ATH-tool

Adaptive threshold hunting for psychophysical and neurophysiological thresholds

Description:

ATH-tool is for determining psychophysical (and neurophysiological) thresholds using parameter estimation by sequential testing. The routines applied have been validated for motor thresholds in transcranial magnetic stimulation: P. Julkunen in IEEE in Transactions on Neural Systems & Rehabilitation Engineering in 2019 "Mobile application for adaptive threshold hunting in transcranial magnetic stimulation".

The used algorithm is a modification of the procedure published by H.R. Lieberman and A.P. Pentland in Behavior Research Methods and Instrumentation in 1982 "Microcomputer-based estimation of psychophysical thresholds: The Best PEST". Some of the modifications were made based on the procedure published by F. Awiszus in Transcranial Magnetic Stimulation and Transcranial Direct Current Stimulation (Supplements to Clinical Neurophysiology, Vol. 56) in 2003 "TMS and threshold hunting".

Usage:

The application works in the range of 0-100 (initial values can be set between 20 and 80). Hence the parameters estimated should be adjusted by appropriate transform in advance to work within the working range.

The procedure can be set to cut-off after a certain number of trials. The cut-off for the routine will be enable after confidence limits of the routine reach 5% of the full working range, or after the minimum number of trials has been reached. Other adjustments to the ATH-tool performance should be made with caution. The procedure can be logged and saved to file (bestPEST_log -folder in your default file storage) or easily send via e-mail. The logs can also be save in local database in you device for later preview or sharing.

User can start the routine using different initialization modes, i.e. determining threshold without prior knowledge of the threshold value (between 20 and 80) using the default initial value determined in the settings, or by setting an initial value based on a good guess. The latter will reach confidence of 5% of the full working range a bit quicker. User can also restart the procedure or toggle a previous response. It is also possible to set the procedure to work so that only long-clicks can be used to mark a response or non-response to avoid accidental clicks.

Properties:

Determines psychophysical threshold

Based on sequential testing

Algorithm is widely used in determining neurophysiological and psychophysical thresholds

Confidence limits calculated based on log-likelihood or full enumeration/Monte-Carlo simulations

Threshold-finding procedure can be logged and saved to a file

Minimum and maximum number of trials can be set in the settings

Default initial value for the threshold can be set in the settings

Works within 0-100 working range

Displays a graph of the process after threshold has been determined with sufficient confidence

Possible to toggle a response during the process

Easy-to-use

Response time-stamps

Share logs via e-mail or compatible software

Store measurements to local database

Free

No adds

Permissions:

Requires permission to write to en external storage (used only for the log files created upon user request)