Research figure: Mary Ainsworth · Written by Artificial Sapiens
Mary Ainsworth and Silvia Bell's 1970 study, “Attachment, Exploration, and Separation: Illustrated by the Behavior of One-Year-Olds in a Strange Situation,” is a foundational empirical analysis of how infants coordinate exploration with attachment-related behavior when a familiar caregiver is present, absent, and reunited with them. Published in Child Development, the paper examined 56 infants approximately one year of age under systematically changing laboratory conditions. Its central scientific finding was that a caregiver's availability helped sustain exploration, whereas separation redirected behavior toward proximity seeking, searching, crying, and renewed contact. The paper also documented avoidance and resistance during reunions, establishing the importance of qualitative behavioral differences within an attachment relationship. The original publication is available through its DOI record and PubMed.
The article is best understood as an investigation of the organization and situational activation of infant behavior. Ainsworth and Bell explored what a short series of separations and reunions reveals about the relationship between attachment and exploration. Their results helped develop an observational framework that subsequently supported research on attachment patterns, caregiver sensitivity, and development. These later research traditions require separate analysis because the 1970 paper's immediate emphasis was the changing balance of observable behaviors within its sample rather than a population-wide taxonomy or clinical diagnostic system.
The original 1970 publication and its scientific identity
The original title was “Attachment, Exploration, and Separation: Illustrated by the Behavior of One-Year-Olds in a Strange Situation.” Its authors were Mary D. Salter Ainsworth and Silvia M. Bell, both associated with Johns Hopkins University in the publication. It appeared in March 1970 in Child Development, volume 41, issue 1, pages 49–67, with DOI 10.2307/1127388 and PMID 5490680. These details are corroborated by the journal issue record, PubMed, and the ERIC bibliographic entry.
The ERIC record identifies an earlier version presented at an American Psychological Association symposium on attachment behavior in humans and animals in San Francisco in September 1968. Accordingly, 1970 is the journal publication date, rather than the date when every underlying observation or argument was first formulated. The paper belongs to the Baltimore research program and should also be situated beside the earlier 1969 chapter by Ainsworth and Barbara A. Wittig. That chapter described the Strange Situation's emerging procedure, while the 1970 work by Ainsworth and Bell presented a detailed analysis of changes in exploration, distress, search, proximity seeking, contact maintenance, resistance, and avoidance. The distinct coauthors and publication histories deserve to remain visible in scholarly accounts.
The paper combined conceptual argument with observational evidence. Its opening sections explained attachment through an ethological and evolutionary framework; its empirical sections examined infants' behavior across structured episodes; and its discussion connected the observations with other human and nonhuman primate research. The authors' closing propositions addressed how attachment can persist when its visible behavioral expressions fluctuate. This organization matters because an empirical episode comparison and a broader theoretical proposition are different kinds of scientific claims. The first describes the observed sample under specified conditions; the second proposes an account extending beyond those observations.
The research problem: why exploration and attachment change together
An infant who plays actively while a familiar caregiver is nearby may show little immediate contact seeking. In an unfamiliar or distressing situation, the same infant may abandon play, turn toward the caregiver, search for that person, or seek physical contact. Ainsworth and Bell asked how these patterns fit together. A theory equating attachment only with currently visible clinging or crying would misinterpret the child who explores freely in the caregiver's presence. The coauthors therefore distinguished the enduring relationship from individual episodes of attachment behavior.
Their framework treated attachment as an organized behavioral relationship whose observable components become more or less prominent depending on circumstances. Exploration involves attending to, manipulating, moving through, and learning about the environment. Attachment-related behavior includes actions promoting access to a familiar person who can provide protection and comfort. Both can be organized adaptively within infancy. The authors proposed that a perceived threat, unfamiliar person, or caregiver's departure shifts the balance, while access to the caregiver can allow exploration to resume. Their argument draws on Bowlby's theoretical work and Ainsworth's observational research, but the 1970 study is specifically attributable to both Ainsworth and Bell Ainsworth and Bell, 1970.
An essential distinction follows: an absence of attachment behavior during play cannot be used as evidence that an infant lacks attachment. Conversely, an infant's crying during one separation provides information about that episode, yet does not by itself establish an attachment classification or diagnose a psychological disorder. This distinction was explicit in the original paper's attempt to understand behavioral organization over time and across contexts.
