Category: States


test with lucid scribe 1.5.5
Thanks to the latest updates, the errors that happen to me are gone. Now REM detection is achieved without a problem, thank you very much for your great work Michael.

I added a new channel that shows when the vREM channel is disabled because large movements have been detected. The minimum and maximum values that define eye movements are now also configurable.

Captured a good REM session starting at 0545.

And here the Toss threshold was reached from eye brow movements during a dream. An extra configuration for the amount of movements above the toss threshold within a timeframe could fix that.

I used the following settings and had only one false positive while settling down to sleep:

Download entry as Lucid Scribe Data (LSD) (605 MB).

The audio track woke me up after just a few seconds. My eyes can be seen opening in the real world after looking around in the dreamworld in the minute featured above. I have since started a shortlist of tracks that play only a few melodies for a couple of seconds that we have been having more luck with: Shortlist.

Download entry as Lucid Scribe Data (LSD) (103 MB).

I rolled into position at 0320 and was dreaming for the better part of the hour. The audio triggered at 0333, again at 0334 and I dreamt through the tracks without waking. I recalled 3 dreams.

Download entry as Lucid Scribe Data (LSD) (127 MB).

Rapid Eye Movements During Dream

The REM-detection algorithm is now ready to be burnt to the standalone version of the sleep lab and can be seen climbing the planes in the minute above.

I could hear Bob’s voice, but had the most annoying itch on my chin that caused major instabilities in NPR. I tried to hang on for dear life, but the itch was so overpowering that I could not hold still for long. When the eye movements subsided a few minutes later, the twitch grew so strong that it almost triggered the next track.

Download entry as Lucid Scribe Data (LSD) (95 MB).

EOG REM Eye Movements

I now have a setup that is fit for a princess! And finally have my wife hooked up… she can be seen sneaking in a dream amidst her beauty sleep in the minute above.

The algorithm had a good time picking up her eye movements – the dark eye-lashes helped it along nicely.

I will add an option to get the lab setup in Heisenberg Blue in her honor!

Download entry as Lucid Scribe Data (LSD).

EOG REM Graph

The vREM channel faired pretty well with the new tweaks to the algorithm. I rolled into position at 05:24 and had an intense dream from 05:48 until 05:51. Then laid dead still for a minute and finally awoke from it and rolled out of it at 05:52. The algorithm differentiated the rolling motions from the eye movements pretty well.

The latest version of Lucid Scribe can now also export the image data so the EOG can be re-viewed remotely! This increases the file size significantly, so I only uploaded 5 am.

Download entry as Lucid Scribe Data (LSD) (134 MB).

Sleep and Consciousness Research Graph

I was able to record eye movements with the camera on the bedside table next to me!

The headband still records more dreams as it trains the camera on your eyes as you move around throughout the night, but is a challenge to sleep with for the casual user. And this setup brings the cost down by 100 and hopefully clears the way for more research!

Download entry as Lucid Scribe Data (LSD) or Comma Separated Values (CSV).

Sleep and Consciousness Research Graph

The outline of the NoIR camera hanging from the halograph headband can be seen in the shadow in the minute above. It also used to create artefacts in previous versions of the algorithm, but appears to be fairing a little better now. My alarm is currently set for 7 but I have been waking up from REM at around 6:30 consistently this week. 

Download entry as Lucid Scribe Data (LSD) or Comma Separated Values (CSV).

Sleep and Consciousness Research Graph

I caught another today… the pillow can be seen in the minute above and doesn’t affect the algorithm anymore. 

I have had a few inquires about using the setup without the headband and this shows that it could work if you can sleep close enough to the night-table or have the camera mounted above the bed somehow. I will try to record a session like that one night…