Recurrent visitations reveal selectivity beyond the 15-Minute City vision
In the transition towards sustainable and equitable urban living, proximity-centred planning has been adopted in cities worldwide. Exemplified by the 15-Minute City (15mC), this planning paradigm often assumes that local amenity provision translates into local use, yet behavioural evidence on recurrent visitation remains limited. To address this gap, we introduce K-Visitation, a scalable, behaviourally informed framework for comparing recurrent visitation with proximity-based expectations, using 18 months of mobile phone data from 720,000 users in Finland. For each individual, K-Visitation compares recurrent destinations ($K_{freq}$) with nearest options ($K_{dist}$), and evaluates travel time differences between them and their neighbourhood proximity baseline. Across Finnish cities, individuals repeatedly bypassed nearby visited options that already exposed them to required daily amenity categories, incurring measurable travel time costs beyond what local opportunities could support. The sharpest divergence appears in amenity-rich urban cores: where the physical conditions for 15-minute living are strongest, recurrent destinations depart most from nearest options. A density-aware null model and destination-level classifier show that this divergence reflects selective behaviour: non-nearest recurrent destinations are not explained by amenity density alone and are often amenity-rich but functionally specialised. Amenity-specific analysis further reveals a hierarchy in which routine anchors adhere more closely to proximity, while specialised and infrastructural functions depend on wider catchments and connectivity. Our findings show that proximity is a necessary but insufficient condition for local living. The 15mC should therefore secure routine local access while using behavioural evidence to identify where proximity is bypassed and where wider connectivity remains necessary.