Research figure: John Bowlby · Written by Artificial Sapiens
John Bowlby's monotropy hypothesis concerns the selectivity of early attachment: infants come to direct protective, proximity-seeking behavior toward familiar, personally significant caregivers, sometimes showing a particular preference among them. The hypothesis became controversial when selectivity was understood as a requirement for one exclusive maternal bond. Historical analysis shows that Bowlby's original 1958 formulation allowed attachment behavior to be directed toward a person or a familiar group, whereas his compressed 1969 description emphasized one especially preferred figure. Contemporary attachment research supports the formation of selective relationships and the importance of dependable care, while documenting attachment networks involving mothers, fathers, relatives, and other caregivers. The strongest scientific account of monotropy therefore distinguishes a child's recognizable attachment figures, their relative priority in particular situations, and the much stronger claim that healthy development requires one biologically specified caregiver. Bowlby (1958); Duschinsky (2020); Bakermans-Kranenburg (2021).
What Is Bowlby's Monotropy Hypothesis?
In attachment theory, monotropy names a tendency for biologically organized social behavior to become directed toward particular attachment figures through experience. In an infant this selectivity is visible when distress, fatigue, unfamiliarity, illness, or separation evokes searching for, signaling to, or seeking contact with familiar caregivers. A stranger who provides ordinary interaction may not have the same meaning as someone whose availability the child has learned to expect. This person-specific organization is central to Bowlby's early account of attachment. Bowlby (1958).
Three propositions have to be separated. First, attachment behavior is selective: children recognize particular people and do not respond as though every available adult were interchangeable. Second, a child may display an attachment hierarchy: at a given age or in certain situations, one figure can be preferred for comfort or protection. Third, an exclusive or mother-only interpretation holds that an optimal attachment system is organized around one person and that other caregivers are necessarily secondary in developmental importance. The first proposition has broad support; the second describes a pattern that can occur, with priorities shaped by context and experience; the third is a much stronger normative and causal hypothesis that contemporary evidence does not establish. Dagan and Sagi-Schwartz (2021); Bakermans-Kranenburg (2021).
The word monotropy can obscure those distinctions because its literal suggestion of orientation toward one object makes singularity sound intrinsic to the concept. Scientific use requires specifying the outcome being predicted. A preference in a reunion episode, a stable bond with one person, the number of people who can provide a secure haven, and the developmental consequences of having several secure relationships are measurable questions with different answers. Treating them as the same claim produces both historical misreadings and faulty practical advice.
Historical Origins: From Psychoanalysis to Selective Attachment
Why Bowlby needed a theory of particular relationships
Bowlby's psychiatric and psychoanalytic training led him to questions about separation, anxiety, and the persistence of emotional bonds. His clinical observations of children exposed to disrupted family care encouraged him to study how actual relationships and changes in a child's environment affect behavior. In his theoretical reconstruction, proximity seeking was not simply a byproduct of feeding: the child had a developing relationship with familiar individuals whose presence could become protective. This shift toward observable social behavior and an evolutionary account of protection supplied the context in which monotropy emerged. Bretherton (1992); Bowlby (1958).
The historical setting mattered. Mid-twentieth-century child psychiatry encountered children affected by wartime separation, hospitalization, evacuation, residential nurseries, and inadequate institutional care. Bowlby and his colleagues worked in a period when some institutions could provide food, shelter, and medical treatment without preserving predictable, emotionally meaningful relationships. These conditions sharpened his concern with continuity and availability. They also created a risk of interpreting several distinct exposures—loss of a familiar person, unstable placements, lack of individualized care, and broader deprivation—as a single phenomenon. Later criticism returned repeatedly to that conflation. Bretherton (1992); Rutter (1972).
Ethology and the significance of familiarity
Bowlby's use of ethology gave selectivity a functional interpretation. For a dependent infant, directing signals toward known protective individuals could increase the likelihood of effective care. Species-typical capacities to follow, cry, cling, orient, and seek proximity would be shaped by individual experience with particular adults. The important point was that attachment behavior became coordinated around familiar relationships. An evolutionary account of the potential value of selective behavior is a theoretical explanation of function; it does not, by itself, demonstrate how many attachment figures a particular child should have. Bowlby (1958); Bakermans-Kranenburg (2021).
The original idea was compatible with the obvious social complexity of human care. Many societies have relied on fathers, grandparents, older siblings, and other familiar adults to protect and nurture children. Evolutionary reasoning about infant survival can support both discrimination between familiar and unfamiliar people and the advantages of a dependable network. Therefore, arguments about adaptation must be paired with actual developmental evidence rather than used as a shortcut to prescribe one household arrangement.
