Content uploaded by Frank C.P. van der Horst
Author content
All content in this area was uploaded by Frank C.P. van der Horst on Nov 06, 2017
Content may be subject to copyright.
Journal of the History of the Behavioral Sciences, Vol. 51(3), 261–284 Summer 2015
View this article online at Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/jhbs.21729
C2015 Wiley Periodicals, Inc.
AINSWORTH’S STRANGE SITUATION PROCEDURE: THE ORIGIN
OF AN INSTRUMENT
LENNY VAN ROSMALEN, REN ´
E VAN DER VEER, AND FRANK VAN DER HORST
The American-Canadian psychologist Mary Ainsworth (1913–1999) developed the Strange
Situation Procedure (SSP) to measure mother-child attachment and attachment theorists
have used it ever since. When Ainsworth published the first results of the SSP in 1969, it
seemed a completely novel and unique instrument. However, in this paper we will show that
the SSP had many precursors and that the road to such an instrument was long and winding.
Our analysis of hitherto little-known studies on children in strange situations allowed us
to compare these earlier attempts with the SSP. We argue that it was the combination of
Ainsworth’s working experience with William Blatz and John Bowlby, her own research
in Uganda and Baltimore, and the strong connection of the SSP with attachment theory,
that made the SSP differ enough from the other strange situation studies to become one
of the most widely used instruments in developmental psychology today. C2015 Wiley
Periodicals, Inc.
“So powerful is this technique in evoking behavioral changes that it is likely to be used
with increasing frequency in studies of mother-infant interaction.” (Ainsworth & Bell,
1970, p. 52)
Attachment theory is one of the better-known theories of contemporary developmental psy-
chology. Its basic theme is that human infants need a consistent nurturing relationship with
one or more sensitive caregivers in order to develop into healthy individuals. Inadequate rela-
tionships are considered to contribute to aberrant behavior and in combination with other risk
factors to psychopathology. The theory was gradually developed by the British psychiatrist
and psychoanalyst John Bowlby and comprehensively formulated in his trilogy (1969, 1973,
1980), after which attachment theory steadily became more influential. The theory has now
reached textbook status and its adherents have published thousands of studies inspired by this
theoretical framework. Ethologist and Nobel Prize winner Niko Tinbergen has praised Bowlby
for his “pioneering work” (Tinbergen in a letter to Ursula Bowlby, October 20, 1981), and
Stephen Suomi, comparative psychologist and former PhD student of American Psychologist
Harry Harlow stated that “the ideas about attachment that Bowlby developed into a formal
theory are still in the mainstream of developmental psychology and child psychiatry, and are
considered highly relevant in several other fields of clinical study. [ ...].Attachment theory
LENNY VAN ROSMALEN is a PhD student and Lecturer at the Centre for Child and Family Studies at
Leiden University, The Netherlands. The article presented here is part of her doctoral thesis on the early
work of Mary Ainsworth.
REN´
E VAN DER VEER is Professor of History of Educational Thinking at Leiden University, The
Netherlands. His research addresses the work of key educational thinkers such as Gal’perin, Janet,
Piaget, Vygotsky, Werner, and Wallon. Among his English-language books are Understanding Vygotsky
(1991, with Valsiner), The Social Mind (2000, with Valsiner), Lev Vygotsky (2007), and The Cambridge
Handbook of Cultural-Historical Psychology (2014, with Yasnitsky and Ferrari).
FRANK C.P. VA N D E R HORST is a Lecturer at the Centre for Child and Family Studies at Leiden
University, The Netherlands. His research addresses the roots of Bowlby’s attachment theory. He is the
author of John Bowlby: From Psychoanalysis to Ethology (2011).Correspondence concerning this article
should be sent to Lenny van Rosmalen, Department of Education and Child Studies, Faculty of Social
and Behavioural Sciences, Centre for Child and Family Studies, Leiden University, Wassenaarseweg 52,
Leiden, the Netherlands; lrosmalen@fsw.leidenuniv.nl.
261
262 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
has basically stood the test of time over the past 50 years, and I believe it will continue to do
so well into the future” (Suomi, Van der Horst, & Van der Veer, 2008).
During the 70s and 80s, Bowlby was generally seen as the (sole) founder of attachment
theory. However, as knowledge about the origin and development of the theory expanded over
the past 25 years, Mary Ainsworth (1913–1999) increasingly gained credit as the cofounder
of attachment theory. Ainsworth, an American-Canadian developmental psychologist, worked
with Bowlby in London from 1950 to 1953. After her return to the United States, Ainsworth
and Bowlby carried on their collaboration until Bowlby’s death in 1990. The importance of her
contribution to attachment theory has been discussed in many recent publications. Van Dijken
(1998), for instance, traced the core of attachment theory to Bowlby’s early childhood and de-
scribed the development of Bowlby’s ideas in his early work while clearly stating Ainsworth’s
contribution. Bretherton (1992, 2003) described the theoretical development of attachment
theory and argued that the fundamental expansion of attachment theory was made possible by
Ainsworth’s insights. Karen (1990, 1994) gave a compact overview of the history of attachment
theory and Ainsworth’s part in developing it, while Isaacson (2006) gave an extended overview
of the work of Ainsworth and Bowlby in her study on the development of attachment theory.
Van der Horst (2011) devoted a separate chapter to Ainsworth’s contribution to attachment
theory in his biography of Bowlby. Kenny (2013), in her book on principal theories of infant
development, named both Bowlby and Ainsworth as key figures in the development of attach-
ment theory. Recently, a special issue of Attachment & Human Development was dedicated to
Ainsworth’s contribution to attachment theory (Grossmann et al., 2013).
One of Ainsworth’s major contributions was the development of a laboratory procedure to
measure and classify the child’s attachment to his or her caregiver. Many authors have pointed
out that this procedure, the Strange Situation Procedure (SSP), played a fundamental role in
the increasing acceptance of attachment theory. Holmes (1993), for instance, stated that the
SSP was essential in providing empirical evidence for Bowlby’s ideas. Rather than observing
caregiver-child patterns for lengthy periods of time, it now seemed possible to typify the
caregiver-child relationship within the time frame of 20 minutes. Researchers were quick to
see this advantage and today the SSP is widely used—a simple search in Google Scholar gives
over 14,000 citations (more than 4,000 of which since 2010) for Ainsworth et al.’s (1978) book
Patterns of Attachment, which provides the most complete guide to scoring the SSP. Ainsworth
herself has explained in interviews and articles how her early research in Toronto, where she
designed instruments to measure security in young adults under the guidance of William Blatz,
and her later work in Uganda and Baltimore researching the bond between mother and child,
eventually led her to design the SSP (e.g., Ainsworth, 1983, 1988, 2010; Ainsworth & Bowlby,
1991; Ainsworth & Marvin, 1995; Myers, 1969; Rudnytsky, 1997; Stevenson, 1998). The finer
details of this development, however, remained unclear.
In the present article, we will add a new perspective to the historiography of the SSP
and show that the situation was much more complicated than these accounts would make us
believe. We will shed light on similar research that was being conducted all around Ainsworth,
and had been for decades. We will show that there were many “strange situation” studies, long
before Ainsworth designed her SSP, and we will discuss the possible reasons why Ainsworth’s
SSP turned out to be more successful than the instruments designed by her colleagues.
THE ORIGIN OF ATTACHMENT THEORY
The roots of attachment theory lie in the early twentieth century when Sigmund Freud’s
ideas became known in the Anglo-Saxon world and various experts promoted the study of the
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 263
child. G. Stanley Hall initiated the Child Study Movement and managed to recruit teachers
and well-to-do mothers to collect knowledge about the development of children (Ross, 1972).
For the first time, mothers were active as “co-researchers” and visited Mothers’ Clubs or
attended Mothers’ Congresses to discuss “optimal” ways to raise children (Hulbert, 2003).
Inevitably and increasingly, mothers were held responsible for the “outcome” of their efforts.
Psychoanalysis, which rapidly became popular in the United States after Hall had invited
Freud to go on a lecture tour in that country (Rosenzweig, 1994), contributed to this idea.
According to psychoanalytic theory, in the pressure cooker of the nuclear family the exact
timing of the mother’s interventions could mean success or failure for the child. Premature or
delayed weaning and early or late toilet training could cause dreaded “fixation” and subsequent
character malformation. Thus, according to this view, the origin of many later mental problems
lay in early childhood, a view that psychoanalysis shared with behaviorism (Beekman, 1977;
Stearns, 2003). The Mental Hygiene Movement, started in 1909 by the former mental patient
Clifford W. Beers and the psychiatrist Adolf Meyer, shared much of this thinking (Richardson,
1989). By identifying early signs of maladaptation in childhood, experts hoped to prevent or
cure mental illness. This led to the foundation in the early 1920s of so-called Child Guidance
Clinics for the treatment of maladjusted children. Child Guidance Clinics employed clinical
teams made up of a psychiatrist, a psychologist, and a psychiatric social worker. Inspired by
psychoanalytic theory and, to a lesser extent, behaviorism, most professionals believed that
the causes of deviant behavior and delinquency were to be found in the social environment
and not so much in the child’s genes. Sir Cyril Burt, for example, believed that “nearly
every tragedy of crime is in its origin a drama of domestic life” (cited in Wooldridge, 1994,
p. 99).
Burt introduced the idea of Child Guidance Clinics in England and it was while working
at the London Child Guidance Clinic as a psychiatrist that Bowlby was able to learn about and
put into practice many of the newer ideas. His belief that separation from the mother or mother
substitute is detrimental to the child he saw confirmed in many of the events connected to
World War II (e.g., children who were evacuated without their parents, children who lost their
parents) and in contemporary social policies (e.g., the limited possibility or even prohibition
to visit sick children in hospitals) (Van der Horst, 2011). Combining clinical observations with
theoretical insights and empirical evidence from the fields of psychology and ethology, Bowlby
gradually developed his attachment theory as published in his well-known trilogy (Bowlby,
1969, 1973, 1980). His general message that children need a stable and preferably full-time
attachment relationship with a loving mother (or mother substitute) was well met after the war
when many mothers did not work and raising children was viewed as a female responsibility
(Wootton, 1959; Riley, 1983).
