Journal of Consulting and Clinical Psychology1996. Vol.64. No. 1.64-73

Copyright 1996 by the American Psychological Association, Inc.0022-006X/9643.00

Attachment Relationships Among Children With Aggressive BehaviorProblems: The Role of Disorganized Early Attachment Patterns

Karlen Lyons-RuthHarvard Medical School

This article reviews recent attachment-related studies of early aggression. That aggressive behavior

toward peers is related to disorganized or controlling patterns of attachment behavior toward parentsbut not to avoidant or ambivalent patterns. Longitudinal attachment studies indicate that risk factorsidentified in cross-sectional studies of aggressive school-age children, such as family adversity, paren-

tal hostility, parental depression, and child cognitive deficits, are already evident in infancy andpredictive of later aggression, before the onset of coercive child behavior. In infancy, these risk factorsare associated with disorganized attachment behaviors toward the caregiver characterized by signs of

fear or dysphoria, irresolvable conflict between opposing behavioral tendencies, and elevated cortisollevels after separation. Disorganized attachment behaviors, in turn, predict aggression in school-agechildren with other family factors controlled.

In this article, I review the major findings from research on

aggressive behavior disorders outside the attachment area, and

then I move to a more in-depth review of recent work investi-

gating the early attachment relationships of aggressive children.

Because attachment studies have focused on the infant and pre-

school periods, the attachment literature extends the emerging

longitudinal picture of aggressive behavior disorders, both de-

velopmenlally into the earliest years of life and contextually into

an expanded view of the family relational context as it is expe-

rienced and represented during early development.

Aggressive behavior disorders are denned by two diagnostic

categories in the Diagnostic and Statistical Manual of Mental

Disorders (3rd ed., revised; DSM-IH-R; American Psychiatric

Association, 1987): oppositional defiant disorder (ODD) and

conduct disorder (CD). A diagnosis of ODD is based on less

severe and less clearly antisocial behaviors than CD, such as

noncompliance toward authority figures, tantrums, argumen-

tativeness, and provocative behavior, whereas a diagnosis of CD

requires several more severe aggressive or antisocial behaviors

such as fighting, truancy, stealing, lying or firesetting. Thus, ag-

gressive children under age 6 rarely qualify for a diagnosis of

CD, but most children with CD have histories of ODD (Loeber,

Lahey & Thomas, 1991). Epidemiologic studies of the general

population estimate that about 6% of school-age children dis-

play CD (see Robins, 1991), and all studies converge on the

conclusion that by school age, CDs are several times more fre-

quent among boys than among girls, with an incidence of 5% to

10% among boys and 1% to 3% among girls (Stewart, DeBlois,

Meardon, & Cummings, 1986).

A version of this article was presented November 17-18, 1992, atthe National Institute of Mental Health Workshop on Attachment andPsychopathology, Bethesda, Maryland.

Correspondence concerning this article should be addressed to Kar-len Lyons-Ruth, Department of Psychiatry, Cambridge Hospital, Har-

vard Medical School, 1493 Cambridge Street, Cambridge, Massachu-setts 02139.

In a recent developmentally oriented review, Greenberg,

Speltz, & DeKlyen (1993) offered a four-factor model of etio-

logical contributors to aggressive behavior disorders. These four

factors included family stressors, discipline, child characteris-

tics such as temperament or neurobiological problems, and at-

tachment relationships. Indices of family stress, such as marital

conflict, single parenthood, low income, low education, and

overcrowding seem to be associated with increased rates of

childhood disorder generally. However, cumulative indices of

family adversity may be particularly strongly related to CD. In

a community study of 300 German children, Blanz, Schmidt,

and Esser (1991) found that at age 13, CD, but not attention

deficit-hyperactivity disorder (ADHD), was predicted by a

family's scores on Rutter's Cumulative Family Adversity Index

assessed at age 8. With regard to parental psychopathology, chil-

dren with CD or combined CD and ADHD are more likely to

have parents with a diagnosable disorder, particularly antisocial

personality disorder, major depression, or substance abuse, than

are children with ADHD alone (Biederman, Munir, & Knee,

1987).

One of the best documented findings in the area of child psy-

chopathology is the consistent relation between harsh and in-

effective parental discipline and aggressive behavior problems

(for a review, see Loeber & Dishion, 1983). This relation has

been reported as early as 2 and 3 years of age (Campbell, 1991)

and has been emphasized in most theories of the etiology of

CDs(e.g., Patterson & Bank, 1989).

School-age children with serious CD also often exhibit early

reading problems and deficits in verbal skills. Moffitt (1993),

in a review of 47 published studies, concluded that one of the

most robust findings in the study of antisocial behavior is an IQ

deficit of 0.50 standard deviation, or about 8 points, compared

with nondeviant peers. This deficit is manifest primarily on ver-

bal subtests, so that performance IQs consistently exceed verbal

IQs among delinquent groups. Causal relations among neuro-

psychological functioning, school achievement, family environ-

ment, and aggressive behavior remain unclear, with complex

64

SPECIAL SECTION: AGGRESSIVE BEHAVIOR PROBLEMS 65

transactional effects most likely to be involved (Hinshaw, 1992;Moffitt, 1993). The research literature on the role of geneticfactors in the etiology of aggressive behavior disorders is stillinconclusive, although a shared environmental effect has beenparticularly striking in twin studies of CD (Plomin, Rende, &Rutter, 1991). The strong gender effect on the incidence of ag-gressive disorders suggests that the disorder is likely to have bio-logical as well as environmental contributors, however.

The antisocial component of externalizing disorders appearsto be quite stable over time. Olweus (1979), reviewing over adozen longitudinal studies of school-age children to adults con-cluded that measures of antisocial behavior have about the sametest-retest reliabilities as measures of childhood intelligence(approximately .50). About half of children with early prob-lems continue on a chronically aggressive trajectory, and thesechronically aggressive children tend to be those who first displaysymptoms during the preschool years, who exhibit greater fre-quency and variety of early symptoms in more settings, includ-ing both home and school, who also display early hyperactivebehavior and who develop covert antisocial behaviors, such asstealing and lying, during the early school years. Recently, inter-generational stability of aggressive behavior and the negative pa-rental behaviors associated with it has also been reported. Hues-man, Eron, Ledfkowitz, and Walder (1984) found that aggres-sive 8-year-olds repeated the poor parenting practices of theirown parents when seen again at age 30.

Given the early onset of aggressive behavior in a sizeable pro-portion of cases, Patterson and Bank (1989) have divided boyswith adolescent CD into two groups: early starters and late start-ers. The attachment literature currently has more to offer inconceptualizing the family environments of early starters,among whom aggressive behavior begins during the preschoolperiod. In Patterson and Bank's model of early starters, at Step1, early coercive interaction between parent and child, charac-terized by scolding and explosive, irritable and inconsistent dis-cipline, leads to escalating child aggressive behavior, which, inStep 2, produces peer rejection, failure in school, and depressedmood. These developments at Step 2 are followed, in turn, byincreased involvement in delinquent acts, deviant peer groupsand substance abuse, as well as failures at work.

In Patterson and Bank's (1989) model, the traits or disposi-tions thought to place the family at risk for initiation of the co-ercion process include parental lack of social competence, pa-rental antisocial trait (self-reported aggression and motor vehi-cle violations) and child difficult temperament (assessed byparent report). In analyzing the relations among these familyfactors, maternal antisocial trait emerged as the primary inde-pendent contributor to coercive interaction. Patterson and Bank(1989) noted more anecdotally, however, that the coercive in-teractions occurred in a larger family context in which the co-ercive child had not learned the prosocial skills necessary toform close relationships with parents or peers, was not taughtto engage in disciplined effort, and had not learned to acceptcritical feedback, although such observations were not system-atically researched. Patterson and Bank further pointed out thatthese observations applied particularly to early starters, inwhom the early onset of this overall family process resulted inearly peer rejection and the omission of critical peer learning

experiences in how to socialize and form intimate relationshipswith equals.

