Conspiracy theorists less likely to believe in misinformation if they experience rewarding romantic connections

Not really, all of the participants were in long-term heterosexual relationships.

Relationships averaged 13.3 (SD=10.2) years in length (130 dating, 36 engaged, 389 married).

They did not investigate how often these couples slept together. However, they do state the following:

Participants in the experimental condition did not report significantly greater trust in their romantic partner, as compared to control participants, likely because participants were already, unsurprisingly, highly trusting of their romantic partners.

So I don't think it is fair to assume that these people were not getting laid.

As for 'crazy people', the paper does not call anyone crazy. They also do not apply an absolute negative connotation to the word, they state:

Like moths to flames, conspiracy theorists gravitate to any whisper of information that fuels their suspicions and distrust. While such inclinations might be advantageous when the government is actually behaving duplicitously, they are less advantageous in the context of a global health pandemic where resistance to public health advice directly compounds death rates (Robertson, 2021; VoPham et al., 2020).