The 1958 Formulation and the 1969 Shift
The original definition in The Nature of the Child's Tie to His Mother
In his 1958 paper, The Nature of the Child's Tie to His Mother, Bowlby introduced monotropy as the direction of instinctual responses toward a particular individual or a group of individuals rather than an unrestricted range of people. The inclusion of a group is crucial. It identifies selective orientation as a process of differentiation by familiarity and experience, and it leaves room for several significant figures. Contemporary historiography emphasizes this wording because the popular equation of monotropy with precisely one attachment figure does not capture the entire original definition. Bowlby (1958); Duschinsky (2020).
Bowlby also associated this process with the way experience can make behavior increasingly directed toward familiar individuals. Once an infant has learned expectations about an adult's responses, the relationship becomes personally specific. This specificity can persist even when another adult performs the same physical task. A familiar attachment figure's value cannot be reduced to calories supplied, hours counted, or adult identity alone; it includes the history and organization of the relationship.
How Attachment and Loss changed the emphasis
The first volume of Attachment and Loss, published in 1969, gave a shorter account of monotropy that emphasized a child's tendency to attach particularly to one figure. That description made a hierarchy or dominant individual sound central. Historian Robbie Duschinsky shows that the 1969 condensation altered the impression conveyed by the 1958 text and influenced subsequent criticism. An accurate history must therefore retain both formulations: the original allowance for a familiar group and the later emphasis on a specially preferred person. Duschinsky (2020); Duschinsky (2025).
Neither wording alone determines whether a particular attachment preference is universal, whether it remains fixed across development, or whether one secure relationship is a sufficient account of the child's social world. These are empirical propositions. Later readers often treated Bowlby's emphasis on primacy as equivalent to maternal exclusivity, partly because his other writings advocated the central importance of mother–child relationships. Evaluating the hypothesis requires checking the wording, historical audience, and observational evidence separately.
Monotropy and the 1951 WHO report are related but distinct
Bowlby's 1951 World Health Organization report Maternal Care and Mental Health addressed the importance of consistent, emotionally supportive care and the harms associated with disrupted or inadequate early environments. The maternal deprivation debate drew together questions about separation, institutional quality, parental loss, and developmental risk. Monotropy, in contrast, is a specific proposition about how attachment behavior becomes selectively directed and, in one interpretation, prioritized among familiar people. A critique of a simplistic mother-only reading of the report does not automatically refute selective attachment; evidence for selectivity does not establish every causal claim historically advanced about deprivation. Rutter (1972); Bretherton (1992).
A Behavioral-System Account of Preference
Activation, proximity, and a secure haven
Bowlby's attachment behavioral system is organized around protection. When alarm is low, a child may explore the environment and play without constant contact. When threat, fatigue, pain, or unfamiliar circumstances increase, the child can direct signals toward attachment figures. The immediate aim is access to someone expected to provide safety and regulation. The same child can consequently appear relatively independent during exploration and strongly proximity seeking under distress. The system's organization emerges from the pattern across circumstances rather than one behavior observed in isolation. Bretherton (1992).
If several caregivers are familiar, a child's preference may depend on current availability, history of comfort in that situation, distance, sensory conditions, and the child's state. One adult may usually provide comfort after injury, another may organize sleep, and another may accompany play. The child can know each relationship differently. A hierarchy that emerges in one episode should not automatically be treated as a fixed ranking of emotional importance across all circumstances.
Selectivity develops through experience
An infant does not begin life with a complete list of attachment figures and immutable ranks. Repeated interactions provide information about who appears, responds, comforts, and helps restore safety. Developmental maturation increases the infant's capacity to recognize people, anticipate their availability, and coordinate more than one relationship. Attachment selectivity is therefore best understood as a developmental achievement shaped by social experience. Its organization can change when caregiving arrangements change, when relationships improve, or when a child grows older.
This account places the quality of interaction at the center of the mechanism. The amount of physical time spent together can create opportunities for familiarity, but time alone does not determine the quality of the response. Sensitivity involves detecting signals, interpreting their meaning, and responding appropriately. A caregiver can be consistently present yet chronically unresponsive; another can become a reliable source of comfort through attentive and predictable interactions. A large quantitative review finds a meaningful, moderate association between sensitivity and secure child–caregiver attachment across mothers and fathers. Madigan et al. (2024).
Internal expectations and relationships with several people
Bowlby's account of internal working models provides a route from accumulated experience to expectations about relationships. A child may expect one adult to respond readily while having different expectations of another. These representations can be partially relationship specific, even when the child's broader history influences interactions across the network. The possibility of differing attachment qualities with different caregivers follows naturally from this perspective. An attachment network contains relationships that may be interconnected yet individually meaningful. Dagan and Sagi-Schwartz (2021).