An extensive developmental literature in nonclinical samplescorroborates the relations reported by Patterson and Bank(1989) between coercive discipline and childhood aggression.This literature also makes clear that coercive parental behavioris part of a broader parental stance in which positive techniquesof motivation and guidance are not being used. For example, ina recent meta-analysis of 47 studies, Rothbaum and Weisz(1994) studied the relation of five major parental control vari-ables to child externalizing behavior (anger, noncompliance,tantrums), including praise for desirable behavior (approval),clear tracking and direction around desirable behavior(guidance), use of positive incentives (motivation setting), giv-ing suggestions and choices (noncoercion), and responsivenessto child-initiated behavior (synchrony). These parental controltechniques all related significantly to externalizing behavior andall loaded on a broad acceptance-rejection factor, rather than anorthogonal permissive-restrictive factor, and were part of alarger pattern of affectionate and responsive, or at the negativepole rejecting and coercive, parenting. These results place coer-cive parenting practices into a larger context of rejecting paren-tal behavior, characterized by the absence of warmly approving,autonomy respecting, and contingent parental responsiveness.

Recent work on hostile attributional biases has further broad-ened our understanding of the cognitive processes accompany-ing aggressive behavior. Dodge, Pettit, McClaskey, and Brown(1986) initially reported that aggressive boys in middle child-hood exhibited biases toward attributing hostile intentions topeers in ambiguous provocation situations. Examining familycorrelates of these biases, Dix and Lochman (1990) found thatmothers of aggressive boys also made more hostile intention at-tributions to videotaped child misbehavior sequences. Mothersof aggressive boys attributed child misbehavior more to nega-tive personality dimensions, attributed more responsibility tothe child for the misbehavior, felt that they would be more upsetby the child's misbehavior and endorsed more forceful disci-pline responses. Pettit, Dodge, and Brown (1988) further foundthat children's aggression was correlated with mothers' tenden-cies to make hostile attributions to vignettes involving their ownchildren. Thus, recent developmental studies expand on Patter-son and Bank's (1989) model with evidence that coercive par-ent-child cycles are embedded in a larger context of rejectingparental behavior and that this rejecting parental stance is likelyto include characteristic perceptual and information processingbiases that may be transmitted from parent to child.

Early Attachment Relationships and Aggressive

Behavior Disorders

Attachment research is now indicating that many of the fam-ily correlates of aggressive child behavior can be identified ininfancy, before the onset of coercive cycles. In addition, disor-ganized infant attachment behaviors appear to be precursorsto the later appearing coercive childhood behavior. As definedinitially by John Bowlby (1969), the term attachment does notrefer to all aspects of the parent-infant relationship. Instead,the attachment behavioral system includes those infant behav-iors that are activated by stress and that have as a goal the re-

66 LYONS-RUTH

duction of arousal and reinstatment of a sense of security, usu-

ally best achieved in infancy by close physical contact with a

familiar caregiver. The parental caregiving system has received

less empirical investigation but theoretically encompasses the

reciprocal set of parental caregiving behaviors that monitor in-

fant safety and security and cooperate with the infant's attach-

ment behaviors to promote optimal regulation of infant arousal

and security. Parental behaviors such as teaching, limit setting,

and playing are less central to the attachment function than

more general aspects of parental availability and sensitive re-

sponsiveness to infant signals. Sensitive responsiveness involves

not only prompt response to distress or negative affect but also

a more general stance of open responsiveness to infant commu-

nications and intentions that serves to prevent the excessive ex-

perience of negative affect. The attachment behavioral system

is also understood to operate in a reciprocal fashion with the

exploratory behavioral system, so that maintenance of a sense

of security is hypothesized to support optimal exploration and

mastery of both social and nonsocial environments. Because

the attachment behavioral system is a system activated by expe-

rienced stress, attachment behaviors may not be evident in fa-

miliar, low-stress environments, so the assessment paradigm of

choice has involved two brief separations in an unfamiliar

setting.

Attachment research has developed in two waves. From 1970

to 198S, investigators focused on replicating and extending the

original discovery of Ainsworth and her colleagues (Ainsworth,

Blehar, Waters, & Wall, 1978) that three organized patterns of

infant behavior toward the caregiver, termed secure, avoidant

and ambivalent patterns, were identifiable at 1 year of age in

response to the mild stress generated by brief separations from

the parent in an unfamiliar laboratory setting (termed the

Strange Situation assessment). From 1985 to the present, as at-

tachment researchers increasingly began to study high-risk fam-

ilies, it became apparent that the behaviors of some infants did

not fit any of the three previously described patterns, leading

Main and Solomon (1990) to codify a fourth infant attachment

category, labeled disorganized (D) attachment behavior. A re-

cent meta-analysis of attachment studies including the disorga-

nized group yielded a distribution of 55% secure, 23% avoidant,

8% ambivalent, and 15% disorganized (van IJzendoorn, 1995).

Because ambivalent patterns are infrequent and have not been

related to later aggressive behavior disorders, they are not dis-

cussed here. During the first wave of attachment research, se-

cure, avoidant and ambivalent patterns of infant behavior were

empirically related to previous differences in maternal care-

giving behavior observed at home, with mothers of infants clas-

sified "secure" rated more sensitive and responsive than moth-

ers of infants in the other two groups (Ainsworth et al., 1978).A series of subsequent studies demonstrated that infants dis-

playing secure patterns of attachment behavior also exhibited

more positive social behaviors toward both parents and peersthroughout the preschool years (for review, see Bretherton,

1985). In the worldwide literature to 1985, the distribution of

attachment patterns was 65% secure, 20% avoidant, and 14%

ambivalent (van IJzendoorn & Kroonenberg, 1988).

A recent advance in our understanding of the caregiving

context of infant attachment behavior has come from Main and

Goldwyn's (in press) introduction of the Berkeley Adult At-

tachment Interview (AAI). On the basis of an hour-long, semi-

structured interview probing parents' coherence in narrating

and evaluating their own early attachment-related experiences,

parents' states of mind regarding attachment issues can be reli-

ably classified into categories corresponding to the secure, am-

bivalent and avoidant patterns observed in infancy. A fourth

category labeled unresolved corresponds to the infant disorga-

nized classification to be discussed later. A meta-analysis of 18

studies investigating the correspondence between infant attach-

ment patterns and maternal interview classifications yielded a

significant three-category agreement rate of 70%, even when

maternal interviews were carried out before the birth of the in-

fant. Four-category correspondence, based on nine studies, was

a reliable 63% (van IJzendoorn, 1995).

