Robust statistical inference for small-sample HCI : Longitudinal case studies from martian analog campaigns

Details

Supervisors

Pircalabelu, Eugen ; Vanderdonckt, Jean

Faculty

Faculté des sciences

Degree label

Master [120] en science des données, orientation statistique, à finalité spécialisée

Abstract

enWhen crewed missions to Mars will move further from Earth, communication delays will require astronauts to rely more on autonomous interactive systems. In the isolated, confined, and extreme (ICE) environments they will live, interface usability becomes an important safety factor. Indeed, wearable gesture interfaces offer a hands-free alternative to touchscreens. But their long-term user experience under realistic operational conditions remains poorly understood. This thesis investigates the longitudinal User Experience (UX) of the Tap Strap 2 and Tap XR gesture interfaces but also drone gesture interaction at the Mars Desert Research Station (n = 8 in 2024; n = 7 in 2025). UX was assessed throughout the missions using UEQ+ forms, investigating adaptation, extravehicular activities and post-mission recovery. Three complementary statistical methods were used to account for the small sample size. Results show that pragmatic usability remained stable despite the simulated Martian conditions and that changes were mainly observed in subjective and affective dimensions. The Tap Strap 2 was consistently perceived as more attractive than the Tap XR and reliability was valued more than novelty. In addition, a significant improvement in affective ratings was also observed one month after the missions. These findings highlight the importance of simple and reliable interfaces and multisensory feedback for future SpaceCHI systems.