Examining Child Maltreatment Througha Neurodevelopmental Lens: ClinicalApplications of the Neurosequential

Model of Therapeutics

BRUCE D. PERRYChildTrauma Academy, Houston, Texas, USA and Department of Psychiatry and Behavioral

Sciences, Northwestern University, Chicago, Illinois, USA

This article provides the theoretical rationale and overview of aneurodevelopmentally-informed approach to therapeutic workwith maltreated and traumatized children and youth. Rather thanfocusing on any specific therapeutic technique, the Neurosequen-tial Model of Therapeutics (NMT) allows identification of the keysystems and areas in the brain which have been impacted byadverse developmental experiences and helps target the selectionand sequence of therapeutic, enrichment, and educational activ-ities. In the preliminary applications of this approach in a varietyof clinical settings, the outcomes have been positive. More in depthevaluation of this approach is warranted, and is underway.

Over the last 30 years, key findings in developmental neurobiologyhave informed and influenced practice in several clinical disciplines, includ-ing pediatrics, psychology, social work, and psychiatry. Despite this influ-ence, the capacity of these large clinical fields to incorporate and translatekey neurobiological principles into practice, program, and policy has beeninefficient and inconsistent. The purpose of this article is to present prelimin-ary efforts to integrate core concepts of neurodevelopment into a practicalclinical approach with maltreated children. This neurosequential model oftherapeutics (NMT) has been utilized in a variety of clinical settings suchas therapeutic preschools, outpatient mental health clinics, and residentialtreatment centers with promising results (Perry, 2006; Barfield et al., 2009).

Received 23 March 2009; accepted 28 April 2009.Address correspondence to Bruce D. Perry, ChildTrauma Academy, 800 Gessner, Suite

230, Houston, TX 77024, USA. E-mail: [email protected]

Journal of Loss and Trauma, 14:240–255, 2009Copyright # Taylor & Francis Group, LLCISSN: 1532-5024 print=1532-5032 onlineDOI: 10.1080/15325020903004350

240

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

CONTEXT AND CURRENT STATUS

Development is a complex and dynamic process involving billions of interac-tions across multiple micro (e.g., the synapse) and macro domains (e.g.,maternal-child interactions). These interactions result in a unique expressionof an individual’s genetic potential and create a miracle of dynamic organiza-tion in the trillions of component parts (e.g., neurons, glia, synapses) compris-ing the human brain. Maltreatment disrupts this hardy process; trauma,neglect, and related experiences of maltreatment such as prenatal exposureto drugs or alcohol and impaired early bonding all influence the developingbrain. These adverse experiences interfere with normal patterns of experi-ence-guided neurodevelopment by creating extreme and abnormal patternsof neural and neurohormonal activity. The resulting negative functional impactof impaired or abusive caregiving on the developing child has been well docu-mented (e.g., Malinosky-Rummell & Hansen, 1993; Margolin & Gordis, 2000).As expected, in any brain-mediated function examined—from speech tomotorfunctioning to social, emotional, or behavioral regulation—developmentaltrauma andmaltreatment increase risk of dysfunction (see also Perry & Pollard,1998; Bremner & Vermetten, 2001; Perry, 2001, 2002; Anda et al., 2006).

In the United States alone, there are millions of maltreated children andyouth in the educational, mental health, child protective, and juvenile justicesystems (Fitzpatrick & Boldizar, 1993; Graham-Berman & Levendosky,1998). The majority of these children do not receive adequate mental healthservices; indeed, most are not even known to be maltreated or traumatized.While current policy efforts to create trauma-informed practices and programsare a welcome start, for children and youth, focusing on trauma alone isinsufficient. Practice, program, and policy must become substance abuse,attachment, and neglect informed as well; we must become fully ‘‘develop-mentally informed’’ to understand and address the range of problems relatedto maltreatment. The following sampling of some principles of neurodevelop-ment illustrates the value of this broader view for clinical practice.

