Procedures - Method

Patterns of Attachment: A Psychological Study of the Strange Situation - Mary D. Salter Ainsworth 2015



The subjects come from white, middle-class, Baltimore-area families, who were originally contacted through pediatricians in private pratice. They were observed in the strange situation at approximately 1 year of age. The total N of 106 infants is comprised of four samples that were observed in the course of four separate projects. (See Table 1.)

Sample 1

TABLE 1 Description of the Four Samples

Table 1 Description of the Four Samples

Sample 1 was observed in the course of a longitudinal study of the development of infant—mother attachment throughout the first year. Twenty-six infants were visited at home at intervals of 3 weeks from 3 to 54 weeks of age. In lieu of or to supplement the second-to-last visit at 51 weeks, the babies were brought into the laboratory for the strange situation. Three subjects were dropped from the strange-situation sample, the first because he was 64 weeks of age by the time the strange situation was devised, the other two because they were ill when brought to the strange situation.

Sample 2

Sample 2 was observed in the course of short-term longitudinal research into the development of the concept of the object as related to infant—mother attachment (Bell, 1970). Thirty-three babies were given cognitive tests at home three times between the ages of 8½ and 11 months, and 1 week after the third testing session they were observed in the strange situation in the laboratory.

Sample 3

Sample 3 was especially assembled for a study of the effect on strange-situation behavior of repetition of the procedure with an interval of 2 weeks between the first and second session. Twenty-four babies were observed, but one was discarded from the sample because, after he had cried throughout the entire first session, it was discovered that he was terrified of an electric fan in the experimental room, and, indeed, of all noise-producing appliances. Because this was a specific fear, and not merely fear of the strange, his record was considered atypical and was discarded.

Sample 4

Main (1973), in a short-term longitudinal study on the relation between infant—mother attachment and later exploration, play, and cognitive function, used Sample 3 (first session) and added to it 27 infants who constitute Sample 4.

The infants of Samples 1 and 2 were observed at home before the strange-situation procedure was experienced, whereas those of Samples 3 and 4 were not, except for one home visit that served also to explain the procedure to the mother and to gain her informed consent. For each sample, however, the strange situation was the occasion for the first visit to the laboratory. It may be noted in Table 1 that the four samples differ somewhat in median and range of age. For all major analyses of data, differences among the four samples were tested for significance.

The Strange-Situation Procedure

The strange situation consists of eight episodes presented in a standard order for all subjects, with those expected to be least stressful occurring first. After a brief introductory episode, the baby was observed with his mother in the unfamiliar, but not otherwise threatening, environment of the experimental room, to see how readily he would move farther away from her to explore a novel assembly of toys. While the mother was still present, a stranger entered and made a very gradual approach to the baby. Only after this did the mother leave, because it was anticipated that separation from her would constitute a greater stress than the presence of a stranger and/or of an unfamiliar environment per se. After a few minutes the mother returned and the stranger slipped out. The mother was instructed to interest her baby in the toys again, in the hope of restoring his exploratory behavior to the baseline level characteristic of when he was previously alone with his mother. Then followed a second separation, and this time the baby was left alone in the unfamiliar environment. As some check on whether any increased distress was a response to being alone rather than to have been separated a second time, and also to ascertain whether separation was more distressing than the presence of a stranger, the stranger returned before the mother finally returned.

There are undoubtedly other sequences of episodes that would be interesting to study, and there may be others that would have been equally or more effective in evoking the responses for which the situation was designed. But as it turned out, the sequence just summarized was very powerful both in eliciting the expected behaviors and in highlighting individual differences. The sequence of episodes is described in more detail as we proceed.

The Physical Situation

Two adjacent rooms were employed for the experimental room and the observation room, connected by two one-way-vision mirror windows. The experimental room was furnished, not bare, but was so arranged that there was a 9-by-9 foot square of clear floor space. For the first 13 subjects of the Sample 1, the floor was covered by a braided rug, but for the last 10 subjects and for all subsequent samples, the mastic tile floor was bare but marked off into 16 squares to facilitate recording of location and locomotion. For Samples 1 and 2, the furnishings approximated those of a university office, with desk, chair, and a bookcase at one side of the room. Bright postcards were tacked around the periphery of the mirror windows. In the period between Samples 2 and 3, the office furniture was moved out and replaced by metal storage cabinets. The postcards had been removed, but colorful posters were tacked to three walls of the room.

