The present study examined testosterone (T) and cortisol (Cort) in fathers engaged with caregiving. We collected saliva samples in the mornings and evenings of two consecutive days in 150 fathers of 1‐ to 5‐year‐old children. Fathers completed questionnaires on socioeconomic status, family structure and life, sleep characteristics and body mass index (BMI), and reported on their engagement in childcare. Fathers used smartphone‐based experience sampling throughout 1 week to sample ongoing activities with their children, including times of supervision, joint play, rough‐and‐tumble play, and cuddling episodes. External observers rated father–child attachment during a home visit.
We began by testing for widely characterized covariates of T and excluded seasonal variations and known predictors associated with lowered T, such as older fathers and those with multiple and young children, lower BMI, shorter sleep duration, and sexual activity before sampling. Most interestingly, however, fathers’ engagement in childcare and attachment to the child appeared more pronounced the greater the diurnal decline in T. Cuddling predicted a similar negative association, whereas joint play and rough‐and‐tumble play (RTP) showed enhancing effects on declining T. Interestingly, all fathering behaviors (except RTP) were positively related to lower Cort. In contrast, supervision was ineffective on both Cort and T.