PRINCIPLES OF NEURODEVELOPMENT

There aremanywell-documented and emerging findings regarding the genetics,epigenetics, and experience-determined elements of neurodevelopment. Only afew are listed below to serve as examples of how these facts and concepts caninform our understanding of maltreated children and therapeutic work. Morecomplete reviews are available elsewhere (e.g., Perry, 2001, 2002, 2006, 2008).

Sequential Development

The brain is organized in a hierarchical fashion with four main anatomicallydistinct regions: brainstem, diencephalon, limbic system, and cortex. During

Examining Child Maltreatment Through a Neurodevelopmental Lens 241

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

development the brain organizes itself from the bottom up, from the least(brainstem) to the most complex (limbic, cortical) areas. While significantlyinterconnected, each of these regions mediates distinct functions, with thelower, structurally simpler areas mediating basic regulatory functions and thehighest, most complex structures (cortical) mediating the most complex func-tions. Each of these main regions develops, organizes, and becomes fully func-tional at different times during childhood. At birth, for example, the brainstemareas responsible for regulating cardiovascular and respiratory functionmust beintact for the infant to survive, and any malfunction is immediately observable.The neural networks involved, therefore, must be mostly organized in utero inorder to become functional at birth. In contrast, the cortical areas responsible forabstract cognition have years before they will become fully organized andfunctional. Each brain area has its own timetable for development. Micro-neurodevelopmental processes such as synaptogenesis will be most active indifferent brain areas at different times and, thereby, be more sensitive toorganizing or disruptive experiences during these times (sensitive periods).

As the brain is developing from the bottom to the top, the process isinfluenced by a host of neurotransmitter, neurohormone, and neuromodula-tor signals. These signals help target cells migrate, differentiate, sprout den-dritic trees, and form synaptic connections. Some of the most important ofthese signals come from the monoamine neural systems (i.e., norepine-phrine, dopamine, and serotonin). These crucial sets of widely distributedneural networks originate in the lower brain areas (brainstem and dience-phalon) and project to every other part of the developing brain. This archi-tecture allows these systems the unique capacity to communicate acrossmultiple regions simultaneously and therefore provide an organizing andorchestrating role during development and later in life. Due to their wide dis-tribution throughout the brain, and their role in mediating and modulating ahuge array of functions, impairment in the organization and functioning ofthese monoamine neurotransmitter systems can result in a cascade of dys-function from lower regions (where these system originate) up to all of thetarget areas higher in the brain. If the impairment occurs in utero (e.g., pre-natal exposure to drugs or alcohol) or in early childhood (e.g., emotionalneglect or trauma), this cascade of dysfunction can disrupt normal develop-ment. Simply put, the organization of higher parts of the brain depends uponinput from the lower parts of the brain. If the patterns or incoming neuralactivity in these monoamine systems is regulated, synchronous, patterned,and of ‘‘normal’’ intensity, the higher areas will organize in healthier ways;if the patterns are extreme, dysregulated, and asynchronous, the higher areaswill organize to reflect these abnormal patterns.

The clinical implications of this principle speak to the importance of thetiming of developmental experience; the very same traumatic experience willimpact an 18-month-old child differently than a 5-year-old. Similar traumaticexperiences occurring at different times in the life of the same child will

242 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

influence the brain in different ways; in many cases, the previous exposurehas sensitized the child, making him or her more vulnerable to future events.And so it is with the timing of positive experience; the developmental stageof a child has a profound impact on how an educational, caregiving, ortherapeutic experience will influence the brain; somatosensory nurturing,for example, will more quickly and efficiently shape the attachment neuro-biology of the infant in comparison to the adolescent.

A more subtle clinical implication is that in order to most efficientlyinfluence a higher function such as speech and language or socioemotionalcommunication, the lower innervating neural networks (e.g., locus coeruleusnorepinephrine systems) must be intact and well regulated. An overanxious,impulsive, dysregulated child will have a difficult time participating in, andbenefiting from, services targeting social skills, self-esteem, and reading,for example. The field of restorative neurology has for many years empha-sized the positive impact of repetitive motor activity in cognitive recoveryfrom stroke. This principle suggests that therapeutic massage, yoga, balan-cing exercises, and music and movement, as well as similar somatosensoryinterventions that provide patterned, repetitive neural input to the brainstemand diencephalon monoamine neural networks, would be organizing andregulating input that would likely diminish anxiety, impulsivity, and othertrauma-related symptoms that have their origins in dysregulation of thesesystems. Our preliminary findings, and those of others (B. van der Kolk,personal communication, June 2008) with maltreated children with suchself-regulation problems, suggest that this is the case (Barfield et al., 2009).