Importantly, a young child's distress when separated from one caregiver does not demonstrate that another familiar caregiver has no attachment significance. Nor does calm behavior with a second caregiver prove that the first was dispensable. How children regulate themselves depends on developmental stage, situation, learned expectations, and the available social network.
Mary Ainsworth's Contribution and the Limits of Attribution
Mary Ainsworth's field observations in Uganda and Baltimore, followed by the Strange Situation research program, gave attachment theory an influential observational language. Her work illuminated the use of a caregiver as a secure base for exploration and a source of comfort during stress, and her research tradition developed patterns of secure, avoidant, and resistant or ambivalent attachment behavior. Ainsworth's methodological contribution must be credited separately from Bowlby's original monotropy proposition. Later researchers added the disorganized or disoriented classification; that classification was not an original claim of monotropy. Bretherton (1992).
A Strange Situation classification describes an observed organization of behavior in a particular relationship under a standardized procedure. It is not a count of all the child's attachments, a direct measure of permanent personality, or a psychiatric diagnosis. Results from a mother–child procedure cannot automatically establish the quality of the same child's relationship with the father, a grandparent, or another familiar caregiver. This measurement constraint is essential when asking whether the strongest predictor is one primary attachment or the combined configuration of several relationships. Dagan and Sagi-Schwartz (2021).
Ainsworth's emphasis on caregiver sensitivity helped make behaviorally testable predictions about quality of care. Yet sensitivity is neither a trait restricted to biological mothers nor a guarantee of a particular classification. Context, child characteristics, measurement method, and conditions of family life affect the associations that studies detect. Bowlby's broad theory and Ainsworth's empirical operationalization should therefore be treated as connected contributions with distinct evidentiary bases.
Early Empirical Criticism: Schaffer and Emerson
Schaffer and Emerson's 1964 study of the development of social attachments in infancy became an important empirical challenge to a strict one-person interpretation. Their observations of infants in Glasgow documented the emergence and expansion of attachment relationships across more than one familiar person. The work helped establish that human infants can discriminate among significant caregivers and develop several attachments, rather than following an obligatory sequence in which a single exclusive bond must be completed before all others become meaningful. Schaffer and Emerson (1964).
The study's contribution is historical and conceptual as well as empirical. It directed attention away from the assumption that a child's social bond could be inferred solely from a presumed primary caregiver and toward the actual pattern of interactions with people in the child's environment. This is consistent with the selective component of the 1958 formulation and conflicts with a universal exclusivity rule. It also raises a methodological question: how should researchers identify attachment figures when opportunities for contact and cultural expectations differ?
Like other early observational programs, the Glasgow study had limits. Its sample was small and geographically and historically specific; observations and caregiver reports were made in conditions different from later standardized attachment assessments. Its results support the existence of multiple early relationships, but they cannot by themselves establish the optimal size of an attachment network or identify causal effects of particular family structures. Those further questions require broader samples, repeated observations, and relevant developmental outcomes.
Michael Rutter and the Critique of Maternal Exclusivity
Michael Rutter's reassessment of maternal deprivation challenged the habit of using a single label for separations of very different kinds. Separation from a familiar caregiver, loss of a specific attachment relationship, chronic lack of responsive care, repeated placement disruptions, and exposure to social adversity can have distinct pathways and consequences. Their effects may overlap, but a theory that attributes all risk to the absence of the biological mother cannot separate those pathways. Rutter (1972); Rutter (1979).
This criticism changed how evidence should be interpreted. If children in severely understaffed institutions show developmental difficulties, the finding is relevant to deprivation and relational instability; it does not isolate an effect of having multiple familiar caregivers. If a child is upset by abrupt separation from a well-known adult, the response confirms the importance of that relationship; it does not show that another sensitive, established caregiver is developmentally harmful. The experimental contrast implicit in a strict monotropy claim would require comparing reliable patterns of care differing in attachment-network structure while attending to other relevant conditions.
Rutter's work also cautioned against deterministic readings of childhood. Early adversity can carry serious risks, but developmental outcomes arise from combinations of experience, individual differences, subsequent relationships, and circumstances over time. The scientific question is how specific exposures alter risk, resilience, and adaptation, rather than whether a child has violated an imagined universal rule of one-caregiver development.
