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

Instructions for Coding and Tabulating Frequency of Behaviors
Appendix II

Instructions for Coding

General

Coding is undertaken for each 15-second time interval, for each episode separately. The frequency of the behavior is measured in terms of the number of time intervals in which it occurs.

When the account of more than one observer is used, it may be found that what one observer reports at the end of one time interval a second observer may report at the beginning of the next time interval. The rule for coding is to assign the score to the time interval in which it was first reported by one observer.

In general, if a behavior item is reported by one observer but not by another, it is assumed to have occurred and will be coded. The chief discrepancies between reports tend to be in respect to direction of visual regard and facial expression, behaviors that one observer may be able to see but not the other. Sometimes, however, an observer intent on one aspect of the baby’s behavior may not see a relevant bit of behavior that another observer picks up, or may see it but not squeeze it into his verbal description. There are occasional discrepancies between reports that cannot be dealt with by this rule of thumb, however. In these instances the coder must exercise judgment about which report to use in his coding. The following rules will be observed in regard to demarcation of the episodes. Episode 2 begins when B has been put down (whether or not the observer has left). Episodes 3 and 7 begin as soon as the stranger enters, and Episodes 5 and 8 begin when the mother actually comes in the door. Episodes 4 and 6 begin when the mother actually goes out the door. Mother crossing the room belongs to Episodes 3 and 5. Episode 8 ends when someone comes in to tell the mother the situation is over. Code until that point. In the case of an incomplete time interval at the beginning or end of an episode, behavior occurring in it will be coded: (1) as part of the coding of the time interval immediately following (or, in the case of the end of the episode, preceding) if the part interval is clearly less than half of a 15-second standard interval; or (2) as a separate interval if the part interval is half or more of a standard 15-second interval.

Preparation of the Coding Sheet

An 11″ × 8½″ page of squared paper will be divided into columns, headed:

1. Episode and time interval
2. Locomotion
3. Body movement^*
4. Body posture^*
5. Hand movements
6. Visual regard
7. Location
8. Contact (Adult)^*
9. Contact (Baby)^*
10. Crying
11. Vocalization
12. Oral behavior
13. Smiling
14. Remarks

Each row will be devoted to the coding of behaviors occurring in one time interval. The time intervals will be numbered from the beginning of each episode. The end of each episode will be indicated by a line drawn across the page. The three sections of Episode 3 will also be divided by lines drawn across the page.

The coding symbols and instructions for using them are presented in the following sections. Note: It is helpful to refer to the tabulation rules while coding.

Coding of Locomotion

The mode of locomotion will be indicated as:

The objective of the locomotion (if any) will be indicated by an arrow; W → M = walk to mother. The objectives will be indicated as:

Avoidance behavior will also be coded, when B clearly moves away from a person—whether the stranger or the mother in the reunion episodes—not merely happening to move away from the person in the course of approaching a toy. The person avoided will be indicated by an arrow, thus: W S →. In Episode 3 the baby may at the same time avoid the stranger and approach the mother. This will be coded as follows: W S → M.

To prepare for tabulation, we distinguish exploratory locomotion from other locomotion. Exploratory locomotion is locomotion in the course of which a baby crawls, creeps, walks, or hitches about, either to explore the physical environment including the toys, or merely to engage in the activity itself. Included are such items as: Cp → T; Cp (E); Cp → D (in Episodes 2, 3, 5, or 8); Cp → T; or Cp → T_s, provided that the approach seems to be to the toy that the adult is manipulating and not to the adult herself.

All locomotions other than exploratory locomotions will be circled, so that they may subsequently be identified easily and excluded from the tabulations of exploratory locomotion. To be excluded are approach and following behavior (e.g., Cp → M; Cp → S), search behavior in the separation episodes (e.g., Cp → D or Cp → Ch_M), or random locomotion when a baby is acutely distressed.

Occasionally an observer may have failed to describe locomotion, but locomotion may be inferred from the coding of B’s location. Thus, if B is said to be in square 5 in one time-interval and in square 7 in another, it may be assumed that there was locomotion, unless, of course, he was moved by M or S. Inferred locomotion will also be noted.

Body Movement

Body movement refers to movement not involving locomotion. Coding of such movements are included for descriptive completeness but, to date, have not been used in statistical comparisons. Suggested coding symbols are:

For some movements the objective will also be indicated. Thus L_M = leans toward mother.

