The Study
How it operated.
The mechanics of the program, as far as they are publicly known: how subjects were identified, how the separation was executed, how the home visits were scheduled, what was measured, and how the resulting data was kept hidden from the people it described.
Subject identification
The study’s subjects were identical twins and triplets surrendered for adoption through Louise Wise Services. The agency, which served Jewish birth mothers and Jewish adoptive parents, was one of a small set of metropolitan adoption agencies handling such placements in the period. Identical multiples are statistically rare in any cohort of adoptions; the study collected its subjects opportunistically, as the agency encountered them, over many years.
Eligibility was established at the agency level. Once the agency had a set of identical siblings to place, the placements were coordinated with the research program. The children entered the study not by anyone’s consent but by virtue of how their adoptions were structured.
The separation protocol
The defining methodological feature of the program was the separate-placement policy. Identical siblings were placed in different adoptive households — deliberately, and without informing the adoptive parents that there was a sibling. The adoptive families came in believing they were taking on a single infant from a single birth mother.
The separations were not random in their structure. Adoptive families were paired across the program in ways that varied at least one notable environmental factor — the public record on this is partial, but in some publicly known cases the families differed in socioeconomic status, household composition, or parenting style. The pairings were not always symmetrical. Triplets were split three ways into homes that, in the publicly known case, differed substantially from one another.
The Disclosed Cover Story
- What families were told
- That their child was part of a routine longitudinal “child development study” conducted by the agency.
- What families were not told
- That the child had an identical twin or triplet placed in another household, and that the “study” was a comparative observation across siblings.
- Who knew the full design
- The senior research staff at the Child Development Center and the supervising staff at Louise Wise Services. Adoptive families, biological mothers, and the children themselves did not.
The home-visit cycle
Once a child was placed, researchers from the Child Development Center conducted recurring home visits. The cadence varied across the years a subject was followed but is generally described, in subjects’ recollections and the available reporting, as approximately annual through early childhood, with additional structured assessments at developmental milestones.
Visits combined informal observation in the home with structured tasks set up in the room. Researchers brought materials with them — toys, picture cards, simple cognitive instruments. Adoptive parents were interviewed about the child’s milestones, behavior, and family life. The child was filmed during portions of each visit. Older children were observed during age-appropriate testing.
What was measured
The program collected a longitudinal record on each child that combined three kinds of data:
- Standardized cognitive and developmental assessment. Age-appropriate intelligence and developmental measures were administered. Subjects who have reviewed redacted portions of their own records describe testing of the kind that was standard for child psychiatric assessment in the period.
- Structured behavioral observation. The child was filmed and observed at scripted tasks chosen to elicit specific reactions — frustration, persistence, problem-solving, attachment behavior. The films were retained.
- Parent interview material. Adoptive parents answered questions about the child’s temperament, milestones, sleep, and behavior, and about the family’s own dynamics. These interviews were transcribed and retained.
The defining feature of the data was not its individual components — those were standard — but its structure. Each subject’s record could be placed alongside the record of an identical sibling whose existence the subject did not know about. That was the dataset.
What the design did not include
The program did not include several features that any contemporary protocol would require. There was no informed consent of any kind. There was no institutional review board oversight in the modern sense; IRBs in their present form did not exist for most of the program’s life. There was no debrief: no provision, ever, for telling the subjects or their adoptive families what they had been part of. The program was not, before the events of 1980 forced exposure, planned to inform its subjects.
That last omission is the methodological feature most worth dwelling on. The decision not to debrief was not a contingency. It was structural. The study was designed to be unknown to its subjects, and the institution that conducted it took the position that the subjects’ ignorance was part of the data — that any contamination from awareness would have compromised what was being observed. From the inside of the design, this was a methodological commitment. From the outside, it is the part that is hardest to forgive.