An extensive literature has addressed the issue of whether se-

cure and insecure attachment patterns should be viewed as

emerging from patterns of caregiver-infant interaction or from

endogenous dispositions of the infant. Four categories of evi-

dence, in addition to the repeated relation between responsive

caregiving and secure infant behavior reviewed by Bretherton

(1985), have pointed to the influence of relational patterns. First,

the attachment strategy shown with one parent is not strongly

associated with the attachment pattern shown to the other parent

(e.g., Fox, Kimmerly, & Schafer, 1991). Second, as noted earlier,

in a sizeable majority of cases, the infant attachment strategy

shown to the primary caregiver at 12 months of age is predictable

from the caregiver's state of mind with regard to attachment is-

sues, even when assessed before the birth of the infant (e.g., Fo-

nagy, Steele, & Steele, 1991). Third, the attachment strategy dis-

played toward the primary caregiver is more predictive of later

social adaptation than the strategies shown toward other caregiv-

ers (Main, Kaplan, & Cassidy, 1985; Main & Weston, 1981;

Suess, Grossmann, & Sroufe, 1992), even when the primary

caregiver is not biologically related (Oppenheim, Sagi. & Lamb,

1988). Finally, various measures of infant temperament have

predicted distress to separation but have not predicted whether

the distressed or nondistressed behavior pattern is classified se-

cure or insecure (Belsky & Rovine, 1987; Vaughn, Lefever, Seifer,

& Barglow, 1989). Other scattered relations between tempera-

ment and aspects of attachment behavior have appeared in the

literature but have not yielded a replicable or clearly interpret-

able body of data.

Avaidant Attachment Patterns and Aggressive Behavior

During the first wave of attachment studies, before the iden-

tification of disorganized attachment behaviors, researchers re-

ported longitudinal relations between insecure attachment re-

lationships in infancy, particularly avoidant attachments, and

angry, noncompliant behavior toward parents or peers during

infancy, toddlerhood, and the preschool period (e.g. Ainsworth

et al., 1978; Fagot & Kavanagh, 1990; Matas, Arend, & Sroufe,

1978; Erickson, Sroufe, & Egeland, 1985). Infants in the avoid-

ant group do not show protest or distress to the caregiver's de-

parture. Instead, they displace their attention away from her

exit, explore actively and are often friendly to the unfamiliar

adult in the room. However, when the caregiver returns, infants

in this group displace attention away from her entrance, fail togreet her, and initially move away if she approaches them. Nu-

SPECIAL SECTION: AGGRESSIVE BEHAVIOR PROBLEMS 67

merous laboratories have documented relations between infant

avoidance and mothers' suppressed anger, lack of tenderness in

touching and holding, insensitive intrusiveness, and rejection of

attachment behavior (Belsky, Rovine, & Taylor, 1984; Gross-

mann, Grossmann, Spangler, Suess, & Unzner, 1985; Lyons-

Ruth, Connell, Toll, & Stahl, 1987; Main, Tomasini, & Tolan,

1979; Matas, Arend & Sroufe, 1978). Main (1990) character-

izes the underlying organized strategy of avoidant infants as one

of restricting the communication of anger and distress by dis-

placing attention onto the inanimate environment, away from

cues that might intensify the desire to seek comfort from a par-

ent who rejects attachment behavior. In support of this inter-

pretation, Spangler and Grossmann (1993) demonstrated that

avoidant infants and secure infants exhibited equivalent hear-

trate increases during the second separation, although secure

infants were significantly more likely to signal their distress. In

addition, the heart rates of avoidant infants remained elevated

while they apparently attended to play objects, whereas those

of secure infants decreased while they attended to objects as

expected.

Whereas most of the early longitudinal attachment studies

focused on middle-income families, the Minnesota High Risk

Study conducted by Egeland and Sroufe was unusual in follow-

ing a large community sample of impoverished mothers and in-

fants from birth into adolescence. The documented prediction

from early maternal and infant behavior to child maladaption

and aggressive behavior during the school years has been a ma-

jor contribution of the study. For example, Egeland, Pianta, and

O'Brien (1993), comparing 45 mothers rated high on intrusive-

ness at 6 months with nonintrusive mothers, found that intru-

sive mothers reported more anxiety and suspiciousness, less ap-

preciation of the complexity of child rearing and the need for

reciprocity with the infant, and a more idealized view of babies

in general. They were also less likely to be living with a partner.

By 12 months, the attachment behavior of infants of intrusive

mothers was more likely to be classified as insecure-avoidant. At

24 and 42 months, mothers' quality of assistance and supportive

presence on teaching tasks were less optimal, and by 42 months

children of intrusive mothers were more negative, noncompli-

ant, avoidant, and hyperactive in laboratory observations. In

first grade, they were rated higher by teachers on both Interna-

lizing and Externalizing scales of the Child Behavior Checklist

(CBCL).

Investigating early predictors of first through third grade

teacher-rated aggression and passive withdrawal in the same

sample, Renken, Egeland, Marvinney, Mangelsdorf, and Sroufe

(1989) found that an avoidant attachment pattern in infancy

continued to predict both aggression and passive withdrawal

among boys but not among girls, whereas maternal hostility at

age 3.5 predicted aggressive behavior in both sexes. After enter-

ing child attachment and affect and maternal hostility into the

regression equation, maternal socioeconomic status (SES),

stress, and social support contributed no further variance to

prediction. The maternal longitudinal predictors of childhood

aggression identified in this study, including hostility and intru-

siveness evident early in the first year, lack of a stable partner,

and a more deviant personality profile, converge well with the

concurrent maternal correlates identified in cross-sectional

studies of school-age, conduct-disordered children.

Despite the impressive overall prediction demonstrated in the

Egeland and Sroufe study, investigators studying more econom-

ically advantaged families than those in the Egeland and Sroufe

sample have not consistently documented the specific relation

between avoidant attachment strategies and later aggressive be-

havior at a clinically significant level (Bates, Bayles, Bennett,

Ridge, & Brown, 1991; Fagot & Kavanagh, 1990; Goldberg,

Perrotta, Minde, & Corter, 1986; Lewis, Feiring, McGuffog &

Jaskir, 1984). In the Bates et al., (1991) .Goldberg et al. (1986),

and Fagot and Kavanagh (1990) studies, no relation between

infant attachment security and later problems were found,

whereas in the Lewis et al. (1984) study, predictive effects oc-

curred only for internalizing symptoms among boys. Thus,

avoidant strategies in low-risk contexts do not appear to consti-

tute a risk factor for early aggression. However, in high-risk fam-

ily settings, early avoidant behavior appears to be prominent

among children who later become aggressive, as discussed later.

Disorganized Attachment Patterns and Aggressive

Behavior

Attachment studies after 1985 increasingly began to focus on

infants at social or psychological risk and to code for the pres-

ence of disorganized behavior as well as the three organized in-

fant attachment patterns. These newer studies indicate that

avoidant infant behavior in high-risk family settings is likely to

be displayed in disorganized rather than organized form, in con-

trast to the organized forms of avoidant infant behavior seen

in low-risk family settings. In the more recent studies, infant

disorganization, rather than avoidance per se, is associated with

later highly aggressive behavior. Whereas approximately 15% of

infants in two-parent, middle-class families display disorga-

nized attachment behavior, the incidence of D behavior in-

creases under attachment-relevant family risk conditions such

as maternal alcohol consumption, maternal depression, adoles-

cent parenthood, or multiproblem family status to a high of

82% among maltreating families (for a review, see Lyons-Ruth,

Repacholi, McLeod, & Silva, 1991). In contrast, serious infant

medical conditions or chronic disabilities do not result in ele-

vated rates of disorganized attachment patterns (van Uzen-

doorn, Goldberg, Kroonenberg, & Frenkel, 1992).

Disorganized infant attachment behavior does not show the

consistency across individuals and the comprehensible strategic

organization characterizing the avoidant and ambivalent, as

well as secure, attachment patterns. Instead, the term disorga-

nized refers to the apparent lack of, or collapse of, a consistent

strategy for organizing responses to the need for comfort and

security when under stress. The term does not refer to mental

disorganization or to behavioral disorganization more generally.

