Behavioural Brain Research 308 (2016) 83–93

Contents lists available at ScienceDirect

Behavioural Brain Research

journal homepage: www.elsevier.com/locate/bbr

Research report

Disorganized attachment in infancy predicts greater amygdala

volume in adulthood
K. Lyons-Ruth a,f,∗ , P. Pechtel b,f , S.A. Yoon c , C.M. Anderson d,e,f , M.H. Teicher d,f
a
Biobehavioral Family Studies Research Program, Cambridge Hospital, United States
b
Center for Depression, Anxiety and Stress Research, McLean Hospital, United States
c
Behavioral and Cognitive Neuroscience Program, Department of Psychology, The Graduate Center of the City University of New York, United States
d
Developmental Biopsychiatry Research Program, McLean Hospital, United States
e
Brain Imaging Center, McLean Hospital, United States
f
Department of Psychiatry, Harvard Medical School, United States

h i g h l i g h t s

• Attachment disturbance in infancy predicts larger left amygdala volume in adulthood.

• Prediction is independent of later maltreatment and later attachment quality.
• Attachment disturbance also predicts adult dissociation and limbic irritability.
• Left volume mediates the relation between early attachment and limbic irritability.
• Disturbed infant attachment may affect adult amygdala volume and psychopathology.

a r t i c l e i n f o a b s t r a c t

Article history: Early life stress in rodents is associated with increased amygdala volume in adulthood. In humans, the
Received 21 October 2015 amygdala develops rapidly during the first two years of life. Thus, disturbed care during this period may
Received in revised form 9 March 2016 be particularly important to amygdala development. In the context of a 30-year longitudinal study of
Accepted 28 March 2016
impoverished, highly stressed families, we assessed whether disorganization of the attachment relation-
Available online 6 April 2016
ship in infancy was related to amygdala volume in adulthood. Amygdala volumes were assessed among 18
low-income young adults (8 M/10F, 29.33 ± 0.49 years) first observed in infancy (8.5 ± 5.6 months) and
Keywords:
followed longitudinally to age 29. In infancy (18.58 ± 1.02 mos), both disorganized infant attachment
Amygdala
Attachment behavior and disrupted maternal communication were assessed in the standard Strange Situation Proce-
Maternal care dure (SSP). Increased left amygdala volume in adulthood was associated with both maternal and infant
Early life stress components of disorganized attachment interactions at 18 months of age (overall r = 0.679, p < 0.004).
Limbic irritability Later stressors, including childhood maltreatment and attachment disturbance in adolescence, were not
significantly related to left amygdala volume. Left amygdala volume was further associated with dis-
sociation and limbic irritability in adulthood. Finally, left amygdala volume mediated the prediction
from attachment disturbance in infancy to limbic irritability in adulthood. Results point to the likely
importance of quality of early care for amygdala development in human children as well as in rodents.
The long-term prediction found here suggests that the first two years of life may be an early sensitive
period for amygdala development during which clinical intervention could have particularly important
consequences for later child outcomes.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction 30–70% of the population attributable risk for depression, suicide

attempts, anxiety disorders and substance abuse [1]. One cen-
Early life stress is increasingly recognized as a risk factor for tral component of early adverse environments is impairment in
psychopathology. Childhood adversity has been associated with parental regulation provided to the infant [2], with meta-analyses
indicating that insecure and particularly disorganized attachment
relationships in infancy predict internalizing and externalizing
∗ Corresponding author at: Department of Psychiatry, Cambridge Hospital, 1493 problems in childhood [2–5]. Impaired regulation of early stress
Cambridge St. Cambridge, MA 02139, United States. due to poor quality care may increase risk through excessive release
E-mail address: [email protected] (K. Lyons-Ruth). of glucocorticoids and associated epigenetic modifications that

http://dx.doi.org/10.1016/j.bbr.2016.03.050
0166-4328/© 2016 Elsevier B.V. All rights reserved.
84 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93

Fig. 1. Amygdala volume in adulthood and infant attachment disturbance.

Amygdala volume was normalized and adjusted for total grey matter volume and race. Left (r = 0.679, p = 0.004) but not right (r = −0.048, p = 0.860) amygdala volume was
related to overall attachment disturbance.

alter critical developmental processes such as neurogenesis, synap- hemisphere of the amygdala. Adults exposed to maltreatment in
togenesis and myelination [6]. childhood have shown predominately right-sided differences in
The amygdala may be particularly vulnerable to such effects of amygdala volume as adults [18–20,38]. Findings based on neu-
early stressors due to high glucocorticoid receptor density [7] and to roimaging in childhood among children exposed to adversity
a postnatal developmental trajectory characterized by rapid initial (institutional rearing, maternal depression, maltreatment) have
growth and gradual pruning [8,9]. In accord with this hypothesis, varied, with some showing overall differences or right-sided dif-
translational studies show that manipulating type and timing of ferences [24,25,39] and others showing predominately left-sided
early stressors leads to persistent alterations in amygdala develop- effects [21–23].
ment and function [10–12]. Both psychological stressors and stress Several factors may contribute to these inconsistencies. First,
hormone administration stimulate dendritic arborization and for- the timing of adversity may be critical [17,30,38]. Given the amyg-
mation of new spines in the amygdala, increasing volume [13,14]. dala’s developmental trajectory, it may be particularly sensitive
This pattern is opposite to stress-induced hippocampal atrophy and to structural changes during early childhood when it is growing
less reversible when the stressor is removed [15]. In particular, at a rapid rate, and again during preadolescence when growth
animal models indicate that low maternal responsiveness (LMR) peaks and pruning takes over, as observed in the hippocampus
during infancy is a potent stressor, associated with a host of alter- [47]. Second, stress-related effects on the amygdala may be cumu-
ations in infant development, including amygdalar effects, that lative, as well as non-linear, with amygdala enlarging in response
persist into adulthood [10–12,16]. In rodent models, LMR has been to early life stress but shrinking over time in the face of contin-
indexed by natural or provoked variations in the responsiveness of ued or overwhelming later stress [48]. Third, the amygdala may
maternal care [10,11,16]. show a differential response to different types of stressors, enlarg-
Studies are now examining the relation between varieties of ing in the face of neglect or insufficient human interaction, as with
early life stress and amygdala volume among human children and prolonged institutional deprivation, or shrinking with exposure to
adults. However, the consequences of early life stress on the human the types of intense abuse often reported by individuals with bor-
amygdala and the underlying causes (e.g., dendritic arborization) derline personality or dissociative identity disorders. Thus, while
remain inconclusive [17]. there is increasing literature to suggest that the human amygdala
First, a number of studies have found volumetric differences in is affected by stressful developmental experiences, it is likely that
relation to severe life events, including childhood maltreatment, factors including timing of stressor [38], age of assessment [23],
institutional rearing, and poverty [18–26]. However, other studies development of associated psychopathology [23], genetic loading
have reported no differences in amygdala volume following similar [22], and extent of stressful life events [22] will be important to the
types of adversity [27–37]. patterning of effects on the amygdala.
Second, adult volumetric differences related to childhood mal- While most human studies have focused on stressful events
treatment have most often involved increased hemispheric volume after the infancy period, a few studies have examined amygdala
([18,19,38]; but see [20] for reduced volume). Increased volumes structure or function in relation to the effect of low maternal
have also been reported among children exposed to institutional responsiveness in the first two years of life, including pro-
rearing or maternal depression [24,25,39]. However, in other stud- longed institutional deprivation or rearing by depressed mothers
ies assessing amygdala volumes in children or adolescents, effects [21,24,25,39]. While most studies found volumetric increases
have involved reduced volumes [21–23]. In addition, smaller amyg- [24,25,39,49] and increased activity in the amygdala [50–52], at
dala volumes in adulthood were reported among individuals with least one study found volumetric decreases [21] and one study
childhood trauma and diagnoses of Borderline Personality [40,41] reported no differences [53].
or Dissociative Identity Disorders [42,43]. Aberrant amygdala vol- To date, however, most studies have inferred low maternal
ume and function have also been reported in other psychiatric responsiveness from more distal indices such as institutional care
disorders marked by affective dysregulation [41,44–46]. or maternal depression. One study has involved direct assessment
Third, volumetric differences have varied in whether they of infant behavior, but maternal behavior was not assessed [49]. To
occurred in both hemispheres or more strongly in the left or right more carefully parse how the quality of maternal care during the
 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93 85

Fig. 2. Amygdala volume and adulthood psychopathology.