Film records were made of the last 10 subjects of Sample 1, as well as of the babies of Sample 3. For the purpose of filming, a glass-covered photography port was put in the wall opposite the observation windows, and sun-gun lights were introduced high in the room. To obscure the noise of the camera, as well as to counter the heat from the bright lights, an electric fan was placed on the bookcase (later on top of the cabinets).

At one end of the experimental room (see Figure 1) was a child’s chair heaped with and surrounded by toys. Near the other end of the room in square 16 was a chair for the mother, and on the opposite side in square 13 was a chair for the stranger. The baby was put down on the line between squares 14 and 15, facing the toys, and left free to move where he wished.

Figure 1 Sketch of the physical arrangements of the strange situation. (Adapted from Bretherton & Ainsworth, 1974.)

FIGURE 1 Sketch of the physical arrangements of the strange situation. (Adapted from Bretherton & Ainsworth, 1974.)

This much attention has been paid to a description of the physical arrangement of the experimental room because even minor variations seemed to affect babies’ behavior. For example, the desk and bookcases attracted more exploratory interest in Samples 1 and 2 than did the cabinets in Samples 3 and 4. More important, it seems likely that the position of the door on the stranger’s side of the room may have affected the likelihood of a baby’s approaching it when the stranger was present. Furthermore, the arrangement of the room in orientation to the observation windows obviously affected what sequences of behavior and facial expression the observers were able to see most clearly. They had a good view of a baby’s face as he approached either the mother’s or stranger’s chair, a profile view (at best) of a baby oriented to the door or to a person entering, but only a back view when the baby was approaching the child’s chair and the heap of toys.

Either one or two observers (more frequently and preferably two) dictated a play-by-play account into Stenorettes of what the baby did, and as much as possible also of what the adult(s) did. The Stenorette microphones also picked up the sound of a buzzer that marked off 15-second time intervals. The observers wore earphones that both enabled them to hear what went on in the experimental room and prevented them from hearing each other’s dictation. An intercom system also made sounds from the experimental room audible in the observation room. This system was not reversed to give instructions to the adults in the experimental room, lest a disembodied voice alarm the infants. Predetermined signals were given by knocks on the wall. On the few occasions when special intervention was necessary, someone went to the door of the experimental room to deliver the message directly.


The usual number of personnel included two observers (O1 and O2), a stranger (S), and an experimenter (E). It was E’s task to time the episodes and to give cues to the mother and stranger that determined their entrances and exits. Whenever possible a fifth person received the mother and baby upon their arrival, reviewed the instructions (of which the mother had a copy and that had previously been discussed with her at a home visit), and introduced them to the experimental room; otherwise either O2 or E did this. The irreducible minimum of personnel (used in Sample 2) was one observer and a second person to act as both E and S.

A necessary complication of the procedure is that separation episodes were curtailed if a baby became so distressed that he clearly would continue to cry throughout an episode of standard duration. Although it is obviously undesirable to allow a baby to become unduly distressed, an effort was made not to curtail episodes unnecessarily, for some babies may protest briefly and then settle down either to play or to search for the mother, or both. Sometimes it is also desirable to prolong an episode. Thus, for example, the first reunion episode was sometimes prolonged so that a baby could fully recover from distress occasioned by the first separation and settle down again to play. Furthermore, should a baby make contact with his mother just before a signal is due for her to leave, the episode may be somewhat prolonged so that the mother’s departure does not constitute a direct rebuff to the baby. The responsibility for deciding when episodes should be curtailed or prolonged was usually delegated to E, if he were experienced enough, so as not to distract O from his primary task of observing.


The original set of toys used for Samples 1 and 2 were selected at a local toy shop and supplemented by other attractive objects, such as bangles, a shiny pie plate, and a long red tube. (See lists in Table 2.) For the two sessions that Sample 3 was to undergo, the original set of toys was divided in half, and each half was supplemented by new toys, mostly Creative Playthings, so that there was an entirely different array of toys in Session 2. Although it was likely that some of the toys were duplications of toys a baby had at home, it was assumed that the total array of toys would be novel enough to activate exploration.

Because so many of the toys were noise-makers, and because so many babies played banging games, it proved not feasible to tape the vocalizations of mother and baby in the experimental room. The observers could distinguish crying from noncrying vocalizations better at first-hand than from the tape. Thus the chief information that was lost by not making taped records was the precise content of some of the adults’ speech, which the observers found difficult to include exactly while dictating an account of all the action.