The particular forms and combinations of disorganized behav-

iors tend to be idiosyncratic from child to child, but include

apprehensive, helpless, or depressed behaviors, unexpected al-

ternations of approach and avoidance toward the attachment

figure, and other conflict behaviors, such as prolonged freezing

or stilling, or slowed "underwater" movements, with aspects of

the three organized strategies often mixed in unpredictable

ways (see Main & Solomon, 1990, fora full description). Often

the outlines of a "best-fitting" or "forced" secure, avoidant, or

ambivalent strategy can be discerned within the larger context

68 LYONS-RUTH

of the infant's disorganized attachment behavior, and coders are

encouraged to subclassify the behavior of each disorganized in-

fant as to the best fitting organized strategy. In support of the

view that disorganized infant attachment behavior constitutes

the collapse of organized strategic behavior, Spangler and

Grossmann (1993) demonstrated that cortisol levels assessed

30 min after the assessment of attachment behavior remained

significantly elevated among infants with disorganized strate-

gies, compared with infants with secure strategies, whereas cor-

tisol levels of avoidant infants were intermediate in value. Span-

gler and Grossmann (1993) interpreted these data as consistent

with animal data indicating that the adrenocortical system is

only activated when adequate behavioral strategies cannot be

applied.

The literature to date also suggests that there may be mean-

ingful subgroups within the disorganized category. Disorga-

nized infant behavior in low-risk, middle-SES samples has been

predominantly of the forced-secure subtype, in which the infant

seeks contact with the caregiver without marked avoidance or

ambivalence and is soothed by her presence but shows other

unusual signs of hesitation, confusion, apprehension, dysphoria

or conflict in relation to her. George and Solomon (1993) have

described a helpless maternal stance as characterizing mothers

in this group. Disorganized infant behavior in samples with se-

rious social risk characteristics has been predominantly of the

forced-avoidant subtype, in which prominent avoidant behav-

ior occurs in unexpected combination with distress, contact

seeking, resistance, or other apprehensive or conflict behaviors.

Disorganized forms of avoidant infant behavior have been asso-

ciated with more severe maternal psychosocial problems than

disorganized-secure behavior, including current maternal de-

pression, maternal childhood histories of violence or abuse, ma-

ternal inpatient psychiatric histories, and documented child

maltreatment (see Lyons-Ruth, Repacholi, McLeod, & Silva,

1991, for review). In addition, parents of infants who display

such disorganized-avoidant behavior are more intrusive, nega-

tive, and role reversing than parents of disorganized-secure in-

fants, who display more withdrawal from the infant (Lyons-

Ruth, Bronfman, & Parsons, 1994).

As disorganized infants and toddlers make the transition into

the preschool years, a developmental reorganization occurs for

many of these children. The signs of conflict, apprehension, or

helplessness characteristic of disorganized attachment strate-

gies in infancy often give way to various forms of controlling

behavior toward the parent, forms that emphasize caregiving

behavior, directing and organizing behavior, or coercive behav-

ior. This developmental shift toward reversal of roles with the

parent has been documented in 6-year-old follow-up studies of

two middle-income samples (Main & Cassidy, 1988; Wartner,

Grossmann, Fremmer-Bombik, & Suess, 1994). However, Cic-

chetti and Barnett (1991), studying maltreated children at 30,

36, and 48 months of age, found that the pronounced conflict

behaviors seen in infancy remained more prevalent than con-

trolling patterns among maltreated children over this age range,

so more longitudinal data are needed.

Several studies now document a relation between disorga-

nized attachment patterns and childhood aggression. Lyons-

Ruth, Alpern, and Repacholi (1993), examining five summary

measures of maternal and infant functioning in a low-income

sample at 18 months, as well as measures of cumulative demo-

graphic risk and gender, found that three measures predicted

deviant levels of hostile aggression toward peers in kindergarten:

infant security of attachment, serious maternal psychosocial

problems, and maternal hostile-intrusive behavior toward the

infant at home. Preschoolers with highly hostile behavior were

six times more likely to have been classified as disorganized in

their attachment relationships in infancy than to have been clas-

sified as secure. For a majority of infants, the disorganized be-

havior included a high level of avoidance. In contrast, infants

classified in the organized avoidant category were not more

likely to be aggressive.

Maternal psychosocial problems, particularly the presence of

chronic depressive symptoms, also contributed to the predic-

tion of childhood hostile behavior. If, at 18 months, the infant

displayed disorganized attachment behavior and the mother

had psychosocial problems, a majority of infants (56%) exhib-

ited deviant levels of hostile behavior in kindergarten, com-

pared with 25% of low-income children with only one of these

risk factors, and 5% of low-income children with neither risk

factor (see also Alpern & Lyons-Ruth, 1993). Maternal psycho-

social problems accounted for variance in childhood aggression

primarily because mothers with psychosocial problems were

more hostile and intrusive in early interactions with their in-

fants (Lyons-Ruth et al., 1993). In addition to this direct effect,

however, maternal depression and hostile-intrusive behavior

also influenced childhood aggression indirectly because both

variables were also associated with an increased incidence of

disorganized attachment behavior in infancy (Lyons-Ruth,

Connell, Grunebaum, & Botein, 1990; Lyons-Ruth et al.,

1991).

The presence of a disorganized attachment strategy also ac-

counted for variance in infant mental development scores at 18

months, independent of variance related to maternal behavior

and maternal IQ (Lyons-Ruth et al., 1991). Infant physical de-

velopment quotients were not lowered, producing a pattern of

"mental lag" or disparity between mental and performance

scores among a subset of disorganized infants, similar to the 8-

point disparity in verbal and performance IQ scores seen among

older conduct-disordered children (Moffitt, 1993). This link

between disorganized attachment strategies and less effective

cognitive functioning has also been demonstrated in an Ice-

landic cohort followed from ages 7 to 17 (Jacobsen, Edelstein,

&Hofmann, 1994).

Hann, Castino, Jarosinski, and Britton (1991), studying 67

infants and their impoverished adolescent mothers, also found

that disorganized attachment relationships were predictive of

later aggression. Sixty-two percent of infants of adolescent

mothers displayed disorganized attachment patterns, whereas

only 15% displayed insecure but organized patterns. The major-

ity of the disorganized infants (60%) were exhibiting disorga-

nized forms of avoidant or resistant strategies. Disorganized at-

tachment status at 13 months was related to the form of conflict

negotiation between mother and child at 20 months. Mothers of

disorganized toddlers were observed to initiate more interaction

but to agree less frequently with their children's initiatives, in

comparison with organized dyads. They also showed less affec-

tion that other mothers. Disorganized toddlers showed the low-

est frequency of social initiatives and tended to show more re-

SPECIAL SECTION: AGGRESSIVE BEHAVIOR PROBLEMS 69

fusals of their mothers' initiatives. They were also more likely toinitiate conflict by aggressive behavior. The overall picture thatemerged was one of interactive imbalance, mutual rejection ofovertures, and increased aggressive conflict tactics by the childby 20 months of age. One interpretation of the interactive datais that disorganized infant attachment strategies and coercivetoddler behavior emerge when more appropriate and organizedchild initiatives are thwarted.

In a follow-up study of 45 of these children at 54 months ofage, mothers' reports of child behavior problems on the CBCLrevealed that externalizing behavior among girls was related toearlier disorganized attachment classification. The absence offindings for boys was probably related to the high rate of selec-tive attrition among disorganized boys (Hubbs-Tait et al.,1991).