Amygdala volume was normalized and adjusted for total grey matter volume and race. Left amygdala volume was strongly related to both limbic irritability (r = 0.770,
p = 0.001) and dissociative symptoms (r = 0.586, p = 0.017) in adulthood, controlling for total grey matter volume and race.

first two years of life may affect the development of limbic stress was directly observed and coded with validated inventories for dis-
systems, it will be important to directly assess quality of care during rupted maternal behavior and disorganized infant behavior at 18
infancy. months of age.
For human infants, the most widely validated paradigm for A previous report assessed amygdala volume in relation to child-
direct observation of the quality of the mother-infant attach- hood maltreatment severity among the longitudinal participants
ment relationship is the Strange Situation Procedure (SSP) [2,3,54]. studied here, as well as a group of healthy cross-sectional con-
This paradigm includes two mildly stressful 3-min separations of trols [38]. Interviewed in adulthood, the longitudinal participants
mother and infant followed by 3-min reunions during which the reported more severe exposure to maltreatment in childhood than
degree of the infant’s security or conflict behavior in approaching controls and reported lower levels of maternal care on the Parental
the mother for care is coded. This assessment yields three classifica- Bonding Instrument than controls (p = 0.002). Both left and right
tions of attachment behavior prevalent in middle class samples that amygdala volumes were larger among longitudinal participants
are considered organized and adaptive, including secure, ambiva- than controls, with least square mean adjusted amygdala volume
lent, and avoidant attachment patterns. These organized patterns 3.8% greater among the longitudinal group. Severity of childhood
are related to variations in maternal sensitivity [55,56] but have not maltreatment was significantly related to right hemisphere vol-
been consistently associated with serious family risk factors. umes but not to left volumes, with severity of maltreatment at
However, some infants do not exhibit the consistent pattern- 10–11 years of age the most important predictor of right amygdala
ing of behavior characteristic of these organized classifications, volume.
showing instead a variety of confused or conflicted behaviors However, no healthy control group was available for assess-
upon reunion with the mother. These infants are described as ing the longitudinal questions of interest in the present report.
disorganized in attachment behavior, and infant attachment dis- Here we examine a separate and orthogonal set of questions
organization has been meta-analytically associated with serious regarding infancy and childhood predictors of adult amygdala
family risk factors, including poverty, maternal psychiatric dis- volumes within the longitudinal group only. Based on previous
order, and maltreatment, as well as later externalizing behavior animal models, we predicted that the quality of the mother-child
problems [2,3]. Disorganized behavior on the part of the infant is attachment relationship in infancy might have a unique relation to
also associated with directly observed disturbed maternal inter- adult amygdala volumes, independent of later contributors includ-
action [57]. In addition, infants with disorganized attachments ing childhood maltreatment. To evaluate this prediction, we first
demonstrate a more prolonged cortisol response to stressors assessed whether overall attachment disturbance, including both
[58–60] and atypical patterns of diurnal cortisol secretion [61]. This disorganization of infant attachment behavior and disruption in
association between disorganized attachment and increased stress mother-infant interaction, was related to amygdala volume in
responses in infancy suggests the further hypothesis that disturbed adulthood. Second, we evaluated whether later stressors in child-
attachment relationships may be associated with differences in hood and adolescence, including maltreatment, could account for
amygdala volume. Given the large body of work confirming the any relations between quality of early care and adult amygdala vol-
later negative child outcomes associated with disorganized attach- umes. Finally, to explore whether any observed differences had
ment, it is important to assess the aspects of brain morphology that functional significance in adulthood, we assessed whether adult
may be affected by such early disturbed care and serve as potential psychiatric symptoms were related to amygdala volumes, includ-
mediators of later psychopathology. ing anxiety disorders, depressive disorders, substance dependence,
To explore effects of early attachment disturbance, as well as and dissociative symptoms. We also assessed symptoms of limbic
later stressors, on amygdala volumes, we recruited subjects from system irritability, which have been strongly related to child-
a low-income cohort first seen in infancy. More than half of the hood adversity [62–65]. Limbic system irritability is assessed by
longitudinal sample experienced clinically significant levels of dis- the Limbic System Checklist-33 LSCL-33; [64] which evaluates
turbed care in infancy as judged by area service providers (see the frequency with which subjects experience symptoms often
Material and Methods 2.1). Quality of mother-infant interaction encountered as phenomena of ictaltemporal lobe epilepsy, such as
86 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93

Table 1
Overview of measured variables.

Procedure Measured variables

Infancy visit (Age 18 months)

Strange Situation Procedure (SSP) Infant attachment disorganization
Maternal disrupted communication
Overall attachment disturbance
Maternal background interviews Family income/person/week
Maternal psychosocial risk

Adolescent visit (Age 20)

Adolescent-parent conflict discussion Disorganized and controlling
attachment interactions

Adult visit (Age 29)

Magnetic resonance imaging Brain gray matter volume in amygdala,
hippocampus, caudate and thalamus;
total grey matter volume
Maltreatment and Abuse Chronology ofExposure Scale (MACE) Severity of childhood maltreatment
Structured Clinical Interview for DSM Axis I Anxiety disorders
(SCID I) Major depressive disorder
Substance dependence
Dissociative Experiences Scale (DES) Dissociative symptoms
Limbic System Checklist-33 (LSCL-33) Limbic irritability

brief hallucinatory events [66]. Table 1 provides a listing of study II psychopathology on the SCID in adulthood (␩ = 0.045–0.032,
measures. p = 0.753 −0.845). Participants were marginally more likely than
Consistent with previous research [24], the hippocampus, cau- non-participants to have been referred for clinical services in
date, and thalamus were selected a priori as contrast or control infancy ␸ = 0.219, p = 0.093 and to have been classified in the
structures. The hippocampus is also highly stress susceptible, with Strange Situation Procedure as having a disorganized attachment
sensitive periods in early childhood and during the pubertal period, in infancy (␸ = 0.235, p = 0.084).
but effects of stress are less persistent [15,24,47]. The remaining
structures are less susceptible to early stress due to their develop-
2.2. Attachment assessments in infancy
mental trajectories and/or lower glucocorticoid receptor densities
[67–69].
Mothers and infants were videotaped in the well-validated
Strange Situation Procedure (SSP) at 18.58 (±1.02) months infant
2. Material and methods age [2,56]. In the SSP, the infant is videotaped in a playroom during
a series of eight structured 3-min episodes in which the mother
2.1. Participants leaves and rejoins the infant twice. The procedure is designed
to be mildly stressful in order to activate the infant’s attach-
Participants were 18 young adults (8 M/10F, 29.33 ± 0.49 years) ment behavior. Infant attachment behavior was reliably coded
first recruited as infants (8.5 ± 5.6 months) as part of a longitudinal by coders naïve to all other data [70]. The bivariate classification
study of the effects of social risk factors on child development [70]. for organized versus disorganized attachment was used in this
The study was approved by the Harvard Medical School, Cambridge report, following earlier precedent [2]. Maternal behavior was reli-
Hospital, and McLean Hospital IRBs. Subjects provided informed ably coded over all episodes of the SSP using the well-validated
written consent and were reimbursed $100 for their time. Atypical Maternal Behavior Instrument for Assessment and Clas-
The larger study from which these participants were recruited sification (AMBIANCE) by coders naïve to all other data [74]. The
consisted of 76 families who were at or below 200% of federal AMBIANCE coding protocol yields a scaled score [1–7] for over-
poverty levels. 52.6% of families had been referred for clinical all Level of Disrupted Communication that takes into account five
help in parenting their infants during the first year of life. Rela- subtypes of maternal disrupted communication: (1) affective com-
tions between infancy risk factors and maladaptive developmental munication errors (e.g. giving contradictory cues; non-response
outcomes in the larger longitudinal cohort have been well charac- or inappropriate response to clear infant cues), (2) role confusion
terized from infancy to adulthood [70–73]. (e.g. self-referential or sexualized behavior), (3) negative-intrusive
At age 29, 33 participants were relocated and screened for behavior (e.g. negative attributions about the infant; mocking or
inclusion in the current study. Eighteen met inclusion criteria and teasing the infant; physical intrusiveness), (4) fearful-disoriented
participated in the MRI study. The remaining subjects did not meet behavior (e.g. appearing frightened by the infant; disoriented wan-
MRI safety criteria or reported substance abuse in the past six dering in infant’s presence), and (5) withdrawal (e.g. fails to greet
months or had a significant medical or neurological condition. All infant; interacts silently; backs away from infant approach). Moth-
participants except one were right-handed. Seventy-two percent ers who are rated at 5 or above on the overall Level of Disruption
of study participants had been referred for parent-infant clinical Scale are classified as Disrupted. Validity of maternal AMBIANCE
services during the first 18 months of life. classification has been confirmed by meta-analysis in relation to
Participants in the MRI study were representative of the larger infant attachment disorganization [57] and, in individual studies, in
longitudinal cohort from which they were recruited. They did not relation to adolescent psychiatric outcomes [72,75,76]. Test-retest
differ from the remainder of the cohort in family demographic data indicate substantial stability in maternal behavior over periods
characteristics effect sizes: family income ␩ = 0.205, p = 0.130; male ranging from 8 months to 5 years, meta-analytic stability coefficient
gender ␸ = −0.136, p = 0.234; mother single parent ␸ = −0.022, t = 0.56 (N = 203) [57].
p = 0.867; mother high school only ␸ = −0.026, p = 0.848; ethnic The above infant and maternal measures are theoretically and
minority status ␸ = 0.052, p = .652; severity of childhood mal- empirically related [57]. To capture the combined effect of overall
treatment (␩ = 0.100, p = 0.460); or extent of Axis I or Axis attachment disturbance, we also created a summary attachment
 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93 87

measure as follows: 0 = No maternal or infant attachment distur- somatic disturbances, brief hallucinatory events, visual phenom-
bance in the Strange Situation Procedure; 1 = Disturbance on either ena, automatisms, and dissociative experiences. Psychometric
maternal or infant assessment in the Strange Situation Procedure; studies have shown that the Limbic System Checklist-33 has high
2 = Disturbance on both maternal and infant assessments in the test-retest reliability (r = 0.92, N = 16) and scores are strongly influ-
Strange Situation Procedure. enced by childhood adversity [64].