Attachment Networks: The Contemporary Research Framework
From a single dyad to a relational configuration
Much classic attachment research assessed one dyad at a time, commonly a child with the mother. This approach made controlled observation and comparison possible, but it also narrowed the unit of analysis. Contemporary attachment-network research asks how a child relates to more than one caregiver and how the combination of relationships predicts later outcomes. A configuration might include two secure relationships, one secure and one insecure relationship, or different patterns of organization across caregivers. Each dyad remains individually significant while the network becomes a subject of study. Dagan and Sagi-Schwartz (2021).
The network perspective involves at least four distinguishable dimensions: the number of familiar attachment figures; the quality of each relationship; the stability and coordination of caregiving over time; and the availability of care during stress. Counting adults without examining these dimensions provides limited information. The developmental contribution of an attachment network may arise from access to several reliable relationships, the compensatory potential of one relationship when another is strained, or more complex patterns of mutual influence. These are hypotheses that need direct testing, not assumptions that any larger network is automatically better.
Dagan and Sagi-Schwartz review the historical reasons for dyad-centered studies and propose that competing network models be evaluated rather than treated as interchangeable. The goal is to determine which configurations offer predictive information beyond the quality of a single relationship. Their account also makes a practical distinction between a theoretical attachment figure and any adult who happens to provide care. An occasional visitor and a consistently responsive caregiver cannot be assumed to serve the same role merely because both interact with the child. Dagan and Sagi-Schwartz (2021).
Evidence on behavioral and emotional outcomes
An individual participant data meta-analysis by Dagan and colleagues assembled data from nine studies involving 1,097 children to examine how attachment configurations with mothers and fathers related to internalizing and externalizing behavioral problems. It assessed patterns involving both relationships rather than assigning all explanatory weight to one parent. The analyses indicated that configurations of security and disorganization could contribute to associations with later adjustment. Such results favor studying the relational configuration while leaving room for heterogeneity and additional developmental causes. Dagan et al. (2021).
The research uses observational attachment assessments and developmental outcomes collected under differing conditions. Its pooled findings cannot determine whether a particular caregiver caused an outcome or whether modifying one bond in isolation would produce a specified change. Family stress, parental mental health, socioeconomic conditions, child temperament, and measurement quality may contribute to observed associations. A statistically detectable pattern is a statement about groups within studied samples, not a clinical forecast about an individual child.
Evidence on language development
A subsequent preregistered individual participant data meta-analysis drew on seven studies and 719 children to test mother–child and father–child attachment configurations in relation to language competence. Children with secure relationships to both parents scored somewhat higher in language competence than those with one or no secure parent relationship, with a small effect (d = 0.26). The same paper found an association for having two organized versus one organized attachment relationship (d = 0.23). Mother–child and father–child attachment quality did not show a differential association with language competence in these analyses. Dagan et al. (2024).
These findings are especially relevant to a strong monotropy claim because they examine more than one relationship and give no evidence that only the relationship with the mother deserves consideration. They also require precise interpretation. The sample was largely White, and the data were observational; average differences were modest, and the paper did not establish that making a second relationship secure necessarily causes better language development. The results support a relational-network framework within the studied populations while highlighting the need for more diverse samples and stronger causal designs.
Maternal and paternal sensitivity
In a 2024 meta-analysis combining 174 samples and 22,914 child–caregiver dyads, Madigan and colleagues reported a pooled association of r = 0.25 between caregiver sensitivity and child attachment security, with a 95% confidence interval from 0.22 to 0.28. The estimated association for maternal sensitivity was r = 0.26 and for paternal sensitivity r = 0.21; the investigators described them as equivalent in magnitude within the available evidence. The paternal evidence base was much smaller, which matters for precision. Madigan et al. (2024).
The result is consistent with an attachment system responsive to caregiving behavior across parent roles. Its effect size also sets a limit on interpretation: sensitivity is important at the group level, but it is only one contributor to attachment variation. The evidence does not establish that all caregivers are behaviorally identical, that one can replace another without emotional consequences, or that family composition alone determines security. It shows that responsiveness merits study with both mothers and fathers and that maternal biological status is insufficient as a scientific explanation of relationship quality.
Is There an Optimal Number of Attachment Figures?
A strict monotropy reading assumes that one special relationship is optimal by virtue of its singularity. A naive network reading could make the opposite assumption that the largest possible number of caregivers is always preferable. Neither proposition follows from the evidence. Children benefit from opportunities to develop recognizable, responsive, and sufficiently stable relationships; very large, constantly changing care systems may offer too little continuity for any particular bond to develop. The limiting variable may be predictable and meaningful interaction rather than an abstract numerical ceiling. Bakermans-Kranenburg (2021).