A potential weakness of the previously discussed studies lies inthe use of rating scales rather than diagnostic criteria for denningaggressive behavior disorders. However, the preschool analogue ofinfant disorganized behavior, termed controlling-disorganized at-tachment patterns, have also been related to aggressive behavioramong two cross-sectional studies of clinic-referred preschoolers,one of which used DSM-III-R criteria for diagnosing opposi-tional defiant disorder (Greenberg, Speltz, DeKlyen, & Endriga,1991; Speltz, Greenberg, & DeKlyen, 1990). In both studies, op-positional children were significantly more likely than those in thecontrol group to show insecure attachment patterns, with a ma-jority of oppositional children classified in the controlling-disorga-nized category. Oppositional children were also more likely toshow separation distress and search behavior during the separa-tion, indicating that behavioral disruption appeared outside thecontext of immediate interaction with mothers. Teachers alsorated oppositional children higher on both internalizing and exter-nalizing behaviors in preschool. Clinic mothers were more likelyto be classified in insecure attachment categories on the Adult At-tachment Interview, with a majority of insecure clinic mothersclassified in the unresolved category, the adult counterpart of thecontrolling-disorganized classification. Seventy percent of childrendisplayed the same type of attachment strategy as their mothers(DeKlyen, 1992).

Given the comparatively recent development of attachmentassessments for high-risk populations and for older children,few studies have explored the attachment behaviors displayed atschool age by aggressive or conduct-disordered children. How-ever, Lyons-Ruth, Easterbrooks, and Cibelli (submitted forpublication) found that a deviant level of externalizing behaviorat school at age 7 was correctly predicted in 87% of cases frominfancy assessments. Disorganized attachment behavior in in-fancy predicted externalizing behavior at age 7 but only amongthe subgroups of disorganized infants who had also displayedlowered mental development scores in infancy (range, 81 to95). Eighty-three percent of highly externalizing children werein the disorganized-low mental development group at 18months of age, compared with only 13% of nondeviant children.In contrast, infant performance scores of highly externalizingchildren were not lowered in comparison with other children,nor were their mothers' verbal scores lowered. Thus, the relativeverbal deficits repeatedly noted among conduct-disordered chil-dren may be evident and predictive of disorder as early as 18months of age. These deficits occurred primarily in the context

of disorganized mother-infant attachment relationships,however.

Intergenerational Context of Disorganized Attachment

Behavior

Additional studies unrelated to aggressive behavior provide afurther picture of the intergenerational family context of disor-ganized attachment behavior. Disorganized attachment behav-ior is also reliably associated with the parent's lack of resolutionof a previous loss or trauma, as revealed in the Adult Attach-ment Interview (van Uzendoorn, 1995). Unresolved loss ortrauma on the Adult Attachment Interview is indexed by lapsesin the monitoring of reasoning or discourse that occur duringloss or trauma-related portions of the interview (see Main &Goldwyn, in press). Because these lapses may be present in onlya few sentences of the interview transcript, Main, Hesse, andvan Uzendoorn (1994) have related such lapses to a tendencyto dissociate, that is, to segregate loss or trauma-related mate-rial in an encapsulated system of consciousness separate fromnormal processing. However, recent studies of personality dis-order (Patrick, Hobson, Castle, Howard, & Maughn, 1994) andmaritally violent adults (Holtzworth-Munroe, Hutchinson, &Stuart, 1994) indicate that Adult Attachment Interview proto-cols of adults in clinical samples often contain more pervasiveindicators of incoherence that place them in rare and less well-described coding categories of the Adult Attachment Interview(e.g., cannot classify or overwhelmed by trauma) in additionto or in place of categorization in the unresolved group. Theseunusual adult states of mind have not yet been studied in rela-tion to parental caregiving and infant attachment behavior butmay be relevant to coercive or violent families. Therefore, fur-ther description is needed of states of mind regarding attach-ment among aggressive parents and their children.

Theoretically, Main and Hesse (1990) have hypothesized thatdisorganization of infant attachment strategies is related to pa-rental unresolved fear, transmitted through behavior that isfrightened or frightening to the infant. Such parental behaviorshould place the infant in an unresolvable paradox because theparent's presence would both heighten the infant's fear andneed for soothing contact and make such contact fear-arousingrather than comforting. Direct studies of frightened or frighten-ing parental behavior have not yet been conducted, however. Inindirect support of the role of fear in the genesis of disorganiza-tion, Lyons-Ruth and Block (1993) have demonstrated that dis-organized infant attachment behaviors occurred predomi-nantly in the context of maternal childhood experiences of fam-ily violence or abuse, but not maternal childhood experiencesof neglect alone. In addition, the severity of family violenceor physical abuse was related to increased hostile and intrusivematernal interaction with her infant. These findings reiteratethe potential multigenerational context of coercive parent-child interactions identified by Huesman et al. (1984).

Only a few other studies of mother-child interaction amongdisorganized dyads have been reported. Main, Kaplan, and Cas-sidy (1985), studying middle-income families, reported thatmothers of disorganized infants as well as mothers of avoidantinfants exhibited low fluency and balance in discourse with thesame children at age 6, with mothers of disorganized infants

70 LYONS-RUTH

particularly dysfluent (see also Strage & Main, 1985). Spieker

and Booth (1988) also found that mothers of disorganized in-

fants differed from mothers of secure infants in having lower

adult conversational skills and less positive perceptions of their

infants' temperaments, echoing Patterson's observations of the

lowered social skills and negative perceptions of child tempera-

ment among coercive mothers. Lyons-Ruth, Bronfman, & Par-

sons (1994) have also found that affective communication er-

rors, in which the mother either displays confusing cues to the

infant (e.g., extends arms toward infant while backing away) or

doesn't respond appropriately to clear infant cues (e.g., laughs

while infant distressed) were particularly discriminative of

mothers whose infants displayed disorganized strategies. These

communication errors were associated with increased infant

distress at home, but not with increased anger or resistance at

home. DeMulder and Radke-Yarrow (1991), studying middle-

income unipolar and bipolar women and their well controls,

concluded that children classified disorganized or controlling-

disorganized were exposed to the most negative affective envi-

ronments, with their mothers displaying a high level of down-

cast affect, more than one negative affect at a high level, and a

low level of tenderness and affection.

Summary and Directions for Future Work

This extended body of attachment studies, taken as a whole,

reiterates the larger literature on CD in emphasizing the impor-

tance of parental social skills and parental hostile and rejecting

behavior in the developmental trajectory leading to early onset

aggressive behavior disorders. However, the attachment litera-

ture extends this picture in several directions.

First, maternal hostile-intrusive behaviors predicting aggres-

sive behavior disorders are observable as early as the sixth

month of infancy, before the onset of coercive cycles (Egeland

etal., 1993; Lyons-Ruth etal., 1993).

Second, the maternal behaviors associated with disorganized

infant attachment behavior in high-risk samples are character-

ized by a frequent lack of responsiveness to appropriate infant

cues and by maternal initiatives that often override clear infant

communications and goals, rather than by the coercive re-

sponses to child coercion most often described in the literature

on older children(Hann etal., 1991; Lyons-Ruth etal., 1994).

Third, the child's aggressive behavior is likely to be preceded

by serious disturbances in the regulation of attachment-related

affects in infancy, with disorganized or controlling attachment

patterns more strongly related to aggressive disorders than the

previously studied avoidant attachment patterns common in

middle-classsamples(Greenbergetal., 1991,1993;Hubbs-Tait

atal., 1991; Lyons-Ruth etal., 1993). However, consistent with

Egeland et al.'s (1993) earlier data, the disorganized attachment

behavior shown by infants who later become highly aggressive

is likely to include pronounced avoidant behavior toward the

parent (Hann el al., 1991; Lyons-Ruth etal., 1991).

