Across cultures, parents intuitively sing to their infants, and experimental work shows that infant-directed singing (IDS) captures attention and regulates affect. Less is known about its cumulative effects in everyday family life or about scalable methods to test these effects in naturalistic settings. We present three randomized controlled trials that examine the long-term impact of IDS, each using daily smartphone-based ecological momentary assessment (EMA) to capture infant mood and caregiver experiences in real time.- Study 1 (worldwide, 2022-23) used a 10-week offset-randomized design with a 4-week music enrichment intervention (N = 110 parent-infant dyads; infant Mage = 3.67 months; 53% female). Families were randomly assigned to receive the simple online music intervention either in Weeks 2-5 (n = 54) or Weeks 7-10 (n = 56).- Study 2 (New Zealand, 2023-December 2025) is a near-direct replication with 92 families (infant Mage = 3.7 months; 54% female) using the same design but an enhanced smartphone-based music intervention (Weeks 2-5, n = 48; Weeks 7-10, n = 44).- Study 3 (U.S., 2025-April 2026) is an eight-month, four-arm RCT testing higher-intensity music exposure over 24 weeks. Families (N = 216; infant Mage = 3.67 months; 49% female) are randomly assigned to one of four conditions: (a) singing (music plus parent-infant interaction), (b) passive music listening (music without parent-infant interaction), (c) book reading (parent-infant interaction without music), or (d) a general control. In the first two RCTs, daily EMA compliance averaged ~72%, demonstrating the feasibility of long-term EMA in infancy. The music intervention reliably increased parents’ frequency of IDS overall (Study 1: β = 0.26, SE = 0.12, p = .02; Study 2: β = 0.04, SE = 0.008, p < .001) and particularly in response to infant fussiness. This increase in IDS yielded a significant group-by-time interaction in infant mood (Study 1: β = 0.18, SE = 0.05, p < .001; Study 2: β = 0.14, SE = 0.06, p = .02), providing robust evidence for a causal effect of IDS on infant affect. No consistent effects emerged for parent mood, stress, or infant arousal. Preliminary analyses of Study 3 indicate similarly high engagement and will test whether longer-term, higher-intensity enrichment produces broader developmental benefits (results will be available at the conference). Collectively, these three studies provide converging experimental evidence that music interventions can increase the use of IDS, with positive causal effects on infant mood. They also show that smartphone-based music interventions and EMA are scalable tools for early childhood music research.