Bakermans-Kranenburg's review explicitly contrasts monotropy with kinship networks and alloparenting. In settings with shared care, several relatives or community members may play protective roles. Such systems can provide redundancy if one caregiver becomes unavailable. In an institution characterized by rotating personnel, by contrast, many adult contacts may coexist with few reliable individual relationships. The same numerical count can therefore describe profoundly different social experiences. Scientific comparisons should attend to duration, contingency, opportunity for emotional familiarity, and the caregiving ecology.
Another complication is age. A younger infant may express attachment differently from a toddler who remembers absent people, anticipates returns, and coordinates expectations across settings. Older children can sustain attachment representations while spending time away from caregivers. The number and practical organization of attachment relationships can change as cognitive and social capacities develop. Results from one age group and one procedure cannot define a universal limit across infancy and childhood.
Multiple Caregivers, Kinship Care, and Cultural Context
Human caregiving occurs within social arrangements shaped by work, kinship, household structure, migration, and community practice. Families may include two parents, one parent, grandparents, adoptive or foster parents, siblings, and other adults who contribute in different ways. Some arrangements distribute care through extended kin; others rely on a small number of adults. To study attachment scientifically, researchers must investigate the relationships children actually experience and the conditions under which those relationships are formed. Bakermans-Kranenburg (2021).
One of the strongest critiques of maternal exclusivity is its dependence on a historically specific view of domestic labor. In the social context of Bowlby's early writings, mothers were often assumed to provide nearly all continuous intimate infant care. When a theory treats that arrangement as a biological necessity, it risks mistaking a social distribution of caregiving for the uniquely natural form of attachment. The empirical question concerns protection and responsiveness in real relationships, not conformity to a particular adult role.
At the same time, recognizing caregiving diversity requires care in generalizing evidence. A substantial share of attachment studies comes from Western or comparatively affluent populations, and measurement procedures were not all developed for diverse social settings. Similar visible behaviors may have different contextual meanings where infants are held frequently, cared for collectively, or separated briefly according to different norms. Research programs comparing attachment networks should explicitly consider these differences instead of assuming that household labels identify relationship mechanisms.
Fathers and Other Familiar Adults
The existence of father–child attachment relationships supplies a direct empirical test of mother-exclusive assumptions. Research documents individual variation in the organization of attachment with fathers and explores how paternal sensitivity relates to security. The 2024 sensitivity meta-analysis found a positive association for fathers, and analyses of network configurations found that relationships with both parents can contribute relevant information about developmental outcomes. Madigan et al. (2024); Dagan et al. (2024).
This evidence should not be reduced to a competition about which parent matters more. Caregiver identity and care practices can covary with employment schedules, cultural expectations, division of labor, and opportunities for interaction. A child's preference in one situation may reflect who has routinely assisted in similar situations. A parent who is less frequently selected during separation distress may still provide secure-base support, playful engagement, or dependable comfort in other circumstances. Research on several relationships is needed to identify those contributions rather than infer them from titles such as mother or father.
Grandparents, adoptive caregivers, foster caregivers, and other familiar adults can also become important attachment figures when they participate in stable, responsive care. The developmental issue is whether the child has been able to form expectations of their availability and protective response. Kinship and legal categories may affect opportunities, resources, and continuity, yet attachment is a property of the lived relationship. A policy judgment about who should care for a child cannot be derived mechanically from the historical word monotropy.
Day Care and the Difference Between Several Figures and Rotating Care
Day-care arrangements illustrate why number and continuity must be distinguished. A child may spend time with a parent and also develop a familiar, trusting relationship with a consistent educator. This offers more than one context for care, but the quality of each relationship depends on availability, responsiveness, and stability. Where personnel change repeatedly and there is little opportunity for individualized interaction, the child may struggle to establish predictable access to a particular adult. These situations should not be coded as equivalent solely because both involve care outside the home. Bakermans-Kranenburg (2021).
The historical observation that children can suffer in emotionally impoverished institutions is therefore compatible with evidence for multiple attachments. Harm associated with impersonal or unstable care does not show that a second secure caregiver is harmful. It shows why institutions, residential services, and educational settings must be evaluated in terms of the social continuity they make possible. Such evaluations may include staffing stability, consistent key caregivers, prompt responses to distress, family access, and opportunities to maintain existing attachment relationships.
Care transitions should be studied in the same manner. A move between caregivers can be stressful because an established relationship is interrupted, the environment is unfamiliar, or predictable routines disappear. The existence of several supportive adults may sometimes help a child navigate transitions, but the buffering potential depends on the actual relationship and circumstances. General statements about day care, family employment, or short separations cannot be derived from attachment theory without specifying age, exposure, quality of care, and relevant outcomes.
