Michael Krause

The following tracking task is implemented for mobile phones (Android), therefore we called it Mobile Tracking Task (MTT). The task of the test subject is to balance a blue cross in the middle of the screen (yellow dot). The performance is tracked with Root Mean Square Values (RMSE). It can be seen as a modification of the one dimensional classical Critical Tracking Task (CTT) [cf. 1]. Please note, that some people using different Continious Tracking Tasks and also abbreviate these CTT, which can be confusing. The original Critical Tracking Task is an instability, the Continous Tracking Tasks use for example the superposition of different sinus signals to generate a (pseudo) random behavior that the test subject should track.

The app has a start screen. From this screen you can ‘Start’ the experiment of ‘Configure’ the settings. The experiment is stopped with the ‘screen off’ or ‘home’ button of the phone. The results of the last trial are shown on the start screen. All results (and configurations) are also logged to a folder (‘MobileTrackingTask’). Please note, if you do an experiment and directly connect your phone to a PC afterwards, maybe not all files are visible; some phones need a restart before. During the experiments make sure the phone is completly booted for some minutes and doesn’t run updates etc.

The center (yellow dot) is the origin (0|0). The x/y-axis are scaled to (-100 to +100) each, independend of screen size and aspect ratio. You can teach in a ‘neutral position’ or use ‘flat’ (on table) as neutral position. Deviations from this neutral poition are used to control the blue cross. How much the cross is influenced by these control signals can be adjusted with the ‘sensitivity’. The drifting from the blue cross to the borders (difficulty), can be controlled with ‘instability’.

The task can be switched between one-dimensional/two-dimensional. For devices with strange accelerometer placements and mappings (typically low price tablets), the axis can be swapped and inverted.

Transmitting the visual signal from the phone to a screen (HMI adaptor) can be beneficial. We used the task in student thesises and compared it to an AAM driving track in a driving simulator mockup. It not completly fulfilled the expectations, but there where indications that a config setting of sensitivity:15, instability:20 is probably easier than the driving simulator track. So we don’t make here any statement or recommendations about replacing driving simulations with artifical primary tasks. The application is provided for free, maybe it is usefull in another experiment as primary or secondary task, or you can tailor the code to your needs.