TABLE 2 Toys Used by Different Strange-Situation Samples

Samples 1 & 2

Sample 4 and First Session of Sample 3

Second Session of Sample 3

Large red ball

Chime ball

Large red ball

Chime ball

Plastic butterfly ball

Racing-car pull toy

Toy telephone

Musical clown

Raggedy Andy doll

Plastic shapes & sticks

Hammer-shaped rattle

Plastic milk bottle containing small objects

Silver bangles

Foil pie plate

Racing-car pull toy

Toy telephone

Raggedy Andy doll

Plastic shapes & sticks

Hammer-shaped rattle

Silver bangles

Foil pie plate

Thumper drum

Clutch ball

Baby shapes

Wooden chicken

Hammer-peg toy

Plastic butterfly ball

Musical clown

Plastic milk bottle containing small objects

Long red tube

Peg bus

Shape box


Baby doll

Toy iron

Wooden grasshopper

Wooden hand “mixer”

Long red tube

Episodes of the Strange Situation

The episodes of the strange situation are delineated in the following general instructions to the personnel—the observers, stranger, and experimenter. (Separate instructions were given to the mother in advance of her arrival at the laboratory, and are shown in Appendix I.) A summary of the episodes is given in Table 3.

Episode 1: Mother, Baby, and Experimenter. This is a very brief, introductory episode. M and B are introduced to the experimental room. M is shown where to put the baby down and where she is to sit after having put him down. M has been instructed to carry the baby into the room. Meanwhile the O notes the B’s response to the new situation from the safety of M’s arms. E leaves as soon as he has completed his instructions.1

Episode 2: Mother and Baby. M puts B down midway between S’s and M’s chairs (on the line between squares 14 and 15), facing the toys. She then goes to her chair and reads (or pretends to read) a magazine. It is expected that B will explore the room and manipulate the objects in it, especially the toys. M has been instructed not to initiate an intervention, although if B obviously wants a response from her, she is to respond in whatever way she considers appropriate.

TABLE 3 Summary of Episodes of the Strange Situation

Number of Episode

Persons Present


Brief Description of Action


Mother, baby & observer

30 secs.

Observer introduces mother and baby to experimental room, then leaves.


Mother & baby

3 min.

Mother is nonparticipant while baby explores; if necessary, play is stimulated after 2 minutes.


Stranger, mother, & baby

3 min.

Stranger enters. First minute: Stranger silent. Second minute: Stranger converses with mother. Third minute: Stranger approaches baby. After 3 minutes mother leaves unobtrusively.


Stranger & baby

3 min. or lessa

First separation episode. Stranger’s behavior is geared to that of baby.


Mother & baby

3 min. or moreb

First reunion episode. Mother greets and/or comforts baby, then tries to settle him again in play. Mother then leaves, saying “bye-bye.”


Baby alone

3 min. or lessa

Second separation episode.


Stranger & baby

3 min. or lessa

Continuation of second separation. Stranger enters and gears her behavior to that of baby.


Mother & baby

3 min.

Second reunion episode. Mother enters, greets baby, then picks him up. Meanwhile stranger leaves unobtrusively.

a Episode is curtailed if the baby is unduly distressed.

b Episode is prolonged if more time is required for the baby to become re-involved in play.

For 2 minutes M will direct B’s attention neither to the toys nor to other objects in the room. If, after 2 minutes, B has not begun to explore the toys, a signal is given to M (a knock on the wall) for her to take him to the toys and to try to stimulate his interest in them. One minute is allowed for this stimulated exploration. Meanwhile E times the episode, beginning when M puts B down. He signals M when 2 minutes are up if, in his judgment, B needs stimulation. When 3 minutes are nearly up, he cues S to go to the experimental room.

The focus of the observation is on the amount and nature of B’s exploration of the strange-situation—locomotor, manipulatory, and visual—and on the amount and nature of his orientation to M.

Episode 3: Stranger, Mother, and Baby. S (who has never met B before) enters and says to M: “Hello! I’m the stranger.” She immediately seats herself in S’s chair and remains silent for 1 minute. She may watch B, but should not stare at him if B seems apprehensive of her. At the end of 1 minute, E knocks on the wall to signal S to begin a conversation with M. M, meanwhile, has been instructed not to begin talking until S initiates interaction with B. At the end of another minute, S is signaled to initiate interaction with B. At the end of 3 minutes, E knocks to signal the end of the episode. At this signal M leaves the room unobtrusively, leaving her handbag behind on her chair and choosing a moment to leave when B seems occupied either with S or with the toys.

The focus of the observation is on how much and what kind of attention B pays to S, in comparison with the attention he pays to M or to exploration, and on how B accepts S’s advances.

