Research figure: John Bowlby · Written by Artificial Sapiens
John Bowlby's Forty-Four Juvenile Thieves Study was a historical clinical investigation of children referred for stealing to a London child guidance clinic. Published in two parts in 1944 and issued as a short book in 1946, it compared 44 children with histories of stealing against 44 other clinic patients. Bowlby reported a striking association between prolonged early separations and a subgroup he labeled “affectionless.” The investigation helped redirect child psychiatry toward real experiences of loss and caregiving disruption. Its design, however, could establish clinical patterns and generate hypotheses rather than demonstrate that separation caused delinquency or a fixed personality disorder. Bowlby, 1944; Duschinsky, 2020; Follan and Minnis, 2010.
Why Bowlby Undertook the Study
The study belongs to the formative years of John Bowlby's career, when child guidance psychiatry, psychoanalytic case interpretation, social work, and attempts at systematic clinical comparison occupied the same intellectual terrain. Bowlby had become concerned that descriptions of children's stealing or emotional detachment were inadequate without an inquiry into their caregiving histories, separations, placements, and the quality of family relationships. He investigated the circumstances in which symptoms developed rather than treating the visible offense as a sufficient explanation of the child. Wellcome Collection, London Child Guidance Clinic archive; Duschinsky, 2020.
His attention to the topic predated the mature theory of attachment. In this early research he worked within the language of character, psychoanalytic development, emotional deprivation, and juvenile delinquency. The later concepts of attachment behavioral systems, internal working models, secure bases, and organized patterns of infant attachment should therefore be read as subsequent theoretical developments rather than variables already measured by the Forty-Four Juvenile Thieves investigation. The study is historically important precisely because its preliminary observations became part of a longer process of theoretical reconstruction. Bretherton, 1992; Duschinsky, 2020.
Bowlby's clinical setting also shaped the question. The London Child Guidance Clinic encountered children referred by schools and other agencies; children whose problems reached such services were not a random cross-section of the population. The social consequences of theft, differences in family resources, the response of schools, and decisions to refer a child into psychiatric care all shaped which cases became available for study. Recognition of this selection process is fundamental to understanding what his comparison can and cannot show.
Original Publications and Historical Chronology
The original research appeared as “Forty-four juvenile thieves: Their characters and home-life” in the International Journal of Psycho-Analysis, volume 25 (1944), in two installments, commonly cited at pages 19–53 and 107–128. It was subsequently published as a stand-alone 56-page volume by Baillière, Tindall and Cox in London in 1946, with later reprints. Bibliographic records confirm that the book republishes work first issued in the 1944 journal. These are versions of the same research, not two independent replications. Bibliothèque Sigmund Freud catalog; Open Library, 1946 edition; WorldCat, 1946 edition.
An earlier presentation on habitual pilfering in 1937, preserved in the Bowlby archival material, helps place the investigation within his continuing research on stealing and emotional life. The clinic records and Bowlby's later reminiscences indicate that experiences of children separated during hospitalization and other disruptions sharpened his attention to the phenomenon. His 1951 World Health Organization report on maternal care belongs to a later phase of the intellectual history and must not be substituted for the actual data of the 1944 paper. Duschinsky, 2020; Wellcome Collection archive.
Research Question and Study Design
The question behind the comparison
Bowlby asked whether recurrent stealing among children arriving at a child guidance clinic was associated with characteristic emotional difficulties and particular early experiences. He was interested in whether a recognizable clinical subgroup shared histories of disrupted affectional relationships. His investigation combined a comparison of groups with detailed histories of individual cases, an approach better described as a retrospective clinical case series with a clinic comparison group than as an experiment. The data concerned children who had already developed problems and had already been selected for psychiatric assessment.
The 44 referred children
The principal group consisted of 44 consecutive cases referred after stealing. This consecutive-case feature reduced some discretion in selecting striking examples within the clinic referral stream, but the stream itself was selective. The group encompassed varied patterns of behavior and varied circumstances. It would be inaccurate to depict every referred child as having the same psychological condition, the same family history, or persistent criminal involvement. In contemporary language, an act or history of stealing is a behavior, not a stand-alone psychological diagnosis. Duschinsky, 2020.
The 44 comparison cases
Bowlby also examined 44 children attending the clinic for problems other than stealing. The comparison group provided a reference against which observed histories and clinical character descriptions could be considered. Both groups were clinic-based: even an apparently striking difference cannot be interpreted as a difference between all children who steal and all children in the wider community. The comparison did not eliminate differences in the original routes to referral, family context, traumatic experiences, or clinicians' awareness of the reason for attendance. Follan and Minnis, 2010.