Bowlby was a theoretician, who collected many retrospective data about the possible
detrimental effects on the child of a suboptimal mother-child relationship but he had no
instruments to do prospective research. The bulk of his conclusions were based on the effects of
gross separations from the mother and the finer details of inadequate mother-child relationships
could only be unearthed in time-consuming clinical research. It was here that Ainsworth’s SSP
came as a godsend. The SSP is a laboratory-based procedure (in this case a standardized
simulation of a stressful situation) intended to reveal patterns of caregiver-child attachment.
It quickly gained popularity among attachment researchers and is regarded to be a reliable
instrument with good predictive validity (Solomon & George, 2008). The SSP is widely used
these days and it has even been adapted to assess the bond between dogs and humans (Rehn,
McGowan, & Keeling, 2013). Although numerous articles and chapters have been written
about the SSP, little is known of its origin and history. Story has it that the SSP was thought up
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
264 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
by Ainsworth and her assistant, Barbara Wittig, in about 20 minutes (Inge Bretherton, personal
communication, March 2, 2013). It may well have been. We will see that it did not, however,
come out of thin air. Ainsworth’s extensive experience in researching security and development
of attachment, combined with a social background of increasing attention for mother-child
relationships and their effect on child development, paved the way for the construction of the
SSP. The wish to intervene in the social environment of the problematic child created the need
for a tool to measure the quality of the mother-child relationship and facilitated the quick and
broad acceptance of the SSP.
DESCRIPTION OF THE SSP
According to attachment theory virtually all children become attached, but the quality
of their attachment relationship differs and insecure attachment may result in developmental
problems. The SSP, by prompting attachment behavior in the child, allows for classification
of attachment security. The SSP as it is used today is basically the same as it was when
Ainsworth first used it in her Baltimore Study (Ainsworth & Wittig, 1969; Ainsworth, Bell,
& Stayton, 1971) and consists of eight episodes. In episodes 1–3, the child (in the company
of the caregiver) is first confronted with a strange environment (a play room) and then with a
stranger (an unknown research assistant). During the fourth episode, the caregiver leaves the
room and the infant is left with the stranger. The caregiver returns during the fifth episode
and the stranger leaves. The caregiver then leaves again (episode 6), which means the infant is
alone in the room. The stranger returns (episode 7), and eventually the caregiver also returns
(episode 8). In order to avoid effects of different parental behavior during the SSP as much as
possible, the caregiver is asked to respond to the child only when necessary and not to initiate
any interaction. Originally, the SSP was observed through a one-way window and observations
were dictated simultaneously by at least two observers. Nowadays, the behavior of the child is
captured on film and coded afterwards.
The three components of the SSP (the strange environment, the stranger, and the sep-
arations from the caregiver) make it stressful for children and prompt attachment behavior.
Special attention is paid to the episodes in which the caregiver is reunited with the child after
the brief separations. During these episodes (5 and 8) it is estimated how much the child
trusts the caregiver by looking at the child’s behavior and at how long it takes before the
balance between exploration of the environment and focus on the parent or caregiver has been
restored. The way in which the child approaches the caregiver at the reunion and seeks contact,
or tries to avoid contact, is angry, or acts in a disorganized way, is decisive for the attachment
classification. Children’s attachment to their mother can be classified as secure (B), insecure
avoidant (A), insecure ambivalent (C), or disorganized (D).
Strong claims regarding attachment’s continuity over time, its impact on later develop-
ment, and claims regarding maternal sensitivity as the most important precursor of strange
situation behavior have not been unanimously supported by the scientific literature of the last
40 years, however (cf. Lamb et al., 1984; Vicedo, 2013; Groh et al., 2014). Ainsworth herself
later expressed regret at the fact that the SSP had ended up as a stand-alone instrument, often
being used as a shortcut method, instead of being used in combination with home observations
(Ainsworth & Marvin, 1995). Nevertheless, regardless of its possible limitations, attachment
researchers value the SSP as an important instrument and have embraced it as a prime measure
of attachment. In addition, many of the adult attachment researchers that started to design their
own instruments in the 1980s used the SSP as a model to work from (Van Rosmalen, Van
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 265
IJzendoorn, & Van der Veer, 2014). If we want to explore the historical roots of this widely
used instrument we need to go back a century.
PREHISTORY OF THE SSP
Against the backdrop of increasing interest in the mental health of children in the early
twentieth century, a considerable amount of research was being done regarding fear in children.
Researchers tried to find out what caused children to be afraid and studied their reactions or
tried to find methods to overcome these fears (Jersild & Holmes, 1935). Hagman (1932), for
instance, studied the overt behavior of children taken into a room, left alone, and subjected to
a phonographic recording of artificial thunder as a possible fear stimulus. The psychoanalytic
focus on separation anxiety caused interest in the influence of an unknown environment on
the child’s behavior and/or the effects of separation from the mother. Nancy Bayley (1932)
reported on an observation of 61 infants that underwent a range of tests, during some of which
it was necessary to be briefly separated from the mother. Bayley put their crying down to,
among other things, the “strangeness of place and persons” (p. 316).
The first time we see a detailed description of the reactions of children specifically to a
strange person in a strange room, and the behavioral patterns in which the fear is manifested,
is in an unpublished study conducted in the early 30s called “The behavior of the child in
strange fields” by F. Wiehe, a student of Kurt Lewin (Lewin, 1935). Lewin, famous for,
among other things, his force field analysis that looks at the factors that influence a situation,
stated in 1933 that “the presence or absence of the mother changes the total structure of
the psychological environment very essentially, especially the child’s feeling of security or
insecurity” (Lewin, as cited in Bretherton & Munholland, 1999, p. 97). Wiehe studied children’s
behavior toward a stranger as they were taken into a strange room, sometimes accompanied
by their mother, sometimes alone, or a strange person was brought into the child’s home.
He then observed the actions of the child toward the stranger by noting down the presence
or absence of 15 possible actions (listen to, look at, turn bodily toward, smile at, speak to,
address to, express wishes, give or throw something, make bodily contact, stay nearby, ask
personal questions, demonstrate ability, show off, make demands, affective reactions) at six
different degrees of “strength of the social field.” This degree of strength was a function of
the spatial distance to the stranger, the duration of his presence, and his conduct. Wiehe found
the strongest pressure resulting in the child becoming motionless, and a weaker degree of
pressure causing the child to cry or to run away or toward his mother. The less pressure on
the child, the more natural free behavior was shown. Wiehe’s study was the first that did not
just note that strangeness caused fear in the child, but also paid attention to how this fear
was expressed in the child’s exploratory behavior. Being a student of Lewin, he analyzed the
child’s behavior in topological terms (i.e., in terms of forces and valences present in a specific
situation or “field”), just like Arsenian, another student of Lewin, would do several years
later.
In the early 40s, Prichard and Ojemann (1941) noted that “insecurity” was frequently
listed among the possible causes of behavioral problems and called for clarification of the
terms “security” and “insecurity,” and for a uniform way to measure them:
Careful analysis of the literature reveals that there is little agreement as to the meaning
of these terms and that methods of identification and measurement of insecurity are
not well developed. [ ...] It would be most helpful if we could develop more precise
methods of classifying children with respect to the strength or frequency with which this
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
266 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
desire motivates their behavior. We need methods by which we can discriminate between
the relatively secure and the relatively insecure children. (Prichard & Ojemann, 1941,
p. 114)
By comparing two groups of preschool children aged 2–5, one group labeled “secure”
and one labeled “insecure” by their respective teachers, Prichard and Ojemann developed a
security rating scale based on the behavior of children in a preschool environment.
In the late 1930s, Mary Shirley and Lillian Poyntz were curious to the effects of modern
American life on children, since an increasing number of mothers were employed outside the
home and nursery schools spread rapidly, when at the same time they felt that “the young
child’s sense of security and well-being is founded upon his experience of mother love and
care” (Shirley & Poyntz, 1941, p. 251). They published a detailed description of the effects
of the absence of the mother on her child during health and developmental examinations
as part of a longitudinal study into child growth. The 199 children concerned, aged 2–8,
underwent half-yearly routine medical checks at the Harvard School of Public Health between
1936 and 1940. Most of these children would normally spend their days at home with their
mothers. As they were left by their mothers at the Center for Research on Child Health and
Development for the whole day, this allowed for issues of separation to be studied. Apart
from looking at the behavior of the child throughout the day, special attention was paid to
the separation and reunion episodes, and what they might imply. As the authors argued: “The
real significance of the mother’s good or poor handling of the parting, however, lies in the
fact that it is usually an accurate indicator of the entire mother-child relationship” (Shirley
& Poyntz, 1941, p. 268). The questions they asked about the reunion were very similar to
the ones asked decades later in the SSP: “How do they greet the mother at the end of the
day? What type of mother-child relationship do these reactions imply?” (Shirley & Poyntz,
1941, p. 253). Psychoanalytic influence was clearly visible: the authors remarked that a child
might be quiet in the mother’s absence, but that this apparent quietness “may hide devastating
undercurrents of fears and doubts” (Shirley & Poyntz, 1941, p. 252). Just like Ainsworth would
do in her Uganda and Baltimore studies, Shirley and Poyntz ascribed part of the children’s
behavior during their mother’s absence to previous adequate or inadequate maternal behavior
and aspects of the home situation. In their words: “the child’s self-confidence and independence
depends upon his having experienced warm and wise maternal care” (Shirley & Poyntz, 1941,
p. 282).