Intellectual setting: Bowlby, the Baltimore observations, Wittig, and Bell
John Bowlby developed the theoretical framework that connected infant proximity seeking and separation responses with evolutionary and developmental functions. Mary Ainsworth contributed extensive naturalistic and laboratory evidence, beginning with observational work in Uganda and extending to longitudinal home observations in Baltimore. Her earlier intellectual education also included William E. Blatz's security theory. Within this developing research network, the Strange Situation became a structured opportunity to observe behavioral systems under conditions more standardized than spontaneous home visits.
The 1970 publication explains that its laboratory situation grew out of an ongoing longitudinal study of infant–mother relationships. Naturalistic observation offered richly contextual evidence but provided relatively few predictable instances of unfamiliarity, separation, and reunion. Ainsworth and Wittig's 1969 methodological account preceded the 1970 paper, and Ainsworth and Bell then reported the comparative behavior of 56 infants in the laboratory. As later historical analyses clarify, the Strange Situation arose through a research program rather than a single invention moment Bretherton, 1992; Duschinsky, 2020.
Silvia Bell's contribution was substantive and documentable. The 1970 paper identifies a subsample of 33 infants investigated within a separate project associated with Bell and states that Bell served as the sole observer for this subsample during the relevant laboratory observations. Both authors participated in developing the reported behavioral analysis and coauthored the publication. Historical descriptions that credit the journal article solely to Ainsworth obscure this record. Barbara Wittig retains the distinct coauthor credit for the 1969 Strange Situation chapter, and neither contribution should be retroactively reassigned.
Participants, sampling, and observational design
The study reported 56 family-reared infants from White, middle-class families contacted originally through pediatricians in private practice. Its first subsample contained 23 infants who had been followed longitudinally from birth and entered the laboratory procedure at approximately 51 weeks. The second subsample contained 33 infants from a separate project, observed around 49 weeks. Thus, the sample consisted of two recruitment and study streams rather than one homogeneous longitudinal cohort. These features are reported directly in the paper's methods section Ainsworth and Bell, 1970.
The study was neither a randomized trial of parenting nor a representative sample of all infant–caregiver relationships. Its experimental strength lay in repeated, structured observations of the same infant across changes in the immediate social environment. The order of episodes was standardized, permitting comparison of mother-present, stranger-present, separated, and reunited conditions. Because those conditions occurred in a sequence, later episodes also involved accumulated exposure, fatigue, and possibly increasing arousal. Interpretation of episode differences therefore requires attention to time and order, not only to the caregiver's location.
The paper's original terminology refers to mothers and uses the social categories of its period. Contemporary developmental research examines attachment to mothers, fathers, and other consistent caregivers and analyzes family, cultural, and socioeconomic variation. Expanding the population of interest is a task of later research; it does not change the historical makeup of the 1970 sample. For historical accuracy, the original participant characteristics should be reported as such rather than silently generalized.
How the Strange Situation was arranged
The laboratory room had toys and a defined space for observing infant movement. A female stranger entered gradually and followed an instructed role; the mother also received instructions. The procedure was designed so that the room was sufficiently unfamiliar to encourage investigation without overwhelming the infant at the outset. The authors emphasized observation of responses to an unfamiliar adult, brief separations, and reunion, with the more stressful conditions placed later. The original study's eight episodes are summarized here because their sequence explains the findings; the detailed history and full coding method belong to the Strange Situation's own methodological literature.
The eight episodes in the original investigation
Episode 1 introduced the mother and infant to the observation room, after which the accompanying observer departed. In episode 2, the infant was free to investigate the room and toys while the mother sat quietly, responding if approached. Episode 3 introduced the stranger, who first sat, then conversed with the mother, then approached the infant gradually. The mother departed at the end of that episode. These phases established the mother-present baseline and the beginning of increasing novelty.
In episode 4, the infant remained with the stranger after the mother left. The stranger's response depended on the infant's state, including attempts to engage, distract, or comfort a distressed infant. Episode 5 brought the mother back and allowed observation of spontaneous reunion behavior; the stranger withdrew. When the infant was sufficiently settled, the mother departed once more after saying goodbye. Episode 6 placed the infant alone for a brief interval, which was shortened if distress became substantial. The stranger returned in episode 7, and the mother returned for the final reunion in episode 8. The study documents small procedural adjustments within its early cases, an important reminder that historical method development preceded later standardization Ainsworth and Bell, 1970.