Sources of information
The inquiry drew on clinical case histories, interviews and reports about home life, psychiatric assessment, and accounts of significant separations. These sources gave the study the depth typical of early child guidance research. They also exposed the investigation to imperfect recall, gaps in records, retrospective reconstruction, and interpretation by clinicians who knew the presenting problems. Independent prospective measurement of caregiving quality, repeated blinded ratings, and random assignment were absent. No child was experimentally assigned to separation, and ethical research would not impose such adversity to test a hypothesis.
What Bowlby Reported
The 'affectionless' subgroup
Fourteen of the 44 children referred for stealing were placed in a category Bowlby described as an “affectionless” character type. In his historical classification this designation concerned conspicuous emotional remoteness or restricted concern for close relationships and social obligations. It reflected clinical judgments of his period and is best retained only with explicit historical attribution. The designation is neither a validated contemporary diagnostic category nor evidence that the children were incapable of affection for life. Duschinsky, 2020; Follan and Minnis, 2010.
The figure of 14 is often stated as approximately 32% of the group referred for stealing. It describes Bowlby's judgments within this particular clinical series; it is not the prevalence of a disorder in young offenders, children separated from caregivers, or the population at large. His comparative design was concerned with how this subgroup differed from other referred children, and the analysis depended heavily on the credibility and consistency of that grouping.
Early separations
Twelve of the 14 children assigned to the “affectionless” group had histories of prolonged separation from their primary caregivers early in life. The proportion is about 86%. Later historical research summarizes early separation among the 44 comparison cases at roughly 10%. The comparison attracted attention because it suggested a clinically meaningful association between certain childhood experiences and particular problems in emotional relationships. Duschinsky, 2020; Follan and Minnis, 2010.
These percentages require their denominators. The 86% figure is 12 of 14 members of Bowlby's selected subgroup, not 86% of all 44 children who stole, and the 10% comparison concerns other clinic cases rather than a population-based sample. A high proportion inside a category defined after assessment does not by itself measure the future risk associated with separation. It also cannot identify whether separation, instability in the caregiving environment, maltreatment, or other correlated experiences best account for the observed pattern.
Why the findings were psychologically provocative
The clinical association gave Bowlby a way to formulate the importance of actual social events in a child's developmental history. His interpretation invited clinicians to ask whether relationships marked by continuity, trust, and responsive care might influence emotional development and behavior. In the psychoanalytic setting of the 1940s this attention to documented childhood experiences helped create a research agenda around separation, placement changes, and protective caregiving arrangements. Bretherton, 1992; Duschinsky, 2020.
How the Original Interpretation Worked
Bowlby interpreted the separation histories as a possible developmental explanation of emotional remoteness and, in some children, stealing. His reasoning linked disruption of close early relationships with later limitations in affection, empathy, and the capacity to participate in enduring interpersonal obligations. These claims should be identified as his clinical etiological hypotheses. The observed association did not establish an unbroken causal chain from early separation to a particular character structure and then to unlawful behavior.
The central issue is the exposure being named. Separation means a period during which a child and familiar caregiver are physically apart. Deprivation concerns an insufficiency in the qualities or availability of care needed for development. Instability can involve repeated disruptions of placements or caregiving relationships. Maltreatment includes abuse and neglect. These experiences may overlap in an individual history, but they have distinct mechanisms and implications. A child cared for safely and responsively by another familiar adult during a temporary absence is in a different situation from a child experiencing harsh treatment, chronic neglect, or repeated placement breakdowns.
The historical term “maternal” reflects the vocabulary and social assumptions of Bowlby's early writing. Modern attachment research examines relationships with primary and additional caregivers, including fathers and other committed adults. The child’s opportunity for reliable and responsive relationships is more relevant to present scientific interpretation than a categorical rule requiring uninterrupted physical presence of one biological mother. This distinction also helps explain why later readers have critically revisited the inference from hospitalization or placement to a presumed absence of meaningful care. Duschinsky, 2020; Fox, Nelson, and Zeanah, 2017.
Reading the Study's Numbers Correctly
The study is sometimes summarized as a simple proof that 44 children stole because they had been separated from their mothers. This compresses several steps that the data cannot justify. Forty-four was the size of the referral group. Fourteen was the subgroup assigned a particular historical character description. Twelve was the number within that subgroup reported to have experienced prolonged early separation. The comparison group contributed another 44 clinic cases. None of those counts is an experimental estimate of the effect of separation on later offending.