Episode 4: Stranger and Baby. E begins to time the episode as soon as M leaves the room. M, meanwhile, comes to the observation room. As soon as M has gone, S begins to reduce interaction with B, so that B has a chance to notice that M has gone, if, indeed he had not already noticed. If B resumes exploring, S retreats to her chair and sits quietly as M did previously, although she is to respond to any advances B may make. We are primarily interested in the amount of exploring B will undertake in contrast with the amount he did when he was alone with M.

If, however, B cries, S will intervene, trying to distract B with a toy; if this fails to calm him, S will attempt to comfort B by picking him up if he permits and/or by talking to him. If S is successful in comforting B, she then puts him down and again attempts to engage his interest in the toys.

Three minutes are allowed for this episode, although it may be curtailed should B become highly distressed and unresponsive to S’s efforts to distract or comfort him. Just before 3 minutes are up (or sooner if the episode is to be curtailed), E cues M to return to the experimental room.

We are interested in the amount and nature of B’s exploration in contrast with earlier episodes. We are also interested in B’s response to M’s departure—crying, search behavior, and any acute distress. B’s response to the stranger is also of importance, including his response to being picked up and put down, and any clinging that he does.

Episode 5: Mother and Baby. M approaches the closed door and speaks outside, loudly enough that B can hear her voice. She pauses a moment, opens the door, and pauses again, to allow B to mobilize a response to her if he is going to. M is instructed to make the baby comfortable, finally settling him on the floor, and interesting him in the toys. Meanwhile S leaves unobtrusively. After 3 minutes, or when it is judged that B is settled enough to be ready for the next episode, M is signaled to leave. She picks a moment (if possible) when B seems cheerfully occupied with the toys, gets up, puts her handbag on her chair, and goes to the door. At the door she pauses and says “bye-bye” to B and leaves the room, closing the door securely behind her.

In general in this episode we are interested in observing B’s response to M after her absence and their interaction after her return.

Episode 6: Baby Alone. E begins timing when M leaves. Three minutes are allowed for B to explore the room while he is alone. If he cries when M departs, he is given a chance to recover in the hope that he may do some exploring, but if he cries hard for a full minute the episode is curtailed.

We are interested, of course, both in B’s exploratory play (if any) when he is left alone in an unfamiliar situation and in his reaction to his mother’s departure—crying, search behavior, grumbling vocalizations, tension movements, and so on.

Episode 7: Stranger and Baby. Just before the end of the 3 minutes (or upon a decision to curtail Episode 6), E cues S to return. S approaches the closed door and speaks outside, loudly enough that B can hear her voice. She pauses a moment, opens the door, and pauses again, to allow B to mobilize a response if he is going to do so. E begins timing Episode 7 as soon as S enters.

If B is crying, S will first attempt to soothe him, picking him up if he will permit it. When and if he calms, she will put him down and attempt to engage him in play. If he gets interested in the toys and begins to play, S will gradually retreat to her chair. If B is not distressed at the time S enters, she invites him to come to her. If B does not come, she approaches B and attempts to initiate play. If he becomes interested in the toys and begins to play with them himself, S will gradually retreat to her chair. In either case, if B signals that he wants interaction or contact with S, she will respond to his wishes, and in general she is to gear her behavior to B’s behavior.

In this episode we are interested primarily in B’s response to S—how readily he is soothed by her, whether he seeks or accepts contact, whether he will interact with her in play—and in how this response compares with B’s response to M in the reunion episodes. Also we are interested to see whether the pull of the toys is strong enough that B permits S to become nonparticipant.

Episode 8: Mother and Baby.2 Just before the end of 3 minutes (or upon a decision to curtail Episode 7), E cues M to return. M opens the door and pauses a moment before greeting B, giving him an opportunity to respond spontaneously. She then talks to the baby and finally picks him up. Meanwhile S leaves.

The Stranger and Her Behavior

Each of our four samples had a different stranger, and Sample 3 had a different stranger in Session 2 from Session 1. On occasion, substitutes were necessary, so that there were 10 strangers in all. All strangers were female.

The role of stranger is a difficult one. On the one hand, she is expected to refrain from undue intervention in order to permit the baby to play, search for his mother, or even display distress spontaneously. On the other hand, she is instructed in Episode 3 to approach the baby and to attract his attention away from the mother and to the toys, and in the separation episodes to distract or comfort the baby if he is distressed. All strangers to some extent geared their behavior to that of the baby, but they had individual styles in approaching the baby, in interacting with him in play, and in attempting to comfort him if he was distressed. A baby’s cries were distressing to all strangers, and consequently it was especially difficult to control, through instruction, just how they should behave in separation episodes.3 Should these differences in stranger behavior affect substantial differences in infant behavior, this should be reflected in intersample differences in the relevant episodes.