2.3. Other risk factors in infancy, childhood, and adolescence

2.5. Imaging data acquisition and processing
Sociodemographic and psychosocial risk factors in infancy were
coded from maternal interviews at study entry in infancy to yield
Data were collected using a 3T TIM Trio scanner (Siemens
measures of per person weekly family income and maternal psy-
AG, Erlangen, Germany) with a 32-channel head coil using
chosocial risk, which was coded positive if any of the following
a T1-weighted Magnetization Prepared Rapid Gradient Echo
three risk factors were present: 1) mother had a history of psy-
(MPRAGE) pulse sequence (TE = 2.25 ms; TR = 2100 ms; FA = 12;
chiatric hospitalization, 2) mother had a history of state protective
FOV = 256 mm; slice number = 128; voxel size = 1.0 × 1.0 × 1.3 mm;
service involvement for maltreating a child, and 3) mother reported
slice thickness = 1.33 mm) in the sagittal plane (scan duration:
clinically significant depressive symptoms on the CES-Depression
6 min). Gray matter volume (GMV) in amygdala, hippocam-
scale, using the established cutpoint of ≥ 16 [77].
pus, caudate and thalamus and total grey matter volume were
Severity of exposure to childhood maltreatment was assessed
assessed using FreeSurfer 5.1 (http://surfer.nmr.mgh.harvard.edu)
at age 29 using the Maltreatment and Abuse Chronology of Expo-
[84]. Although this program originated as a method for assessing
sure Scale MACE; [1,38,65]. The participant reports on exposure
cortical structure [85–88], it has evolved to include robust tools
to ten types of maltreatment during each year of childhood from
for subcortical volume analyses [84,89]. Voxels within subcorti-
ages 6–18, including childhood sexual abuse, parental verbal abuse,
cal regions are labeled using an elaborate process based on both
parental non-verbal emotional abuse, parental physical abuse, wit-
a subject-independent probabilistic atlas derived from a hand-
nessing of intra-parental physical violence or violence toward
labeled training set and on subject-specific measures [89]. This
siblings, peer verbal abuse, peer physical abuse, parental emotional
procedure has been found to label the brain in a manner that
neglect, and parental physical neglect. Items within each category
is statistically indistinguishable from those provided by experi-
were selected using item response theory, and category scores were
enced manual raters [84]. Overall, FreeSurfer provides one of the
summed to provide a total score, which has shown excellent test-
most reliable automated brain segmentation methods for assess-
retest reliability across age (r = 0.91, n = 75).
ing the amygdala and these measures correlate highly with expert
Quality of attachment in adolescence was reliably assessed at
hand tracings [90]. Automated segmentation eliminates differences
age 20 using the validated Goal-Corrected Partnership in Adoles-
between studies and facilitates replication by other investigators
cence Coding System (GPACS; [78]) applied to a videotaped conflict
[19,91]. The authors (CMA, PP) visually inspected all T1-weighted
discussion task in which parent and young adult discussed a pres-
and automated images. Manual adjustments were not required.
elected topic of conflict in their relationship for 10 min. The GPACS
Regional volumes and total GMV were extracted and exported into
coding system includes the rating of each videotape on 10 five-
SPSS 19.0 and R [92] for statistical analysis.
point scales. All coders were naïve to other study data. Coding of
all scales was reliable across raters, ri = 0.75–0.96 (N = 16). Confir-
matory Factor Analysis of the ten scales yielded one factor indexing
Collaborative/Organized interaction and three factors indexing dis- 2.6. Statistical analysis
organized forms of interaction, including Punitive, Disoriented, and
Role-Confused interaction. Good validity has been demonstrated in Regional GMV was centered and scaled for each region to pro-
relation to concurrent psychopathology and relations to romantic vide an arbitrary mean of 100 and SD of 10, to facilitate comparisons
partners, as well as to disorganized attachment in infancy [78]. between regions. Our primary hypothesis was that greater attach-
ment disturbance would be associated with greater amygdala
2.4. Adult psychopathology volume. Therefore our analysis strategy focused on assessing lin-
ear effects through partial Pearson and point-biserial correlations.
The Structured Clinical Interview for DSM Axis I was used to Hypothesis testing of relations between neuroimaging measures
assess the presence of Axis I disorders (SCID [79]). The SCID was and aspects of developmental history and adult psychopathology
administered in the laboratory by a trained clinician. The SCID was conducted using partial correlations controlling for relevant
yields reliability kappa’s around 0.61 for current diagnosis and 0.68 covariates (SPSS 22.0). Multiple imputation was employed to esti-
for lifetime diagnoses, comparable to other structured diagnostic mate missing data [93], using the Markov Chain Monte Carlo
interviews [80,81]. Diagnoses with adequate frequency of occur- procedure [94] in SPSS software. The percentages of missing data
rence for analysis in the current sample included anxiety disorder, ranged from 0 to 11.1%, well within the recommended ranges for
major depressive disorder, and substance dependence. imputation procedures. Mediation analyses were carried out using
Dissociation was assessed with the Dissociative Experiences bootstrapping techniques with INDIRECT [95].
Scale [82]. The DES is a 28-item questionnaire assessing the extent We predicted that more severe early attachment disturbance
of dissociative experiences. Respondents indicate how much of would be associated with increased GMV in the amygdala but
the time they experience particular dissociative phenomena on a would not be associated with similar increase in GMV in hippocam-
scale from 0 to 100%. A meta-analysis has demonstrated convergent pus, caudate, or thalamus. Thus, we did not adjust the significance
validity with other measures of dissociation, predictive validity in of amygdala measures for these planned negative control compar-
relation to Dissociative Identity Disorder, and robust test-retest isons. In addition, we had hypothesized a priori that the amygdala
reliability (␣ = 0.93; [83]). would be most sensitive to observed quality of care in infancy, and
The Limbic System Checklist-33 (LSCL-33; [64]) evaluates that this relation would not be accounted for by effects of more
limbic irritability, that is, the frequency with which subjects distal risk factors in infancy, childhood, and adolescence. There-
experience symptoms often encountered as phenomena of ictal- fore, we also did not adjust the significance of amygdala measures
temporal lobe epilepsy [66]. These items consist of paroxysmal for these additional planned negative control comparisons.
88 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93