In a retrospective clinical study, investigators begin with an existing outcome, review reported histories, and compare exposures across cases or groups. This can reveal patterns worth testing. A causal estimate requires a much stronger basis for deciding what would have happened under a different childhood environment. The study did not establish temporal trajectories through prospective observation, did not independently balance potential confounders, and did not provide a defensible population sampling frame for claims about children in general.
The distinction matters for clinical ethics. Descriptions that predict a child's enduring character or criminality from separation history would overstate Bowlby's data and risk stigmatizing people who were adopted, hospitalized, fostered, or cared for by multiple adults. A historical association within a selected clinic cohort is a reason to investigate a possible developmental vulnerability, not a destiny assigned to individual children.
Methodological Strengths
Detailed attention to developmental history
Bowlby's work represented an effort to put concrete childhood circumstances at the center of clinical formulation. The case material recognized that children sharing the behavior of stealing could have very different histories and emotional presentations. Historical records of caregiving disruption, when carefully interpreted, can generate hypotheses that narrow symptom-only approaches would overlook. His work also documented a comparison group rather than resting exclusively on a series of memorable illustrations.
A bridge between clinical practice and research questions
The repeated appearance of certain experiences suggested questions that could be studied more rigorously: Which types of disruption matter? At what ages? Under what caregiving conditions? How do abuse, neglect, separation, and the stability of replacement care interact? Which developmental outcomes are affected, and which children remain resilient? Asking such questions was an important contribution even though this particular investigation could not settle them.
Historical value of the archived case material
The surviving archive and later scholarship make possible a reconstruction of what Bowlby measured and omitted. The Wellcome Collection catalog locates documentation on child guidance and delinquency within Bowlby's professional papers. Archival work can trace changing questions and interpretations across decades; it also helps historians distinguish what the 1944 publication claimed from the more systematic attachment theory developed later. Wellcome Collection archive; Duschinsky, 2020.
Principal Methodological Limitations
Selection and referral bias
A child referred to a psychiatric clinic after stealing differs in multiple possible respects from a child whose stealing never becomes a clinical referral. Referral thresholds may reflect adults' concerns, school practices, family stress, and local institutional resources. Both the focal group and the comparison group were drawn from people who reached a clinic, limiting their representativeness. Comparing two clinical groups is valuable for clinical pattern recognition; it cannot supply population prevalence or population risk by itself.
Retrospective information and recall
Caregiving histories may be incompletely documented or reconstructed from memory after a child has developed difficult behavior. Parents, other informants, and clinicians may recall or interpret the past differently when later symptoms are known. These threats can create apparent links or obscure genuine links. More reliable exposure assessment requires records compiled before the outcome or prospectively gathered observations from multiple informants.
Diagnostic circularity and observer expectations
The “affectionless” category involved clinical interpretation, and the investigator knew the reason for referral. A child placed in a subgroup on the basis of behavior that was already judged unusual might receive greater scrutiny for adverse experiences than another child. The research reported no modern standardized blinded diagnostic protocol for the category. Its descriptive labels therefore cannot be handled as interchangeable with contemporary validated instruments.
Confounding by maltreatment and home adversity
Caregiver separation can coexist with poverty, serious illness, family conflict, neglect, physical abuse, institutional living, and changes in placement. These circumstances may affect development through several pathways. If histories of adversity are not adequately measured and distinguished, an apparent effect attributed to separation may partly reflect harm occurring before, during, or after it. In his historical reassessment, Robbie Duschinsky identified evidence of maltreatment in a substantial proportion of the case material and described Bowlby's later acknowledgment that the original paper underrepresented abuse. Duschinsky, 2020.
Direction of explanation
A difficult social environment might contribute both to placement disruption and later behavior. A child's existing behavioral or developmental difficulties might also influence care decisions and make separations more likely. The same experiences may involve reciprocal processes over time. Because the study started from clinical cases and reconstructed earlier circumstances, it could not confidently separate all such explanations. Its correlations should therefore remain descriptive unless supported by independent designs.
Historical social and diagnostic assumptions
The early twentieth-century classifications of “psychopathy,” character, and maternal role were created in a social and clinical context far removed from contemporary diagnostic practice. Their language also carries a risk of moral judgment. The study should be analyzed as an original historical document, with respect for the children whose lives were interpreted by adults and without imposing the category as a present-day clinical fact.
