<p>In the current study, we assessed the role of caregiving in testosterone reactivity by examining whether and how testosterone changed in romantic partners as they jointly cared for an “infant simulator,” a lifelike doll designed to mimic many of the behaviors of a real infant. Although testosterone often declines as a function of caring for a child, especially when parents feel able to meet the child’s needs, less is known about joint caregiving with romantic partners. Couples without children (<i>N</i> = 30; 60 participants) engaged in a joint caregiving task in the laboratory before and after taking care of the infant at home, allowing us to examine how testosterone reactivity changed over time and with caregiving experience. We found that, on average, both female and male participants showed significant post-caregiving declines in testosterone following the second, but not the first, lab session. These findings were not mediated by participants’ feelings about or performance during the at-home caregiving. Results were, however, moderated by participants’ comfort with babies and their expectations about parenting: Participants who were less comfortable with babies or had less positive parenting expectations showed larger differences in reactivity across sessions, suggesting that caregiving experience may have contributed to testosterone declines in the second session. Our findings suggest that those with more caregiving experience or who had more positive expectations about parenting were more easily able to lean into nurturance, perhaps leading to testosterone declines. Taken together, our findings reflect the nuanced, situation-dependent influences on testosterone in dyadic contexts.</p>