What is Smart Timer?
Many lucid dreamers are familiar with the concept of setting alarms to go off in the middle of the night in order to practice lucid dream induction. Smart Timer is an attempt at playing reminders closer to the expected REM period by:
- Analyzing past historical data and creating a list of potential REM periods throughout the night
- Adjusting the next expected REM period based on user events and activity earlier in the night
Together with observation of the past history, smart timer delivers REM reminders significantly more accurately than the previously used method of REM induction. This means that:
- Smart Timer does not depend on when you go to sleep.
- Smart Timer does not depend on how long it takes for you to fall asleep – it adjusts itself.
- Smart Timer compensates for the differences between REM episodes caused by wakefulness due to writing in a dream journal, bathroom breaks, etc.
Smart timer in action:
How it works
Hypnogram is a simplification of many different sources of data available for sleep research. Based on that, sleep can be classified into 5 distinct stages + awakening. Here’s a hypnogram example that I could find. Since it closely matches what I’m seeing on my sleep graphs, I’m using it as a basis of my research. Here’s another example of a hypnogram, along with a description of what it is.
Here’s the same hypnogram, imported into the Lucid Dreaming App and superimposed with the dream and awakening events that I entered into the app. The light blue line is an example of actigraphy data, as it aligns with the hypnogram. Here the hypnogram is shifted forward by 20 minutes for clarity purposes. It is evident that dreams cluster above the expected REM periods
A problem with this method is that not all activity on all days aligns with the one example hypnogram that I have:
Here enters the smart timer. Since Lucid Dreaming App does not have the luxury of EEG, EMG or other ExGs, we have to work with what we have – a history of dream events (orange triangles), awakenings (black circles), hypothetical REM episodes (Red line) and sleep score levels reported by the app (colorful spikes)
Smart Timer analyzes this data and produces two predictions:
- First, it predicts when the next REM period (red line segment) will occur throughout the night. This is important, as you may start the app at different times in the evening, and it may take you longer/shorter to fall asleep. Smart Timer attempts to compensate by these factors by looking at activity spikes within expected REM window and user events entered through gestures.
- Second, when a predicted REM cycle begins, the SmartTimer starts to closely look at the activity level. When the sum of activity level since the start of the REM period reaches a certain threshold (default is 10, you can change this in preferences), the smart timer plays the reminder, delayed by 2 minutes. This means that an REM episode with no activity will produce the reminder at the end, while a series of small activity spikes will trigger the reminder (hopefully before the awakening)
And now is when I say that the smart timer has to be configured to become truly smart! To configure the smart timer you will need some historical data – at least 3-4 nights. It is best if it has user events with dreams recorded immidiately upon awakening.
Start to analyze your sleep data by looking at your sleep score graphs. They may look rather chaotic, but after looking at graphs like these for over a month, I see some patterns.
Here, there’s a consistent peak at around 3:00 hours after going to bed. This is when I wake up to use the bathroom, because I drink some water before going to bed. Every single day I see a spike at around that time. This is a good place to start the analysis. Large spikes with activity scores over 15 (for my particular bed, with sensitivity of 0.0033) indicate that I probably got out of bed and returned within a couple minutes. Smaller spikes, between 10 and 15 may indicate writing to a dream journal while in bed. A yet smaller spike, around 5 indicates significant activity, likely being awake or interacting with the screen lightly – making the “awake” or “dream” gesture and then going back to sleep.
Activity levels below 5 probably indicate minor activity, such as turning around. This kind of analysis allows me to distinguish between purposeful and involuntary activity. Involuntary activity is what is interesting for dream research.