The 2010 Reappraisal by Follan and Minnis
Michael Follan and Helen Minnis returned to Bowlby's clinical material in a peer-reviewed analysis published in Child: Care, Health and Development. They examined the 14 historically designated “affectionless” cases against the other 74 children in Bowlby's combined series and compared their characteristics with a contemporary clinical sample relevant to attachment disorders. Their reconsideration retained the marked association with separation but questioned whether separation was the best causal interpretation of the cases. Follan and Minnis, 2010.
The authors reported that 86% of Bowlby's subgroup had experienced prolonged early separation and multiple care placements, compared with about 10% of the other clinical cases. They also emphasized maltreatment, neglect, and the cumulative complexity of adversity. Their interpretation proposed that experiences of maltreatment could have been a central factor in both historical and modern clinical presentations, and that developmental outcomes arise from interacting environmental and other vulnerabilities. This was a reinterpretation and comparison of case material, not a new prospective experimental replication of Bowlby's original claim. Follan and Minnis, 2010.
The reappraisal illustrates a productive feature of scientific history: a carefully preserved case series can remain informative even when its original causal conclusion is judged too narrow. Later researchers can recode evidence, examine previously neglected variables, and ask what alternative mechanisms fit the histories. These retrospective reconstructions have limits of their own, because archives seldom contain all the measures a modern investigator would wish to collect.
Does the Study Establish Reactive Attachment Disorder?
No. The Forty-Four Juvenile Thieves investigation preceded modern diagnostic definitions of reactive attachment disorder, or RAD, by decades. Bowlby's historical “affectionless” descriptions can be compared with certain contemporary phenomena, but retrospective similarity is not a valid clinical diagnosis. RAD concerns a specific pattern of substantially inhibited or absent attachment behavior associated with markedly insufficient care, assessed under present developmental and diagnostic standards. Disinhibited social engagement disorder concerns a distinct pattern of socially disinhibited behavior. Neither can be diagnosed from the fact that a child stole or experienced a caregiver separation. Zeanah and Gleason, 2015.
The later Follan and Minnis comparison is best understood as an exploration of historical resemblance and etiological questions. It does not demonstrate that all 14 children would meet a current RAD definition, and it provides no reason to convert an older character label directly into a contemporary psychiatric category. Modern diagnosis requires developmentally appropriate assessment, evidence of the requisite clinical pattern, and careful consideration of the child's caregiving history and alternative explanations. Follan and Minnis, 2010; Zeanah and Gleason, 2015.
A further distinction concerns infant attachment classifications. Secure, avoidant, resistant, and disorganized classifications refer to patterns observed in specified research procedures and developed through later attachment scholarship, notably Mary Ainsworth and subsequent investigators. They are neither synonymous with Bowlby's “affectionless” category nor themselves diagnoses of psychiatric disorder. Their scientific histories and methods belong to related but independent research literatures. Bretherton, 1992.
What Later Developmental Research Actually Supports
Contemporary research supports the developmental importance of consistent, responsive, and protective caregiving and shows that severe psychosocial deprivation can affect social relationships and mental health. The Bucharest Early Intervention Project, for example, studied children raised in institutions and included a randomized comparison of foster-family intervention and care as usual. This design offered substantially stronger evidence about the effects of changing care environments than Bowlby's retrospective clinic study. Yet its findings concern severe institutional deprivation and a particular intervention context; they should not be generalized to every brief or necessary caregiver absence. Fox, Nelson, and Zeanah, 2017; Wade and colleagues, 2022.
A multilevel analysis of that intervention followed 136 previously institutionalized children and reported improved cognitive, physical, and psychiatric outcomes for the randomized foster-care group compared with care as usual. The interpretation is that opportunities for nurturing family care can alter developmental pathways following severe early deprivation. It does not demonstrate that all adverse outcomes arise through one mechanism, nor that replacing a caregiver temporarily is itself a comparable exposure. Bucharest Early Intervention Project analysis.
Quantitative research on caregiver sensitivity and attachment provides another relevant perspective. A large 2024 meta-analysis reported a moderate positive association between caregiver sensitivity and children's attachment security across maternal and paternal samples. Such associations support the role of caregiving quality while leaving room for individual variability, measurement issues, contextual influences, and additional mechanisms. They cannot be retroactively treated as proof of Bowlby's specific claim that early separation caused affectionlessness and theft. Madigan and colleagues, 2024.
