Myndlift's thresholds are designed to provide users with the most stable thresholds possible to enable learning, while putting safety measures in place to prevent frustrating or boring sessions.
The threshold is calculated dynamically for the first 3 minutes of each session. During this three minute period, clients will get immediate feedback, and shouldn't notice that calibration is occurring in the background.
As soon as 3 minutes of clean EEG are recorded, the algorithm sets a threshold that will remain stable during the round except for extreme cases.
The extreme cases which will trigger recalibration are:
When there's too much noise in the signal, causing the overall success rate to reach above the overall-maximum or below the overall-minimum
The overall-maximum is calculated as an overall session success rate of above 85% or the target success goal set + 15%, whichever is higher.
The overall-minimum is calculated as an overall session success rate of 30% or the target success goal set - 8%, whichever is lower.
The same algorithm is set per frequency as well.
When a recalibration occurs, as a result of one of the extreme cases detailed above, the threshold is recalculated once based on the previous 90 seconds of clean EEG. The new threshold will remain the same for the remainder of the session.
In order to prevent thresholds from changing too frequently, the overall success rate for the sessions needs to be above 35% and below 80% to deploy a program.
For more information about success rates or setting up a custom protocol, please see this article: Manage Protocols
Still need help? Contact us at support@myndlift.com for assistance.