Preference, Separation Distress, and Clinical Interpretation
Infants may cry, protest, search, or cling when separated from a preferred caregiver. Such reactions are developmentally intelligible expressions of an activated attachment system. They do not prove that the child possesses only one bond, and they do not establish a diagnosis. A child may tolerate separation better when a familiar second caregiver is available, or may still express distress despite the reassuring presence of another adult because the absent person's particular relationship matters. Both patterns are compatible with an organized attachment network.
A preference for one caregiver likewise should not be used to diagnose an insecure attachment relationship with others. Attachment classifications are based on defined observations of relationship behavior; diagnostic disorders require distinct clinical criteria, developmental context, and an appropriate assessment process. Monotropy is a historical explanatory hypothesis rather than an instrument for diagnosing a child or assigning blame to a family. Clinically meaningful distress warrants attention to intensity, duration, developmental appropriateness, impairment, and the child's wider circumstances.
For the same reason, a secure attachment to one caregiver cannot be inferred merely because that person spends the most time with the child or receives the most requests for assistance. High frequency of contact may reflect routine, access, and practical division of caregiving. Security concerns the organization of expectations and behavior under relevant conditions. Research should avoid treating physical proximity, emotional security, exclusivity, and overall development as interchangeable measures.
The Main Scientific Critiques of Monotropy
Conceptual ambiguity
The first problem is definitional. The 1958 formulation encompassed specific individuals or groups; the 1969 emphasis suggested a singularly preferred figure. Subsequent popular accounts often collapsed specificity, hierarchy, and exclusivity. A hypothesis can be tested only when these are operationalized distinctly. Observing that a child recognizes a particular caregiver is weak evidence for the existence of an attachment hierarchy and no evidence for a rule that only one figure should matter. Duschinsky (2020).
Confounding selectivity with maternal status
The second problem is assuming that the person most often selected must be the biological mother. Selection depends on the child's history with available caregivers. Different arrangements can produce different patterns, and other adults can acquire attachment significance. Evidence that maternal sensitivity predicts security is evidence about responsive caregiving within a mother–child dyad, not a demonstration that maternal biology uniquely causes security. Studies of paternal caregiving directly undermine the necessity of mother-exclusive interpretation. Madigan et al. (2024).
Treating historical case material as causal proof
A third problem concerns evidence quality. Early clinical and institutional observations identified important developmental suffering but frequently involved selection biases, retrospective histories, changing standards of care, and overlapping adversities. A child living without stable, responsive relationships differs from a child supported by several known caregivers. Historical cases can suggest mechanisms and motivate research; they cannot independently resolve whether a single attachment figure is necessary or how many secure bonds are optimal. Rutter (1972).
Undermeasuring the whole network
A fourth limitation arises when researchers assess one child–caregiver relationship and treat the result as a description of the child's entire attachment world. A dyadic measurement is valuable for its intended question, but it leaves other relationships unmeasured. The development of network models and individual participant data synthesis makes it possible to ask whether combinations of attachment security predict outcomes beyond individual dyads. Preliminary findings show the importance of asking that question, while the empirical literature still needs larger and more diverse datasets. Dagan and Sagi-Schwartz (2021); Dagan et al. (2024).
Overgeneralizing into parenting prescriptions
A final criticism concerns the move from a scientific mechanism to universal family advice. A theory of selective comfort seeking does not determine employment choices, prescribe one parent as the sole caregiver, or rank family forms. Developmental science can evaluate relationships, care environments, and risks, while policy decisions also involve resources, equality, cultural context, and practical constraints. The historical concept of monotropy becomes misleading when presented as an absolute moral standard for mothers or other adults.
What the Evidence Establishes—and What Remains Open
Evidence is well established that infants develop selective attachment relationships with familiar caregivers and that the organization of care and caregiver responsiveness are relevant to attachment security. Multiple attachment relationships are documented, and neither mother-only exclusivity nor an immutable one-person bond is required to describe the range of observed child behavior. Schaffer and Emerson (1964); Dagan and Sagi-Schwartz (2021); Madigan et al. (2024).
Evidence is more qualified on the developmental contribution of specific configurations. The available individual participant data analyses identify group-level associations involving security with mothers and fathers, including modest associations with language competence. The results strengthen the case for studying networks and challenge one-relationship sufficiency as a research assumption. However, observational associations do not prove causation, and network findings may vary by age, outcome, culture, and measurement. Dagan et al. (2021); Dagan et al. (2024).
Evidence is limited for any universally optimal number of attachment figures or fixed hierarchy valid across cultures and developmental periods. Large social networks may offer support, while very unstable contact can undermine opportunities for contingent, emotionally meaningful relationships. The relevant variables are likely relational and contextual, including caregiver stability, responsiveness, coordination, and access when the child needs protection. These hypotheses require stronger comparative, longitudinal, and cross-cultural research. Bakermans-Kranenburg (2021).