The sequence created contrasts that can be studied behaviorally: mother alone with the infant, mother with a stranger, stranger without mother, reunions, complete temporary separation, and renewed stranger contact. The infant's reunion behavior became especially informative because the same individual could communicate desire for closeness, accept consolation, resist interaction, or divert attention away from the returning caregiver. Episodes were shortened when distress warranted, reflecting the researchers' stated intention to avoid an unnecessarily disturbing experience.
What the researchers actually measured
The authors observed behavior through a one-way window and used continuous narrative accounts, transcriptions, and subsequent coding. They described a system recording exploratory movement, manipulation, visual attention, and crying within 15-second intervals. For a standard three-minute episode, each of these frequency scores could reach 12 intervals; shorter or longer episodes were adjusted for duration. These scores captured how often a behavior appeared during a segment, rather than treating every behavior as an undifferentiated sign of attachment.
A second coding strategy used seven-point ratings of five behavior classes. Proximity and contact seeking included approaching, reaching, and other actions oriented toward gaining the caregiver. Contact maintenance included clinging or holding on after contact had been achieved. Proximity or interaction avoidance included turning away, ignoring, or avoiding engagement in circumstances where a response might otherwise be expected. Contact resistance included rejecting or struggling against contact or attempted interaction. Search behavior during separation included looking toward the exit, approaching the door, or attending to the caregiver's empty chair. These separate dimensions allowed the investigators to observe apparently conflicting actions, including seeking contact while simultaneously resisting it Ainsworth and Bell, 1970.
The paper reported high agreement in selected coding checks: across eight randomly selected cases, Pearson correlations for independent coding were .99 for exploratory locomotion, .93 for manipulation, .98 for visual exploration, and .99 for crying. For 14 randomly selected cases, rank-order coefficients for the five interaction and search dimensions ranged from .93 to .97. These figures document agreement under the original scoring conditions. They are not a modern, multisite validation study, and their small reliability-check subsamples should remain visible when evaluating measurement strength.
The researchers also distinguished behaviors directed toward the mother from those directed toward the stranger. This relational specificity is essential. A child could approach a stranger under one circumstance and seek the mother under another; the meanings of contact, avoidance, and comfort depend partly on the person, episode, and preceding separation experience. Coding a behavior without identifying its social target would discard information central to the paper's research question.
Principal finding: caregiver presence supported exploration
The most prominent pattern was the organization of exploration around caregiver availability. When infants were initially with their mothers, they investigated objects and the surroundings. The stranger's entry was followed by reduced locomotor, manipulatory, and visual exploration. Exploration remained relatively low during the first separation and increased following the mother's return. When the mother left again, exploratory activity declined once more; the lowest levels appeared when the stranger was present after the infant had also experienced being left alone. The authors examined episode changes using binomial comparisons, with the reported main differences generally reaching their specified thresholds Ainsworth and Bell, 1970.
This result supplied operational support for the secure-base interpretation: caregiver accessibility may make engagement with a novel environment possible, while separation makes the caregiver's absence a more urgent concern. The concept concerns a dynamic relationship. It does not assert that every securely attached infant must explore constantly, nor that less play necessarily means poorer development. Novelty, temperament, physical state, previous experience, and cultural practices can all affect observable activity.
The temporal order of the procedure also matters. Later episodes did not simply reproduce the first baseline with one independent factor changed. Once a child had undergone separation, reunion could involve lingering activation of attachment-related behavior. The 1970 discussion itself noted that exploration did not always return immediately to its original level after the mother came back. That carryover is theoretically meaningful, while also complicating any simple causal reading of individual episode contrasts.
Crying, separation distress, and searching for the caregiver
Crying was generally low at the initial mother-present baseline and did not show a conventionally significant increase when the stranger first appeared while the mother remained present; the paper reports a comparison of p = .068 for that early change. Crying increased when the mother left, declined after reunion, and increased especially strongly when the infant was left alone later in the procedure. The stranger's return did not reliably reverse this separation-related distress. The pattern indicated that an unfamiliar adult's mere presence and the absence of an attachment figure have distinguishable behavioral consequences.