Coded items that are clearly social in their implications, or are clearly related to separation anxiety, to response to reunion, or that otherwise imply emotional disturbance will be circled in order to draw attention to them. Thus may indicate that B is rocking in distress, or bouncing with delight upon mother’s return.

Body Posture

Body posture will be recorded for time intervals in which neither locomotion nor body movement is recorded. Suggested symbols are:

Again, coded items that are clearly related to emotional disturbance will be circled. Thus: .

Hand Movements

Hand movements with respect to persons or objects will be coded, and those related to persons will be represented by circling the coded symbol. Suggested symbols are:

The chief coding difficulties are encountered in distinguishing between fine and gross manipulatory movements—that is, between FM and GM. The distinction is difficult to make for 1-year-olds, although it tends to be clear enough with older children. When in doubt, code GM.

It is assumed that T, F, G, FM, GM, and R refer to toys unless otherwise specified. If the object is a person, this should be indicated. Thus:

Unusual objects, such as the doorstop, books, and so on, may be indicated by a jotted note; or if they are clearly part of the whole pattern of exploratory manipulation of the environment, they may be considered equivalent to toys, and no subscript or note added.

Hand movements with a subscript referring to persons will be circled, for these are generally considered social and are not included in the statistical frequency count of exploratory manipulation. Fingering of one’s own body or clothing may frequently be considered a tension movement; if this is the case, the item is circled.

Some babies suck and chew toys or otherwise put them to their mouths. When they grasp such an object to take it to or from the mouth, this is not to be coded G, but rather to be considered part of the oral activity and coded in another column.

It may be noted that only uncircled R, G, GM, and FM are included in the statistical count for exploratory manipulation. T and F are excluded. Nevertheless they are retained in the coding for descriptive purposes—they suggest that B is being very tentative or muted in his attempts to explore manually.

Visual Regard

The critical distinction here is whether the visual regard is directed toward a toy or some other aspect of the physical environment in an exploratory sense, or whether it is directed toward a person or some aspect of the physical environment, such as the door or mother’s chair during her absence, which may be assumed to be associated with the person.

The coding of visual regard sometimes depends on inference. Thus if the baby is moving toward a toy, or manipulating it, it is usually a reasonable inference that he is also looking at it, or if he is approaching a person he is also looking at her. Special subscripts will be used, however, when a person directs the baby’s visual regard:

The distinctions between T_M and M_M between T_S and S_T are sometimes difficult to make, but they are important. M_T and S_T will be used when the baby seems to be looking both at the adult and at what the adult is doing with a toy, and not merely at the toy to which the adult is trying to attract attention. Subsequently, M_T will be tabulated as visual regard toward the mother and S_T as visual regard toward the stranger. T_M and T_S will not, however, be tabulated as visual exploration, for it is assumed that the visual regard of the toy was induced by the action of the adult and therefore lacks the spontaneity characteristic of visual exploration. (Similarly, induced visual regard directed toward some other feature of the physical environment, as, for example, when the adult distracts an unhappy baby by showing him a poster on the wall, will be given a subscript—E_S—so that it is not later counted as visual exploration.)

If the baby looks at a toy at the instigation of the stranger, therefore, it is coded as T_S in the relevant 15-second time interval; but if the baby continues to look at it in a subsequent time interval, after the stranger has ceased manipulating it, then the coding is T. For example, a baby may look at a ball that the stranger is about to roll (T_S), but then may turn to look at the ball after the stranger has rolled it away (T). Or a baby may look at a toy that his mother offers (T_M), take the toy, and continue to look at it as he holds it himself (T).

If a baby looks both at an adult manipulating a toy and also clearly at the toy that she is manipulating, both S_T or M_T and T_S and T_M may be coded in the same time interval.

Although, as stated, it is often necessary to infer the direction of visual regard from what the observer reports the baby as otherwise doing, inference must not be stretched too far. Thus, for example, it cannot be inferred that a baby is looking at an adult because he is being held by her. The adult may be holding the baby up over her shoulder, or on her lap facing out, but even when the position is not specified, it can usually be assumed that the baby is not looking at the adult who is holding him. Therefore visual regard directed toward M or S when the baby is being held is coded only if the observer had specifically reported that the baby looked up at the adult or that he was held in a face-to-face confrontation.

