A multimodal dataset for authoring and editing multimedia content: The MAMEM project

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Rehabilitation Engineering

Background:

• The MAMEM project aims to enable individuals with motor disabilities to create multimedia content using mental commands and gaze.
• Existing human-computer interaction (HCI) frameworks require robust datasets for developing assistive technologies.

Purpose of the Study:

• To present a novel multimodal dataset combining biosignals and eye-tracking for HCI research.
• To support the development of advanced HCI systems for individuals with severe motor impairments.

Main Methods:

• Collected electroencephalography (EEG), electrooculography (EOG) via eye-tracking, galvanic skin response (GSR), and heart rate (HR) signals.
• Acquired data from 34 participants (18 able-bodied, 16 with motor impairments).
• Recorded data during interaction with a specialized web browsing and multimedia manipulation interface, and during imaginary movement tasks.

Main Results:

• A comprehensive dataset integrating diverse biosignals and precise eye-tracking data was successfully compiled.
• The dataset captures neural and physiological responses during complex HCI tasks and imagined actions.

Conclusions:

• The dataset serves as a valuable resource for advancing HCI research, particularly for assistive technologies.
• This work facilitates the development of systems to enhance societal integration for individuals with motor disabilities.