Table 2 bance at 18 months of age was strongly predictive of adjusted

Descriptive data.
left amygdala volume in adulthood (partial r = 0.679, df = 14,
Mean ± SD or N (%) p = 0.004, 95% CI = 0.31–0.87; Fig. 1). This association was signifi-
Adult characteristics cantly stronger than the correlation between overall attachment
Age (yrs.) 29.3 ± 0.5 disturbance and adjusted right amygdala volume (right partial
Female 10 (57%) r = −0.048, p = 0.860, 95% CI = –0.5–0.43; Meng’s test for depen-
White 14 (78%) dent correlations Z = 2.343, p = 0.020) [96]. This asymmetry was also
Single 11 (61%)
reflected in an increased laterality index (r = 0.513, df = 16, p = 0.030,
College degree 2 (11%)
95% CI = 0.06–0.79). Overall attachment disturbance accounted for
Psychosocial risk factors in infancy
14.8% of variance in volume of the left amygdala.
Family income/person/week $45.3 ± 22.6
Mother any psychosocial risk 12 (66.7%)
As expected, overall attachment disturbance was not signif-
icantly related to hippocampal (left partial r = 0.177, p = 0.512,
Childhood and adolescent risk factors
95% CI = −0.32–0.59; right partial r = 0.400, p = 0.125, 95%
Childhood maltreatment
Severity of maltreatment in childhood 26.8 ± 14.5 CI = −0.08–0.73), caudate (left partial r = 0.410, p = 0.115; right
Attachment disturbance in adolescence partial r = 0.400, p = 0.125, 95% CI = −0.07–0.74), or thalamic
Punitive interaction 2.2 ± 0.6 (left partial r = 0.210, p = 0.435, 95% CI = −0.29–0.62; right partial
Caregiving interaction 2.2 ± 0.9 r = 0.052, p = 0.848, 95% CI = −0.43–0.51) volumes. However, the
Disoriented interaction 1.5 ± 0.6
size of the relations to right hippocampus and left and right
Adult symptomatology caudate suggests that further work may be warranted in larger
Anxiety disorder 4 (22.2%)
samples.
Major depressive disorder 4 (22.2%)
Substance dependence 7 (38.8%) Given these initial findings, overall attachment disturbance
Dissociative symptoms 7.1 ± 4.6 was decomposed into the separate variables for maternal dis-
Limbic irritability 16.9 ± 9.2 rupted interaction and infant disorganization. Both maternal
Note. N = 18. (partial r = 0.595, p = 0.015, 95% CI = 0.18–0.83) and infant (partial
r = 0.543, p = 0.030, 95% CI = 0.1–0.81) assessments were strongly
predictive of left amygdala volume. Relations to right amyg-
3. Results
dala volume were not significant (maternal partial r = −0.272,
p = 0.308, 95% CI = −0.66–0.22; infant partial r = 0.146, p = 0.590, 95%
3.1. Covariates
CI = −0.34–0.57). 1
Given the significant prediction from maternal disrupted com-
Total grey matter volume (TGMV) was correlated strongly with
munication found above, we also examined the five component
amygdala volume and was controlled in all analyses to reduce
scores for types of maternal disrupted interaction. Consistent
error variance. Given limited degrees of freedom, we restricted
with animal data on low maternal responsiveness, maternal
additional covariates to those showing at least a moderate relation-
withdrawing behavior and contradictory communications made
ship (r ≥ 0.25) to amygdala volume. Age, gender and race were also
the strongest contributions to left amygdala volume (withdraw-
assessed in relation to amygdala volumes and only race (white = 0,
ing partial r = 0.397, p = 0.128, 95% CI = −0.09–0.73; contradictory
black = 1) was retained as a covariate due to its moderate associ-
communications partial r = 0.405, p = 0.120, 95% CI = −0.08–0.73).
ation with left amygdala volume (r = 0.34). Age and gender were
Though not reaching significance with this sample size, these mod-
eliminated due to weak relationships once TGMV was controlled.
erate effect sizes suggest that further work on maternal withdrawal
Hence, all partial correlations below are adjusted for TGMV and
and contradictory cues to the infant would be fruitful. Other types of
race.
disturbed interaction were negligible or negative in effect, includ-
ing negative intrusive behavior (partial r = 0.111, p = 0.682, 95%
3.2. Descriptive data for predictor variables
CI = −0.38–0.55), role confusion (partial r = −0.280, p = 0.294, 95%
CI = −0.66–0.22), and disorientation (partial r = −0.190, p = 0.481,
In infancy, 67% (n = 12) of infants were classified as disorganized
95% CI = −0.60–0.30).
in their attachment behavior toward mother and 61% (n = 11) of
mothers were classified as disrupted in their responses to their
infants’ attachment cues. On the three-level variable for overall
attachment disturbance, 16.7% (n = 3) displayed no disturbance
on either infant or maternal assessments, 38.9% (n = 7) displayed
disturbance on one assessment, and 44.4% (n = 8) displayed distur-
bance on both assessments. Descriptive data on other measures are
shown in Table 2. turbance on left amygdala volume, F (2, 13) = 5.97, p = 0.031. Post-hoc comparisons
revealed that the group with both maternal and infant attachment disturbance (95%
3.3. Overall attachment disturbance in infancy CI 1461.8–1604.3) had greater left amygdala volume compared to the groups with
no attachment disturbance (95% CI 1363.3–1575.3; p = 0.008) and with either mater-
nal or infant attachment disturbance alone (95% CI 1591.0–1729.6; p = 0.029). The
Controlling for relevant covariates, we first assessed whether group with either maternal or infant attachment disturbance alone did not differ sig-
greater overall attachment disturbance was associated with greater nificantly from the group with no attachment disturbance (p = 0.314). However, the
amygdala volume, as hypothesized.1 Overall attachment distur- very low power for examining subgroup differences should be taken into account in
interpreting the post-hoc comparisons. Similarly, for maternal disrupted interaction
there was a significant between-group effect of disrupted interaction on later left
amygdala volume, F (1,14) = 7.72, p = 0.017 (not-disrupted 95% CI = 1414.6–1567.7;
1
Given low power, our primary analysis strategy focused on assessing a linear disrupted 95% CI 1566.1–1685.3), as well as a significant between-group effect of
effect through Pearson and point biserial partial correlations. However, we further infant attachment disorganization on left amygdala volume, F (1, 14) = 5.95, p = 0.031
tested the robustness of these effects by treating overall attachment disturbance, (organized 95% CI = 1391.3–1574.3; disorganized 95% CI = 1554.2–1669.6). For right
maternal disrupted communication, and infant disorganization as categorical vari- amygdala volume, there were no between-group effects of overall attachment dis-
ables in an ANCOVA, with TGMV and race as covariates. Consistent with the previous turbance (F (2, 13) = 0.06, p = 0.940), maternal disrupted interactions (F (1, 14) = 1.22,
analyses, there was a significant between-group effect of overall attachment dis- p = 0.328), or infant disorganization (F (1, 14) = 0.36, p = 0.597).
 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93 89

3.4. Other risk factors in infancy, childhood, and adolescence Table 3

Relations between Overall Attachment Disturbance and Psychiatric Symptoms in
Adulthood.
We then assessed whether the obtained relation between over-
all attachment disturbance and left amygdala volume could be Overall attachment
disturbance
explained by other risk factors in infancy, childhood, and adoles-
Pearson partial r (95% CI)
cence. In infancy, risk factors such as low family income or maternal
psychosocial risk might serve as third variables explaining the Anxiety disorder 0.111 (−0.38–.55)
Major depressive disorder −0.119 (−0.56–.37)
above relations. However, neither left or right amygdala volumes
Substance dependence 0.422x (−0.06–.74)
were significantly related to family income (left partial r = −0.054, Dissociative symptoms 0.586** (0.16–0.83)
p = 0.843, 95% CI = −0.51–0.42; right partial r = −0.374, p = 0.154, Limbic Irritability 0.606** (0.19–0.84)
95% CI = −0.72–0.11) or presence of serious maternal psychoso- Note. X p < 0.10, ** p < 0.01.
cial risk (history of psychiatric hospitalization, involvement with
state protective services, or high depressive symptoms; left partial
r = −0.003, p = 0.991, 95% CI = −0.47–0.46; right partial r = −0.218, Left
p = 0.417, 95% CI = −0.62–0.28). In addition, controlling for these risk amygdala
.39(.11)** volume .80(.44)†
factors did not alter the significant relation of overall attachment a b
disturbance to left amygdala volume, Fchg (1,10) = 6.22, p = 0.032,
R2chg = 0.090. Overall .32(.26) Limbic
attachment
Because later difficulties in family relationships might also serve c’ irritability
disturbance
as third variables accounting for prediction from overall attach-
ment disturbance to later amygdala volume, we then assessed
whether severity of childhood maltreatment might account for the Overall .63(.21)* Limbic
obtained association between amygdala volume and attachment attachment
c irritability
disturbance. Severity of maltreatment was unrelated to left amyg- disturbance
dala volume (left partial r = −0.006, p = 0.982, 95% CI = −0.47–0.46).
In addition, controlling for severity of maltreatment did not alter Fig. 3. Left amygdala volume mediating the relation between infant attachment
disturbance and limbic irritability in adulthood.
the relation of overall attachment disturbance to left amygdala vol-
The association between attachment disturbance and limbic irritability in adulthood
ume, Fchg (1,12) = 7.07, p = 0.020, R2chg = 0.090. was mediated by adjusted left amygdala volume. The values in the figure are r coef-
In relation to right amygdala, however, and consistent with ficients (bootstrap standard error) for each path. Path a, direct effect of attachment
other literature, severity of maltreatment was moderately asso- disturbance on left amygdala volume. Path b, direct effect of left amygdala volume
ciated with increased volume (partial r = 0.377, p = 0.150, 95% on limbic irritability. Path c’, direct effect of attachment disturbance on limbic irri-
tability. Path c, total effect of attachment disturbance on limbic irritability. Indirect
CI = −0.11–0.72). The effect did not reach significance with this
effect ab = 0.481 [95% CI = 0.175–0.977], p < 0.01.
sample size, but the effect size was similar to those that did reach †p < .10, *p < 0.05, **p < 0.01.
significance in studies with larger samples [38].
Finally, we assessed whether later attachment disturbance in
adolescence might account for the link between early attach- CI = −0.49–0.44; substance dependence: partial r = −0.154, p = 0.569,
ment disturbance and left amygdala volume. However, none of 95% CI = −0.58–0.34).
the three forms of disturbed attachment in adolescence were In contrast, left amygdala volume was strongly related to
significantly related to left or right amygdala volume (punitive both dissociative symptoms (partial r = 0.586, p = 0.017, 95%
left partial r = 0.151, p = 0.577, 95% CI = −0.34–0.58; right partial CI = 0.16–.83) and limbic irritability (partial r = 0.770, p = 0.001, 95%
r = −0.179, p = 0.507, 95% CI = −0.6–0.31; disoriented left partial CI = 0.47–0.91). There were no corresponding associations between
r = −0.006, p = 0.982, 95% CI = −0.47–0.46; right partial r = −0.237, right amygdala volume and either limbic irritability (partial
p =0.377, 95% CI = −0.63–0.26; role-confused left partial r = 0.209, r = 0.032, p = 0.906, 95% CI = −0.44–0.49) or dissociative symptoms
p = 0.437, 95% CI = −0.29–0.62; right partial r = −0.293, p = 0.271, (partial r = −0.310, p = 0.242, 95% CI = −0.68–0.18) (Fig. 2).
95% CI = −0.20–0.67). In addition, controlling for all three forms of Similar to left amygdala volume, early attachment disturbance
attachment disturbance in adolescence did not alter the relation was also strongly related to both dissociation and limbic irritability
of early attachment disturbance to left amygdala volume, Fchg [1, but did not reach significance for the three psychiatric disorders
11] = 12.50, p = 0.005, R2chg = 0.15. above (Table 3). However, the moderately strong relation between
attachment disturbance and substance dependence (r = 0.422) indi-
cates a fruitful direction for further work.
3.5. Psychiatric symptoms in adulthood Finally, we assessed whether left amygdala volume might
mediate the relation between early attachment disturbance and
The last set of analyses assessed whether enlarged left later dissociation and limbic irritability. Mediation analyses, using
amygdala volume had functional significance for psychiatric recommended bootstrapping methods [95], indicated that the pre-
outcomes in adulthood. The three psychiatric diagnoses with diction from overall attachment disturbance in infancy to limbic
sufficient frequency for analysis in this cohort (anxiety disor- irritability in adulthood was mediated by increased volume of the
der, major depressive disorder, lifetime substance dependence) left amygdala (bias-corrected confidence interval did not contain
were not significantly related to left amygdala volume (anxiety zero (CI 2.05–13.27); Fig. 3). In regards to dissociation, mediation
partial r = 0.428, p = 0.098, 95% CI = −0.05–0.75; major depres- by left amygdala volume of the relation between early attachment
sion, partial r = −0.111, p = 0.682. 95% CI = −0.55–0.38; substance disturbance and later dissociation could not be confirmed with this
dependence, partial r = 0.404, p =0.121, 95% CI = −0.08–0.73). How- modest sample size (CI −0.25–4.52).
ever, again, the effect sizes for anxiety disorders and substance
dependence were moderate and warrant further investigation 4. Discussion
in larger samples. Right amygdala volume was unrelated to
these disorders (anxiety disorder: partial r = −0.090, p = 0.740, 95% Congruent with animal studies on poor maternal care [10,11]
CI = −0.39–0.53; major depression: partial r = −0.031, p = 0.909, 95% and human data on institutional care [21,24,39], the present find-
90 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93

