Temporary crating (TC) of lactating sows has been proposed as an alternative to permanent crating (PC), which are widely used. Research on these effects has primarily focused on piglet growth and mortality, whereas positive welfare indicators such as affiliative behaviours including social interactions have received less attention. Moreover, existing studies often lack longitudinal perspectives or employ limited methodological approaches. Therefore, the aim of this study was to investigate both short- and long-term effects of PC and TC on affiliative social nosing behaviour as nosing behaviour contributes to social bonding and positive experiences in pigs. The experiment included 12 sows (6 control in PC, 6 experimental in TC). Observations focused on nosing behaviour of the sow and piglets during two periods: (P1) 24 hours after crate opening, and (P2) the days before weaning. Behavioural coding was conducted using the BORIS software, with observations recorded for a total of 120 minutes per sow in each period. This consisted of eight of 15-minute blocks distributed between 12:00 a.m. – 8:00 p.m. using a continuous sampling method. Data were analysed using generalized linear models (PROC GLIMMIX, SAS). All interaction types (piglet-to-sow, sow-to-piglet, piglet-to-piglet) were analysed jointly, then separate datasets were constructed for each interaction type. Nosing interactions were summed across the full 120-minute period for each individual. Preliminary analyses revealed a significant effect of system x period interaction (p = 0.0036) when analysing all interaction types together. Piglet-to-sow nosing showed a significant effect of system x period interaction (p < 0.0001) with similar interaction rates in P1 (TC: 0.6699; PC: 0.6153), but higher interaction rates in TC than PC in P2 (TC: 0.8723; PC: 0.03938). Sow-to-piglet nosing only showed a strong effect of period (p < 0.0001) indicating a decline from P1 to P2 (TC: -0.2008 RIGHTWARDS ARROW -0.8679; PC: -0.1945 RIGHTWARDS ARROW -1.2988). Piglet-to-piglet nosing demonstrated significant effects of system x period interaction (p < 0.0001) with higher interaction rates in PC than TC in P1 (TC: 1.7155; PC: 1.9784), but lower interaction rates in PC vs TC in P2 (TC: 1.7634; PC: 1.2221). The results indicate that temporal changes in behaviour differed between housing systems and periods. In particular, TC appears to support more sustained and piglet-initiated social interactions throughout lactation. Since affiliative social nosing is considered an important form of pig behaviour and may contribute to social cohesion and positive affective states, these findings suggest potential benefits of TC for positive welfare. However, the small number of sows represents a limitation and future research should be conducted on larger sample size.