 7000 Unit has two channels, such that you can control two pairs of electrodes with one device. As we are using 3 TENS units, we can use 6 pairs of electrodes in total. Whereas for a simple demonstration of communication between EMG and EMS can be done with just one TENS unit and two EMG devices, it become more interesting using the other electrodes on other parts of the body, potentially simulating what it would feel like in multiple regions - this gets more into the question of power dynamics and control.

Version 1

This version uses one EMG device on my left hand, whereas Treyden's arms are both connected to one TENS unit. This is

The relay board serves as a switch that either allows there to be a connection between the TENS uni and the electrodes or not based on a certain threshold value that has to be surpassed by the EMG device. When the relay opens a terminal, this means that the connection between EMS electrode and TENS unit is established, leading to an electric stimuli.

Version 2

We later on developed a system in which you can rewire the wires, such that there are more channels controlled by the relay. Besides, in order to give the machine a bit more of character, we decided to print a case for it, using Prusa's 3D printer.

199 views0 comments

Recent Posts

See All


bottom of page