Fourth, a pattern of "mental lag," similar to the disparity in

verbal and performance scores observed among older conduct-

disordered children, is evident among a subset of infants dis-

playing disorganized attachment behaviors (Lyons-Ruth et al.,

1994).

Fifth, infant data indicates substantial phenotypic disconti-

nuity in the behaviors of aggressive children from infancy to

preschool or school age, with the disturbance in regulation of

attachment-related affects and behaviors in infancy more

strongly characterized by indicators of conflict, apprehension,

helplessness, dysphoria, and unpredictable behavior sequences,

than by coercive behavior per se (Main & Solomon, 1990).

Finally, a heightened incidence of distress and dysphoria

seems to characterize many children with early onset aggression

from infancy onward, rather than developing as a secondary re-

sponse to peer rejection. This is seen in the helpless, apprehen-

sive, or depressed infant behaviors that contribute to the disor-

ganized attachment classification, in the heightened incidence

of distress among disorganized infants when observed at home

(Lyons-Ruth et al., 1994), in the heightened distress observed

among disorganized or controlling children around separation

and reunion episodes with parents in the laboratory (Greenberg

et al., 1991, 1993), and in the mixed patterns of internalizing

and externalizing symptoms displayed by oppositional children

in preschool (Greenberg et al., 1991,1993). Attachment theory

would trace this dysphoria to the disruptions in functioning of

the child's primary attachment relationships, which leave the

child without an organized, relationally oriented strategy for

regulating felt security.

Given that many correlates of later childhood aggression are

already evident in infancy, more research is needed to delineate

the infant and preschool developmental trajectories of children

at high risk for early onset aggressive behavior disorders. A se-

ries of longitudinal studies from infancy through elementary

school is now needed using a subset of common family and child

assessment procedures. Recent longitudinal studies have devel-

oped a core set of assessments to evaluate critical developmental

tasks from infancy to school age, but none of these studies was

specifically designed to study the emergence of serious conduct

problems (e.g., Belsky et al., 1984; Egeland et al., 1993; Main

et al., 1985; Wartner et al., 1994; Waters, Wippman, & Sroufe,

1979).

Continued refinement of developmental assessment proce-

dures specific to very high-risk populations will be needed as

part of such studies, including procedures for assessing attach-

ment relationships among high-risk school-age children. The

recent process of refining the infant attachment assessment to

include a disorganized category has demonstrated the recursive

process of instrument refinement that is likely to be necessary

as procedures developed for assessing adequately functioning

children are applied to high-risk samples. Thus, investigators

need to be encouraged to include both theory-driven and ex-

ploratory descriptive analyses in high-risk studies. Document-

ing that young children at risk for CD show less optimal re-

sponses on a variety of developmental tasks is unlikely to be

useful without accompanying close description of how the devi-

ant behavior is organized, what underlying goals or functions it

may serve, and what emerging meaning systems support and

direct it.

More research is also needed describing the family relational

contexts in which aggressive behavior disorders develop, beyond

descriptions of the coercive process itself. Future studies should

include a more comprehensive view of parental affect and be-

havior toward the child, increased information about parental

relationship histories and their derived representational models

SPECIAL SECTION: AGGRESSIVE BEHAVIOR PROBLEMS 71

for guiding caregiving behavior, more explicit study of cognitive

processing strategies and biases among both parents and chil-

dren, and more in-depth specification of concurrent parental

psychiatric symptomatology, including dissociative processes,

Axis II disorders, depression, substance abuse, and posttrau-

matic symptoms. The influence of family adversity will need to

be understood in relation to how adversity affects representa-

tional processes, personality structure, psychological symp-

toms, and family process.

The important continuities in development revealed by at-

tachment research occurred because of the move to a more or-

ganizational view of the coherence or goal-corrected nature of

behavior in relationships. More open debate is needed regarding

whether the current analytic models, which tend to use vari-

ables as units of analysis rather than individuals and which em-

phasize causal, linear relations among variables, are well suited

to identifying and describing organizational coherence in indi-

vidual behavior (see also Magnusson & Bergman, 1988). Re-

cent developmental research on longitudinal and intergenera-

tional continuity in parental behavior suggests the need for a

move to a more conceptually grounded and integrated view of

the central functions of caregiving relationships and of the po-

tentially limited number of ways that affect, cognition, and be-

havior are patterned in relation to those functions.

Finally, several recent attachment-oriented studies have indi-

cated that the early attachment relationship can be positively

influenced by well-targeted family and parenting supports

(Jacobson & Frye, 1991; Lieberman, Weston, & Pawl, 1991;

Lyons-Ruth et al., 1990). Therefore, broadly conceived and

carefully designed early intervention studies now seem indi-

cated. Randomly assigned intervention designs constitute one

of the few vehicles available for exploring what kinds of change

in adversity-related family process is causally related to change

in disruptive child behavior. Given that problems in the early

attachment relationship appear to precede overt disorder, inter-

ventions beginning early in infancy would seem warranted.

References

Ainsworth, M. D. S., Blehar, M., Waters, E., & Wall, S. (1978). Patterns

of attachment. Hillsdale, NJ: Erlbaum.

Alpern, L., & Lyons-Ruth, K. (1993). Preschool children at social risk:

Chronicity and timing of maternal depressive symptoms and child

behavior problems at school and at home. Development and Psycho-

pathology, 5,371 -387.

American Psychiatric Association. (1987). Diagnostic and statistical

manual of mental disorders (3rd ed., rev.). Washington, DC: Author.

Bates, J. E., Bayles, K., Bennett, D. S., Ridge, B., & Brown, M. M.

(1991). Origins of externalizing behavior problems at eight years of

age. In D. J. Pepler & K. H. Rubin (Eds.), The development and treat-

ment of childhood aggression (pp. 93-120). Hillsdale, NJ: Erlbaum.

Belsky, J., & Rovine, M.( 1987). Temperament and attachment security

in the Strange Situation: An empirical rapprochement. Child Devel-

opment. SS. 787-795.

Belsky, J. Rovine, M., & Taylor, D. (1984). The Pennsylvania Infant

and Family Development Project III: The origins of individual

differences in infant-mother attachment: Maternal and infant con-

tributions. Child Development, SS, 718-728.

Biederman, J., Munir, K., & Knee, D. (1987). Conduct and opposi-

tional disorder in clinically referred children with attention deficit

disorder: A controlled family study. Journal of the American Acad-

emy of Child and Adolescent Psychiatry, 26, 724-727.

Blanz, B., Schmidt, M. H., & Esser, G. (1991). Familial adversities and

child psychiatric disorder. Journal of Child Psychology and Psychia-

try, 32, 939-950.

Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. New York:

Basic Books.

Bretherton, I. (1985). Attachment theory: Retrospect and prospect. In

I. Bretherton &E. Waters (Eds.), Growing points of attachment the-

ory and research. Monographs of the Society for Research in Child

Development, 50( 1-2, Serial No. 209), 66-104.

Campbell, S. B. (1991). Longitudinal studies of active and aggressive

preschoolers: Individual differences in early behavior and in outcome.

In D. Cicchetti & S. L. Toth (Eds.), Internalizing and externalizing

expression of dysfunction: Rochester Symposium on Developmental

Psychopalhology(Vol 2, pp. 57-89). Hillsdale, NJ: Erlbaum.

Cicchetti, D., & Barnett, D. (1991). Attachment organization in mal-

treated preschoolers. Development and Psychopathology, 3, 397-411.

