Characterizing the temporal changes of biotic communities and disentangling the importance of their driving mechanisms are central themes in ecology and environmental management. Non-native species have multiple adverse impact on native communities and species. However, there is a dearth of information on how non-natives influence the long-term dynamics of native communities. In this study, we compared the role of non-native species alongside various local and regional factors in the long-term dynamics of stream fish assemblages in the catchment area of Lake Balaton, Hungary, the largest lake in Central-Europe. Although, we found no consistent trend in species re-ordering between native and non-native species, native fish assemblages indicated significantly lower temporal stability with increasing relative abundance of co-occurring non-natives. Structural equation modelling revealed that assemblage dynamic patterns were also determined by a habitat degradation gradient. Moreover, habitat degradation and the presence of fishponds increased the relative abundance of non-native species, further affecting native assemblages through indirect interactions. Fish assemblage dynamics showed also some degree of finer-scale spatial structure. Non-metric multidimensional scaling further reinforced the destabilizing effect of non-native species but highlighted that native fish assemblages could generally be described by non-directional gradual or saltatory changes over time, regardless of the relative abundance of non-natives. Our results thus revealed that the studied native fish assemblages may possess some degree of resilience against biological invasions. However, increasing temporal variability induced by non-natives makes native assemblages more vulnerable to environmental stochasticity threatening their long-term persistence. This is especially worrying, since the frequency and intensity of environmental disturbances are expected to rise in the future due to climate change and increasing anthropogenic pressure on aquatic ecosystems.