Activity-Dependent Organization: Use-Dependent Modification

The brain organizes in a use-dependent fashion. In the developing brain,undifferentiated neural systems are critically dependent upon sets of environ-mental and micro-environmental cues (e.g., neurotransmitters, cellular adhe-sion molecules, neurohormones, amino acids, ions) in order for them toappropriately organize from their undifferentiated, immature forms (forreviews, see Perry, 2001, 2008). The molecular cues that guide developmentare dependent, in part, upon the experiences of the developing child. Thequantity, pattern of activity, and nature of the activation from these neuro-chemical and neurotrophic factors depend upon the presence and the natureof the total sensory experience of the child. When the child has adverseexperiences—loss, threat, neglect, and injury—there can be disruptions ofneurodevelopment leading to compromised functioning (see below).

This principle has many clinical implications. Primary is the role thisprinciple plays in psychopathology. Use-dependent changes in the brainare the origin of neuropsychiatric symptoms related to exposure to threat,fear, chaos, stress, and trauma. The monoamine systems mentioned earlierare crucial components of the stress-response neural networks in the brain.

Examining Child Maltreatment Through a Neurodevelopmental Lens 243

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

When a child (or an adult) is threatened and activates this stress response inan extremely prolonged or repetitive fashion, the neural networks involvedin this adaptive response will undergo a ‘‘use-dependent’’ alteration. Thevery molecular characteristics of individual neurons, synaptic distributions,dendritic trees, and a host of other microstructural and microchemicalaspects of these important neural networks will change. And the end effectis an alteration in the baseline activity and reactivity of the stress responsesystems in the traumatized individual. The brain will ‘‘reset’’—acting as ifthe individual is under persistent threat. The details of this process have beenwell described elsewhere (Perry & Pollard, 1998; Perry, 2001).

The principle of use dependence is at the heart of effective therapy.Therapy seeks to change the brain. Any efforts to change the brain or systemsin the brain must provide experiences that can create patterned, repetitiveactivation in the neural systems that mediate the function=dysfunction thatis the target of therapy. In many cases, this will mean (as mentioned above)that the target of the intervention should be the innervating neural systemsand not the area or system that is the final mediator of the function=dysfunction (e.g., physical exercise helps stroke victims recover speech).This is a significant problem in the conventional mental health approach tomaltreated children; many of their problems are related to disorganized orpoorly regulated networks (e.g., the monoamines) originating lower in thebrain. Yet, our clinical interventions often provide experiences that primarilytarget the innervated cortical or limbic (i.e., cognitive and relationalinteractions) regions in the brain and not the innervating source of the dys-regulation (lower stress-response networks). Even when targeting the appro-priate systems in the brain, we rarely provide the repetitions necessary tomodify organized neural networks; 1 hour of therapy a week is insufficientto alter the accumulated impact of years of chaos, threat, loss, and humilia-tion. Inadequate ‘‘targeting’’ of our therapeutic activities to brain areasthat are not the source of the symptoms and insufficient ‘‘repetitions’’ com-bine to make conventional mental health services for maltreated childrenineffective.