The strict claim that healthy development requires exclusive attachment to a single biological mother is scientifically unsupported. That conclusion leaves intact the serious importance of individual relationships and the potential harm of disrupting them. An attachment figure can be irreplaceable to a child at a given moment without being the child's only meaningful attachment figure.
Methodological Questions for Future Evaluation
A rigorous test of monotropy should begin by stating which version is under study. If the hypothesis concerns selectivity, investigators need repeated observations distinguishing familiar from unfamiliar adults in relevant situations. If it concerns relative priority, they need to assess how choices change with threat, fatigue, access, age, and the history of responsiveness. If it concerns exclusivity, they need data on whether multiple secure relationships coexist and whether their presence predicts advantages, disadvantages, or neither after accounting for caregiving conditions. One observation protocol cannot automatically answer every version.
Attachment-network research also needs to specify whose relationship is measured, the developmental period, and the relevant outcomes. Studies of infant behavior during short separation episodes, teacher reports of later adjustment, language tests, and adolescent self-reports cannot be merged into one indistinct outcome of attachment. Analysts should examine measurement invariance, information from several caregivers, clustering within families, missing data, cultural variation, and possible confounding. Individual participant data meta-analysis can harmonize measures and compare configurations, but it remains dependent on the representativeness and quality of the underlying samples.
Research designs that follow children through changes in their social networks may clarify how secure relationships emerge, persist, and reorganize. Particularly useful questions concern the extent to which a second established relationship buffers disruption to another, how responsive new caregivers become attachment figures, and what continuity is needed when children enter foster care, day care, or other settings. These questions produce testable predictions without presupposing a universal one-person ideal.
Implications for Child Welfare and Clinical Practice
The most defensible practical implication is to protect meaningful relationships and organize care so that children can come to know dependable adults. A child in foster care, hospital care, residential care, or a reorganizing family may need support for continuity and sensitive responses during change. Policies that allow familiar caregivers to remain involved, avoid avoidable turnover, and recognize existing bonds are consistent with the scientific importance of attachment. Policy details should still be evaluated with direct evidence about the population and setting concerned. Rutter (1979); Bakermans-Kranenburg (2021).
For parents and other caregivers, a single preference episode says little about a child's whole relational development. Consistent responsiveness, opportunities for comfort, and respect for the child's established relationships are more informative than attempts to force the child to choose one adult. Supporting several reliable caregivers can be compatible with attachment selectivity because distinct relationships remain personally significant. Professionals should avoid inferring pathology from family structure or from the presence of multiple caring adults.
Clinical interpretation must assess the child's circumstances rather than turn the historical hypothesis into a contemporary treatment protocol. When children experience persistent distress, disrupted placements, severe neglect, or other serious difficulties, assessment should address the complete developmental and caregiving context. Attachment-related observations may contribute information, but diagnosis, differential assessment, and intervention selection require the standards appropriate to the presenting clinical question.
Monotropy in the History of Attachment Theory
Monotropy remains historically significant because it compresses a fundamental discovery and a contested interpretation into one word. Bowlby's account made the familiar caregiver a distinct person in the child's behavioral world, challenging explanations that treated affection as a derivative of feeding. That insight helped establish selective attachment as an object of scientific investigation. His later emphasis on an especially preferred figure contributed to a public narrative that could be read as prescribing maternal exclusivity. The distinction becomes clear when the original publications are read chronologically rather than through simplified textbook summaries. Bowlby (1958); Duschinsky (2020).
The intellectual development from Bowlby to later network research is also a methodological history. Bowlby constructed the theoretical framework; Ainsworth developed influential behavioral assessments; researchers such as Schaffer and Emerson examined multiple early bonds; Rutter differentiated kinds of adversity; and contemporary teams investigate mother–child and father–child configurations with harmonized individual-level datasets. Each stage changed the questions that could be asked and the evidence available to answer them. These contributions belong to a developing scientific field rather than to a single unaltered doctrine.
An evidence-based account can preserve the enduring significance of a preferred, familiar caregiver while rejecting the claim that developmental security has one biologically predetermined source. Children's attachment relationships are selective, relationally specific, historically formed, and capable of being studied both individually and as a network. That is the lasting research question at the center of monotropy.
Frequently Asked Questions
Did Bowlby say babies can attach to only one person?
His 1958 definition of monotropy explicitly included direction toward a particular individual or a group of individuals. His 1969 summary placed more emphasis on one especially preferred figure. Therefore, the strongest one-person-only statement is an inadequate representation of his complete historical account. The careful interpretation separates selectivity, possible preference, and exclusive maternal necessity. Duschinsky (2020).