Michael Rutter's historical reassessment of maternal deprivation also helped shift the discussion from singular experiences of separation to more differentiated developmental pathways, relational processes, individual differences, and resilience. Researchers must ask what form of care a child received, what adversity occurred, what protective relationships remained available, and which outcomes were measured. These distinctions are now central to defensible interpretation of early adversity. Rutter, 1979.
From the Thieves Study to Attachment Theory
The thieves paper was an early point in Bowlby's movement toward a general account of affectional relationships. It placed childhood separation and the reality of care experiences within his clinical explanatory framework. Subsequent observations of hospitalized children, collaboration with James Robertson, his work for the World Health Organization, dialogue with ethology, and the research of Mary Ainsworth contributed to a larger theory concerned with protection-seeking and the regulation of proximity to caregivers. These were later theoretical and empirical developments with their own sources, concepts, and limits. Bretherton, 1992; Duschinsky, 2020.
Mary Ainsworth's systematic studies of caregiving and infant behavior were a distinct empirical contribution, rather than an operationalization already contained in the 1944 case series. Likewise, James Robertson's observation of children in hospital settings and the Robertsons' work on separation form a separate observational program. The intellectual relationship between these researchers matters because accounts of the history sometimes merge different decades of work into an oversimplified founding event.
A historically precise sequence recognizes an initial clinical hypothesis, a broadening concern with caregiver availability and separation, the emergence of an ethological theory of attachment, and increasingly systematic studies of early relationships. The mature model of an attachment behavioral system was articulated much later than the Forty-Four Juvenile Thieves investigation. The 1944 paper supplies historical motivation and observational material, rather than a modern experimental validation of the entire theory.
Implications for Child Welfare and Clinical Reasoning
One enduring practical lesson is to examine the context of a child's behavior: patterns of caregiving, changes in placements, neglect, abuse, disruptions of contact, and the availability of trusted adults can matter profoundly. Such inquiry is an ordinary part of developmentally informed clinical assessment. A formulation should distinguish a description of behavior from the causal explanation eventually supported by multiple sources of evidence.
When children require hospital care, temporary alternative care, or protection from unsafe conditions, the aim is to preserve relationships and supply responsive, stable caregiving while addressing the reasons for separation. Preventing harmful disruption is a valuable child welfare goal, but forcing a child to remain in a dangerous environment in the name of continuity would confuse physical proximity with protective care. Modern evidence on family-based intervention after institutional deprivation demonstrates that improving care can support recovery. Fox, Nelson, and Zeanah, 2017; Wade and colleagues, 2022.
A child who steals, appears emotionally distant, has experienced placement changes, or lives with a single parent must be assessed on their own circumstances. The original paper provides no individual diagnostic rule and no evidence for blaming mothers, foster caregivers, adopters, or children. Clinical concern depends on distress, impairment, development, multiple sources of information, observed relationships, and the possibility of alternative explanations. Contemporary attachment disorders require their own evaluation; ordinary insecure attachment, separation-related distress, and disruptive behavior do not automatically meet such criteria. Zeanah and Gleason, 2015.
Scientific Legacy and Present Evidentiary Status
The central historical contribution of Forty-Four Juvenile Thieves is the demonstration of a research problem that deserved sustained investigation: how early care, relationship disruption, and adversity become entangled with emotional development and difficult behavior. The study supplied richly described clinic material and an influential hypothesis. Its most familiar numerical finding concerned a selected subgroup rather than the entire group of young people referred for theft.
The separation association within the original cases is established as a reported historical observation. The stronger explanation that separation was the specific cause of a distinctive, enduring character defect is contested and underdetermined by that design. The later recognition of abuse, neglect, placement instability, and relational quality requires a broader model. Modern intervention and longitudinal research makes a compelling case for supportive caregiving, while leaving individual outcomes probabilistic and context-dependent. Duschinsky, 2020; Follan and Minnis, 2010; Rutter, 1979.
The proper place of the study in the history of psychology is therefore as a formative and revisable scientific investigation. Its value includes the questions it raised, the evidence later researchers reexamined, and the methodological lessons supplied by the gap between clinical description and causal proof. That gap helped shape the research agenda that eventually moved beyond the original clinic and beyond the historical assumptions surrounding the word “maternal.”
Questions About the Forty-Four Juvenile Thieves Study
What was the main aim of Bowlby's investigation?
Bowlby examined whether children referred for stealing had distinctive emotional characteristics and early home experiences, particularly separations from caregivers. The investigation combined clinical descriptions with comparison against children referred for other difficulties. It raised an etiological hypothesis that subsequent developmental research needed to test with stronger designs.