The origins of—and therapeutic recovery from—neglect are manifesta-tions of the principle of use dependence as well. Neglect, from a neurodeve-lopmental perspective, is the absence of the necessary timing, frequency,pattern, and nature of experience (and the patterns of neural activationcaused by these experiences) required to express the genetic potential of acore capability (e.g., self-regulation, speech and language, capacity forhealthy relational interactions). Neglect-related disruptions of experience-dependent neural signals during early life lead to a range of abnormalitiesor deficits in function (see Perry, 2001, 2006). As discussed above, themalleability of the brain shifts during development, and therefore the timingand specific ‘‘pattern’’ of neglect influence the final functional outcome. Achild deprived of consistent, attentive, and attuned nurturing for the first 3

244 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

years of life who is then adopted and begins to receive attention, love, andnurturing may not be capable of benefiting from these experiences with thesame malleability as an infant. In some cases, this later love is insufficient toovercome the dysfunctional organization of the neural systems mediatingsocioemotional interactions. With little appreciation of neurodevelopment,neglect-related problems in maltreated children are missed (in over 80% ofchildren under the age of 6 removed by child protective services, there aresignificant developmental problems, yet this population rarely receives adevelopmental assessment in most states), ignored (a minority of childrenin child protective service care with mental health, learning, speech andlanguage, or developmental problems receive consistent services), orlumped into the overinclusive current label of ‘‘complex’’ trauma or, worse,bipolar disorder. Even when children do receive mental health services,neglect-related issues are rarely appreciated as having a distinct pathophy-siology and pathogenesis related to but different from trauma.

Disproportional Valence of Early Childhood Experience

The sequential development of the brain and the activity dependence ofneurodevelopment create times during development when a given neuralsystem is more sensitive to experience than others. In healthy development,that sensitivity allows the brain to rapidly and efficiently organize in responseto the unique demands of a given environment to express from its broadgenetic potential those characteristics that best fit the child’s world; differentgenes can be expressed, and different neural networks can be organizedfrom the child’s potential to best fit that family, culture, and environment.We all are aware of how rapidly young children can learn language, developnew behaviors, and master new tasks. The very same neurodevelopmentalsensitivity that allows amazing developmental advances in response topredictable, nurturing, repetitive, and enriching experiences makes thedeveloping child vulnerable to adverse experiences.

The simple and unavoidable conclusion of these neurodevelopmentalprinciples is that the organizing, sensitive brain of an infant or young childis more malleable to experience than a mature brain. While experiencemay alter the behavior of an adult, experience literally provides the organiz-ing framework for an infant and child. Because the brain is most plastic(receptive to environmental input) in early childhood, the child is most vul-nerable to variance of experience during this time. Again, the clinical, prac-tice, and policy implications are profound. Early identification and aggressiveearly interventions are more effective than reactive services. Despite solidresearch documentation on early intervention and effective therapeuticservices targeting young mothers, infant mental health, home visitationprograms, and high-quality child care programs, support for these programsis scant and inadequate.

Examining Child Maltreatment Through a Neurodevelopmental Lens 245

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

Relational Mediation of Major Developmental Experiences

Life is full of novelty, unpredictability, challenges, stressors, and, often,trauma. There are individual differences in how we cope with and overcomestress and trauma. Much is yet to be understood; genetic factors, for example,appear to influence hardiness or sensitivity. Yet, one recurring observationabout resilience and coping with trauma is the power of healthy relationshipsto protect from and heal following stress, distress, and trauma. This relationalmodulation of stress is mediated by two interrelated and broadly distributedsystems in the human brain: the stress response systems and neural networksinvolved in bonding, attachment, social communication, and affiliation. Tobest understand the intimate interdependence of these systems in the brain,it is useful to examine the conditions into which the human brain evolved.

For the vast majority of the last 200,000 years, humans have lived inhunter-gatherer clans in the natural world. The size of our living groupswas small—40 to 60 people. These multigenerational, multifamily groupswere the main source of safety from the dangers of the world. Our survivaldepended upon the ability to communicate, bond, share, and receive fromother members of our family and clan. Without others, the individual couldnot survive in the natural world. Then, and today, the presence of familiarpeople projecting the social-emotional cues of acceptance, compassion,caring, and safety calms the stress response of the individual: ‘‘You are oneof us, you are welcome, you are safe.’’ This powerful positive effect ofhealthy relational interactions on the individual—mediated by the relationaland stress-response neural systems—is at the core of relationally basedprotective mechanisms that help us survive and thrive following traumaand loss. Individuals who have few positive relational interactions (e.g., achild without a healthy family=clan) during or after trauma have a muchmore difficult time decreasing the trauma-induced activation of the stressresponse systems and therefore will be much more likely to have ongoingsymptoms (i.e., there will be more prolonged and intense activation ofthe stress response systems and, hence, a ‘‘use-dependent’’ alteration in thesesystems). This capacity to benefit from relational interactions is, in turn,derived from our individual developmental experiences.