Search behavior complemented crying. Infants sometimes oriented toward the door through which the mother had left, moved toward the exit, or attended to the location associated with her. Such actions have behavioral specificity: they are organized around the absent caregiver rather than simply representing undirected agitation. Search and crying together offered evidence that separation could activate proximity-promoting behavior even when the caregiver was not physically available for contact. The findings are consistent with the original authors' attachment-based interpretation while remaining observations of this particular study rather than universal response laws.
The 1970 authors also compared short laboratory separations with research on longer separations and nonhuman primates. Those comparisons are part of their theoretical interpretation, not direct evidence that a brief, ethically curtailed laboratory episode produces the same developmental consequences as prolonged separation, hospitalization, or deprivation. Duration, context, prior history, opportunity for alternative care, and repeated exposure all shape the meaning of separation. Conflating these very different settings would misstate the scope of the data.
Reunion behavior: proximity, comfort, resistance, and avoidance
Reunion observations were among the paper's most consequential features. The return of the mother often elicited active attempts to approach, reach, or maintain contact. Contact seeking after a brief separation could remain elevated even when crying had begun to diminish. The authors also reported striking combinations: some infants sought contact yet simultaneously struggled against interaction, and others avoided approach or oriented away from the mother. This combination showed that attachment-related behavior is qualitatively varied and can include competing behavioral tendencies.
Approximately one third of the original sample displayed some degree of contact resistance during the first reunion, and about one half displayed some degree during the later reunion. The authors interpreted the coexistence of strong contact maintenance and resistance in some cases as ambivalent behavior under attachment activation. Proximity avoidance was also observed during reunion in part of the sample. These are descriptive findings about the incidence of coded behavior in particular episodes, not interchangeable with later percentages of children assigned to attachment classification categories Ainsworth and Bell, 1970.
One important implication is that reunion behavior cannot be described adequately through crying alone. Two infants may cry at similar levels during separation but organize reunion differently. One may seek and accept comfort, another may avoid engagement, and another may seek contact while resisting it. Investigating those organized patterns became central to later classification research. Yet the 1970 article explicitly emphasized sample-wide behavior in its main findings and noted that detailed individual differences were examined elsewhere. It should not be presented as if every later subgroup criterion had already been fully established in this single paper.
Attachment itself versus observable attachment behavior
The paper's theoretical conclusion was that attachment persists across periods when its immediately recognizable behaviors are weak or absent. A child who is absorbed in toys while the caregiver remains accessible may be able to explore precisely because the relationship supports a sense of security. Observable clinging, crying, and searching rise or fall with the child's appraisal of the surrounding situation. The enduring organization of attachment and its momentary behavioral activation are analytically distinct constructs.
This distinction is relevant to scientific measurement. A single behavior sampled in one episode cannot be taken as a direct measurement of the underlying relationship's strength. The researchers recommended investigating qualitative patterns, the circumstances eliciting them, and their antecedents instead of assigning simplistic quantities to the intensity of attachment. In contemporary terms, the validity of a relational interpretation depends on the full pattern and context of observation, the quality of the coding system, and independent evidence beyond a single response.
The 1970 discussion can also be read as a methodological warning against overinterpretation. Laboratory behavior becomes more informative when compared with naturalistic observations and with theory-driven predictions. The authors situated their results alongside other experimental and field evidence, while acknowledging that short laboratory samples require interpretation. The enduring significance lies partly in this insistence that observational categories be connected to a coherent research question.
Where the 1970 findings fit within attachment classification history
Mary Ainsworth's research program subsequently became closely associated with the secure, avoidant, and resistant patterns conventionally labeled B, A, and C. The 1969 paper with Wittig and later work developed early distinctions in reunion behavior. In 1978, Mary Ainsworth, Mary C. Blehar, Everett Waters, and Sally Wall published Patterns of Attachment, a systematic monograph integrating observational measures, caregiver sensitivity, and classifications. The 1970 Ainsworth–Bell article is an important precursor and empirical companion, not a substitute for the full 1978 methodological and classification account Ainsworth et al., 1978.