ings underscore the potential importance of the quality of early and 11 years of age [38], but show left amygdala enlargement here
maternal care for the development of the amygdala. In addition, in relation to quality of early care. These differential effects within
these findings add to the evidence base that human amygdala vol- the same subjects, using well-validated assessments of both early
ume is sensitive not only to very atypical rearing conditions, such care and later maltreatment, suggest some differential hemispheric
as institutional care, but also to serious disturbances in maternal sensitivity to type and timing of childhood stressors.
care observed among infants reared at home [21,25]. The profile of maternal behavior associated with left amyg-
The high rate of infant attachment disorganization in this sam- dala enlargement was similar to that described in the translational
ple (67%) points to a high level of disturbance, compared to a rate literature on low responsive mothers, with mothers’ withdraw-
of 15% among infants in middle-income samples [2], 25% of infants ing behaviors (e.g. failure to greet, backing away from the
in low-income families [2], and up to 85–90% of maltreated infants infant, interacting from a distance, interacting silently) and con-
[97–100]. Cyr et al. [100] also found that families with multiple tradictory communications (failure to respond appropriately to
social risk factors, but no reported maltreatment, were as likely clear infant cues; mixed communications such as sweet voice
to have disorganized infants as maltreating families. As noted ear- but negative message) showing the largest associations with
lier, this was a degree of risk that was evident to pediatric nurses amygdala volume (r = 0.397 and 0.405, respectively). In contrast,
and other service providers in the community and, in many cases, mother’s negative-intrusive, disoriented, and role-confused behav-
resulted in referral to infant mental health services. iors showed negligible to negative associations to left amygdala
Notably, attachment assessments are among the best validated volume (r = 0.11, −0.19, and −0.28, respectively). Behavioral studies
assessments of early risk in developmental science. Supporting have confirmed that, even in the context of overall disorganiza-
large scale studies and meta-analyses have confirmed their rela- tion, maternal withdrawing behavior is associated with higher
tion to concurrent risk factors in infancy and their predictive value levels of infant approach behavior, while maternal negative- intru-
for behavior problems in childhood [2–4,54,101,102]. In addition, sive, disoriented, and role-confused behaviors are associated with
these studies have consistently pointed to disorganized attach- increased infant avoidance of the mother [74]. Thus, the mother’s
ments as the forms of insecure attachment that carry the greatest relative unavailability to the infant may be important in stimulat-
risk for later psychopathology [2–4,101,102]. Thus, these results ing infant hypervigilance to the mother’s whereabouts and infant
suggest that further work exploring the neural mechanisms asso- approach motivation, mediated in part by the left amygdala.
ciated with deviations in early attachment relationships may be Consistent with this thinking, the infant’s disorganized behav-
particularly fruitful for understanding developmental trajectories ior is thought to indicate that the infant cannot organize a strategy
toward pathology. of consistently avoiding or consistently approaching the caregiver
The specificity of the effect of early disturbance on regional and, instead, exhibits confused, contradictory or aborted attempts
brain volumes was also important. We hypothesized that the amyg- to approach the mother for comfort. While consistent avoidance
dala would be most affected by quality of early care compared of the mother at reunion has been associated with maternal irri-
to hippocampus, thalamus, or caudate, based on evidence that tation, anger, rejection, and interfering behavior [56,74,111–113],
the amygdala has high glucocorticoid receptor density [7] and is infant disorganized behavior has been associated with mixed signs
developing more rapidly shortly after birth [9]. In accord with of both maternal hostility and withdrawal [57,74,114]. The pattern-
expectations, these other regions were not strongly or significantly ing of maternal data here suggests that the mother’s withdrawal
related to early attachment disturbance. However, as noted, the and contradictory communications may be particularly important
medium effect sizes for left and right caudate [103] and right hip- to the left amygdalar effects related here to disorganized attach-
pocampus, while not as strong as those for the amygdala, may ment, in contrast to the right hemisphere effects recently associated
warrant further study in larger samples. with infant organized avoidant behavior [49].
In addition, the effect of early care on amygdala volume in Finally, intriguing findings from rodent models indicate that
adulthood was not explained by intervening family disturbances low maternal responsiveness (LMR) is related to differential left
in childhood and adolescence, including both severity of childhood hemisphere responding in medial prefrontal cortex (mPFC). Ani-
maltreatment and observed disturbances in attachment in adoles- mals exposed to LMR develop excessive hemispheric asymmetry
cence. Thus, this work adds to the evidence from animal models in mPFC DA stress responsivity, such that peak DA stress responses
that early stress-related dendritic growth in the amygdala may be in left mPFC of LMR animals are two to three times greater
resistant to change [10–12,15]. than those of controls [115]. Although mPFC and amygdala have
The results also contribute additional intriguing evidence extensive bidirectional connections and often function in tandem
regarding the possibility of differences in left versus right amygdala [110,116,117], these mPFC findings on greater left responding
sensitivity to the types and timing of critical stressors in infancy under conditions of deprived care are suggestive but have yet to
and childhood. While both left and right amygdala develop shortly be extended to amygdala structure or function.
after birth, the left amygdala develops more rapidly, while the right In relation to the current findings, one potential hypothesis
amygdala has a more prolonged developmental period [9]. In addi- is that low levels of maternal care differentially activate the left
tion, left amygdala has been shown to be particularly responsive to amygdala and promote infant hypervigilance to the whereabouts
maternal stimuli in childhood [104], as well as to the discrimination of the mother, possibly also dampening early right amygdala activ-
of family faces more generally [105]. Such findings are consis- ity [115]. In this view, the amygdala may be lateralized to respond
tent with the motivational hypothesis emerging from EEG studies differentially to different evolutionary threats to survival, and this
that the left hemisphere is more involved in motivating approach laterality may be enhanced under threat of abandonment. Models
behavior and the right hemisphere is more involved in motivating of prototypic threat have focused on threat of attack, with proto-
avoidance behavior [106]. Thus, right amygdala may be differ- typic responses of fight, flee, or freeze. An equally potent survival
entially responsive to negative or threatening stimuli [107–109], threat to the infant is the threat of neglect or abandonment by the
and alterations in structure [38,109,110] of the right amygdala caregiver. Fight, flight, or freeze responses would be counterpro-
in adulthood have been particularly related to exposure to child- ductive in response to maternal withdrawal and might decrease
hood maltreatment. Also notably, compared to cross-sectional survival. When the mother is withdrawn or inattentive, the adap-
controls, the same participants studied here exhibited right amyg- tive response is to increase the intensity of the distress signal to the
dala enlargement in relation to the severity of maltreatment in parent and to pursue proximity and contact with the parent, that
middle childhood, with peak effect from maltreatment between 10 is, to seek and squeak rather than to fight or flee.
 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93 91