In 1942, Shirley was the first to actually use the term “strange situation” as a factor to be
measured. Just like in her 1941 study with Poyntz, children were observed at the Center for
Research on Child Health and Development. This time Shirley classified the responses of the
children to being separated from their mother for the day. The classifications ran from “least
mature” (crying and resisting on arrival) to “most mature” (arriving at the center eagerly) in
eight steps. Again, Shirley made a point of analyzing the children’s behavior on the basis of the
home situation and the mother’s attitude toward child care. Children had been coming to the
Center from birth, and during each visit many potentially relevant matters were recorded, start-
ing with the behavior of the parent and the child when the child was picked up from home or
dropped off by the parent. Children’s comments were noted down during their day at the Center,
and the parent was interviewed at the end of the day, before picking up the child. This informa-
tion was compared to the child’s adjustment on the day of separation. Shirley firmly believed
that a child’s upbringing could tell us something about their behavior in a strange situation:
“The brief histories given above point in the direction of that well-established clinical hypoth-
esis that family relationships are a factor of paramount importance to the child’s adjustment in
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 267
any situation” (Shirley, 1942, p. 210). The question posed in the introduction of her paper, “Is
the adjustment of a child to an unusual event related to factors within his home experiences –
to his relations of confidence and affection with his parents?” (Shirley, 1942, p. 202), was
answered in the affirmative in the summary:
A child’s level of adjustment depends [ ...]muchmoreupon the wholesomeness of his
up-bringing in the home, and the security and affection given him by his parents. A secure
and wholesomely loved child goes forth to meet a new experience in a spirit of adventure,
and comes out triumphant in his encounters with new places, new materials, and new
friends, old and young. (Shirley, 1942, p. 217)
Shirley did not mention William Blatz (see below), but this observation matched the
description of what Blatz called “an independently secure child” in several of his childcare
manuals (Blatz, Millichamp, & Fletcher, 1936; Blatz, 1944). Shirley thus believed that children
who are secure and wholesomely loved behave with confidence in a strange situation.
Ainsworth’s line of thinking on the behavior of children in a strange situation bears great
resemblance to that of Shirley, but for reasons unknown Ainsworth never mentioned Shirley
in relation to her own strange situation work. We know that Ainsworth was aware of at least
some of Shirley’s earlier work because in an unpublished paper on methodological problems
in parent-child interaction research she mentioned a 1933 study by Shirley (Ainsworth, 1964),
and it is likely that she saw the references to Shirley’s work in Arsenian’s important 1943
study to which she referred on numerous occasions. However, when Ainsworth talked about
patterns of attachment behavior for the first time at a meeting of the Tavistock Mother-Infant
Interaction Study Group in London in 1961, the earliest researcher she mentioned in this
respect was Arsenian (Ainsworth, 1963).
Jean Arsenian (1943) was troubled by the fact that even though the concept of security was
now mentioned frequently by various researchers (e.g., by William Blatz, Ruth Horowitz, and
Lois Barclay Murphy), there did not seem to be an instrument with which to measure it: “Plainly
the problem of the origin of individual differences in security can be investigated constructively
only after child psychologists have agreed on the behavioral evidences of security and insecurity
in young children” (Arsenian, 1943, p. 225). She was the first to explicitly state that a strange
situation may be used as a diagnostic instrument: “In any situation the specific evidence of
security is assumed to be the appearance of positively adaptive patterns of behavior; conversely,
negatively adaptive or emotional forms of behavior will indicate insecurity” (Arsenian, 1943,
p. 225). She conducted an experimental laboratory study and was the first to systematically
vary the absence and presence of the mother in a situation unknown to the child, which enabled
her to separate the effects of the strange situation per se from the effects of separation from
the mother. Four experimental groups were formed out of a total of 24 children aged between
11 and 30 months. They were somewhat unusual in that they grew up in a reformatory with
their mothers (cf. Spitz’s sample of prostitutes in Van Rosmalen, Van der Horst, & Van der
Veer, 2012), who either could not give their child exclusive care as they worked as “helpers” in
the nursery and had many children to take care of, or worked in distant parts of the institution
and were only allowed to see their child during visiting hours. Arsenian started off with two
groups: the Alone group, which consisted of 16 children entering the strange situation alone,
and the Mother group, which consisted of eight children who were accompanied by their
mothers (or, in cases where the mothers of the children worked elsewhere in the institution and
were not available, by a “substitute” mother or a nursery helper) when placed in the strange
room. Six children from the Alone group were on later trials accompanied by their mother and
so became the Alone-Mother group, and five children from the Mother group were on later
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
268 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
trials left alone in the experimental room and became the Mother-Alone group. The children
were observed during 10 or 11 trials consisting of five minutes in the strange room on alternate
days. If they were accompanied by their mother or substitute, “the adult sat near the entrance
to the strange room and was instructed to remain as impassive as possible” (Arsenian, 1943,
p. 229).
All behavior of the children was classified into five categories, three of them indicating
goal-directed and adaptive behavior (play, locomotion, talking), and two of them indicating
signs of distress (crying, and “autistic gestures” [cf. Krout, 1935] like thumb sucking, waving
arms, stamping feet, etc.). The average duration of each type of behavior was determined.
Children in the Alone group spent most of the first four trials crying and engaging in autistic
behavior, but the children stopped crying in later trials. In the Mother group, adaptive behav-
iors dominated and at all times they showed more adaptive behavior than children from the
Alone group. The children showed different patterns of approach and withdrawal activity. Ar-
senian eventually distinguished 10 different patterns of behavior (from less to more adaptive:
nonmotile withdrawal and crying, agitated movement, retreat with crying, attack, regressive
encapsulation, nonmotile withdrawal without crying, encapsulation in play, approach with
conflict, and free approach) and continued to rate the 10 patterns on a security scale ranging
from −5 to +5. Using these scale values, Arsenian computed security scores for the children:
“Security of any child on a given trial was found by multiplying the scale-value of each pattern
which he displayed by the percentage of time during which he exhibited it” (Arsenian, 1943,
p. 237). Average scores for the Mother group and the Alone group were computed and it
was found that children in the Mother group behaved much more securely, and that security
increased over time in both groups—security indicating specific behavior in context, not a
specific mother-infant relationship. Of course, there were enormous individual differences.
Lewin’s influence was clearly visible when Arsenian stated that “insecurity is formulated as a
function of the unfamiliarity, or unstructuredness, of the environment in relation to the child’s
feeling of power in it” (Arsenian, 1943, p. 248).
It is clear that Arsenian introduced important techniques and notions into the study of a
strange situation’s effects on children. She specifically chose a strange room and the removal or
introduction of a mother or caregiver to study their effect on children’s feeling of security. She
rated the behavior of the children in the strange situation on a 10-point scale, and translated
the behavior into a degree of security. However, unlike Ainsworth would do two decades later,
Arsenian did not classify child behavior in terms of attachment patterns, nor did she explicitly
study exploration or pay attention to the reunion with the mother (as Shirley had done).
Even though the studies of Shirley and Arsenian were mentioned on numerous occasions
during the 40s and 50s in reviews and articles discussing a diversity of topics (motivation,
curiosity, group dynamics, and methods of research, to name but a few) and authors expressed
the need for a test that could aid in diagnosis of security of personality, this period did not see
many researchers conducting strange situations of their own. An exception was Glen Heathers
(1954), who partly replicated and tried to improve on Shirley’s (1942) study. As part of the
Fels Longitudinal Study, a child development study started in 1929 in Ohio and designed to
study physical growth, maturation, and psychological development, Heathers wanted to create
a measure of emotional upset in children during a “novel social situation” (Heathers, 1954,
p. 147) in order to relate this measure of upset to measures of social adequacy in school play,
and to see if there was a relationship between measure of upset during the novel situation and
certain aspects of home atmosphere and maternal behavior. The subjects were 31 two-year-
olds. Heather’s strange situation consisted of the child being parted from the mother at the door
of their home, and taken by a stranger (the Trip Observer) to a strange car that would take him
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 269
to nursery school at the Fels Research Institute for the first time. The procedure was repeated
for the next four days, to measure adjustment. The Trip Observer filled in an 18-item behavior
checklist (compared to the eight-step behavior rating scale Shirley had used) to make the
record of upset responses “more specific and more objective” (Heathers, 1954, p. 148). Next,
the child was observed during play at the nursery school and behavior scored on 14 categories,
paying attention to types of play activities and “various forms of dependent and independent
responses” (Heathers, 1954, p. 151). Prior to the first trip to nursery school, a home visitor
conducted a two-hour visit to the child’s home to observe mother-child interaction and to
interview the mother about her child’s experiences and behavior. The Fels Parent Behavior
Rating Scales were used to measure harmony in the home, sociability of the family, maternal
warmth, and maternal indulgence. Strict habit training was less en vogue now; the atrocities
of World War II had suggested that unconditional obedience may not be the best parental goal
and paved the way for the more permissive attitude exemplified by Benjamin Spock’s Baby
and Child Care (Beekman, 1977, p. 195). The results of the Fels study showed a positive
correlation for trip upset and social insecurity or inadequacy in school play. No relation was
found between trip upset and maternal indulgence, but Heather found home influences and
maternal warmth to be associated with low trip upset. However, correlations were generally
low and Heather concluded that “much more extensive observation of home influences would
be required to provide adequate data for testing the relations between the child’s experiences
at home and his behavior in situations outside the home” (Beekman, 1977, p. 157).