The later disorganized or disoriented D category is associated with research by Mary Main and Judith Solomon. It should be attributed to that later extension rather than credited to Ainsworth and Bell's 1970 study. Likewise, adult romantic attachment research uses different populations, tasks, and instruments. Infant reunion behavior observed at approximately one year of age cannot be equated with adult self-reported attachment anxiety or avoidance. These distinctions protect the historical record and prevent a laboratory study of infancy from being turned into an all-purpose personality explanation.
The paper's specific contribution can be summarized more precisely. It showed how observable shifts in exploratory and proximity-oriented behavior could be elicited, recorded, and interpreted across repeated, standardized social contexts. It introduced detailed evidence concerning caregiver availability, separation, stranger contact, reunions, ambivalence, and avoidance within the attachment research program. Later classifications systematized additional individual differences through cumulative work by multiple researchers.
Methodological strengths and the evidentiary limits of the original sample
The study's main strengths were direct behavioral observation, repeated within-infant contrasts, explicit episode descriptions, differentiated behavioral dimensions, and reports of coding agreement. Because the authors distinguished visual, manipulative, and locomotor exploration from crying and several forms of proximity-oriented behavior, their analysis could test more specific predictions than an undifferentiated measure of distress. The two-part sample also brought a larger observational base than the 23-child longitudinal subset alone, though the projects differed in their observation histories.
The limitations are equally specific. All 56 infants were drawn from White, middle-class families in a particular historical and institutional setting. Recruitment through private pediatricians narrowed the sampling frame. The first 23 infants and the additional 33 were not identical in their longitudinal exposure. Several coding agreement checks relied on small subsamples, and the researchers themselves described procedural adjustments in early sessions. The repeated sequence involved cumulative exposure and possible carryover. These circumstances limit population generalization, purely causal inference, and claims of invariant behavior across cultures or caregiving environments.
Laboratory observation changes what is seen. A child may respond to a stranger, novel toys, a filming or observation environment, and transitions differently from how the same child would behave at home. Caregiver instructions can also alter spontaneous interaction. Ethical curtailment of a distressing episode changes observation duration in the very cases of greatest distress, a humane feature with measurement implications. None of these points invalidates the original findings; each specifies what was measured and what conclusions require further evidence.
The 1970 paper was also framed by mother–infant dyads. Later research expanded questions of father–child attachment and other caregiving relationships, and compared classifications across countries and family circumstances. Contemporary use of the procedure requires training, attention to cultural and developmental appropriateness, and recognition that a relational classification is not a psychiatric diagnosis. One session does not determine a child's personality, a parent's worth, or a fixed developmental destiny.
What subsequent systematic evidence establishes
A major 2023 meta-analysis by Sheri Madigan and colleagues synthesized more than 285 Strange Situation studies involving 20,720 child–caregiver dyads. Across this modern literature, the investigators reported distributions of 51.6% secure, 14.7% avoidant, 10.2% resistant, and 23.5% disorganized classifications, with meaningful variations by geographic setting and risk context. Those pooled frequencies describe the studies included in a modern international literature, not the 56 infants in the 1970 Ainsworth–Bell paper. Their value here is to demonstrate how a procedure that began in a small historical research program generated a much larger empirical record Madigan et al., 2023.
That synthesis found cross-regional differences and elevated rates of some classifications in samples exposed to substantial social and developmental adversity. Variations in distributions do not automatically establish that each behavior carries an identical meaning in every culture. Differences in routine caregiving arrangements, separation familiarity, environment, and assessment context warrant substantive study. A separate 2023 meta-analysis of modified Strange Situation classifications in preschool and early childhood samples also reported variations related to method and risk exposure Deneault et al., 2023.
Another 2024 meta-analysis examined associations between maternal and paternal sensitivity and children's attachment security across many studies. The synthesis supports a meaningful, moderate-to-modest relationship between sensitive caregiving and security while also demonstrating that sensitivity explains only part of the variation. This evidence is relevant to the wider Baltimore research program and to questions about conditions supporting secure-base exploration. It does not establish that Ainsworth and Bell's 1970 procedure experimentally isolated caregiver sensitivity or demonstrated a single sufficient cause of security Madigan et al., 2024.
