Rory Spiteri

Human in Multi-Factor Authentication Research Project

Design a risk-adaptive MFA decision model that adjusts authentication requirements according to defined behavioural and contextual risk indicators. Furthermore, deploy and test the prototype within a controlled Cyber Range environment and collect relevant authentication and security telemetry using appropriate monitoring tools.

Resarch Multi-Factor Authentication Psychology

This project examines how a risk-adaptive Multi-Factor Authentication (MFA) model implemented in Keycloak affects the balance between usability, accessibility and security, using simulated users in a controlled environment. It begins with a review of at least 20 relevant publications on MFA approaches, human factors in authentication, the usability–security trade-off, MFA fatigue and adaptive authentication, which establishes the research gap and informs the experimental design.

The project will then design a decision model that calculates a risk score for each login attempt based on three indicators: failed attempts, use of a new device, and a change in the login country.

The project makes four main contributions. It provides a controlled Keycloak testbed in which MFA configurations can be tested safely without affecting real users or systems. It defines a transparent risk-adaptive policy model that links contextual indicators to authentication requirements. It offers a synthetic persona and scenario framework for representing differences in user behaviour. Finally, it documents an evaluation procedure that records both security and friction outcomes, so the tests can be repeated and extended.