Dive Deeper: Good Enough Is the Finding
Five studies behind this week’s post — what they found, why they matter, and where the evidence has limits.
This is the research companion to Good Enough Is the Finding. Read the main post first.
Dive Deeper: Good Enough Is the Finding
Post 7 argues that "good enough" parenting is not a consolation prize, it is the empirical finding. Woodhouse showed that secure base provision matters more than global sensitivity. Tronick showed that caregiver-infant pairs are mismatched 70% of the time. These five studies deepen the case that repair, not perfection, is the mechanism that builds secure attachment.
1. 174 Studies, 22,914 Children. Sensitivity Is Moderate, Not Perfect
Madigan, S., Deneault, A.-A., Duschinsky, R., Bakermans-Kranenburg, M. J., Schuengel, C., van IJzendoorn, M. H., Ly, A., Fearon, R. M. P., Eirich, R., & Verhage, M. L. (2024). Maternal and paternal sensitivity: Key determinants of child attachment security examined through meta-analysis. Psychological Bulletin.
DOI: 10.1037/bul0000433
What they found: Across 174 studies and more than 22,000 children, caregiver sensitivity is associated with attachment security at r = .25. That is a real and reliable effect, and it is moderate, not large. It means sensitivity explains about 6% of the variance in attachment security. The rest is explained by other factors: child temperament, family context, the specific quality of the repair, and things we have not yet measured. Notably, paternal sensitivity (r = .21) and maternal sensitivity (r = .26) were statistically equivalent). So, fathers matter as much as mothers.
Why this matters for FCD: This is the empirical backbone of "good enough." If perfect sensitivity produced perfect attachment, the correlation would be much higher. The moderate effect size is not a failure of the science. It is the finding. Sensitivity matters, but it is one input among many. The system is designed to produce security through "good enough" caregiving (i.e., consistent response to distress, repair after rupture) not through optimization.
What it doesn't answer: A correlation of .25 means a lot of variance is unexplained. What other caregiver behaviors, child characteristics, or contextual factors account for the rest is one of the biggest open questions in attachment research. Woodhouse's secure-base finding is one piece. The full picture is still being assembled.
2. Mismatch Is the Norm, Repair Is the Mechanism
Feldman, R. (2007). Parent-infant synchrony and the construction of shared timing; physiological precursors, developmental outcomes, and risk conditions. Journal of Child Psychology and Psychiatry, 48(3-4), 329-354.
DOI: 10.1111/j.1469-7610.2006.01701.x
What they found: Feldman's research documents that parent-infant pairs spend the majority of their interaction time out of sync. Mismatch is not an error. It is the normal state. What predicts healthy development is not the proportion of time spent in synchrony, but the pattern of disruption and recovery, the continuous oscillation between mismatch and repair. This pattern of temporal coordination predicts self-regulation, empathy, symbolic competence, and attachment security across childhood and into adolescence.
Why this matters for FCD: This is the scientific basis for the repair argument in Post 7. If 70% mismatch is the norm even in healthy pairs, then the goal was never sustained attunement. The goal was always repair. Your child is not learning from your perfection. They are learning from the rhythm of rupture and return. That rhythm is the teaching. Don’t berate yourself or your child for the mismatch; you are FAR from alone.
What it doesn't answer: Synchrony research captures moment-to-moment dynamics. How these micro-interactions aggregate into long-term developmental trajectories, and what distinguishes "healthy mismatch with repair" from "problematic mismatch without repair" at a quantitative threshold, is still being defined.
3. Attachment Is Not Fixed by Early Experience
Thompson, R. A. (2000). The legacy of early attachments. Child Development, 71(1), 145-152.
DOI: 10.1111/1467-8624.00153
What they found: Thompson reviewed the evidence on attachment continuity and discontinuity and concluded that early attachment security is not a permanent trait. It is revisable. Children who are insecurely attached in infancy can become securely attached later if caregiving quality improves. Children who are securely attached can become insecure if caregiving deteriorates. Attachment security is not a stamp applied in the first year. It is a living assessment that the child continuously updates based on the relationship they are actually experiencing.
Why this matters for FCD: This dismantles the last piece of the perfectionism trap. Not only is "good enough" sufficient but even "not good enough" in the early months does not lock in a permanent outcome. The system is revisable. Later caregiving matters. Repair works not just within interactions but across developmental time. This is liberation, not a license. The science says both that early experience matters and that later experience can revise it.
What it doesn't answer: The conditions under which attachment revises (how much change in caregiving quality, for how long, at what developmental stage) are not precisely quantified. We know revision happens. We are less certain about the dose required.
4. Enhanced Sensitivity Causes Better Attachment — 25 RCTs
van IJzendoorn, M. H., Schuengel, C., Wang, Q., & Bakermans-Kranenburg, M. J. (2023). Improving parenting, child attachment, and externalizing behaviors: Meta-analysis of the first 25 randomized controlled trials on the effects of Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline. Development and Psychopathology, 33(2), 468-490.
DOI: 10.1017/S0954579421001462
What they found: Across 25 randomized controlled trials involving more than 2,000 families, the Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline (VIPP-SD) improved parenting behavior (d = 0.37) and child attachment security (d = 0.47). The correlation between parenting improvement and attachment improvement was r = .48 , strong evidence for a causal pathway. When you improve caregiver sensitivity through targeted intervention, attachment security improves as a consequence. The effect is not merely correlational. It is causal.
Why this matters for FCD: This closes the loop on the "good enough" argument. Sensitivity matters (Madigan showed r = .25). It is trainable (VIPP-SD shows d = 0.37). And training it improves attachment (d = 0.47). The intervention is brief, typically 6 sessions of video feedback. You do not need to become a different person. You need to become slightly more attuned, slightly more responsive, and the system registers it. "Good enough" is reachable. Try it.
What it doesn't answer: VIPP-SD is effective but has not consistently reduced child externalizing behavior (r = .07). Why sensitivity improvements translate to attachment gains but not to behavioral improvements is an important gap and may reflect the fact that behavior is influenced by many factors beyond the attachment relationship.
5. Predictability, Not Perfection; The Brain's Actual Requirement
Pollak, S. D., & Gunnar, M. R. (2025). What developmental science has to say about caregiving. Daedalus.
DOI: 10.1162/daed_a_02122
What they found: Pollak and Gunnar synthesized decades of caregiving research and identified three features of optimal caregiving: 1) timing (responding when the child needs it); 2) predictability (the child can anticipate that the response will come); and 3) consistency (the response pattern holds over time). The child's brain is a statistical learning engine. It builds its model of the world from the patterns it detects. What it needs is not perfect attunement. It needs a pattern reliable enough to learn from. Predictability is the neurobiological requirement. Perfection is not.
Why this matters for FCD: This reframes "good enough" in neurobiological terms. The child's brain is not recording each interaction as pass/fail. It is extracting a pattern. The pattern it needs is: "When I signal distress, someone responds. Not every time. Not perfectly. But enough that I can predict it." That prediction is what security is built on. Not the quality of any single interaction, the reliability of the pattern.
What it doesn't answer: "Predictability" is a useful organizing concept, but the precise statistical parameters (how much consistency is "enough," how many misses break the pattern) are not yet specified. The concept points in the right direction without providing a specific threshold.
Coming Up
Next week's Dive Deeper accompanies "Your Voice Is a Drug," the evidence that a parent's physical and vocal presence changes a child's stress biology in real time. We look at what oxytocin, cortisol, and the amygdala tell us about why presence is not a metaphor.
Every Functional Child Development post comes with a Dive Deeper companion. Subscribe for the full picture.