The larger research record thus supports the ongoing scientific utility of structured attachment observation, while showing why historical sample characteristics, contextual moderators, and measurement design must be considered. Established evidence includes the paper's archival existence, coauthorship, documented sample, episode structure, and original observed patterns. Broad causal claims about future psychopathology from any one infant reunion are unsupported by the 1970 design. Claims of exact cultural universality of interpretation remain subject to qualification and research.
Scientific meaning of caregiver sensitivity, without rewriting the 1970 question
Caregiver sensitivity entered Ainsworth's wider research program through observations of the timing, appropriateness, and contingency of responses to infant signals. It provides one possible relational context for secure-base behavior. Nevertheless, the 1970 paper's primary empirical contrasts concerned mother presence, stranger entry, separation, and reunion, while different parts of the Baltimore program investigated caregiving behavior across homes and time. Separating these levels prevents the mistaken impression that a single paper measured every hypothesized mechanism of attachment.
The relationship between a child's exploration and a caregiver's availability also includes the possibility of reciprocal influence. Infants differ in activity, sensitivity to novelty, developmental timing, health, and experiences. Caregivers respond to those characteristics in varying contexts. Population associations between sensitivity and attachment do not imply that every separation response is caused by caregiving quality. A coherent developmental account therefore combines relational measures, child characteristics, environmental conditions, and repeated observations.
Cultural questions and the interpretation of unfamiliarity
In some settings, infants regularly encounter multiple caregivers or frequent separations from a particular caregiver; in other settings, brief separation from a primary caregiver is uncommon. Expectations about exploration, bodily contact, proximity, and stranger interaction may also differ across families and cultural contexts. The Strange Situation creates a common procedure, but the psychological meaning and stressfulness of its episodes can vary. This is why cultural research examines observed behavior and the caregiving ecology together, rather than inferring developmental outcomes from category distributions alone.
The original sample gives no direct estimate of how infants across the world respond under the same conditions. Its importance is historical and methodological: it enabled researchers to pose comparable questions and helped generate later evidence. Subsequent literature identifies differences between regions and between risk contexts, while continuing to investigate the meaning of patterns across circumstances Madigan et al., 2023. Stronger universal claims require converging evidence from diverse populations and multiple forms of observation.
Ethical and clinical interpretation
The authors explicitly planned short separations and allowed early termination of episodes when distress warranted. That ethical feature needs to be preserved in historical descriptions. Modern research with young children requires appropriate consent, sensitive handling of distress, properly trained observers, and methods suitable for the child's developmental and clinical context. The original procedure was created for research on behavioral organization in relationships, not for diagnosing an infant with a mental disorder.
An attachment classification summarizes an observed relationship-specific pattern under structured conditions. It is conceptually distinct from a symptom, a stable personality trait, a screening questionnaire, or a clinical diagnosis. Children's reactions to novelty and separation are influenced by multiple conditions, and a single observed avoidance or resistance behavior is not evidence of a specific disorder, trauma history, or future mental illness. Clinical decision-making, when needed, requires its own comprehensive standards and qualified assessment.
The study also should not be invoked to recommend deliberate separation as a parenting intervention or as a do-it-yourself test. Its scientific value comes from a controlled, ethically governed research design and from interpretation within a wider longitudinal and comparative literature. Parents and caregivers gain more useful guidance from evidence on responsiveness, stable relationships, and children's broader developmental needs than from trying to assign a label based on one reunion.
The place of Ainsworth and Bell in the history of developmental psychology
The 1970 paper advanced a research strategy in which theoretically meaningful behavioral systems were made observable through structured, repeatable changes in context. It joined conceptual arguments about attachment's persistence to carefully described measurements of exploration, crying, search, approach, contact maintenance, resistance, and avoidance. The combination made a complex interpersonal relationship empirically accessible while preserving the distinction between behavioral episodes and the relationship they express.
Its place in the historical sequence is clear. Bowlby's attachment framework provided the broader theoretical background. Ainsworth's Uganda and Baltimore observations developed the field and longitudinal basis. The 1969 Ainsworth–Wittig chapter described the evolving Strange Situation; the 1970 Ainsworth–Bell paper presented a detailed study of the attachment–exploration balance; and the 1978 Ainsworth–Blehar–Waters–Wall monograph integrated a mature research program. Main and Solomon later introduced disorganized/disoriented coding within another stage of the field's development. Each publication contributes a distinct form of evidence and an identifiable pattern of intellectual labor.
