Developing Driving Systems for ASD Patients
With Virtual Reality and BCI’s
Imagine this you have just turned sixteen in the state of California and you are off to take your driving test to finally get your license. You are now one step farther away from the boundaries that covered you towards your full independence.
You can now finally drive to that one local coffee shop without asking and waiting for a driver to take you. Feels good right? Well patients with ASD are unfortunately unable to do the same and scientists are trying to research more into how they are then independently to do the same actions.
What is ASD?
Want to read this story later? Save it in Journal.
One in every sixty-eight people has the force pressure of having to figure emotions and try to figure the ability to comfort and share the same emotions when they are mentally unable to. Generally, patients with ASD or Autism Spectrum Disorder tend to misread situations and respond instead with emotions that are not typical within a certain situation. It’s hard for patients with this costly neurodevelopmental disorder to manage social interactions as it’s noticed there are deficits in their social communication skills.
Even when researchers try to study ways for young patients to improve social skills, language development, and emotional recognition many individuals would still tend to fail to achieve the typical rates to relate to adult independence.
The ability to drive in particular ties with the behaviors that relate to daily activities. Patients with this disease face difficulty learning and when leading to the idea of learning how to drive there’s difficulty with identifying general driving hazards.
The Proposal
Growing numbers of studies have been investigated with virtual reality-based interaction with ASD as it provides a potential where an immersive, interactive, and reactive realistic environment could benefit behavioral learning.
It provides a real-world situation that could be introduced to the patient and be figurative to how they may solve it out how they could solve it. Researchers have come up with the idea to make a VR based driving system to address the driving skills of adolescents with ASD.
A proposed goal is to incorporate an EEG-based Brain Computer interface within the VR-based driving system to have a non-invasive method to estimate the levels of engagement, emotional state, and mental workload. This could be used to order and facilitate an individualized adoption for driving skill learning for young ASD patients.
BCI applications have generally been developed as new control and communication for individuals undergoing motor disabilities, but a new emergence of commercial within the Neuromarketing realm. BCI devices like NeuroSky Mindwave and Emotiv EPOC provide the different choices of BCIs in different applications… like video games.
The System’s Description
The system that the researchers are proposing and looking into have five different modules:
- VR Driving Module (VDM)
- Physiological Data Acquisition Module (PDM)
- EEG Data Acquisition Module (EDM)
- Gaze Data Acquisition Module (GDM)
- Observer-based Assessment Module (OAM)
Communication among these modules are based on the users architecture through a local network. The VDM provides a virtual environment where participants could operate a virtual vehicle within the environment using a steering wheel and pedalboard. The environment consists of a driving game that has six difficulty levels. Each difficulty would also provide three assignments which are then composed of eight trials. In example:
- Decreasing speed when near a school zone
- Lane merging
- Left/Right turning
- Stopping at a stop sign
EEG signals, physiology signals, eye gaze data, and performance data are all recorded by the EDM, PDM, GDM, and VDM. Where the OAM records behavioral assessments within the rating scale of 0–9; between engagement, enjoyment, frustration, boredom, and difficulty.
With the individual differences within the game; the difficulty of how the game is perceived rather than the whole difficulty level of the actual game as a whole. This could be measured through the difficulty shown through the mental workload of the patient who is going through the session.
The proposed EEG based BCI for the VR based driving system consists of three modules:
- Signal preprocessing module
- Feature generation module
- Classification mode
BCIs serves as an implicit communication channel to enrich the human to computer interaction. Typically raw EEG signals are collected from the scalp of the participants and then fed to a signal processing module to move outliers and correct EOG and EMG artifacts to enhance the signal to noise ratio.
The feature Generation Module Transform the time series signals into a set of meaning features for a classification module to then detect engagement level, emotional state, and the mental workload of the participant. Or in brief; extract features of time series and reduce the number of features by selecting a subset of useful features to represent a lower-dimensional space.
The classification results are sent to the VR driving module system can be used together with physiology signals, eye gaze data, and the performance data to make decisions on how the adaptation of the virtual driving game could aim to work with the behavioral assessment data.
The Experiment’s Design
So far the researchers had studied sixteen male teens with ASD. (There’s a 5:1 ratio between male and female diagnostics of ASD) Where they ranged from 13 to 18 years old were to undergo six experimental sessions in the range of an hour on seperate days.
At the start of each session the driver simulator was to be adjusted to the comfort of the participant. Then an EEG sensor would be placed as well as physiology sensors were to be placed on the head and body of each participant. Then it is followed by a tracker calibration process.
In the first session out of the six, the participant would then watch a short tutorial explaining basic rules and vehicle controls. Before beginning each task, there would be a three minute collaboration to detect baseline data to figure the average of what EEG signals and physiology signals are going to be read. Then the participants were to sit quietly with the stimulator with their eyes still open to generate a baseline for eye movement gaze.
Each driving part of the session constituted three minutes of practice which would be followed up with a driving assignment. The practice mode is designed to familiarize the participant on how they were to control the vehicle in the game.
There are generally. no restrictions on how the participants react during the game and their driving behaviors. In contrast, the assignment mode is directed by GPS instructions where the participants were to still follow traffic rules.
The rest of the sessions are similar selected assignments from the same difficulty whereas the difficulty would be increased form one visit to the next. In each session a therapist observes the participant on their performance in driving from another room. This provides an online rating at each end of each assignment. Providing a periodic rating during each assignment on participants and general states.
TLDR:
- Autism Spectrum Disorder is a developmental disability that causes significant social, communication, and behavioral challenges
- Researchers has proposed a structure of a EEG based BCI for a VR based driving system for ASD patients
- Currently they are still looking for better classification results; such as time domain, synchronicity features and examples of time domain features
