Skip to content

Metrics

in this project, we'll get to record differing parameters.

In order to get an overview, we'll list them in this file as they get decided throughout the research project.

metric type how to calculate/record sensor in-game or data processing
engagement ball flag if a child always shoots the ball immediately away, or is showing no interest at all in the ball VR in-game
brain activity EEG EEG headset EEG in-game start and stop the recording
response read time VR
start of new trial VR
eye tracking of response? EOG

Metrics Transfer

the VR application will be running on the android device. Either we try to send all data wirelessly/via usb cable to the attached computer (Need to open port?)

Or we try to save the data on the VR headset and later transfer it to the server.

I prefer transmitting it via a websocket/MQTT.

POC

Metrics Folder

structure:

  • per child we have a folder
  • in the folder we have:
    • two folders for each block
    • in each block:
      • a metric file for the EOG data
      • a metric file for the EEG data
      • a metric file for the GSR data
      • a response file for the caregiver scorings
      • a setup file for each trial
      • a file from KUL with their questionnaires.

The unity application should be the master, the EEG (python script with LSLMarker stream) and GSR (openframeworks+ LSLMarker stream) are masons.

Decisions

EOG

We put some objects on the Unity_layers#eye. These metrics we store per trial in the metrics file.

Metrics of EOG

Per trial:

  • per contact:
    • when did the child look at the caregiver?
    • for how long?
  • given a minimum time between two consecutive 'looks', how often did the child look at the caregiver?

Per block of trials:

  • do we need some metric? These can be deducted I believe out of the trials