By placing Silvia Bell visibly beside Mary Ainsworth, the article preserves a core principle of scientific historiography: an influential method is developed across observations, collaborators, papers, measurement decisions, and subsequent revisions. The 1970 paper remains valuable for its particular empirical question and the unusually explicit account of how that question was investigated, rather than as a simplified origin story for all later attachment research.
Frequently asked questions about the 1970 study
What exactly did Ainsworth and Bell publish in 1970?
They coauthored “Attachment, Exploration, and Separation: Illustrated by the Behavior of One-Year-Olds in a Strange Situation,” published in Child Development, 41(1), 49–67. The paper combined an attachment–exploration conceptual analysis with observations of 56 infants in a structured laboratory procedure. Its stable identifier is DOI 10.2307/1127388.
How many infants participated, and how old were they?
There were 56 infants: 23 observed in the laboratory at approximately 51 weeks after longitudinal follow-up from birth, and 33 from a separate project observed at approximately 49 weeks. The participants came from White, middle-class families recruited initially through private pediatricians. These details limit how far the findings can be generalized without additional evidence.
What was Silvia Bell's role?
Bell was a named coauthor and her separate project supplied the 33-infant subsample. The study also states that she was the sole observer for that subsample. Her scientific role should be cited independently of Ainsworth's broader career or Wittig's coauthorship of the earlier 1969 chapter.
Did the paper invent the Strange Situation procedure?
The 1970 paper is an important published analysis using the procedure. Its own text identifies an earlier methodological account by Ainsworth and Barbara Wittig in 1969 and reports features of the procedure's development and adjustment. The procedure belongs to a longer empirical program.
Did Ainsworth and Bell diagnose secure or insecure attachment in these infants?
The central published analysis described changes in exploration, separation behavior, and reunion patterns, while noting individual differences investigated elsewhere. Later work formalized the A/B/C classification system. A behavioral attachment classification is an observational research construct, not a diagnosis of psychiatric illness.
Why is the 1970 paper still studied?
It demonstrated how caregiver availability, unfamiliarity, and brief separation were associated with structured changes in behavior. Its differentiated coding methods and conceptual distinction between attachment and overt attachment behavior helped establish a durable empirical approach that was expanded, criticized, and refined in subsequent decades of developmental research.
Related Articles
Mary Ainsworth — A scientific biography and overview of her observational research, intellectual background, coauthors, and contributions to developmental psychology.
Attachment in Infancy and Early Childhood: Security and Development — The wider developmental literature on infant attachment security and its relevance to early relationships.
Mary Ainsworth's Strange Situation Procedure: Design, Episodes, and Interpretation
Mary Ainsworth's Baltimore Longitudinal Study: Methods and Discoveries
Mary Ainsworth and Barbara Wittig's 1969 Strange Situation Chapter
References
Ainsworth, M. D. S., and Bell, S. M. (1970). Attachment, exploration, and separation: Illustrated by the behavior of one-year-olds in a strange situation. Child Development, 41(1), 49–67. https://doi.org/10.2307/1127388
Ainsworth, M. D. S., Blehar, M. C., Waters, E., and Wall, S. (1978). Patterns of attachment: A psychological study of the Strange Situation. Lawrence Erlbaum Associates. https://doi.org/10.4324/9781315802428
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
Deneault, A.-A., Bureau, J.-F., Duschinsky, R., Fearon, P., and Madigan, S. (2023). A meta-analysis of the distribution of preschool and early childhood attachment as assessed in the Strange Situation Procedure and its modified versions. Attachment and Human Development, 25(2), 322–351. https://doi.org/10.1080/14616734.2023.2187852
Duschinsky, R. (2020). Cornerstones of attachment research. Oxford University Press. https://doi.org/10.1093/med-psych/9780198842064.001.0001
Madigan, S., Fearon, R. M. P., van IJzendoorn, M. H., Duschinsky, R., Schuengel, C., Bakermans-Kranenburg, M. J., Ly, A., Cooke, J. E., Deneault, A.-A., Oosterman, M., and Verhage, M. L. (2023). The first 20,000 Strange Situation procedures: A meta-analytic review. Psychological Bulletin, 149(1–2), 99–132. https://doi.org/10.1037/bul0000388
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