The Mother and Her Behavior

Each mother was instructed in advance about the purpose and procedures of the strange situation and about the role she was to play. This instruction almost invariably took place in the course of a home visit. The mother was then given a mimeographed set of instructions. (See Appendix I.) When she arrived at the laboratory, the instructions were again discussed, if she felt uncertain of them, and she was provided with a small card that summarized the episodes and the cues for which she was to be alert. Adequate advance briefing is considered important, so that the mother does not feel anxious or uncertain about her role in the situation.

The instructions were intended to control the mother’s behavior, especially in the preseparation episodes, in which it was desired to see what the baby would do spontaneously and without undue intervention from his mother. It was impossible, however to prevent all individual differences in maternal behavior from manifesting themselves. Indeed, little effort to control maternal behavior was exerted by the instructions covering the reunion episodes. It was recognized that maternal behavior would be much affected by individual differences in infant behavior, that it would be difficult to provide for such contingencies in the instructions, and that in any event mothers would tend to behave in their own characteristic ways in reunion.

There were two ways in which our instructions to the mother made the situation more artificial than we would have liked for a few infant—mother pairs, although this outcome seems unavoidable if there is to be any attempt at standardization. Some mothers reported feeling unduly constrained in the preseparation episodes. To them it seemed unnatural to put the baby down to play without first introducing him to the toys by playing with him briefly, and indeed their infants tended to look puzzled when the mother sat down in a nonparticipant role. For such infants the strange situation was perhaps stranger than for others. Second, the instructions for Episode 5 specified chiefly that the mother was to reinterest the baby in the toys. It was our distinct impression that some mothers would have spent more time in comforting and reassuring their infants had it not been for this instruction.

Despite the fact that individual differences in maternal behavior were somewhat smoothed out by our instructions, the strange situation yields a surprisingly large amount of information about the mother’s role in interaction although it was not intended to do so and although this report does not analyze maternal behavior. On the other hand, differences in maternal behavior, especially in the reunion episodes, required that measures of attachment be based on a scoring system that took into account the contingencies of maternal behavior.

Finally, we must emphasize that no apologies are offered for these difficulties. A tight control of maternal behavior is impossible and indeed undesirable. The compromise represented in our procedures turned out to have effected a reasonable degree of standardization of the situation, while allowing most mothers to behave naturally and fairly comfortably.

Training of Observers

The observers, except for the first two,4 were trained by apprenticeship. They first watched a number of subjects in the strange situation and listened to experienced observers dictate their narrative reports of the action. Then, in turn, each was permitted to act as O2 and later to check their narrative reports against those of O1. Meanwhile, the prospective observers familiarized themselves with the instructions for coding strange-situation behavior, so that they could appreciate the distinctions that need to be made in regard to the behaviors to be reported.

In general, observers were instructed to report in as much detail as possible what the baby did and what interaction he had with the mother and with the stranger. Specifically, they were instructed to report the following:

  • —all locomotion, tracing the baby’s progress by noting the identifying numbers of the squares entered;
  • —the baby’s posture and body orientation and changes therein;
  • —the toys and other objects he reached for, touched, or manipulated; the nature of his manipulation; and whether he put a toy down, dropped it, or threw it;
  • —any touching or rubbing of his own body or clothing;
  • —what he looked at, whether at the mother, stranger, toys, door, or other features of the physical environment; whether he glanced, watched, or stared; and whether he pointedly refused to look, or looked away;
  • —smiles and at what or whom the baby looked when he smiled;
  • —vocalizations and at what or at whom the baby looked when he vocalized;
  • —crying of all kinds and degrees, whether unhappy noises, cryface, fussing, clear-cut crying, or screaming, and when crying ceased;
  • —oral behavior, including putting objects (or fingers or thumb) into the mouth, or sucking or chewing them;
  • —in addition to looking, vocalizing, and smiling, any interaction with either mother or stranger across a distance, such as pointing to toys, showing toys, or offering toys;
  • —any locomotor approach behavior to mother or stranger, whether it was speedy or slow or interrupted, whether it was spontaneous or invited by the adult, whether it was partial or ended in close proximity or actual physical contact;
  • —any avoidance behavior, either of mother or stranger, especially upon the entrance or approach of either figure, including moving away, turning away, or looking away;
  • —if an adult and baby came into physical contact, who initiated the contact; whether the baby merely touched the adult or clambered up or held on; and whether the adult merely steadied the baby or picked him up;
  • —if the baby was picked up, did he cling, sink in, or hold on, or did he resist contact by pushing away, hitting, kicking, stiffening, or squirming to get down;
  • —if he was put down, who initiated the put-down, and did the baby protest it or actively resist release;
  • —initial response to the entrances of the mother or stranger were especially noted—whether the baby smiled, vocalized, cried, reached, leaned, or approached, or whether he ignored or avoided the person; similarly responses to the approach of either mother or stranger were especially noted;
  • —any angry or resistant behavior in any context, whether it is pushing away, throwing away, or dropping toys offered by an adult, or resistance when physical contact is offered, or tantrums;
  • —during separation episodes, whether the baby looked at, approached, banged on, or tried to open the door, and whether he looked at, approached, or touched his mother’s empty chair or her handbag.