Location

This coding refers to the location of the baby in the room with reference to the squares marked off on the floor. It is intended to indicate where a baby goes on his own initiative and when out of contact with an adult. The coding for each time interval is by the square in which the baby is during that time interval for example, 7 or 6/7 if he is on the line between 6 and 7, or

if he is at the point of intersection of the four squares 6, 7, 10, and 11. Movement from one square to another is indicated by arrows. Thus 8 → 11 → 14 indicates that the baby moved from square 8 to square 14 through square 11. Be sure to include all squares through which a baby moved to get from one point to another. If the observer(s) did not include all such squares in their dictation, the path of the baby may be reconstructed assuming that he took a direct line from the square in which he started to the square in which he ended up.

Other locations to be noted in the location column are:

If the baby is being held, that is all that it is required to note. It is not necessary to trace the location of the adult who is holding the baby.

Contact Adult

In this column is indicated the adult’s action in any time interval in which the adult and the baby are in physical contact. This column is descriptively useful, but is not used for any frequency measures. The following symbols may be used:

If there is any ambiguity about the identity of the adult, the subscript M or S may be added to the contact symbol.

Contact Baby

In this column indicate the baby’s action in any time interval in which he is in physical contact with an adult. It is included for descriptive purposes and is not represented in any frequency measure.

The following symbols may be used:

Crying

Six degrees of crying are distinguished and recorded in this column:

Observers may sometimes report “unhappy noises” when perhaps it would be better to have reported fussing or crying. This is probably because at the time they cannot see the baby’s face to notice whether or not there is a cry-face. If there is a single report of an unhappy noise or protest, especially if it is specified that there is no cry-face, code C₆. But if there is a series of unhappy noises, code this as fussing (C₃) or as crying (C₂). It may be noted that C₁, C₂, C₃, and C₄ are tabulated as “real crying,” whereas C₅ and C₆ are counted as “minimal crying.”

Additional codings that are useful for descriptive purposes are:

A lull presents some difficulty if it occupies a whole time interval. Then it is important to indicate whether the lull is a mere diminution in intensity; in this it would be better to code a shift—say, from C₂ to C₃ or from C₁ to C₂. In this way the crying behavior for the time interval in question can be scored as “real crying” even though there has been a reduction in intensity. Sometimes, if a baby is really screaming, he may let out one burst and then hold his breath for a while as though merely gathering his strength for the next burst, with his face meanwhile contorted and usually red. If such a silent period should occupy a whole 15-second time interval, the coding of crying must reflect the context of the sequence and not isolated fragments. Thus there is no sense in a sequence of C₄, C₄, and C₄. This is obviously intermittent crying or fussing, and thus should be recorded in successive time intervals as C₂, C₂, and C₂ or as C₃, C₃, and C₃.

For the purposes of the descriptive summary, in which response to separation is of interest, it is significant to note other manifestations of distress than crying. Although some or all of these may be noted in other columns, it is useful to highlight them by repeating the entry in the crying column. For example, note in conjunction with crying such behaviors as random crawling, hunching into an abandoned little bundle, burying the face against the floor, detached visual regard, or eyes closed. Similarly, “tension movements” such as hunching, stretching, rubbing the ear, rubbing the back of the neck, and the like may give the clear impression that they occur as part of an effort not to cry. If so, mention may be made of these in the crying column.

Vocalization

In this column merely enter a check mark for each time interval in which a baby emits a vocalization not codable as a cry. Wherever possible indicate by a subscript whether the vocalization seems clearly directed to M or to S. Thus ✓ indicates a mere vocalization, whereas ✓ M indicates a vocalization directed to the mother. Vocalizations occasionally include laughs; these are infrequent enough not to require a symbol, but may be represented in a written entry. Coughs will also be noted, but will not be included in the subsequent tabulation of vocalizations; it cannot be assumed that coughs are social or communicative, as vocalizations tend to be.

Oral

In this column indicate oral behavior of the sucking, chewing, mouthing, or mouth-movement variety, but neither vocalization nor kissing nor biting. (Vocalization will be recorded in the “vocalization” column, and kissing and biting are considered “interactive” and should be noted in the “remarks” or contact baby column.) Symbols for oral behavior are:

Do not record as oral putting a finger or toy “to the mouth” or moving it “toward the mouth.” The object has to go into the mouth to be counted.