How many children were studied?
The main referred group contained 44 children with histories of stealing and the comparison group contained 44 other clinic patients, for a combined set of 88 cases. Bowlby's well-known 86% figure applies to 12 of 14 children assigned to one historical character subgroup, not to 86% of everyone studied.
What did “affectionless psychopathy” mean in 1944?
It was Bowlby's historical clinical description of a subgroup he considered conspicuously emotionally remote. The term does not identify a validated modern diagnosis. It should not be used to label current children or to claim that an individual is permanently incapable of affection.
Did Bowlby prove that maternal separation causes stealing?
No. The study reported an association between early prolonged separation and one selected clinical subgroup. Retrospective data, confounding by other adversities, clinic selection, and nonblinded character assessment prevent a causal inference about separation and later theft.
Is the study still relevant to modern attachment research?
Yes, as a primary historical source and a demonstration of how clinical patterns can lead to a wider research program. Contemporary work distinguishes disrupted care, deprivation, abuse, caregiver sensitivity, and clinical attachment disorders with methods and evidence that emerged long after 1944.
Related Articles
John Bowlby — A comprehensive scientific biography and history of Bowlby's major works, collaborators, and the development of attachment theory.
Attachment in Infancy and Early Childhood: Security and Development — An independent overview of empirical attachment development, caregiving, and early relationships.
Is Attachment Theory Evidence-Based? What Research Supports, What It Does Not, and Common Misuses — A contemporary evidence appraisal, distinct from the 1944 historical study.
John Bowlby's Maternal Deprivation Hypothesis and WHO Report — The 1951 WHO report and the scientific reassessment of deprivation.
John Bowlby's Influence on Child Welfare and Institutional Care — The historical impact on child welfare and institutional practices.
References
Bowlby, J. (1944). Forty-four juvenile thieves: Their characters and home-life. International Journal of Psycho-Analysis, 25, 19–53, 107–128. Historical journal record.
Bowlby, J. (1946). Forty-four juvenile thieves: Their characters and home-life. Baillière, Tindall and Cox. WorldCat bibliographic record.
Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759–775. https://doi.org/10.1037/0012-1649.28.5.759.
Duschinsky, R. (2020). John Bowlby and the Tavistock Separation Research Unit. In Cornerstones of Attachment Research (pp. 1–108). Oxford University Press. https://doi.org/10.1093/med-psych/9780198842064.003.0001.
Follan, M., and Minnis, H. (2010). Forty-four juvenile thieves revisited: From Bowlby to reactive attachment disorder. Child: Care, Health and Development, 36(5), 639–645. https://doi.org/10.1111/j.1365-2214.2009.01048.x.
Fox, N. A., Nelson, C. A., and Zeanah, C. H. (2017). The effects of psychosocial deprivation on attachment: Lessons from the Bucharest Early Intervention Project. Psychodynamic Psychiatry, 45(4), 441–450. PubMed.
King, L. S., Guyon-Harris, K. L., Valadez, E. A., Radulescu, A., Fox, N. A., Nelson, C. A., Zeanah, C. H., and Humphreys, K. L. (2023). A comprehensive multilevel analysis of the Bucharest Early Intervention Project: Causal effects on recovery from early severe deprivation. American Journal of Psychiatry, 180(8), 573–583. https://doi.org/10.1176/appi.ajp.20220672.
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., and Verhage, M. L. (2024). Maternal and paternal sensitivity: Key determinants of child attachment security examined through meta-analysis. Psychological Bulletin, 150(7), 839–872. https://doi.org/10.1037/bul0000433.
Rutter, M. (1979). Maternal deprivation, 1972–1978: New findings, new concepts, new approaches. Child Development, 50(2), 283–305. PubMed.
Wade, M., and colleagues. (2022). The Bucharest Early Intervention Project: Adolescent mental health and adaptation following early deprivation. Child Development Perspectives, 16(3), 157–164. https://doi.org/10.1111/cdep.12462.
Wellcome Collection. (n.d.). London Child Guidance Clinic: Bowlby archival material, PP/BOW/C.3. Archival catalog.
Zeanah, C. H., and Gleason, M. M. (2015). Annual research review: Attachment disorders in early childhood—clinical presentation, causes, correlates, and treatment. Journal of Child Psychology and Psychiatry, 56(3), 207–222. https://doi.org/10.1111/jcpp.12347.