Abstract Both social interaction and the lack thereof have been discussed as being beneficial for creativity. Strikingly, in both cases there seems to be an implicit assumption that different creative activities are all influenced in the same way. However, the idea that different creative behaviors are all equally influenced by social interaction seems not reasonable (e.g., poem writing vs. singing a song at a family celebration). The reduction in social contacts in Germany during the coronavirus pandemic offered a unique opportunity for a field experiment with high ecological validity. To explore how different creative activities are each influenced by reduced social interaction, 130 participants were asked in an online survey using a proxy pretest design about the frequency of everyday creative activities before and during the core episode of social distancing. The change in frequency for performing a given creative activity depended on the frequency at baseline, but not in a linear manner. Instead, the relationship was u-shaped: creative activities most frequently performed at baseline showed an increase in frequency, creative activities of medium frequency at baseline decreased, and the least frequent activities changed the least. In particular, activities represented by “Personal Environment Creativity” increased in frequency, while activities represented by “Self-Expressive Creativity” decreased in frequency during lockdown. Recalling the four p’s of creativity, it is accepted that the person aspect differs between domains. It is therefore surprising that we commonly assume press-level features work the same way across all domains, which is challenged by the current study.