Smile

In this column will be recorded a baby’s smiles. In lieu of check marks the following symbols will be used to indicate whether or not a smile was directed toward a person.

A mere “pleasant face” will not be coded as a smile, but “little smiles” and “half-smiles” may be so coded.

Remarks

Because the coding sheet has a generally descriptive use, as well as being a source of frequency tabulations, the use of the “remarks” column is encouraged. In it can be recorded behaviors or aspects of behavior that are not represented in the coding system, or highlights of behavior that were particularly striking.

Instructions for Tabulation

The following behaviors will be tabulated:

1. Exploratory locomotion
2. Exploratory manipulation
   • (a) fine manipulation
   • (b) gross manipulation
   • (c) total manipulation
3. Visual exploration
4. Visual orientation
   • (a) to Mother
   • (b) to Stranger
5. Change of location
6. Crying
   • (a) “Real” crying
   • (b) “Minimal” crying
   • (c) Total crying
7. Vocalization
8. Oral behavior
9. Smiling
   • (a) to Mother
   • (b) to Stranger
   • (c) All smiling

In this analysis we are concerned only with Episodes 2 through and including 8. Episode 1 is omitted because it is so brief.

It is assumed that each of the episodes consists of 12 time intervals of 15 seconds each. If an episode contains 13 or more time intervals (as is often the case, particularly with Episode 5), or fewer than 12 (as is the case if the episode was curtailed), the frequency measure will be prorated. Thus, for example, if Episode 4 lasted 3 minutes and 15 seconds, (i.e., 13 time intervals) the frequency for the episode would be expressed as 12/13 of the actual frequency. Or if Episode 6 lasted only 2 minutes (i.e., 8 time intervals), the frequency would be expressed as 12/8 of the actual frequency.

To prepare the tabulation sheet allow a double column for each of the behaviors listed above. In the left-hand column state the obtained frequency, and in the right-hand column state the prorated frequency (or repeat the obtained frequency if it does not need to be prorated). To assist in the prorating, you will find it convenient to have a column at the extreme left showing the number of time intervals in each episode.

Episode 3 presents a special problem, for it consists of three segments of 1 minute (4 time intervals) each. When prorating, each segment is to be dealt with separately, assuming each to have four time intervals. Thus, for example, if the third segment lasted 1 minute and 15 seconds, the frequency for that segment would be 4/5 of the obtained frequency. Once each segment has thus been dealt with, the corrected frequencies of the three segments are summed.

The details for tabulation of the behaviors are as follows:

Exploratory Locomotion

This is locomotion in the course of which a baby creeps about (or crawls, walks, or hitches), either to explore the physical environment, including the toys, or merely to perform the activity itself.

Included are:

1. luiLocomotion to the toys (e.g., Cp → T), even to one that is being manipulated, offered, or shown by an adult (e.g., W → T_S).
2. Locomotion to any other aspect of the physical environment (except for the door or the mother’s chair in the separation episodes).
3. Locomotion for its own sake—for example, Cp → (E).

Excluded are the following:

1. Approach to a person (e.g., Cp → M, W → S).
2. Following a person (e.g., Cp → M).
3. Search behavior in separation episodes (e.g., W → D, Cp → Ch_M).
4. Moving away from a person in a way that is interpreted as avoidance (e.g., Cp S →).
5. Random locomotion shown by some babies while acutely distressed. This tabulation is easier if all behaviors to be excluded had been circled in the course of coding them.

Exploratory Manipulation

This refers to hand movements that are exploratory—that is, directed to toys or some aspect of the physical environment with manipulatory or exploratory intent (or, indeed, in play); it is intended to exclude hand movements that are part of social interaction or physical contact, or that are merely expressive of affect or are random.

Included in the FM column is:

Included in the GM column are:

The figure in the total column is not the sum of the items in the FM and GM columns; it is the total number of time intervals in the episode in which either FM or GM (or R or G) or both occur.

