Cindy Meston and Penny Frohlich (2003) investigated how residual physiological arousal from a roller‑coaster ride affects perceptions of attractiveness. Participants at an amusement park either just finished or were about to begin a ride. They then rated the attractiveness and dating desirability of an opposite‑gender target photograph.
Those exiting the ride rated the photographed person as significantly more attractive and more desirable for dating than those entering, but only when riding with a non‑romantic partner. The fear‑induced arousal from the ride could get misattributed to attractiveness when the actual source (the ride) isn’t consciously linked to the arousal.
I put very little stock in that study. It’s a small effect (Fig. 3), there’s hints of p-hacking (one of their p-values is p=0.049, and they seem to have done a lot of tests with no bonferroni correction), and at the end the measurement is a questionnaire but questionnaire data can’t always be taken at face value.
Relevant study:
https://labs.la.utexas.edu/mestonlab/files/2016/05/excitation-transfer.pdf
I put very little stock in that study. It’s a small effect (Fig. 3), there’s hints of p-hacking (one of their p-values is p=0.049, and they seem to have done a lot of tests with no bonferroni correction), and at the end the measurement is a questionnaire but questionnaire data can’t always be taken at face value.