The consequences of increased left amygdala volume for adult (NIH Award #UL1 RR 025758) and the Frederick Leonhardt Foun-
symptomatology further suggest the potential clinical implications dation awarded to KLR, MHT, and PP, R01 MH091391 awarded to
of the present findings. The Limbic System Checklist-33 (LSCL-33) MHT, and R01 MH062030 awarded to KLR. The authors would like
was created to test the hypothesis that childhood adversity kin- to thank Nancy Hall Brooks, Sarah Richardt, and Cynthia McGreen-
dled the amygdala and associated limbic system and increased the ery for their invaluable contributions to participant recruitment
occurrence of symptoms characteristically observed in individuals and data collection.
with temporal lobe epilepsy [64]. In previous work, these ‘limbic
irritability’ symptoms were dramatically elevated among adults
with maltreatment histories [33,63,65,118]. In addition, a signif- References
icant inverse correlation has been found between LSCL-33 ratings
and integrity of left fornix [119], integrity of left inferior longitudi- [1] M.H. Teicher, J.A. Samson, Childhood maltreatment and psychopathology: a
nal fasciculus [62], and T2-relaxation time (an indirect measure of case for ecophenotypic variants as clinically and neurobiologically distinct
subtypes, Am. J. Psychiatry 170 (10) (2013) 1114–1133.
relative cerebral blood flow) in the cerebellar vermis [120]. These [2] M.H. van I.Jzendoorn, C. Schuengel, M.J. Bakermans-Kranenburg,
fiber tracts and regions all have specific connections to components Disorganized attachment in early childhood: meta-analysis of precursors,
of the limbic system [121]. Current findings using a prospective concomitants, and sequelae, Dev. Psychopathol. 11 (2) (1999) 225–249.
[3] R.P. Fearon, M.J. Bakersman-Kranenberg, M.H. Van I.Jzendoorn, A. Lapsley,
design extend this work to indicate that limbic irritability is also G.I. Roisman, The significance of insecure attachment and disorganization in
predicted by serious disturbances in the attachment relationship the development of children’s externalizing behavior: a meta-analytic
in infancy. Notably, this prediction is mediated here by increased study, Child Dev. 81 (2) (2010) 435–456.
[4] S. Madigan, L. Brumariu, V. Villani, L. Atkinson, K. Lyons-Ruth,
volume of the left amygdala.
Representational and questionnaire measures of attachment after early
Despite these suggestive findings, the interpretation of morpho- childhood: a meta-analysis of relations to child internalizing and
metric change remains challenging, as multiple factors contribute externalizing problems, Psychol. Bull. 142 (4) (2016) 367–399.
[5] S. Madigan, L. Atkinson, K. Laurin, D. Benoit, Attachment and internalizing
to structural volume (e.g., size of neurons and glial cells, density,
behavior in early childhood: a meta-analysis, Dev. Psychol. 49 (4) (2013)
vascularity) [122]. Based on randomized animal models, possible 672–689.
neurobiological mechanisms relating early attachment disturbance [6] S.J. Lupien, B.S. McEwen, M.R. Gunnar, C. Heim, Effects of stress throughout
to increased left amygdala volume include both early stress-related the lifespan on the brain, behaviour and cognition, Nat. Rev. Neurosci. 10 (6)
(2009) 434–445.
overproduction of new spines in the amygdala and later inadequate [7] A. Peiffer, N. Barden, M.J. Meaney, Age-related changes in glucocorticoid
pruning. In addition, because amygdala volume was assessed in receptor binding and mRNA levels in the rat brain and pituitary, Neurobiol.
adulthood, non-linear developmental change in regional volumes Aging 12 (5) (1991) 475–479.
[8] C. Payne, C.J. Machado, N.G. Bliwise, J. Bachevalier, Maturation of the
and intervening stressful events that were not measured here may hippocampal formation and amygdala in Macaca mulatta: a volumetric
have influenced the relation between quality of early care and left magnetic resonance imaging study, Hippocampus 20 (8) (2010) 922–935.
amygdala volume. Although we examined the intervening factors [9] A. Uematsu, et al., Developmental trajectories of amygdala and
hippocampus from infancy to early adulthood in healthy individuals, PLoS
most prominent in current literature, other factors may have been One 7 (10) (2012) e46970.
involved in mediating the observed relation. In addition, the small [10] P.M. Plotsky, et al., Long term consequences of neonatal rearing on central
sample size makes this a ‘proof of construct’ preliminary study corticotropin-releasing factor systems in adult male rat offspring,
Neuropsychopharmacology 30 (12) (2005) 2192–2204.
that needs to be followed up by larger longitudinal studies of lim-
[11] M.M. Cohen, et al., Early-life stress has persistent effects on amygdala
bic development in relation to quality of care in infancy. We are function and development in mice and humans, Proc. Natl. Acad. Sci. 110
currently conducting one such study. (45) (2013) 18274–18278.
[12] C. Raineki, S. Moriceau, R. Sullivan, Developing a neurobehavioral animal
Although there is current concern regarding replicability of
model of infant attachment to an abusive caregiver, Biol. Psychiatry 67
small sample findings in neuroscience [123] we would argue that (2010) 1137–1145.
small-scale, carefully done exploratory studies play an important [13] R. Mitra, S. Jadhav, B.S. McEwen, A. Vyas, S. Chattarji, Stress duration
role in emerging areas of science, by exploring a broad range of modulates the spatiotemporal patterns of spine formation in the basolateral
amygdala, Proc. Natl. Acad. Sci. 102 (26) (2005) 9371–9376.
possible influences and assessment paradigms to identify those [14] A. Vyas, S. Jadhav, S. Chattarji, Prolonged behavioral stress enhances
that warrant more expensive and long-range studies. Committing synaptic connectivity in the basolateral amygdala, Neuroscience 143 (2)
to large-scale studies too soon in the development of a field runs (2006) 387–393.
[15] A. Vyas, A.G. Pillai, S. Chattarji, Recovery after chronic stress fails to reverse
the risk of foreclosing this exploratory phase and weakening rather amygdaloid neuronal hypertrophy and enhanced anxiety-like behavior,
than strengthening future work. Neuroscience 128 (4) (2004) 667–673.
[16] C. Caldji, et al., Maternal care during infancy regulates the development of
neural systems mediating the expression of fearfulness in the rat, Proc. Natl.
5. Conclusions Acad. Sci. 95 (9) (1998) 5335–5340.
[17] N. Tottenham, M.A. Sheridan, A review of adversity, the amygdala and the
hippocampus: a consideration of developmental timing, Front. Hum.
In conclusion, the results suggest that disorganized early attach- Neurosci. 3 (2009) 68.
ment relationships may have long-term effects on later adaptation [18] C.F. Weems, M. Klabunde, J.D. Russell, A.L. Reiss, V.G. Carrion, Post-traumatic
by promoting increased volume in the left amygdala that con- stress and age variation in amygdala volumes among youth exposed to
trauma, Soc. Cognit. Affect. Neurosci. 10 (12) (2015) 1661–1667.
tributes to increased irritability in limbic pathways. These findings [19] R.A. Morey, et al., Amygdala volume changes in posttraumatic stress
also suggest that the first two years of life may be an early sensitive disorder in a large case-controlled veterans group, Arch. Gen. Psychiatry 69
period during which clinical intervention could preempt long-term (11) (2012) 1169–1178.
[20] I.M. Veer, et al., Evidence for smaller right amygdala volumes in
consequences of early attachment disturbance.
posttraumatic stress disorder following childhood trauma, Psychiatry
Res.-Neuroimag. 233 (3) (2015) 436–442.
[21] J.L. Hanson, et al., Behavioral problems after early life stress: contributions
Conflict of interest of the hippocampus and amygdala, Biol. Psychiatry 77 (4) (2015) 314–323.
[22] D. Pagliaccio, et al., Stress-system genes and life stress predict cortisol levels
Authors report no conflicts of interest. and amygdala and hippocampal volumes in children,
Neuropsychopharmacology 39 (2014) 1245–1253.
[23] S. Whittle, et al., Childhood maltreatment and psychopathology affect brain
Acknowledgements development during adolescence, J. Am. Acad. Child Adolesc. Psychiatry 52
(9) (2013) 940–952.
[24] N. Tottenham, et al., Prolonged institutional rearing is associated with
This research was supported by funding received from the Har- atypically large amygdala volume and difficulties in emotion regulation,
vard Catalyst/Harvard Clinical and Translational Science Center Dev. Sci. 13 (1) (2010) 46–61.
92 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93