Excluded from the tabulation of exploratory manipulation are:

1. Grasping (G) or reaching (R) for a person or a part of her clothing.
2. T = merely touching an object, even a toy; this is considered too tentative to be exploratory manipulation.
3. F = fingering an object, except when it is clearly a mode of exploring a textured surface. Fingering is often a manifestation of tension and not exploratory.
4. GM_A = implies banging, pushing, or throwing done aggressively, in petulance or anger, and not in exploration.
5. Offering or showing toys to another person, or taking toys offered by another person.
6. Any hand movement directed toward the door, the mother’s chair, or mother’s handbag in any separation episode.
7. Any hand movement, whether fingering or actual manipulation, that the baby directs toward his own person or clothing.
8. Any sustained holding on to a toy without manipulating it in any way. (Many babies at this age will hold a small toy in one hand for long periods, without looking at it and without manipulating it, even holding on to it while creeping.)
9. Any hand movement, even G or GM, undertaken while crying hard, for this does not seem to be exploratory.
10. Ge = gesture made with the hand, even while holding a toy, when B is not actually manipulating the toy.

Again, it is obvious that tabulation is facilitated if items to be excluded as exploratory manipulation are circled in the course of coding.

Visual Exploration

This refers to behavior in which the baby spontaneously looks at (explores visually) a toy or some other aspect of the physical environment, such as the doorstop, the furniture, or pictures on the wall, or if he looks around the room generally.

Included are:

Excluded are:

1. Looking at a person.
2. Looking at the door, the chair, or the mother’s handbag in any of the separation episodes.
3. Looking at a toy (T_M or T_S) at the instigation of another person.
4. Looking at the window mirror, the pictures, and the like at the instigation of another person.

Visual Orientation

The interest is in the proportion of time spent in looking at the mother or at the stranger, rather than at the toys or other aspects of the physical environment. The frequency counts are made separately for the mother and for the stranger. A score of 1 is given for each time interval in which the baby looks at the mother, whether or not he has also looked at the stranger, at the toys, or at anything else in the same time block. If he glances twice or more at the mother in the same time block, the score is still 1. A similar procedure is to be followed for tabulating frequency of looking at the stranger. The frequency of looking at the toys or at other aspects of the physical environment is identical with the visual-exploration score that has already been obtained.

Included are:

Excluded are:

Crying

Three columns are provided for crying in the tabulation sheet:

Real crying includes all crying coded as 1,2, 3, or 4. A score of 1 will be given for each time interval in which crying 1, 2, 3, or 4 occurs, even though there may also be a lull (L), stopping crying (S/C), or crying coded as 5 or 6 in the same time interval.

Minimal crying includes crying coded as 5 or 6. A score of 1 is given for each time interval in which such crying occurs, provided there is no cry in the same time interval that has been coded 1, 2, 3, or 4.

Total crying includes all crying—1,2, 3,4,5, and 6. It is the sum of the other two frequency figures.

If an episode was curtailed because the baby was unduly distressed, it may be assumed that he would have continued to cry throughout all remaining time intervals had the episode not been curtailed. Therefore, instead of prorating such episodes, the time intervals eliminated by the curtailment will be counted as “real crying.” Thus if the baby began to cry after 30 seconds had elapsed and continued until the episode was curtailed at 60 seconds, he would receive a score of 10 for crying in that episode.

Vocalization

Because vocalizations at age 1 tend to be discrete rather than continuous (as crying tends to be), and because they are relatively infrequent (in comparison with looks), absolute frequency of occurrence in each episode may be more useful than the frequency measure based on the 15 second time intervals. In either case, include laughs as vocalizations, but not coughs. Score separately vocalizations, directed to the mother, to the stranger, and total vocalizations, including those not clearly directed toward a person.

Smiling

For the same reasons indicated for vocalization, absolute frequency of occurrence of smiling in each episode may be more useful than the frequency measure based on the 15-second time intervals. Score separately smiles to the mother (M), to the stranger (S), and the total number of smiles, including those not clearly directed toward a person (E).

Oral Behavior

This includes oral behavior that is of an autoerotic, tension-reducing variety. It also includes chewing or biting that might be exploratory. In the case of a 1-year-old, it seems age-inappropriate to explore by oral means, and it seems likely that such behavior has at least an element of the autoerotic and tension reducing in it. We have also included here the rare instances in which the baby sucks on the nipple of a bottle or pacifier that has been provided for him.

Score all time intervals, therefore, in which an F, T, P, B, or M has been recorded.

Note

* Behaviors included under these headings have not yet been used by us in frequency measures. Recent ethological studies of young humans suggest that bodily movements and orientation, as well as hand gestures and facial expression, are noteworthy social behaviors. We therefore did not omit body movement and posture from this set of instructions. As for the contact items, they are useful descriptively, even though they represent some overlap with the system for scoring interactive behavior.