Seventeen years after publication, Arsenian’s study was for the first time discussed at
length in an overview of techniques used to study the behavior of children in the lab by Bijou
and Bear (1960). They gave a detailed one-page description of her setup and concluded that
the technique of this study seems to manipulate an important class of variables for
this (and older) age ranges. The response measures used by Arsenian are broad and
general (appropriate for an initial investigation); a refinement of these, and a wide use
of many other responses and tasks, should prove of great significance in experimental
investigations of these children. (Bijou & Bear, p. 171)
The next researcher to pick up on Arsenian’s study was Ainsworth.
HISTORY OF THE SSP
Mary Ainsworth wrote her dissertation in 1939 (Salter, 1939) under the guidance of
William Blatz. It is well known that Ainsworth’s interest in what we now call attachment stems
from Blatz’s teachings. Blatz, often referred to as the “Doctor Spock of Canada” (Wright,
1996), was the first director of the Institute of Child Study at the University of Toronto, and
had started his research career by studying children in the 1920s and 1930s in a nursery
school especially set up for the purpose. A controversial figure in his time, he denounced
authoritarianism and punishment, and promoted freedom for children so that they would learn
from experience. His involvement as Educational Consultant for the Dionne quintuplets from
1935 to 1938 was afterwards viewed as unfortunate. The quints, born in 1934, were separated
from their family by the state to be raised and studied by experts. Paying visitors could watch
the children from behind a one-way window, which generated millions of dollars for the
Canadian state. The quints, however, were just one of Blatz’s many projects and in the end he
became best known for his security theory on which he lectured for years. A written version
of the complete theory had to wait until his last book, Human Security (Blatz, 1966), which
was published posthumously and was based on tape recorded notes.
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
270 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
According to Blatz, a child starts off having to depend on his parents. If the child feels
certain the parent is going to be there for him, no matter what, the dependence is “secure”
and the child feels comfortable to go and explore. The exploration will result in development
toward a state of “independent security,” although Blatz admitted in his later writings that
independent security can probably never be reached completely and that a form of “mature
secure dependency” on friends and/or a partner is possibly the highest achievable goal. In the
meantime, some people will remain “immaturely dependent” or rely on defense mechanisms
(such as denial, dissociation, rationalization, or regression) in order to deal with feelings of
insecurity. Ainsworth’s dissertation was fully based on Blatz’s security theory. She designed
an instrument to measure security in (young) adults by using extensive questionnaires and
comparing them to case histories derived from autobiographies. After her dissertation she
kept expanding and improving on this instrument until 1958 when she published a much more
developed version of her questionnaires (Ainsworth & Ainsworth, 1958; Van Rosmalen, Van
IJzendoorn, & Van der Veer, 2014).
Ainsworth was the first, but by no means the only one attempting to measure Blatz’s
concept of security. In the years that followed many other members of Blatz’s team in Toronto
designed measures for security. Correspondence shows that Ainsworth was aware of these
developments while working at Johns Hopkins, but she later stated that members of the Blatz
team “went on to construct their own [tests] either for children or for infants, along lines that
did not really fit with my interpretation of Blatz’s security theory” (Ainsworth, 1988, p. 6).
Betty Flint designed the Flint Infant Security Scales for children from three to 24 months
(Flint, 1959, 1974) and Ainsworth briefly discussed Flint’s study in her paper on the effects
of maternal deprivation (Ainsworth, 1962). She also used the Flint Infant Security Scales at
the start of the Baltimore study (see below), but discontinued using them (together with some
other measures) after the first few months—the paper work was too time consuming, and the
Flint scales turned out not to be discriminating among the babies of the sample (Isaacson,
2006).
Having obtained her doctoral degree, Ainsworth spent several years lecturing at the
Psychology Department of the University of Toronto before joining the Canadian Women’s
Army Corps in 1942, where she attained the rank of major. During her four years there, she held
different positions in personnel selection, which involved clinical work such as administering
tests, conducting interviews, taking histories, and counseling. She returned to the University of
Toronto in 1946 to work as assistant professor. To prepare herself for the teaching of a class on
personality assessment she studied the Rorschach technique and the Thematic Apperception
Test. A few years later she co-authored a book on the Rorschach test with Bruno Klopfer
(Klopfer et al., 1954) and continued to use the test to validate her security questionnaires (Van
Rosmalen, Van IJzendoorn, & Van der Veer, 2014). Thus, the key components of Ainsworth’s
thinking (the parent as a source of security and as a base from which to explore the environment)
and her skills and experience in the area of personality assessment, were already present when
she arrived in London.
Ainsworth had followed her husband to London in 1950 and started working with John
Bowlby at the Tavistock Clinic, researching the effects on young children of being separated
from their mother. Ainsworth assisted James Robertson in analyzing the detailed field notes
he made while observing young children during separations, for instance in hospitals or in
institutions (Bretherton, 1992; Van der Horst, 2011). By doing so she gained a great deal of
experience in the analysis of observational data that she was able to use in the field research
she conducted several years later, when she followed her husband again, this time to Uganda.
There, in 1954–1955, Ainsworth studied normative development, mother-infant interaction,
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 271
and the development of attachment. She did this by observing 26 mothers with their children
at their homes for a few hours every two weeks, and conducting interviews with the mothers
(Ainsworth, 1967). When studying these mother-child dyads, she saw increasing evidence for
Bowlby’s claim that attachment is based on interaction rather than on the mother providing
food and other basic needs (which was the idea supported by psychoanalysts). Ainsworth left
11 years between finalizing her data collection in Uganda and publishing the book about the
study. She commented that
the full significance of what I observed and recorded in my field reports emerged only
gradually, not merely in the process of analyzing my observations, but also in the course
of reading, discussions with others interested in mother-infant interaction, and further
research into the early development of attachment. (Ainsworth, 1967, p. ix)
One of the places where she discussed the findings of her Uganda study was at the
Tavistock Study Group on mother-infant interaction in London in 1961, organized by Bowlby.
The reception of her work quite exceeded her expectations as she had not “expect[ed] experts to
find it so original and stimulating” (Ainsworth in a letter to Leonard Doob, October 12, 1961).
She proposed specific criteria to determine if the infant had “formed an attachment to his mother
as a special person” that went beyond those of Bowlby (Ainsworth, 1963). Ainsworth found
that only looking at crying, following, and clinging as reactions to (threatening) separation, as
Bowlby (1958) had proposed, was not enough to establish the strength of attachment. Contrary
to popular belief according to which intensity of attachment could only be measured by the
intensity of protest by the child in separation situations, Ainsworth stated:
Yet to judge the strength of the infant’s attachment to his mother solely in terms of the
intensity of behavior reflecting separation anxiety would seem to be a mistake; some of
the infants in this study who seemed most solidly attached to their mothers displayed little
protest behavior or separation anxiety, but rather showed the strength of their attachment
to the mother through their readiness to use her as a secure base from which they could
both explore the world and expand their horizons. (Ainsworth, 1963, p. 103)
After exhaustively examining her field notes Ainsworth came up with the following types
of attachment behavior: differential crying, smiling, vocalization, visual-motor orientation
toward mother, crying when mother leaves, following, “scrambling” over mother, burying face
in mother’s lap, clinging, greeting by lifting arms, clapping hands, or approaching through
locomotion, and exploration from mother as a secure base.
When describing the role of the mother as a secure base, or “haven of safety,” from which
to explore the world, Ainsworth was clearly inspired by Blatz and Arsenian, as we have seen
before, but also by Harlow (1958), who in turn had also been inspired by Arsenian. All three had
reported that the baby used the mother as a secure base from which it would go out and explore,
ready to face external threats without panic. Harlow, who was also present at the Tavistock
Study Group meetings, conducted research with rhesus monkeys and described how the mother
or mother surrogate acted as a secure base for the baby monkey in a strange situation. During
the so-called “open field test,” the baby monkey was put in a strange environment (a room
measuring six by six feet containing some stimuli), and Harlow described how the monkey
would rush to the mother surrogate as soon as possible and cling to her. After a few sessions,
however, the baby would let go of the mother and start to venture out more and more, using her
as a “base of operations” (Harlow, 1958, p. 679). The monkey would “explore and manipulate
a stimulus and then return to the mother before adventuring again into the strange new world”
(Harlow, 1958, p. 679). However, if the (surrogate) mother was not present in the room, the
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
272 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
monkey would remain in one spot, in a crouched position, and show no exploratory behavior
whatsoever (Van der Horst, LeRoy, & Van der Veer, 2008).
When the Uganda book finally appeared in 1967, Ainsworth had gone further than just
describing the development of attachment in general and had divided the children into three
attachment groups, or classifications: secure attached, insecure attached, and nonattached.
Classifications were based purely on extensive observations in the home. Ainsworth was now
confident that a close attachment could develop simultaneously with increasing competence
and independence. She again pointed out, as she had done in 1961 at the Tavistock Study Group
meeting, and in her article published in 1964 on patterns of attachment behavior, that it was the
insecure child who would cling to his mother, refuse to leave her, and for whom maintaining
interaction from a (small) distance, even part of the time, was not enough (Ainsworth, 1964,
1967).
In the meantime, while still analyzing the data for the Uganda book, Ainsworth had started
the Baltimore Study in 1963. She and her colleagues observed mother-infant interaction in 26
families by following infants from three to 54 weeks during three-weekly home visits lasting
four hours each. The study was initially intended as a replication of the Uganda study and
its aim was to investigate the development of patterns of attachment and to systematically
observe maternal behavior (from which the construct of maternal sensitivity would later be
developed). At the end of the first year, a large amount of data had been gathered regarding
the behavior of the children in the home situation. As the study developed, it became clear to
Ainsworth that apart from observing the infants in the home environment, it was necessary
to include an observation in an experimental setting. She wanted to assess the child’s use of
the mother as a secure base outside the home environment, and she felt the need to base her
research on measurable information in order to make her observations acceptable within a
very behaviorist environment (Silvia Bell, personal communication, March 5, 2013). Thus,
Ainsworth together with her assistant Barbara Wittig, designed the SSP.