[25] S.J. Lupien, et al., Larger amygdala but no change in hippocampal volume in [55] M. van IJzendoorn, Adult attachment representations, parental
10-year-old children exposed to maternal depressive symptomatology since responsiveness, and infant attachment: a meta-analysis on the predictive
birth, Proc. Natl. Acad. Sci. 108 (34) (2011) 14324–14329. validity of the adult attachment interview, Psychol. Bull. 117 (3) (1995)
[26] P. Pechtel, K. Lyons-Ruth, C.M. Anderson, M.H. Teicher, Sensitive periods of 387–403.
amygdala development: the role of maltreatment in preadolescence, [56] M.D.S. Ainsworth, M.C. Blehar, E. Waters, S. Wall, Patterns of Attachment: A
Neuroimage 97 (2015) 236–244. Psychological Study of the Strange Situation, Lawrence Erlbaum, Oxford
[27] J.D. Bremner, et al., Magnetic resonance imaging-based measurement of England, 1978.
hippocampal volume in posttraumatic stress disorder related to childhood [57] S. Madigan, et al., Unresolved states of mind, anomalous parental behavior,
physical and sexual abuse–a preliminary report, Biol. Psychiatry 41 (1) and disorganized attachment: a review and meta-analysis of a transmission
(1997) 23–32. gap, Attach. Hum. Dev. 8 (2) (2006) 89–111.
[28] P. Brambilla, et al., Anatomical MRI study of borderline personality disorder [58] M. Nachmias, M. Gunnar, S. Mangelsdorf, R.H. Parritz, K. Buss, Behavioral
patients, Psychiatry Res. 131 (2) (2004) 125–133. inhibition and stress reactivity: the moderating role of attachment security,
[29] A.L. van Harmelen, et al., Reduced medial prefrontal cortex volume in adults Child Dev. 67 (2) (1996) 508–522.
reporting childhood emotional maltreatment, Biol. Psychiatry 68 (9) (2010) [59] G. Spangler, K.E. Grossman, Biobehavioral organization in securely and
832–838. insecurely attached infants, Child Dev. 64 (1993) 1439–1450.
[30] S.L. Andersen, et al., Preliminary evidence for sensitive periods in the effect [60] K. Bernard, M. Dozier, Examining infants’ cortisol responses to laboratory
of childhood sexual abuse on regional brain development, J. tasks among children varying in attachment disorganization: stress
Neuropsychiatry Clin. Neurosci. 20 (3) (2008) 292–301. reactivity or return to baseline? Dev. Psychol. 46 (6) (2010) 1771–1778.
[31] E.B. Ansell, K. Rando, K. Tuit, J. Guarnaccia, R. Sinha, Cumulative adversity [61] M.P. Luijk, et al., Attachment, depression, and cortisol: deviant patterns in
and smaller gray matter volume in medial prefrontal, anterior cingulate, and insecure-resistant and disorganized infants, Dev. Psychobiol. 52 (5) (2010)
insula regions, Biol. Psychiatry 72 (1) (2012) 57–64. 441–452.
[32] R.A. Cohen, et al., Early life stress and morphometry of the adult anterior [62] J. Choi, B. Jeong, A. Polcari, M.L. Rohan, M.H. Teicher, Reduced fractional
cingulate cortex and caudate nuclei, Biol. Psychiatry 59 (10) (2006) 975–982. anisotropy in the visual limbic pathway of young adults witnessing
[33] U. Dannlowski, et al., Limbic scars: long-term consequences of childhood domestic violence in childhood, Neuroimage 59 (2) (2012) 1071–1079.
maltreatment revealed by functional and structural magnetic resonance [63] M.N. Dackis, F.A. Rogosch, A. Oshri, D. Cicchetti, The role of limbic system
imaging, Biol. Psychiatry 71 (4) (2012) 286–293. irritability in linking history of childhood maltreatment and psychiatric
[34] M.D. De Bellis, J. Hall, A.M. Boring, K. Frustaci, G. Moritz, A pilot longitudinal outcomes in low-income, high-risk women: moderation by FK506 binding
study of hippocampal volumes in pediatric maltreatment-related protein 5 haplotype, Dev. Psychopathol. 24 (4) (2012) 1237–1252.
posttraumatic stress disorder, Biol. Psychiatry 50 (4) (2001) 305–309. [64] M.H. Teicher, C.A. Glod, J. Surrey, C. Swett Jr., Early childhood abuse and
[35] M.D. De Bellis, Developmental traumatology. part II: brain development, limbic system ratings in adult psychiatric outpatients, J. Neuropsychiatry
Biol. Psychiatry 45 (10) (1999) 1271–1284. Clin. Neurosci. 5 (3) (1993) 301–306.
[36] M.D. De Bellis, et al., Brain structures in pediatric maltreatment-related [65] M.H. Teicher, J.A. Samson, A. Polcari, C.E. McGreenery, Sticks, stones, and
posttraumatic stress disorder: a sociodemographically matched study, Biol. hurtful words: relative effects of various forms of childhood maltreatment,
Psychiatry 52 (11) (2002) 1066–1078. Am. J. Psychiatry 163 (2006) 993–1000.
[37] V.G. Carrion, et al., Attenuation of frontal asymmetry in pediatric [66] P.A. Spiers, D.L. Schomer, H.W. Blume, M.M. Mesulam, Temporolimbic
posttraumatic stress disorder, Biol. Psychiatry 50 (12) (2001) 943–951. Epilepsy and Behavior: Principles of Behavioral Neurology, FA Davis,
[38] P. Pechtel, K. Lyons-Ruth, C.M. Anderson, M.H. Teicher, Sensitive periods of Philadelphia, 1985, pp. 289–326.
amygdala development: the role of maltreatment in preadolescence, [67] P.D. Patel, et al., Glucocorticoid and mineralocorticoid receptor mRNA
Neuroimage 97 (2015) 236–244. expression in squirrel monkey brain, J. Psychiatr. Res. 34 (6) (2000) 383–392.
[39] M.A. Mehta, et al., Amygdala, hippocampal and corpus callosum size [68] J.N. Giedd, et al., Quantitative magnetic resonance imaging of human brain
following severe early institutional deprivation: the English and Romanian development: ages 4–18, Cereb. Cortex 6 (4) (1996) 551–560.
adoptees study pilot, J. Child Psychol. Psychiatry 50 (8) (2009) 943–951. [69] K.M. Hasan, et al., Multimodal quantitative magnetic resonance imaging of
[40] M. Driessen, et al., Magnetic resonance imaging volumes of the hippocampus thalamic development and aging across the human lifespan: implications to
and the amygdala in women with borderline personality disorder and early neurodegeneration in multiple sclerosis, J. Neurosci. 31 (46) (2011)
traumatization, Arch. Gen. Psychiatry 57 (12) (2000) 1115–1122. 16826–16832.
[41] C.G. Schmahl, E. Vermetten, B.M. Elzinga, J.D. Bremner, Magnetic resonance [70] K. Lyons-Ruth, D.B. Connell, H.U. Grunebaum, S. Botein, Infants at social risk:
imaging of hippocampal and amygdala volume in women with childhood maternal depression and family support services as mediators of infant
abuse and borderline personality disorder, Psychiatry Res.-Neuroimag. 122 development and security of attachment, Child Dev. 61 (1) (1990) 85–98.
(3) (2003) 193–198. [71] K. Lyons-Ruth, M.A. Easterbrooks, C.D. Cibelli, Infant attachment strategies,
[42] E. Vermetten, C. Schmahl, S. Lindner, R.J. Loewenstein, J.D. Bremner, infant mental lag, and maternal depressive symptoms: predictors of
Hippocampal and amygdalar volumes in dissociative identity disorder, Am. internalizing and externalizing problems at age 7, Dev. Psychol. 33 (4)
J. Psychiatry 163 (4) (2006) 630–636. (1997) 681–692.
[43] G. Weniger, C. Lange, U. Sachsse, E. Irle, Reduced amygdala and [72] K. Lyons-Ruth, J.F. Bureau, B. Holmes, A. Easterbrooks, N.H. Brooks,
hippocampus size in trauma-exposed women with borderline personality Borderline symptoms and suicidality/self-injury in late adolescence:
disorder and without posttraumatic stress disorder, J. Psychiatry Neurosci. prospectively observed relationship correlates in infancy and childhood,
34 (5) (2009) 383–388. Psychiatry Res. 206 (2–3) (2013) 273–281.
[44] S.L. Rauch, L.M. Shin, E.A. Phelps, Neurocircuitry models of posttraumatic [73] K. Lyons-Ruth, L. Alpern, B. Repacholi, Disorganized infant attachment
stress disorder and extinction: human neuroimaging research–past, classification and maternal psychosocial problems as predictors of
present, and future, Biol. Psychiatry 60 (4) (2006) 376–382. hostile-aggressive behavior in the preschool classroom, Child Dev. 64 (1993)
[45] K. Saleh, et al., Impact of family history and depression on amygdala volume, 572–585.
Psychiatry Res. 203 (1) (2012) 24–30. [74] K. Lyons-Ruth, E. Bronfman, E. Parsons, Atypical attachment in infancy and
[46] C. Lange, E. Irle, Enlarged amygdala volume and reduced hippocampal early childhood among children at developmental risk. IV. Maternal
volume in young women with major depression, Psychol. Med. 34 (6) frightened, frightening, or atypical behavior and disorganized infant
(2004) 1059–1064. attachment patterns, Monogr. Soc. Res. Child Dev. 64 (3) (1999) 67–96.
[47] S.L. Andersen, M.H. Teicher, Stress, sensitive periods and maturational [75] Z. Shi, J.-F. Bureau, M.A. Easterbrooks, X. Zhao, K. Lyons-Ruth, Childhood
events in adolescent depression, Trends Neurosci. 31 (4) (2008) 183–191. maltreatment and prospectively observed quality of early care as predictors
[48] M.H. Teicher, J.A. Samson, Annual research review: enduring of antisocial personality disorder features, Infant Mental Health J. 33 (2012)
neurobiological effects of childhood abuse and neglect, J. Child Psychol. 55–69.
Psychiatry 57 (3) (2016) 241–266. [76] P. Pechtel, A. Woodman, K. Lyons-Ruth, Early maternal withdrawal and
[49] C. Moutsiana, et al., Insecure attachment during infancy predicts greater non-verbal childhood IQ as precursors for substance abuse diagnosis in
amygdala volumes in early adulthood, J. Child Psychol. Psychiatry 56 (5) young adulthood: results of a 20 year prospective study, Int. J. Cognit. Ther. 5
(2015) 540–548. (2012) 316–329.
[50] F. Maheu, et al., A preliminary study of medial temporal lobe function in [77] L.S. Radloff, B.Z. Locke, The community mental health assessment survey
youths with a history of caregiver deprivation and emotional neglect, and the CES-D scale. In: M.M. Weismann, J.K., Myers, C.E. Ross (Eds.),
Cognit. Affect. Behav. Neursci. 10 (1) (2010) 34–49. Community surveys of psychoatric disorders, Surveys Psychological
[51] E.J. McCrory, et al., Heightened neural reactivity to threat in child victims of Epidemiology, 4 (1986), 177–189.
family violence, Curr. Biol. 21 (23) (2011) 947–948. [78] I. Obsuth, K. Hennighausen, L.E. Brumariu, K. Lyons-Ruth, Disorganized
[52] N. Tottenham, et al., Elevated amygdala response to faces following early behavior in adolescent-parent interaction: relations to attachment state of
deprivation, Dev. Sci. 15 (3) (2011) 307–319. mind, partner abuse, and psychopathology, Child Dev. 85 (1) (2014)
[53] M.A. Sheridan, M.E. Fox, C.H. Zeanah, K.A. McLaughlin, C.A. Nelson, Variation 370–387.
in neural development as a result of exposure to institutionalization early in [79] M.B. First, R.L. Spitzer, M. Gibbon, J.B.W. Williams, Structured Clinical
childhood, Proc. Natl. Acad. Sci. 109 (32) (2012) 12927–12932. Interview for DSM-IV-TR Axis I Disorders, Research Version Patient Edition.
[54] S. Madigan, L. Atkinson, K. Laurin, D. Benoit, Attachment and internalizing (SCID-I/P), Biometrics Research, New York State Psychiatric Institute, New
behavior in early childhood: a meta-analysis, Dev. Psychol. 49 (4) (2013) York, 2002.
672–689.
 K. Lyons-Ruth et al. / Behavioural Brain Research 308 (2016) 83–93 93