Finally, the observers were advised to dictate as quickly as possible and to keep talking even when little new was occurring. Advice to “talk like a sports-caster” seemed helpful to apprentice observers.

Number of Observers and their Reliability

The observers differed from one sample to another. (See Table 4.) For Sample 2 there was only one observer. For the other samples there were always two observers, paired in a variety of combinations. For Sample 3 there were particularly many observers, both because there were two sessions of the strange situation and because emphasis was placed on training graduate students in the strange-situation procedure. Those who were to be observers of the second session for any given infant, however, were not even present at the first session.

A reliability check was undertaken with the four subjects observed jointly by Observers A and D. For this purpose their separately dictated accounts were coded separately for four frequency measures: locomotor exploration, exploratory manipulation, visual exploration, and crying. Product-moment coefficients of .99 were found for each of the exploratory behaviors, and one of .98 for crying. Further tests of interobserver agreement were not made, because (except for Sample 2, which had only one observer) our practice was to base all measures on the consolidated reports of the two observers.


The final transcription of the dictated accounts brought together the accounts of two observers. Each account was divided into separate paragraphs, one for each 15-second time interval, with the paragraphs for the two observers side by side to facilitate comparison. The only problem in preparing the transcriptions was to ascertain when one episode ended and the next began. The following rules were adopted. Episode 5 began with the entrance of the mother and ended when she went out the door. Episode 7 began with the entrance of the stranger and ended with the entrance of the mother.

Film Records

TABLE 4 Observers for the Four Samples


Number of Observers

Identity of Observers



A, B, C, D






A, E, F, G, H, I, J, K, L, M



F, G, H, I, N, O

A series of still photographs were taken of the first 13 infants in Sample 1, shot through the one-way mirrors. Cinematic records were made of the last 10 infants in Sample 1 and all of Sample 3 through a plate-glass photography port. It was impossible, however, to avoid breaks in continuity, for even with a camera equipped with an electric drive mechanism and with a cumbersome chamber holding a 400-foot reel of film, only 10 minutes of a 20-minute situation could be filmed. Furthermore, except when the infant and adult were near to each other, the camera could not pick up the behavior of both participants in an interaction. For these reasons the continuously dictated accounts of the observers were more useful than the film records for research purposes, although the latter are useful for illustrative purposes.

Videotape equipment was not available, but even had it been, dictated accounts would still have been used as a supplement to the videotape records in order to record sequences of interaction that the videotape could not pick up.


1 Here and elsewhere in these instructions, M stands for mother, B for baby, E for experimenter, O for observer, and S for stranger.

2 For the first 13 subjects of Sample 1, Episode 8 was terminated as soon as the initial reunion behavior was observed, perhaps after only 1 minute had elapsed. It soon became apparent, however, that this second reunion episode was important in its own right, and thenceforward observation was continued for a standard 3 minutes for subsequent subjects and samples. The brief duration of Episode 8 for the majority of Sample 1, however, made it impossible to use, in this episode for this sample, all of the measures that were used in the rest of the episodes.

3 Initially, for the first 13 cases of Sample 1, there were somewhat more complex instructions for the stranger in Episode 7 (see Ainsworth & Wittig, 1969). But so many infants were distressed in Episode 6 and continued to be distressed in Episode 7 that we shifted to the aforementioned instructions, which corresponded to the way in which the stranger had actually behaved in Episode 7 in respect to the first 13 infants.

4 Mary Ainsworth and Barbara Wittig.