However, at that time they were by no means the only ones studying the behavior of
children in a strange situation. In the middle and late 60s, we see a sudden wave of experiments
with children in strange situations, many of them inspired by Shirley, Arsenian, and/or Harlow.
STRANGE SITUATIONS IN THE 1960S:PARALLEL DEVELOPMENTS
World War II had stirred the interest in the consequences of mother-child separation
and psychoanalysis had now reached the summit of its popularity. Bowlby’s WHO report
Maternal care and mental health (1952) was the first attempt to chart out the adverse effects of
maternal deprivation. Now, more than ever, perhaps, it was believed that the children’s mental
and social problems were caused by the social environment and, specifically, by the nuclear
family. Researchers went to great pains to discover the social origin of children’s problematic
behaviors.
As part of a psychoanalytically oriented, longitudinal study designed by John Benjamin,
Katherine Tennes and Esther Lampl (1966) reported on the observation of 27 infants aged
between six and 36 months in a strange situation, with the aim of investigating “the contribution
of the instinctual drives, both libidinal and aggressive, to the intensity of infantile separation
anxiety as determined by fear of object loss” (p. 426). The children, in the company of
their mothers and an examiner, were observed in a laboratory setting while playing, doing
developmental tests, and interacting socially with the examiner. After about 45 minutes the
mother was requested to leave the room and stay away for 10–15 minutes. The distress of the
child was rated on a five-point scale, and the experimenters were interested in the infant’s way
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 273
of dealing with his anxiety. In the first instance it was left to the child to reduce his anxiety
by himself. Failing that the examiner would try to offer physical comfort, and if that did not
work an attempt was made to distract the child by offering him a toy. The defense reactions
that were observed were described at length and included a “regressive attempt at kinesthetic
restitution of the missing object” (stop moving and not looking at the stranger); “active mastery
of the environment” (attempts to follow the mother); “inhibition of motor activity” (decrease
in activity and motility); “use of inanimate objects” (offering a toy to the stranger, holding
on to a toy, or banging or throwing a toy). Even though their explanations were all strongly
psychoanalytically orientated, Tennes and Lampl reached the same conclusion Ainsworth had
presented in 1965 (Ainsworth & Wittig, 1969), that is, intensity of separation anxiety cannot
be used as the sole criterion of the child’s attachment to the mother:
The level of libidinal investment of both partners has no predictive value for the intensity
of infantile separation anxiety. Infants who lack a libidinal investment in the mother do
not have separation anxiety, but some children with a very high libidinal investment in
the mother also fail to develop separation anxiety. (Tennes & Lampl, 1966, p. 435)
Berg, Stark, and Jameson (1966) were concerned that the behavior of children, as they
were being observed with their mothers in a child psychiatric clinic, was influenced by the
presence of the interviewing doctor. They wanted to test objectively the effect of the presence
or absence of a stranger on children’s attachment and exploratory behavior and observed
17 preschool children. The setting (see Figure 1) bore a great resemblance to the setting of
the SSP: the child was placed in a playroom with his mother for two subsequent periods of
20 minutes, during one of which a stranger was present. The room contained some toys, and
the mother was asked to stay seated and to try not to influence the behavior of the child. A
four-foot radius circle was drawn on the floor around the mother’s chair, and the amount of
time the child spent inside the circle during each period was measured. Observers also noted
certain types of behavior of the child, like physical contact with mother, movements in the
direction of the toys, approaches to mother, speech, finger sucking, eye rubbing, rhythmical
movements of the limbs, and communication with mother or stranger. The results allowed for
the 17 children to be divided into three groups according to their reactions to the stranger:
group A, consisting of seven children, showed more proximity and contact seeking toward
their mothers in the presence of the stranger; group B, consisting of four children, did not
necessarily seek proximity but showed severe shyness by hardly speaking in the presence of
the stranger, and group C, consisting of six children, did not seem to be bothered much by the
stranger’s presence. In this study, maternal personality was taken into account by asking the
mothers to complete the Eysenck Personality Inventory, the results of which were compared
to the behavior of the child. The study showed that a higher mean maternal neuroticism score
was related to more proximity seeking behavior of the child in the presence of the stranger.
However, the authors hesitated to draw any strong conclusions: some measures of behavior
“were insufficiently precise to distinguish those children who were more obviously affected
by the stranger’s presence” and some children “were apparently unaffected by the stranger
and even showed more exploration and less attachment when she was there” (Berg, Stark, &
Jameson, 1966, p. 249).
A couple of totally different and, according to today’s standards rather horrifying studies
were published by Miriam Rosenthal in 1967, who wanted to test “the suggestion made by
Harlow that some form of dependency behavior toward a mother might reduce the threatening
aspects of unfamiliar, novel situations, thus allowing exploratory behavior to take place”
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
274 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
FIGURE 1.
Setting of Berg, Stark and Jameson (1966)
(Rosenthal, 1967b, p. 357). She manipulated anxiety levels in 64 nursery-school girls aged
3–5 after placing them in a strange environment, and measured their “dependency” by looking
at attention-seeking and proximity-seeking behavior (seeking positive or negative attention,
seeking praise and approval, seeking help, seeking proximity or being in physical contact with
the adult). In her first study, the child’s dependency behavior toward the mother was compared
with the dependency behavior toward a stranger (Rosenthal, 1967a). In her second study,
she looked for effects of the strange situation and level of anxiety on dependency behavior
(Rosenthal, 1967b). The strange environment was a room in an office building filled with
toys and typical playschool material (see Figure 2). The child was taken into the room by an
experimenter, while the mother or stranger were already present in the room, seated at a desk
behind a curtain. The adult was instructed not to initiate any interaction with the child, but if
necessary to respond without rejecting it. Sessions lasted for 30 minutes. In the low-anxiety
condition the room was decorated with pictures of smiling faces, and a tape recording of
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 275
FIGURE 2.
Setting of Rosenthal (1967a)
children’s songs was heard from the room next door. In the high-anxiety condition, however,
the room looked different:
The child, on entering the room, faced a slow-burning alcohol lamp standing on a stainless
steel tray. Next to it was a pair of scissors, a white paper tissue, and a pencil. The pictures
of the smiling faces were replaced with a group of sad faces. (Rosenthal, 1967a, p. 123)
The children’s songs were replaced by sounds of banging on a metal object, a child crying, and
a high-pitched shriek, coming from behind a red door, next to which the lamp was positioned.
Then, after about 12 minutes
and following a loud continuous shriek, the red door opened very slowly (the experimenter
waited until the child was looking in that direction) and a hand in an arm-length black
glove reached slowly in, put out the lamp and withdrew, closing the door once more.
Within two or three minutes a crying sound was heard. (Rosenthal, 1967a, p. 123)
Unsurprisingly, proximity seeking increased in frequency in the high-anxiety situation
(which would probably scare the wits out of the average adult). Frequency of total dependency
behavior towards the mother was significantly higher than toward the stranger.
Roberta Collard (1968) conducted a much milder strange situation study, but from a
different perspective: measuring the effect of birth order on reaction to a stranger in a strange
situation. Presuming that later-born children would have siblings around them a lot of the
time and that their mothers would be less protective and anxious than they had been with
their first child, later borns were expected to show less fear. Collard observed 36 infants
aged 38–56 weeks who were sat on their mother’s lap at a testing table, where a stranger
took position opposite the baby and offered it a toy. Response latency before picking up the
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
276 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
toy was measured, as were exploratory and play responses, and positive or negative social
responses. Collard found that first-born infants indeed took longer to pick up the toy than
later-born infants. The social responses toward the stranger also differed: later borns showed
more positive social responses, like laughing and smiling, than did first borns.
Cox and Campbell (1968), uneasy about the lack of evidence on which to base the
decision to let a mother be present or not during experiments with children, decided to partly
replicate Arsenian’s study. They conducted two experiments: one consisting of 20 mothers
with children aged between 13 and 15 months, and one consisting of 20 mothers with children
aged between 23 and 37 months. Mother and child were taken into an observation room with
a chair for mother in one corner, and a pile of four toys in the corner diagonally opposite. Half
of the mothers (the experimental group) would stay in the room for four minutes, leave the
room for four minutes, and return for the last four minutes. The other half of the mothers (the
control group) would stay in the room for the full 12 minutes. Observers noted the incidence
of eight behaviors of the child: touching mother, holding mother, speech, movement, play,
touching objects, placing object in mouth, and crying. According to the authors, these results
demonstrated that “when young children play in a strange situation their behavior is affected
by the presence or absence of their mothers. Absence of the mother produces a decrease in
talking, movement, and playing with toys” (p. 129).
Conrad Schwarz (1968), inspired by Arsenian (1943) but also by the (at the time still
unpublished) studies of Ainsworth and Wittig (1969) and Rheingold (1969), carried out his
own version of the strange situation. In the introduction he declared: “there appears to be no
experimental evidence which bears directly upon the existence of a unique role for attached
individuals as inhibitors of distress in children over three years” (Schwarz, 1968, p. 314). The
subjects were 16 children aged 4, who were taken to a room with their mother by a stranger.
The stranger started playing games with mother and child. After about five minutes the
experimenter entered and asked either the stranger to step out of the room, or the mother. After
the child was left with the mother or the stranger, the fear stimulus was activated: a remotely
controlled 10-inch tall mechanical toy gorilla, hidden in a box, was made to beat its chest,
walk out of the box, and beat its chest again (cf. Harlow & Zimmerman, 1959). Observations
included the child’s facial expressions, position in the room (the room was marked off in
18-inch squares), and visual orientation. Contrary to what was expected, results showed more
observed fear in the presence of the mother. In company of the stranger the children looked
away from the gorilla sooner and appeared to focus on the toys. Schwarz (1968) suggested that
the mother’s presence “may have facilitated the communication and expression of fear. Being
left in a strange room with a strange person may have induced a general inhibition of motility
and emotional communication” (Schwarz, 1968, p. 321, original italics).