[80] D.L. Segal, M. Hersen, V.B. Van Hasselt, Reliability of the structured clinical [104] N. Tottenham, M. Shapiro, E.H. Telzer, K.L. Humphreys, Amygdala response
interview for DSM-iII-R: an evaluative review, Compr. Psychiatry 35 (4) to mother, Dev. Sci. 15 (3) (2012) 307–319.
(1994) 316–327. [105] S.M. Platek, S.M. Kemp, Is family special to the brain? An event-related fMRI
[81] J. Ventura, R.P. Liberman, M.F. Green, A. Shaner, J. Mintz, Training and study of familiar, familial, and self-face recognition, Neuropsychologia 47
quality assurance with the structured clinical interview for DSM-iV (3) (2009) 849–858.
(SCID-I/P), Psychiatry Res. 79 (2) (1998) 163–173. [106] A.K. Fetterman, S. Ode, M.D. Robinson, For which side the bell tolls: the
[82] E.M. Bernstein, F.W. Putnam, Development, reliability, and validity of a laterality of approach-avoidance associative networks, Motiv Emot. 37 (1)
dissociation scale, J. Nerv. Mental Dis. 174 (12) (1986) 727–735. (2013) 33–38.
[83] M.H. van I.Jzendoorn, C. Schuengel, The measurement of dissociation in [107] S. Schneider, et al., Boys do it the right way: sex-dependent amygdala
normal and clinical populations: meta-analytic validation of the lateralization during face processing in adolescents, Neuroimage 56 (3)
Dissociative Experiences Scale (DES), Clin. Psychol. Rev. 16 (1996) 365–382. (2011) 1847–1853.
[84] B. Fischl, et al., Whole brain segmentation: automated labeling of [108] S. Yoshimura, et al., Self-referential processing of negative stimuli within
neuroanatomical structures in the human brain, Neuron 33 (3) (2002) the ventral anterior cingulate gyrus and right amygdala, Brain Cogn. 69 (1)
341–355. (2009) 218–225.
[85] A.M. Dale, B. Fischl, M.I. Sereno, Cortical surface-based analysis. I. [109] U. Dannlowski, et al., Childhood maltreatment is associated with an
Segmentation and surface reconstruction, Neuroimage 9 (2) (1999) automatic negative emotion processing bias in the amygdala, Hum. Brain
179–194. Mapp. 34 (2013) 2899–2909.
[86] B. Fischl, M.I. Sereno, A.M. Dale, Cortical surface-based analysis. II: Inflation, [110] R.J. Herringa, et al., Childhood maltreatment is associated with altered fear
flattening, and a surface-based coordinate system, Neuroimage 9 (2) (1999) circuitry and increased internalizing symptoms by late adolescence, Proc.
195–207. Natl. Acad. Sci. 110 (47) (2013) 19119–19124.
[87] B. Fischl, A. Liu, A.M. Dale, Automated manifold surgery: constructing [111] J. Belsky, M. Rovine, D. Taylor, The Pennsylvania infant and family
geometrically accurate and topologically correct models of the human development project III: the origins of individual differences in
cerebral cortex, IEEE Trans. Med. Imag. 20 (1) (2001) 70–80. infant-mother attachment: maternal and infant contributions, Child Dev. 55
[88] B. Fischl, A.M. Dale, Measuring the thickness of the human cerebral cortex (1984) 718–728.
from magnetic resonance images, Proc. Natl. Acad. Sci. 97 (20) (2000) [112] K. Lyons-Ruth, D. Connell, D. Zoll, J. Stahl, Infants at social risk: relationships
11050–11055. among infant maltreatment, maternal behavior, and infant attachment
[89] B. Fischl, et al., Automatically parcellating the human cerebral cortex, Cereb. behavior, Dev. Psychol. 23 (1987) 223–232.
Cortex 14 (1) (2004) 11–22. [113] M. Main, L. Tomasini, W. Tolan, Differences among mothers of infants
[90] R.A. Morey, et al., A comparison of automated segmentation and manual judged to differ in security of attachment, Dev. Psychol. 15 (1979) 472–473.
tracing for quantifying hippocampal and amygdala volumes, Neuroimage 45 [114] M. Main, E. Hesse, Parents’ unresolved traumatic experiences are related to
(3) (2009) 855–866. infant disorganized attachment status: is frightened and/or frightening
[91] L. Bergouignan, et al., Can voxel based morphometry, manual segmentation parental behavior the linking mechanism? in: M.T. Greenberg, D. Cicchetti,
and automated segmentation equally detect hippocampal volume E.M. Cummings (Eds.), Attachment in the Preschool Years: Theory, Research
differences in acute depression? Neuroimage 45 (1) (2009) 29–37. and Intervention, University of Chicago Press, Chicago, IL, 1990, pp. 161–182.
[92] R. Development Core Team, R: A Language and Environment for Statistical [115] T.Y. Zhang, P. Chretien, M.J. Meaney, A. Gratton, Influence of naturally
Computing, R Foundation for Statistical Computing, Vienna, Austria, 2010. occurring variations in maternal care on prepulse inhibition of acoustic
[93] J. Schafer, J.W. Graham, Missing data: our view of the state of the art, startle and the medial prefrontal cortical dopamine response to stress in
Psychol. Methods 7 (2002) 147–177. adult rats J. Neurosci. 25 (2005) 1493–1502.
[94] W. Gilks, S. Richardson, D. Spiegelhalter, Markov Chain Monte Carlo in [116] L.J. Gabard-Durnam, et al., The development of human amygdala functional
Practice, Chapman, Hall, London, 1996. connectivity at rest from 4 to 23 years: a cross-sectional study, Neuroimage
[95] K.J. Preacher, A.F. Hayes, Asymptotic and resampling strategies for assessing 95 (2014) 193–207.
and comparing indirect effects in multiple mediator models, Behav. Res. [117] C.A. Burghy, et al., Developmental pathways to amygdala-prefrontal
Methods 40 (3) (2008) 691–879. function and internalizing symptoms in adolescence, Nat. Neurosci. 15 (12)
[96] R. Meng X-l Rosenthal, D.B. Rubin, Comparing correlated correlation (2012) 1736–1741.
coefficients, Psychol. Bull. 111 (1) (1992) 172–175. [118] Y. Ito, et al., Increased prevalence of electrophysiological abnormalities in
[97] V. Carlson, D. Cicchetti, D. Barnett, K. Braunwald, Disorganized/disoriented children with psychological, physical, and sexual abuse, J. Neuropsychiatry
attachment relationships in maltreated infants, Dev. Psychol. 25 (4) (1989) Clin. Neurosci. 5 (4) (1993) 401–408.
525–531. [119] J. Choi, B. Jeong, M.L. Rohan, A.M. Polcari, M.H. Teicher, Preliminary evidence
[98] D. Barnett, J. Ganiban, D. Cicchetti, Maltreatment, negative expressivity, and for white matter tract abnormalities in young adults exposed to parental
the development of type D attachments from 12 to 24 months of age, verbal abuse, Biol. Psychiatry 65 (3) (2009) 227–234.
Monogr. Soc. Res. Child Dev. 64 (3) (1999) 97–118. [120] C.M. Anderson, M.H. Teicher, A. Polcari, P.F. Renshaw, Abnormal T2
[99] D. Cicchetti, F.A. Rogosch, S.L. Toth, Fostering secure attachment in infants in relaxation time in the cerebellar vermis of adults sexually abused in
maltreating families through preventive interventions, Dev. Psychopathol. childhood: potential role of the vermis in stress-enhanced risk for drug
18 (3) (2006) 623–649. abuse, Psychoneuroendocrinology 27 (2002) 231–244.
[100] C. Cyr, E.M. Euser, M.J. Bakermans-Kranenburg, M.H. Van I.Jzendoorn, [121] J. Glascher, R. Adolphs, Processing of the arousal of subliminal and
Attachment security and disorganization in maltreating and high-risk supraliminal emotional stimuli by the human amygdala, J. Neurosci. 23 (32)
families: a series of meta-analyses, Dev. Psychopathol. 22 (1) (2010) 87–108. (2003) 10274–10282.
[101] P. Fearon, J. Belsky, Infant–mother attachment and the growth of [122] J.N. Giedd, Quantitative MRI of the temporal lobe, amygdala, and
externalizing problems across the primary-school years, J. Child Psychol. hippocampus in normal human development: ages 4–18 years, J. Comp.
Psychiatry 52 (7) (2011) 782–791. Neurol. 366 (2) (1996) 223–230.
[102] NICHD Early Child Care Research Network, Child-care and family predictors [123] K.S. Button, et al., Power failure: why small sample size undermines the
of preschool attachment and stability from infancy, Dev. Psychol. 37 (2001) reliability of neuroscience, Nat. Rev. Neurosci. 14 (2013) 365–376.
847–862.
[103] J.R. Villablanca, Why do we have a caudate nucleus? Acta Neurobiol. Exp.
(Warsz.) 70 (1) (2010) 95–105.