The weekend effect hypothesis predicts that weekly cycles of human activity impact animal behaviour and physiology. This hypothesis has been supported in the context of recreational activity in natural environments but it is unknown whether it also applies to urban animals. We tested this hypothesis by comparing the sentinel (territorial vigilance), foraging and vocal behaviours of an urban-dwelling bird, the Rufous Hornero Furnarius rufus, between weekdays and weekends (and holidays) within a university campus in central Brazil. The level of human activity (noise, traffic and pedestrian flow) increases greatly on weekdays on this campus. We predicted that the birds would perceive a greater predation risk and would need to adjust their acoustic signals in response to anthropogenic noise on weekdays. Thus, we expected that the birds would spend more time as sentinels and less time foraging, and would sing for longer periods and at a higher pitch on weekdays than on weekends. We also expected weaker duet responsiveness (answering partner-initiated song) on noisier weekdays than on weekends, assuming that noise would disrupt signal transmission between partners. We found that birds spent slightly more time (~4%) in sentinel behaviour and less time (~2-5%) foraging on weekdays than on weekends, but these effects were small and not statistically significant. Birds were equally likely to sing solos, start duets and answer partner-initiated duets on weekends and weekdays. Finally, phrase duration and acoustic parameters of duets were similar on weekends and weekdays. Our results provide little support for the weekend effect hypothesis, suggesting that these urban-dwelling birds may be habituated or indifferent to periodic variation in human activity levels.