A new platform to test digital attacks in mixed reality training

CAESAR is a multi-user framework for evaluating cognitive attacks in search-and-rescue tasks.

CAESAR is an end-to-end multi-user Mixed Reality (MR) framework. It is used for the systematic evaluation of cognitive attacks in collaborative search-and-rescue training.

Mixed Reality (MR) enables immersive training by coupling virtual content with the physical environment. It supports embodied interaction.

Cognitive attacks can subtly distort visual feedback, virtual content, or system synchronization. These attacks affect users who rely on virtual elements for perception, coordination, and decision-making.

The CAESAR framework features a modular architecture. It integrates four classes of cognitive attacks within a synchronized collaborative scenario.

A multimodal data collection pipeline captures behavioral, performance, and subjective measures. This happens under adversarial conditions.

A controlled user study analyzed a latency-based attack. This study demonstrated measurable effects on task performance and user behavior.

Key facts

  • CAESAR integrates four classes of cognitive attacks.
  • The framework uses a multimodal data collection pipeline.
  • A latency-based attack showed measurable effects on user behavior and task performance.

All sections

Edition