In 1969, the same year Ainsworth published the data on the SSP for the first time, Harriet
Rheingold published two studies on the behavior of infants in a strange environment, and the
effect of the presence or absence of the mother. The data on these studies, just like Ainsworth’s
data on the SSP (Ainsworth & Wittig, 1969), had already been presented in 1965 during the
fourth Tavistock Study Meeting. Ainsworth and Rheingold knew each other quite well, and
even though they designed and conducted their studies independently and were guided by
different theories, they were aware of each other’s work, and were inspired partly by the same
studies (e.g., Bayley, 1932; Arsenian, 1943). Rheingold’s study covered several experiments
in which the effect of a strange environment on the behavior of 10-month-old children was
observed: “Although the effect of a strange person upon the behavior of human infants has
often been studied... the effect of a strange environment has less often been investigated”
(Rheingold, 1969, p.137). In the first experiment, the strange environment was a totally empty
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 277
FIGURE 3.
Setting of Rheingold and Eckerman (1969)
room, in the second one this empty room contained some toys, and in the third one the room
contained a stranger. Each experiment consisted of four or five trials lasting two to three
minutes each, and for half of the subjects the mothers were present in the first few trials, while
for the other half of the subjects the mothers were present in the last few trials (much like
Arsenian’s Mother-Alone and Alone-Mother groups). The floor of the room was divided into
squares (as, incidentally, was the floor of the room Ainsworth would use for the SSP during the
first few years) and for the trials with the mother or the stranger they were instructed to sit in
a specific square in the room and not to talk or play with the child. It was allowed, however, to
look at the child, to smile or to comfort him. Observers recorded vocal responses and locomotor
activity. Rheingold concluded that a strange environment inhibited exploratory behavior and
evoked emotional distress. Presence of the mother, however, supported exploratory behavior,
prevented crying, and evoked nonprotest vocalizations. Putting toys in the room did not make
much difference, whereas putting a strange person in the room caused distress and inhibited
physical activity. Children whose mothers had been with them in the strange room during the
first few trials were slightly more at ease in the later trials without their mother than were the
children who had started off on their own. Children who were in the strange room alone during
the first few trials remained more distressed even when their mothers were present during
subsequent trials.
In her next study, Rheingold (Rheingold & Eckerman, 1969) again looked at exploratory
behavior of 10-month-old infants in a strange environment, this time paying particular attention
to “the process by which the infant detaches himself from his mother and her near environment”
(Rheingold & Eckerman, 1969, p. 272). In this study, the mother was always present and seated
in a small room, together with the child. The small room allowed access to a larger room:
the strange environment (see Figure 3). Sometimes this larger room was empty, sometimes it
contained toys. The main difference with Rheingold’s previous study, in which the children
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
278 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
were placed in the large room by the experimenter, was that this time the infant started from his
mother’s side and could decide himself whether to enter the strange environment or not. All 24
children in the study left their mothers and moved into the strange environment, regardless of
whether there were toys in the room or not. Comparing her two studies, Rheingold remarked
that in the first study the results were “attributed to the strangeness of the room. The present
results suggest that it was not solely the physical properties of the room but the additional
conditions of being placed and left alone that provoked the distress and inhibition” (Rheingold
& Eckerman, 1969, p. 281). It was noted that apart from entering the strange environment,
all children also returned to the small room, and often re-entered the large room. According
to Rheingold, this to-ing and fro-ing may well be illustrating the child’s exploration from the
mother as a “secure base” (Ainsworth, 1963), or “a base of operations” (Harlow, 1958).
Before turning to the first publication on the SSP, let us summarize what these strange
situation studies conducted in the mid- and late 60s have in common that makes them different
from Ainsworth’s SSP. To start with, all these studies were normative and looked at the general
behavior of children as a group, finding that children showed more proximity-seeking behavior
when they found themselves in scary situations, that they used the mother as a secure base
from which to explore, or that they usually preferred their mother over a stranger. The next
important difference is that most of these studies attributed the behavior of the child directly
to the situation and not to past mother-child interactions. Reasons given for a child behaving
in certain ways were that the child was put into a strange environment, a stranger was present,
mother was present or not, or the child felt threatened by fear inducing stimuli. Except for
two studies (Berg, Stark, & Jameson, 1966; Tennes & Lampl, 1966), and unlike Shirley and
Arsenian before them, these researchers did not take the relationship between mother and
child into consideration. And lastly, even though Berg and colleagues divided the children into
three groups based on their reaction to the stranger, none of these studies tried to classify the
children by translating the behavior into a degree of security like Arsenian had done. In other
words, many researchers from different theoretical backgrounds experimented with “strange
situations” to answer various research questions but none of these procedures developed
further into an instrument measuring security or attachment. In addition to the aforementioned
differences, this could partly be due to the fact that they did not fit into a larger research
tradition or paradigm.
ARRIVAL OF THE SSP
Ainsworth presented her findings on the SSP for the first time in 1965, but they were
not published until 1969 (Ainsworth & Wittig, 1969). The article was part of the longitudinal
Baltimore study. When the child was 12 months old, each infant-mother pair was observed in
a final session, the SSP. Special attention was paid to three aspects: first, Ainsworth wanted
to look at how the infant would use the mother as a secure base from which to explore the
surroundings, since she considered this to be one of the most important criteria of a healthy
attachment. Second, Ainsworth was interested in the child’s response to a stranger. At home,
the children had been generally comfortable in the company of strangers, but nothing was
known of the infants’ behavior in unfamiliar surroundings. Thirdly, Ainsworth wanted to look
at the child’s response to the mother’s departure and return. Again, the children had been
mostly comfortable with their mother’s departure from the room when at home in familiar
surroundings, but Ainsworth was interested in their reactions when in unfamiliar surroundings.
Ainsworth stated that:
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 279
The situation was designed to be novel enough to elicit exploratory behavior, and yet
not so strange that it would evoke fear and heighten attachment behavior at the outset.
The approach of the stranger was gradual, so that any fear of her could be attributed to
unfamiliarity rather than to abrupt, alarming behavior. The episodes were arranged so
that the less disturbing ones came first. Finally, the situation as a whole was intended to
be no more disturbing than those an infant was likely to encounter in his ordinary life
experience. (Ainsworth, 1964, p. 53)
Ainsworth also hoped that this study would provide evidence for her notion that insecurely
attached children are the ones that show a high level of anxiety in a minor separation situation,
and that securely attached children would not necessarily show this by their intensity of protest
when separated from their mother, as for instance Schaffer and Emerson (1964) maintained
(Ainsworth & Wittig, 1969).
Fourteen infants were put through the SSP and were rated on the following behavioral
items: exploratory behavior, visual orientation, crying, responses to mother leaving the room,
responses to mother’s return, responses to the stranger’s entrance, responses to being picked
up (by the mother or by the stranger), and responses to being put down (by the mother or by the
stranger). In agreement with Ainsworth’s (1963) findings with the Ganda infants, the results,
when related to the infant’s behavior at home, and taking into account his developmental
history, did support the hypothesis that the child’s ability to use his mother as a secure base
from which to explore the environment is an important part of the child’s attachment to his
mother. Furthermore, the results of the responses to separation in the strange situation allowed
Ainsworth to divide the babies into three tentative groups: group A, consisting of four children
that showed very little disturbance upon separation (the classification that would later develop
into insecure avoidant); group B, consisting of six babies who were clearly upset by the
separation, but at the same time managed to adapt (the securely attached children), and group
C, consisting of four babies that were also clearly upset, but showed distinct maladaptive
behavior (the classification that would later develop into insecure ambivalent). Whereas in
Infancy in Uganda Ainsworth (1967) had still used the classification “non-attached,” she now
stated that “the development of attachment is not easily discouraged; it can be distorted, but it
takes more deprivation than is represented in this sample to discourage its growth altogether”
(Ainsworth & Wittig, 1969, p. 136).
In Ainsworth and Bell’s (1969) report on patterns of mother-infant interaction in the
feeding situation the final classification system was published for the first time. However,
since the focus of this paper was on the effect of feeding practices Ainsworth stated in a
footnote that this classification system, based on a sample of 56 infants, “will be described in
detail in a future publication.”
Ainsworth and Bell (1970) then continued to publish a normative study on the SSP
that described behaviors characteristic of the sample as a whole during each of the episodes
instead of describing the individual differences reported by Ainsworth and Wittig (1969).
The publication started with a summary of the ethological and evolutionary viewpoints of
attachment, which in effect were at the base of Ainsworth’s SSP, and was followed by an
explanation of attachment and attachment behavior. The study itself consisted of observations
of 56 subjects, 23 of which were part of the Baltimore study, and 33 from a study done by Bell
(1970). The main findings were threefold: first, the infants used the mother as a secure base
from which to explore the strange environment. Second, as the mother left the room, exploration
lessened or stopped altogether, and the infant showed proximity and contact-seeking behavior.
Third, when the mother returned, the infant kept showing more proximity and contact-seeking
behavior, and exploration remained at a lower level than before the mother left the room. About
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
280 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
a third to half of the sample showed contact-resisting behavior to some degree after the first or
second reunion. Ainsworth pointed out that these findings were in accordance with findings
of other experimental studies, clinical studies, and field studies. Five propositions were given
for a comprehensive concept of attachment:
1. Attachment is not coincident with attachment behavior.
2. Attachment behavior is heightened in situations perceived as threatening.
3. Attachment behavior is incompatible with exploratory behavior.
4. After prolonged absence from the object of attachment, attachment behavior may
diminish, but is likely to reemerge.
