Shared spaces for active mobility prioritize the safety and comfort of vulnerable road users by segregating them from motorized vehicles. However, the diverse speed regimes of pedestrians and cyclists can lead to encounters that may affect their comfort. In addition, the very perception of comfort may vary across individuals depending on their demographics, and therefore the determinants of comfort and their effects may not be fixed across all individuals. Despite these complexities, there is limited research in understanding the heterogeneous interactions between cyclists and other road users in shared spaces. To bridge this gap, we conducted an intercept survey complemented by an experimental section involving 594 cyclists in Sweden. This study focuses on gaining insights into cyclists' experiences, particularly their comfort levels during 'passing' and 'meeting' events with other road users in shared spaces. We then used the collected data to develop a random effect latent class ordered probit model to scrutinize the determinants of cycling comfort in passing and meeting scenarios. The latent class specification is employed to account for unobserved heterogeneity in the data. Findings reveal that female cyclists generally perceive less comfort compared to their male counterparts in both scenarios. Passing events have a more negative impact on older adults, leading to less comfort compared to younger cyclists. We also found that previous cycling experience increases comfort in shared facilities, particularly for older adults. These results highlight the intricate nature of perceived comfort in interactions, particularly concerning demographic characteristics, contributing to the promotion of user diversity in shared spaces.

Shared spaces for active mobility aim to offer safe and comfortable mobility for vulnerable road users by separating them from motorised vehicles. However, the distinct navigation characteristics of these users may increase the complexity of their interactions. The emergence of e-bikes which are faster and heavier than regular bikes has further increased this complexity. This study aims to shed light on the interdependency of e-bikes and pedestrians behaviours in shared spaces, and investigate how they influence each other's navigation. Through a controlled experiment in Lund, Sweden, data were collected on a total of 1520 trajectories of e-bike and pedestrians, their demographics and cycling experience. A simultaneous equation model was used to quantify the interactions between the participants. Results demonstrate significant correlations among variables, highlighting the model's capacity to effectively capturing the hypothesized interdependencies. The findings can inform the development of level-of-service indices and surrogate safety measures for shared spaces.