Detecting Very Slight Tendon Movements In Wrist

I've always had an intense fascination for lucid dreaming. Finding ways to communicate or control outside devices based off the slight body movements that are possible despite the natural paralysis during sleep is a topic of great interest to me.

I have several LDs (lucid dreams) each night and through much experimentation have found that with a bit of concentration, I could move my fingers enough to trigger very sensitive buttons (if placed perfectly before falling asleep) or to grip myself (if my hand is placed say, near my neck before intentionally entering a dream from waking) such that the sensation of my fingers reaching my neck would wake me.

These experiments were proof enough that if I perhaps had some sort of glove or wrist mounted piezo sensors that are able to detect the individual tendon/finger movements with high enough sensitivity, combined with software that can parse patterns and reliably filter out minor unintentional simultaneous movements for the fingers that are prone to movement when others nearby are moved, it would be possible to establish one way communication with outside devices, the Internet, etc.

Input could be in the form of different patterns or even text (tap code, morse code, or through a hybrid allowing use of each digit of the hands) or musical notes (each finger representing a note of a scale with a finger or two used to change octave or to toggle sustain.) As a musician who "writes" his best music within his dreams, this could be quite handy for getting melodies down before they vanish from memory on waking.

For example, to activate input, you could perform a simple pattern that is unlikely to be triggered unintentionally. It could be as easy as left thumb flex, right thumb flex repeated in that order several times.

From this point, you could use another pattern to choose to trigger an alarm (sound, vibration or electrical shock), or to initiate musical or text recording. There are many different uses that could be explored, and it might even be possible that 2 way communication could be established if a reliable method of transmitting to the dreamer without waking them could be created. This is however beyond the scope of my intentions.

I have suffered from sleep paralysis on a nightly basis for over a decade now which can be extremely frightening and last up to several minutes, and during this time I am able to move my fingers as well as shake my head back and forth (which I need to do to break out otherwise the paralysis will persist until I'm pulled into a nightmare.) My solution to these terrifying experiences was wearing a heart rate monitor with an alarm that goes off past a certain BPM, which only happens when I am frightened. Now that I am so used to these experiences, my heart rate does not increase enough that I can reliably use it as a trigger without also waking myself out of exciting dreams that I'd rather not have prematurely terminated.

I am fairly confident that a device such as this could help with breaking out of my paralysis at will and PERHAPS even be sold commercially to other sufferers of this condition, however niche the market. Giving me control over my sleep paralysis alone would be a highly satisfactory result with significant quality of life improvements and be worth the effort of building this. Waking up partners in bed with my head thrashing is less than ideal!

As an avid software programmer, I do not think that the code involved would be terribly difficult. However, I have very little electrical knowledge and building this without existing plans would be tough and time consuming. It would be enough to have a pressure/flex voltage for each digit transmitted to a Raspberry Pi or via Bluetooth to a computer and the rest could be done through software.

My questions are..

Does a commercial device exist that would be suitable for my purposes and not be terribly expensive? Perhaps a glove?

Is a wrist strap with piezo sensors over each tendon that control the fingers a realistic possibility to achieve my goals? I do not need to know the position of the fingers or whether it is pushing against something. Just that movement is being attempted.

Do existing plans exist to create a similar device?

Thank you!

iceng29 days ago

I would attach two accelerators on both sides of the wrist and only use the difference to trigger the signal you choose ...

This should avoid natural arm movement but trigger when only the wrist moves..

Since you like software large common motions like reposition or blanket move can be totally ignored..

Ril900 (author)  iceng28 days ago

Thanks! Check this out:

Reading finger movements through the tendons in the wrist. This is more what I was looking for.