The adverse eect of ambient noise on humans has been extensively studied in elds like cognitive science, indicating a signicant impact on cognitive performance, behaviour, and emotional state. Surprisingly, the eect of ambient noise has not been studied in the context of mobile interaction. As smartphones are ubiquitous by design, smartphone users are exposed to a wide variety of ambient noises while interacting with their devices. In this paper, we present a structured analysis of the eect of six distinct ambient noise types on typical smartphone usage tasks. The evaluated ambient noise types include variants of music, urban noise and speech. We analyse task completion time and errors, and nd that dierent ambient noises aect users dierently. For example, while speech and urban noise slow down text entry, being exposed to music reduces completion time in target acquisition tasks. Our study contributes to the growing research area on situational impairments, and we compare our results to previous work on the eect of cold-induced situational impairments. Our results can be used to support smartphone users through adaptive interfaces which respond to the ongoing context of the user. CCS Concepts: • Human-centered computing → Empirical studies in HCI; • Human-centered computing → Ubiquitous and mobile computing; • Human-centered computing → Smartphones;