DeKlyen, M. (1992). Childhood psychopathology and intergenerational

relations in the representation of attachment: A comparison of normal

and clinic-referred disruptive preschoolers and their mothers. Unpub-

lished doctoral dissertation, University of Washington, Seattle.

DeMulder, E. K., & Radke-Yarrow, M. (1991). Attachment with affec-

tively ill and well mothers: Concurrent behavioral correlates. Devel-

opment and Psychopathology, 3, 227-242.

Dix, T. H., & Lochman, J. E. (1990). Social cognition and negative

reactions to children: A comparison of mothers of aggressive and

nonaggressive boys. Journal of Social and Clinical Psychology, 9,418-

438.

Dodge, K. A., Pettit, G. S., McClaskey, C. L., & Brown, M. M. (1986).

Social competence in children. Monographs of the Society for Re-

search in Child Development, 51(2, Serial No. 213).

Egeland, B., Pianta, R., & O'Brien, M. A. (1993). Maternal intrusive-

ness in infancy and child maladaptation in early school years. Devel-

opment and Psychopathology, 5, 359-370,

Erickson, M., Sroufe, L. A., & Egeland, B. (1985). The relationship

between quality of attachment and behavior problems in preschool

in a high-risk sample. In I. Bretherton & E. Waters (Eds.), Growing

points of attachment theory and research. Monographs of the Society

for Research in Child Development, 50( 1-2, Serial No. 209), 147-

166.

Fagot, B. I., & Kavanagh, K. (1990). The prediction of antisocial be-

havior from avoidant attachment classifications. Child Development,

61, 864-873.

Fonagy, P., Steele, H., & Steele, M. (1991). Maternal representations

of attachment during pregnancy predict the organization of infant-

mother attachment at 1 year of age. Child Development, 62, 891-905.

Fox, N. A., Kimmerly, N. L., & Schafer, W. D. (1991). Attachment to

mother/attachment to father: A meta-analysis with emphasis on the

role of temperament. Child Development, 62, 210-225.

George, C., & Solomon, J. (1993, March). Internal working models of

caregiving. Paper presented at the meeting of Society for Research in

Child Development, New Orleans, LA.

Goldberg, S., Perotta, M., Minde, K., & Corter, C. (1986). Maternal

behavior and attachment in low birthweight twins and singletons.

Child Development, 57, 34-46.

Greenberg, M. T., Speltz, M. L., & DeKlyen, M. (1993). The role of

attachment in the early development of disruptive behavior prob-

lems. Development and Psychopathology, 5, 191-213.

Greenberg, M. T, Speltz, M. L., DeKlyen, M., & Endriga, M. C.

(1991). Attachment security in preschoolers with and without exter-

nalizing problems: A replication. Development and Psychopathology,

J,4I3-»30.

Grossmann, K., Grossmann, K. E., Spangler, G., Suess, G., & Unzner,

72 LYONS-RUTH

L. (1985). Maternal sensitivity and newbom's orientation responses

as related to quality of attachment in northern Germany: Growing

points of attachment theory and research. Monographs of the Society

for Research in Child Development. 50( 1-2, Serial No. 209), 233-

256.

Hann, D. M, Castino, R. J., Jarosinski, J., & Britton, H. (1991, April).

Relating mother-toddler negotiation patterns to infant attachment

and maternal depression with an adolescent mother sample. In J.

Osofsky & L. Hubbs-Tait (Chairs), Consequences of adolescent

parenting: Predicting behavior problems in toddlers and preschoolers.

Symposium conducted at the biennial meeting of the Society for Re-

search in Child Development, Seattle, WA.

Hinshaw, S. (1992). Externalizing behavior problems and academic un-

derachievement in childhood and adolescence: Causal relationships

and underlying mechanisms. Psychological Bulletin, 111, 127-155.

Holtzworth-Munroe, A., Hutchinson, G., & Stuart, G. L. (1994). Com-

paring the working models of attachment of violent and nonviolent

husbands. Manuscript under review.

Hubbs-Tait, L., Eberhart-Wright, A., Ware, L., Osofsky, J., Yockey, W.,

& Fusco, J. (1991, April). Maternal depression and infant attach-

ment: Behavior problems at 54 months in children of adolescent

mothers. In J. Osofsky & L. Hubbs-Tait (Chairs), Consequences of

adolescent parenting: Predicting behavior problems in toddlers and

preschoolers. Symposium conducted at the biennial meeting of the

Society for Research in Child Development, Seattle, WA.

Huesman, L. R., Eron, L. D., Ledfkowits, M. M., & Walder, L. O.

(1984). Stability of aggression over time and generations. Develop-

mental Psychology, 20, 1120-1134.

Jacobsen, X, Edelstein, W., & Hofmann, V. (1994). A longitudinal

study of the relation between representations of attachment in child-

hood and cognitive functioning in childhood and adolescence. Devel-

opmental Psychology, 30, 112-124.

Jacobson, S. W., & Frye, K. F. (1991). Effect of maternal social support

on attachment: Experimental evidence. Child Development. 62, 572-

582.

Lewis, M., Feiring, C., McGuflbg, C, & Jaskir, J. (1984). Predicting

psychopathology in six-year-olds from early social relations. Child

Development, 55,123-136.

Lieberman, A. F., Weston, D. R., & Pawl, J. H. (1991). Preventive in-

tervention and outcome with anxiously attached dyads. Child Devel-

opment, 62, 199-209.

Loeber, R.,& Dishion, T. (1983). Early predictors of male delinquency:

A review. Psychological Bulletin, 93,68-99.

Loeber, R., Lahey, B. B., & Thomas, C. (1991). Diagnostic conundrum

of oppositional defiant disorder and conduct disorder. Journal of Ab-

normal Psychology, 3, 379-390.

Lyons-Ruth, K., Alpern, L., & Repacholi, B. (1993). Disorganized in-

fant attachment classification and maternal psychosocial problems as

predictors of hostile-aggressive behavior in the preschool classroom.

Child Development, 64, 572-585.

Lyons-Ruth, K., & Block, D. (1993). The disturbed caregiving system:

Conceptualizing the impact of childhood trauma on maternal care-

giving behavior during infancy. In J. Solomon (Chair), Defining the

caregiving system. Symposium conducted at the biennial meeting of

the Society for Research in Child Development, New Orleans, LA.

Lyons-Ruth, K., Bronfman, E., & Parsons, E. (1994). Atypical mater-

nal behavior and disorganized infant attachment strategies. Manu-

script submitted for publication.

Lyons-Ruth, K., Connell, D. B., Grunebaum, H., & Botein, S. (1990).

Infants at social risk: Maternal depression and family support ser-

vices as mediators of infant development and security of attachment.

Child Development, 61. 85-98.

Lyons-Ruth, K., Connell, D., Zoll, D., & Stahl, J. (1987). Infants at

social risk: Relations among infant maltreatment, maternal behavior,

and infant attachment behavior. Developmental Psychology, 23, 223-

232.

Lyons-Ruth, K., Easterbrooks, M. A., & Cibelli, C. Disorganized at-

tachment strategies and mental lag in infancy: Prediction of external-

izing problems at age seven. Manuscript submitted for publication.

Lyons-Ruth, K., Repacholi, B., McLeod, S., &Silva, E. (1991). Disor-

ganized attachment behavior in infancy: Short-term stability, mater-

nal and infant correlates, and risk-related subtypes. Development and

Psychopathology, 3, 377-396.

Magnusson, D., & Bergman, L. R. (1988). Individual and variable-

based approaches to longitudinal research on early risk factors. In M.

