Mar Gonzalez-Franco

Mar Gonzalez-Franco is a computer scientist and neuroscientist at Google, where she leads the Blended Intelligence Research and Devices (BIRD) lab. Her research combines virtual reality (VR), augmented reality (AR), computer graphics and haptics with the study of human perception and behavior. Her work includes avatars, virtual embodiment, interaction devices and interfaces that combine different forms of user input.

Reviewed 21 October 2026. Official professional history, avatar research and questionnaire versions, haptic work and original award record About review dates.

Education and research career

Gonzalez-Franco studied computer science and completed a doctorate in neuroscience and virtual reality at the University of Barcelona in 2014. Before her work at Microsoft Research, she held research positions at University College London, the Massachusetts Institute of Technology, Tsinghua University and Airbus Group. Microsoft's biography for its July 2017 Faculty Summit identifies her as a researcher at the company and describes her work across immersive technology and human perception. Her current Google biography identifies BIRD as her lab and lists immersive interaction, avatars, haptics and artificial intelligence among her research interests.

Measuring avatar embodiment

With Tabitha C. Peck, Gonzalez-Franco developed questionnaires for studying virtual embodiment, the experience of a virtual body or body part as one's own. Their 2018 review proposed 25 questions drawn from earlier experiments, organized around ownership, agency and motor control, touch, body location, appearance and responses to external events. The purpose was to make embodiment measurements more comparable between experiments, rather than assume that every participant experiences an avatar in the same way. The authors presented this as a proposal requiring further validation.

Their follow-up, published in February 2021, analyzed 443 completed questionnaires from nine experiments using all or part of the earlier instrument. It reduced the questionnaire to 16 questions and four interrelated scales: appearance, response, ownership and multisensory experience. The revision incorporated agency and location questions into the other scales; it did not establish that these experiences were unimportant. The authors also identified questions for further investigation, including how the refined questionnaire relates to physiological measurements. These are subjective research measures, with the revised structure derived from the collected studies.

Avatar resources and animation

Gonzalez-Franco was first author of the 2020 paper accompanying the research release of the Microsoft Rocketbox avatar library. It described rigged human models, whose internal skeletons support animation, and explained their use in embodiment experiments, social interaction and crowd simulation. The paper also documented practical limitations of that release: clothing was part of the model mesh, facial animation used bones rather than supplied blendshapes, and extreme changes to body proportions could distort the model. The library provided reusable characters rather than automatically generated likenesses of individual participants. The repository subsequently recorded additions of facial blendshapes in 2022.

She was also first author of MoveBox: Democratizing MoCap for the Microsoft Rocketbox Avatar Library, presented at IEEE AIVR in 2020. MoveBox maps motion capture data from a single Kinect V2 or Azure Kinect depth camera onto avatars in Unity. Its original implementation included recording and playback, animation looping, resizing, simple lip synchronization and hand animation. A separate route recovered motion from existing RGB video offline. The paper reported limitations from occlusion and skeletal tracking, and described the system as producing lower-quality motion than high-end capture equipment.

Haptic and multimodal interaction

Gonzalez-Franco coauthored the 2020 UIST paper on Haptic PIVOT, a wrist-worn research device. Its motorized joint brings a handle into the hand for interaction with a virtual object and retracts it to leave the palm free. While the handle is held, the mechanism can apply forces between the forearm and hand. This approach combines a wearable mechanism with physical contact supplied when needed.

Her later work includes the coauthored IEEE VR 2025 paper EmBARDiment: an Embodied AI Agent for Productivity in XR. Its prototype combined spoken requests with eye tracking to select text from virtual windows as context for an AI assistant. Implemented in Unity on a Quest Pro, it used text recognition and a short contextual memory alongside language-model and speech services. The design assumed that recently viewed text was relevant to the user's question. The authors identified eye-tracking quality and situations where useful information lies outside the user's visual attention as limitations requiring further work.

Recognition

Gonzalez-Franco received the 2022 IEEE VGTC Virtual Reality Significant New Researcher Award.