Does a child who prefers one parent lack attachment to the other?
No inference of that kind follows from a single preference. A child's response may vary with the nature of distress, previous experience, familiarity, routine, and who is available. Relationships with mother and father can each be assessed in their own right, and research has documented significant attachment-related associations for both. Madigan et al. (2024).
Does research show that more caregivers are always better?
Research supports the existence and relevance of multiple relationships, while also emphasizing continuity and responsiveness. A network of familiar, dependable adults differs substantially from a large roster of adults who rarely interact predictably with the same child. The evidence does not supply a universal maximum or minimum number of figures that guarantees developmental health. Bakermans-Kranenburg (2021).
Is monotropy the same as maternal deprivation?
Monotropy concerns the selectivity or prioritization of attachment behavior. Maternal deprivation was a historically broader concept used in debates about absence, separation, loss, institutional care, and inadequate emotional responsiveness. These questions overlap, but their mechanisms and evidence must be distinguished. Rutter (1972).
Can a grandparent, adoptive parent, or childcare professional be an attachment figure?
An adult who participates in stable, familiar, responsive caregiving can become personally significant to the child. The relevant scientific question concerns the observed relationship and expectations of availability. Biological relationship or occupational title alone neither guarantees nor rules out attachment significance. Network research is designed to examine such relationships in the contexts where children actually live. Dagan and Sagi-Schwartz (2021).
Can attachment classification identify a disorder?
An attachment classification obtained from a standardized observational procedure is a research description of relationship behavior in that procedure. It is not itself a psychiatric diagnosis. Any clinical assessment requires the appropriate diagnostic framework, developmental evaluation, and consideration of the child's broader circumstances.
Related Articles
Attachment in Infancy and Early Childhood: Security and Development examines observed attachment security, caregiving relationships, and developmental findings beyond Bowlby's historical formulation.
Is Attachment Theory Evidence-Based? What Research Supports, What It Does Not, and Common Misuses evaluates the broader contemporary evidence base and common extrapolations from attachment research.
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
Bakermans-Kranenburg, M. J. (2021). The limits of the attachment network. New Directions for Child and Adolescent Development, 2021(180), 117–124. https://doi.org/10.1002/cad.20432.
Bowlby, J. (1958). The nature of the child's tie to his mother. International Journal of Psycho-Analysis, 39, 350–373. https://pubmed.ncbi.nlm.nih.gov/13610508/.
Bowlby, J. (1969). Attachment and loss. Vol. 1: Attachment. Basic Books. https://books.google.com/books?id=dvZGAAAAMAAJ.
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.
Dagan, O., and Sagi-Schwartz, A. (2021). Early attachment networks to multiple caregivers: History, assessment models, and future research recommendations. New Directions for Child and Adolescent Development, 2021(180), 9–19. https://doi.org/10.1002/cad.20446.
Dagan, O., Schuengel, C., Verhage, M. L., and colleagues. (2021). Configurations of mother-child and father-child attachment as predictors of internalizing and externalizing behavioral problems: An individual participant data meta-analysis. New Directions for Child and Adolescent Development, 2021(180), 67–94. https://doi.org/10.1002/cad.20450.
Dagan, O., Schuengel, C., Verhage, M. L., and colleagues. (2024). Configurations of mother-child and father-child attachment relationships as predictors of child language competence: An individual participant data meta-analysis. Child Development, 95(1), 50–69. https://doi.org/10.1111/cdev.13998.
Duschinsky, R. (2020). John Bowlby and the Tavistock Separation Research Unit. In Cornerstones of Attachment Research. Oxford University Press. https://academic.oup.com/book/28760/chapter/235146160.
Duschinsky, R. (2025). A coming-of-age story: Attachment research in the 1980s. In Developments in Attachment Research. Oxford University Press. https://academic.oup.com/book/60504/chapter/522599045.
Madigan, S., Deneault, A.-A., Duschinsky, R., and colleagues. (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. (1972). Maternal deprivation reconsidered. Journal of Psychosomatic Research, 16(4), 241–250. https://doi.org/10.1016/0022-3999(72)90005-0.
Rutter, M. (1979). Maternal deprivation, 1972–1978: New findings, new concepts, new approaches. Child Development, 50(2), 283–305. https://pubmed.ncbi.nlm.nih.gov/114367/.
Schaffer, H. R., and Emerson, P. E. (1964). The development of social attachments in infancy. Monographs of the Society for Research in Child Development, 29(3), 1–77. https://doi.org/10.2307/1165727.