Rutter (Ed.), Studies of psychosocial risk (pp. 45-61). New York

Cambridge University Press.

Main, M. (1990). Cross-cultural studies of attachment organization:

Recent studies, changing methodologies, and the concept of condi-

tional strategies. Human Development, 33, 48-61.

Main, M, &Cassidy, J. (1988). Categories of response to reunion with

the parent at age 6: Predictable from infant attachment classifications

and stable over a 1-month period. Developmental Psychology, 24,415-

426.

Main, M., &Goldwyn, R. (in press). Adult attachment rating and clas-

sification systems. In M. Main (Ed.), A typology of human attach-

ment organization assessed in discourse, drawings, and interviews.

New York: Cambridge University Press.

Main, M., & Hesse, E. (1990). Parents' unresolved traumatic experi-

ences are related to infant disorganized attachment status: Is fright-

ened and/or frightening parental behavior the linking mechanism?

In M. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attach-

ment in the preschool years: Theory, research and intervention (pp.

161-184). Chicago: University of Chicago Press.

Main, M., Hesse, E., & van Uzendoorn, M. H. (1994). Unresolved sta-

tus in the Adult Attachment Interview: Related to dissociation, ab-

sorption, and unusual beliefs. Manuscript submitted for publication.

Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, child-

hood and adulthood: A move to the level of representation. In I.

Bretherton & E. Waters (Eds.), Growing points of attachment theory

and research. Monographs of the Society for Research in Child Devel-

opment, 50( 1-2, Serial No. 209), 66-104.

Main, M., & Solomon, J. (1990). Procedures for identifying infants as

disorganized/disoriented during the Ainsworth Strange Situation. In

M. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment

in the preschool years: Theory, research and intervention (pp. 121-

160). Chicago: University of Chicago Press.

Main, M., Tomasini, L., & Tolan, W. (1979). Differences among moth-

ers of infants judged to differ in security of attachment. Developmen-

tal Psychology, 15, 472-473.

Main, M., & Weston, D. R. (1981). The quality of the toddler's rela-

tionship to mother and to father: Related to conflict behavior and the

readiness to establish new relationships. Child Development, 52,932-

940.

Matas, L., Arend, R. A., & Sroufe, L. A. (1978). Continuity of adapta-

tion in the second year: The relationship between quality of attach-

ment and later competence. Child Development, 49, 547-556.

Moffitt, T. E. (1993). The neuropsychology of conduct disorder. Devel-

opment and Psychopathology, 5, 135-151.

Olweus, D. (1979). Stability of aggressive reaction patterns in males: A

review. Psychological Bulletin, 86, 852-875.

Oppenheim, D., Sagi, A., & Lamb, M. (1988). Infant-adult attach-

ments on the kibbutz and their relation to socioemotional develop-

ment four years later. Developmental Psychology, 24, 427-433.

Patrick, M., Hobson, R. P., Castle, D., Howard, R., & Maughn, B.

(1994). Personality disorder and the mental representation of early

social experience. Development and Psychopathology, 6, 375-388.

Patterson, G. R., & Bank, L. (1989). Some amplifying mechanisms for

SPECIAL SECTION: AGGRESSIVE BEHAVIOR PROBLEMS 73

pathologic processes in families. In M. R. Gunnar & E. Thelen

(Eds.), Systems and development: The Minnesota Symposia on Child

Psychology (Vol. 22, pp. 16-20). Hillsdale, NJ: Erlbaum.Pettit, G. S., Dodge, K. A., & Brown, M. M. (1988). Early family expe-

rience, social problem solving patterns and children's social compe-

tence. Child Development, 59, 107-120.Plomin, R., Rende, R., & Rutter, M. (1991). Quantitative genetics and

developmental psychopathology. In D. Cicchetti & S. L. Toth (Eds.),

Internalizing and externalizing expressions ofdysfunaion (Vol. 2, pp.

155-202). Hillsdale, NJ: Erlbaum.Renken, B., Egeland, B., Marvinney, D., Mangelsdorf, S., & Sroufe,

L. A. (1989). Early childhood antecedents of aggression and passive-

withdrawal in early elementary school. Journal of Personality, 57,

257-281.Robins, L. N. (1991). Conduct disorder. The Journal of Child Psychol-

ogy and Psychiatry and Allied Disciplines, 32, 193-212.Rothbaum, E, & Weisz, J. (1994). Parental caregiving and child exter-

nalizing behavior in nonclinical samples: A meta-analysis. Psycho-

logical Bulletin, 116, 55-74.Spangler, G., & Grossmann, K. E. (1993). Biobehavioral organization

in securely and insecurely attached infants. Child Development, 64,1439-1450.

Speltz, M. L., Greenberg, M. T., & DeKlyen, M. (1990). Attachmentin preschoolers with disruptive behavior: A comparison of clinic-re-

ferred and nonproblem children. Development and Psychopathology,

2, 31-46.Spieker, S. J., & Booth, C. (1988). Maternal antecedents of attachment

quality. In J. Belsky & T. Nezworski (Eds.), Clinical implications ofattachment (pp. 300-323). Hillsdale, NJ: Erlbaum.

Stewart, M. A., DeBlois, C. S., Meardon, J., & Cummings, C. (1986).Aggressive conduct disorder children. Journal of Nervous and Mental

Disease, 16S, 604-610.

Strage, A., & Main, M. (1985, April). Attachment and parent-child

discourse patterns. In M. Main (Chair), Attachment: A move to the

level ofrepresentation. Symposium conducted at the biennial meeting

of the Society for Research in Child Development, Toronto, Ontario,

Canada.Suess, G. J., Grossmann, K. E., & Sroufe, L. A. (1992). Effects of infant

attachment to mother and father on quality of adaptation to pre-

school: From dyadic to individual organisation of self. International

Journal of Behavioral Development, 15, 43-65.van Uzendoorn, M. H. (1995). Adult attachment representations, pa-

rental responsiveness, and infant attachment: A meta-analysis on thepredictive validity of the Adult Attachment Interview. PsychologicalBulletin, 117, 387-403.

van Uzendoorn, M. H., Goldberg, S., Kroonenberg, P. M., & Frenkel,

O. J. (1992). The relative effects of maternal and child problems onthe quality of attachment: A meta-analysis of attachment in clinical

samples. Child Development, 63, 840-858.van Uzendoorn, M., & Kroonenberg, P. (1988). Cross-cultural patterns

of attachment: A meta-analysis of the Strange Situation. Child Devel-

opment, 59, 147-156.

Vaughn, B. E., Lefever, G. B., Seifer, R., & Barglow, P. (1989). Attach-ment behavior, attachment security, and temperament during in-

fancy. Child Development, 60, 728.Wartner, U. G., Grossmann, K., Fremmer-Bombik, E., & Suess, G.

(1994). Attachment patterns at age six in south Germany: Predict-

ability from infancy and implications for preschool behavior. Child

Development, 65, 1014-1027.Waters, E., Wippman, J., & Sroufe, L. A. (1979). Attachment, positive

affect, and competence in the peer group: Two studies in construct

validation. Child Development, 50, 821-829.

Received November 22, 1993

Revision received August 29, 1994

Accepted May 5, 1995 •

New Editor Appointed

The Publications and Communications Board of the American Psychological Association announcesthe appointment of Kevin R. Murphy, PhD, as editor of the Journal of Applied Psychology for a six-year term beginning in 1997.

As of March 1, 1996, submit manuscripts to Kevin R. Murphy, PhD, Department of Psychology,Colorado State University, Fort Collins, CO 80523-1876.