5. Attachment relations are qualitatively different from one attached pair to another.
Based on these propositions, Ainsworth and Bell put forward their argument for the
measure of quality of attachment as opposed to quantity:
The qualitative differences, together with the sensitivity of attachment behavior to sit-
uational determinants, make it very difficult to assess the strength or intensity of an
attachment. It is suggested that, in the present state of our knowledge, it is wiser to
explore qualitative differences, and their correlates and antecedents, than to attempt
premature quantifications of strength of attachment. (Ainsworth & Bell, 1970, p. 65)
The detailed description of the final classification system of the SSP was published one
year later (Ainsworth, Bell, & Stayton, 1971). The paper reported on the same study as the
normative paper (Ainsworth & Bell, 1970), but this time the authors looked at individual
differences. Comparing the extensive home observations and developmental history of the
infant-mother relationship with the infant’s behavior during the SSP made it possible to clearly
define the different attachment categories. To test the new classificatory system, an additional
comparison was made between the attachment-exploration balance at home, where the infants
were categorized into five groups according to the ability of the infant to use the mother as
a secure base from which to explore the world, and the infant’s SSP classification. Ainsworth
found “an impressive degree of congruence between a baby’s response to his mother in the
strange situation and the quality of the attachment-exploration balance at home” (Ainsworth,
Bell, & Stayton, 1971, p. 39).
Next, Ainsworth compared maternal behavior with the attachment classifications (cf.
Berg, Stark, & Jameson, 1966). Maternal behavior had been rated during home observa-
tions and was divided into four dimensions: acceptance-rejection, cooperation-interference,
accessibility-ignoring, and sensitivity-insensitivity. There was no doubt that mothers of group
B babies were significantly more sensitive, but Ainsworth hoped to be able to identify what
distinguished A and C mothers by studying the other three dimensions. During the last quarter
of the first year, the four dimensions were rated separately for each home visit and results indi-
cated that group A mothers were significantly more rejecting than group C mothers. Ainsworth
herself was pleased with the results:
In a previous publication (Ainsworth & Wittig, 1969) we reported the strange-situation
findings for the first 14 subjects in our sample, and our impression that individual differ-
ences were related to differences in style of mother-infant interaction throughout the first
year of life. In view of the expensive and very time-consuming nature of longitudinal
research, it is an attractive notion that one might in a 20-minute procedure obtain a rea-
sonably reliable and valid assessment of the nature of the relationship that has developed
between an infant and his mother. (Ainsworth, Bell, & Stayton, 1971, p. 19)
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 281
This statement by Ainsworth provides a partial explanation of the success of her mea-
surement procedure. If one believes with Bowlby and other attachment theorists that social
problems such as juvenile delinquency and pilfering find their roots in inadequate mother-child
relationships and if one hopes to identify suboptimal mother-child relationships in order to
intervene in the family process and remedy the problems, the availability of a simple labo-
ratory procedure to typify that relationship is indeed a godsend. For the first time, instead
of conducting lengthy and repeated clinical interviews with mothers in the Child Guidance
Clinic, instead of repeatedly paying costly visits to the homes of mothers and children, one
could simply run the mother-child dyads through the lab procedure and identify the couples
at risk.
CONCLUSION
In 1969, Ainsworth published her first study on the SSP. She was, however, by no means
the first researcher to look at the behavior of children in a strange situation. The first studies
on this subject emerged in the 1930s. When reviewing the relevant studies published from
the 1930s to the early 1970s, however, we found nothing remotely like a linear progression.
Quite the opposite: researchers were not aware of, or ignored, previous studies using a strange
situation, or chose to change the previous setups to answer their own particular research
questions. For instance, even though Shirley and Poyntz (1941), Shirley (1942), Heathers
(1954) and Berg, Stark, and Jameson (1966) focused on the mother-child relationship in a
similar way Ainsworth did, Ainsworth never mentioned these studies in relation to the SSP.
Some of the studies carried out in the 1960s had in common that they borrowed from Shirley
and Arsenian, but again, researchers appeared to be unaware that similar research was being
conducted elsewhere at around the same time. Most of the procedures discussed in this paper
were not followed up, possibly, because they did not become imbedded in a research paradigm
that generated new research questions with potential social and clinical implications.
Comparing all the strange situation studies available from the beginning of the twentieth
century until the first publications on Ainsworth’s SSP, we argue that the latter was different
for a number of reasons. First, the SSP was based on a blend of Blatz’s security theory
and, more importantly, Bowlby’s attachment theory (which she helped develop) and offered
explanations for the observed behavior from an evolutionary and an ethological standpoint.
This gave Ainsworth’s instrument a solid backbone. Second, she was one of the few to focus
on parent-child relationships instead of traits or behaviors of the child. Whereas most studies
conducted in the 1960s attributed the behavior of the child to the presence of the stranger or
the strange surroundings, Ainsworth presumed the child might feel more or less secure as a
result of his relationship with his caregiver. Third, the SSP was not a normative study but gave
information on developmental differences between individual children. And last but not least,
the SSP allowed researchers to infer or classify the child’s attachment relationship with the
mother from the behavior of the child in the lab. For the first time researchers had at their
disposal a simple and time-saving instrument with which to typify the attachment relationship
of individual children with their caregivers. This satisfied the need to identify problematic
development at an early stage. Or, in the words of attachment researcher Michael Lamb: “The
SSP has become popular in part because of claims that Strange Situation behavior predicts
important aspects of the child’s behavior as much as several years later” (Lamb et al., 1984,
p. 136).
In this contribution we have extended the historiography of attachment theory and of the
SSP in particular, and shown that Ainsworth was not the first, and certainly not the only one
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
282 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
to observe the behavior of a child in a strange situation, with or without his mother. She was,
however, one of the few who translated the behavior of the individual child into a degree of
security or attachment. Supported by the strong connection with Bowlby’s attachment theory
and providing a quick and easy way to measure attachment, the SSP became the instrument of
choice for many attachment researchers.
ACKNOWLEDGMENTS
The authors wish to thank Silvia Bell, Inge Bretherton, Bob Marvin, and Donelda Stay-
ton, who facilitated the preparation of this paper by allowing the first author to interview
them extensively. Inge Bretherton and Marinus van IJzendoorn kindly provided constructive
criticisms of a first draft. Part of this research was made possible by grants awarded to the
third author (van der Horst) by, respectively, the K ¨
ohler-Stiftung and the Dr. J. L. Dobberke
Stichting voor Vergelijkende Psychologie.
REFERENCES
Ainsworth, M. D. S. (1962). The effects of maternal deprivation: A review of findings and controversy in the context
of research strategy. Deprivation of maternal care: A measurement of its effects (Vol. 14). Geneva: World Health
Organization.
Ainsworth, M. D. S. (1963). The development of infant-mother interaction among the Ganda. Determinants of Infant
Behavior, 2, 67–112.
Ainsworth, M. D. S. (1964). Patterns of attachment behavior shown by the infant in interaction with his mother.
Merrill-Palmer Quarterly of Behavior and Development, 10, 51–58.
Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the growth of love. Baltimore: Johns Hopkins
University Press.
Ainsworth, M. D. S. (1983). Mary D. Salter Ainsworth. In A. N. O’Connell & N. F. Russo (Eds.), Models of
achievement: Reflections of eminent women in psychology (pp. 200–219). New York: Columbia University
Press.
Ainsworth, M. D. S. (1988). Security. Unpublished discussion paper prepared for the Foundations of Attachment
Theory Workshop. Convened for the New York Attachment Consortium by G. Cox-Steiner & E. Waters, Port
Jefferson, NY.
Ainsworth, M. D. S. (2010). Security and attachment. In R. Volpe(Ed.), The secure child: Timeless lessons in parenting
and childhood education (pp. 43–53). Charlotte, NC: Information Age.
Ainsworth, M. D., & Ainsworth, L. H. (1958). Measuring security in personal adjustment. Toronto: University of
Toronto Press.
Ainsworth, M. D., & Bell, S. M. (1969). Some contemporary patterns of mother-infant interaction in the feeding
situation. In A. Ambrose (Ed.). Stimulation in Early Infancy. London: Academic Press, 133–170.
Ainsworth, M. D. S., & Bell, S. M. (1970). Attachment, exploration, and separation: Illustrated by the behavior of
one-year-olds in a strange situation. Child Development, 41, 49–67.
Ainsworth, M. D. S., & Bowlby, J. (1991). An ethological approach to personality development. American Psychol-
ogist, 46, 333.
Ainsworth, M. D. S., & Marvin, R. S. (1995). On the shaping of attachment theory and research: An interview with
Mary D.S. Ainsworth (Fall 1994). Monographs of the Society for Research in Child Development, 60, 3–21.
Ainsworth, M. D. S., & Wittig, B. A. (1969). Attachment and exploratory behaviour of one-year-olds in a strange
situation. Determinants of infant behaviour. London: Methuen.
Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. (1971). Individual differences in strange-situation behaviour of
one-year-olds. In Shaffer, H. R. (Ed.), The origins of human social relations. London & New York: Academic
Press.
Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of
the strange situation. Hillsdale, N.J.: Lawrence Erlbaum.
Arsenian, J. M. (1943). Young children in an insecure situation. Journal of Abnormal and Social Psychology, 38, 225.
Bayley, N. (1932). A study of the crying of infants during mental and physical tests. Pedagogical Seminary and
Journal of Genetic Psychology, 40, 306–329.
Beekman, D. (1977). The mechanical baby: popular history of the theory and practice of child raising. Westport, CT:
Lawrence Hill.
Bell, S. M. (1970). The development of the concept of object as related to infant-mother attachment. Child Develop-
ment, 41, 291–311.
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
AINSWORTH’S STRANGE SITUATION PROCEDURE 283
Berg, I., Stark, G., & Jameson, S. (1966). Measurement of a stranger’s influence on the behaviour of young children
with their mothers. Journal of Child Psychology and Psychiatry, 7, 243–250.
Bijou, S. W., & Baer, D. M. (1960). The laboratory-experimental study of child behavior. In P. H. Mussen (Ed.),
Handbook of research methods in child development. New York: Wiley.
Blatz, W. E. (1944). Understanding the young child. London: University of London Press.
Blatz, W. E. (1966). Human security: Some reflections. London: University of London Press Limited.
Blatz, W. E., Millichamp, M. A., & Fletcher, M. (1936). Nursery education: Theory and practice. New York: William
Morrow and Company.
Bowlby, J. (1952). Maternal care and mental health: A report prepared on behalf of the World Health Organization
as a contribution to the United Nations programme for the welfare of homeless children. Geneva: World Health
Organization.
Bowlby, J. (1958). The nature of the child’s tie to his mother. International Journal of Psychoanalysis, 39, 350–373.
Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York: Basic Books.
Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation: Anxiety and anger. New York: Basic Books.
Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss: Sadness and depression. New York: Basic Books.
Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychol-
ogy, 28, 759.
Bretherton, I. (2003). Mary Ainsworth: Insightful observer and courageous theoretician. Portraits of Pioneers in
Psychology, 5, 317–331.
Bretherton, I., & Munholland, K. A. (1999). Internal working models in attachment relationships: A construct revisited.
In J. Casidy & P. R. Shaver (Eds.), Handbook of attachment. Theory, research, and clinical applications (pp.
89–111). New York: The Guildford Press.
Collard, R. R. (1968). Social and play responses of first-born and later-born infants in an unfamiliar situation. Child
Development, 39, 325–334.
Cox, F., & Campbell, D. (1968). Young children in a new situation with and without their mothers. Child Development,
39, 123–131.
Flint, B. M. (1959). The security of infants. University of Toronto Press.
Flint, B. M. (1974). The Flint Infant Security Scale. Toronto: Toronto University Press.
Groh, A. M., Fearon, R. P., Bakermans-Kranenburg, M. J., van IJzendoorn, M. H., Steele, R. D., & Roisman, G. I.
(2014). The significance of attachment security for children’s social competence with peers: A meta-analytic
study. Attachment & Human Development, 16(2), 103–136.
Grossmann, K. E., Bretherton, I., Waters, E., & Grossmann, K. (2013). Maternal sensitivity: observational studies
honoring Mary Ainsworth’s 100th year [Special issue]. Attachment & Human Development, 15, 443–447.
Hagman, E. R. (1932). A study of fears of children of pre-school age. Journal of Experimental Education, 1, 110–130.
Harlow, H. F. (1958). The nature of love. American Psychologist, 13, 673–685.
Harlow, H. F., & Zimmerman, R. R. (1959). Affectional response in the infant monkey. Science, 130, 12.
Heathers, G. (1954). The adjustment of two-year-olds in a novel social situation. Child Development, 25, 147–158.
Holmes, J. (1993). John Bowlby and attachment theory. London: Routledge.
Hulbert, A. (2003). Raising America: Experts, parents, and a century of advice about children. New York: Alfred A.
Knopf.
Isaacson, K. L. (2006). Mary Ainsworth and John Bowlby:the development of attachment theory. Retrieved September
8, 2010 from Proquest on the World Wide Web: http://www.proquest.com/products-services/dissertations
Jersild, A. T., & Holmes, F. B. (1935). Methods of overcoming children’s fears. Journal of Psychology, 1, 75–104.
Karen, R. (1990). Becoming attached. Atlantic Monthly, 265, 35–70.
Karen, R. (1994). Becoming attached: First relationships and how they shape our capacity to love. New York: Oxford
University Press.
Kenny, D. T. (2013). Bringing up baby: The psychoanalytic infant comes of age. London: Karnac Books.
Klopfer, B., Ainsworth, M., Klopfer, W. G., & Holt, R. (1954). Developments in the Rorschach technique, Volume I:
Technique and theory. New York: Harcourt, Brace & World.
Krout, M. H. (1935). Autistic gestures: An experimental study in symbolic movement. Psychological Monographs:
General and Applied, 46, i-126.
Lamb, M. E., Thompson, R. A., Gardner, W. P., Charnov, E. L., & Estes, D. (1984). Security of infantile attachment
as assessed in the “strange situation”: Its study and biological interpretation. Behavioral and Brain Sciences, 7,
127–147.
Lewin, K. (1935). A dynamic theory of personality: Selected papers (D. K. Adams & K. E. Zener, Trans.). New York:
McGraw.
Myers, R. (1969). Interview of Mary Ainsworth. Oral history of psychology in Canada. Retrieved from
www.feministvoices.com/mary-ainsworth/
Prichard, E., & Ojemann, R. H. (1941). An approach to the measurement of insecurity. Journal of Experimental
Education, 10, 114–118.
Rehn, T., McGowan, R. T., & Keeling, L. J. (2013). Evaluating the Strange Situation Procedure (SSP) to assess the
bond between dogs and humans. PLoS One, 8, e56938.
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
284 LENNY VAN ROSMALEN, REN´
E VAN DER VEER, AND FRANK VAN DER HORST
Rheingold, H. L. (1969). The effect of a strange environment on the behavior of infants. Determinants of Infant
Behavior, 4, 137–166.
Rheingold, H. L., & Eckerman, C. O. (1969). The infant’s free entry into a new environment. Journal of Experimental
Child Psychology, 8, 271–283.
Richardson, T. R. (1989). The century of the child: The mental hygiene movement & social policy in the United States
and Canada. Albany: State University of New York Press.
Riley, D. (1983). War in the nursery. Theories of the child and mother. London: Virago Press.
Rosenthal, M. K. (1967a). Effects of a novel situation and of anxiety on two groups of dependency behaviours. British
Journal of Psychology, 58, 357–364.
Rosenthal, M. K. (1967b). The generalization of dependency behaviour from mother to stranger. Journal of Child
Psychology and Psychiatry, 8, 117–133.
Rosenzweig, S. (1994). The historic expedition to America (1909): Freud, Jung and Hall the King-maker. St. Louis:
Rana House.
Ross, D. (1972). The psychologist as prophet. Chicago, London: The University of Chicago Press.
Rudnytsky, P. L. (1997). The personal origins of attachment theory: An interview with Mary Salter Ainsworth.
Psychoanalytic Study of the Child, 52, 386–405.
Salter, M. D. (1939). The concept of security as a basis for the evaluation of adjustment. Unpublished dissertation.
Schaffer, H. R., & Emerson, P. E. (1964). The development of social attachments in infancy. Monographs of the
Society for Research in Child Development, 29, 1–77.
Schwarz, J. C. (1968). Fear and attachment in young children. Merrill-Palmer Quarterly of Behavior and Development,
14, 313–322.
Shirley, M., & Poyntz, L. (1941). The influence of separation from the mother on children’s emotional responses.
Journal of Psychology, 12, 251–282.
Shirley, M. M. (1942). Children’s adjustments to a strange situation. Journal of Abnormal and Social Psychology, 37,
201.
Solomon, J., & George, C. (2008). In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment. Theory, research,
and clinical applications (pp. 383–416). New York: The Guildford Press.
Stearns, P. N. (2003). Anxious parents: A history of modern childrearing in America. New York, London: New York
University Press.
Stevenson, H. (1998). SRCD oral history interview—Mary Ainsworth. Retrieved from www.feministvoices.com/
mary-ainsworth/
Suomi, S. J., Van der Horst, F. C., & Van der Veer, R. (2008). Rigorous experiments on monkey love: An account of
Harry F. Harlow’s role in the history of attachment theory. Integrative Psychological and Behavioral Science,
42, 354–369.
Tennes, K. H., & Lampl, E. E. (1966). Some aspects of mother-child relationship pertaining to infantile separation
anxiety. Journal of Nervous and Mental Disease, 143, 426–437.
Van der Horst, F. C. (2011). John Bowlby: From psychoanalysis to ethology. Chichester, UK: John Wiley & Sons.
Van der Horst, F. C., LeRoy, H. A., & Van der Veer, R. (2008). “When strangers meet”: John Bowlby and Harry
Harlow on attachment behavior. Integrative Psychological and Behavioral Science, 42, 370–388.
Van Dijken, S. V. (1998). John Bowlby: His early life: A biographical journey into the roots of attachment theory.
London: Free Association Books.
Van Rosmalen, L., Van der Horst, F. C., & Van der Veer, R. (2012). Of monkeys and men: Spitz and Harlow on the
consequences of maternal deprivation. Attachment & Human Development, 14, 425–437.
Van Rosmalen, L., Van IJzendoorn, M. H., & Van der Veer, R. (2014). Measuring attachment: The Ainsworth security
questionnaires revisited. Unpublished manuscript.
Vicedo, M. (2013). The nature and nurture of love: From imprinting to attachment in Cold War America. Chicago:
University of Chicago Press.
Wooldridge, A. (1994). Measuring the mind: Education and psychology in England c. 1860-c. 1990. Cambridge:
Cambridge University Press.
Wootton, B. (1959). Social science and social pathology. London: George Allen & Unwin.
Wright, M. J. (1996). William Emet Blatz: A Canadian pioneer. In G. A. Kimble, C. A. Boneau, & M. Wertheimer
(Eds.). Portraits of pioneers in psychology (Vol. 2, pp. 199–212). Washington, DC: American Psychological
Association/Erlbaum.
JOURNAL OF THE HISTORY OF THE BEHAVIORAL SCIENCES DOI 10.1002/jhbs