At birth, the developing stress-response networks in the brain (includingthe monoamine systems mentioned above) are rapidly organizing. Theprimary source of the patterned somatosensory interactions that providethe organizing neural input to the developing stress-response system is theprimary caregiver. The role of the stress response system is to sense distress(e.g., hunger, thirst, cold, threat) and then act to address this challenge tohomeostasis to promote survival: if hungry, eat; if cold, find shelter; if thirsty,drink. Infants are incapable of meeting their needs; they cannot feed, warm,or comfort themselves and depend upon their caregiver to become the exter-nal stress regulator. The primary caregiver, through consistent, nurturing, and

246 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

predictable responsive caregiving, provides the patterned, repetitive neuralstimulation (again, the principle of ‘‘use dependence’’) for the infant’s devel-oping brain required to build in an adaptive and flexible stress-responsecapacity (self-regulation) as well as healthy attachment capabilities. If thecaregiver is depressed, stressed, high, inconsistent, or absent, these two cru-cial neural networks (stress-response and relational) develop abnormally.The result is a child more vulnerable to future stressors and less capable ofbenefiting from the healthy nurturing relational supports that might helpbuffer future stressors or trauma.

Early developmental experiences with caregivers—the infant’s firstexposure to humans—create a set of associations and ‘‘templates’’ for thechild’s brain about what humans are. Are humans safe, predictable? Arethey a source of sustenance, comfort, and pleasure? Or are they unpredict-able and a source of fear, chaos, pain, and loss? These initial caregivingexperiences create the ‘‘template’’ that the child carries into future relationalinteractions, either increasing or decreasing the capacity of the child to ben-efit from future nurturing, caring, and invested adults. Relationships in earlychildhood, then, can alter the vulnerability=resilience balance for an indivi-dual child (do human relational interactions calm you when distressedbecause your ‘‘template’’ is based upon nurturing, or do they make you feelmore anxious and vulnerable because your primary caregiver was inconsis-tent and abusive?).

There is another aspect of the interconnectedness of the stress responseand relational neurobiology. The human experience is characterized byclan-on-clan, human-on-human competition for limited resources. Indeed,the major predator of humans is now, and has always been, other humans.In our competitive, violent past, encounters with unfamiliar, nonclan mem-bers were as likely to result in harm as in harmony. As the infant becomesa toddler and the toddler becomes a child, the brain is making a catalogueof ‘‘safe and familiar’’ attributes of the humans in his clan; the language,the dress, the nonverbal elements of communication, and the skin color ofhis family and clan become the attributes of ‘‘safe and familiar’’ that, in futureinteractions with others, will tell his stress response networks to be calm. Incontrast, when this person interacts with strangers, the stress response sys-tems activate; the more unfamiliar the attributes of this new person, the morethe activation. In some cases, a clan’s beliefs may have exacerbated this; if thechild grows up with ethnic, racial, or religious beliefs and values that degradeor dehumanize others, the stress activation that results in an encounter withdifferent people can be extreme. In this case, relational interactions activateand exacerbate trauma-related stress reactivity. A recent study by Chiao andcolleagues (2008), for example, showed that fear-related social cues fromindividuals from one’s own group=ethnicity have greater ‘‘power.’’ They willinduce greater amygdala activation than similar cues from individuals not inone’s group. Similar group contagion of positive emotional states has been

Examining Child Maltreatment Through a Neurodevelopmental Lens 247

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

documented (e.g., Fowler & Christakis, 2008). All of this points to the power-ful influence of the social milieu on individual neurobiological functioning.

The social milieu, then, becomes a major mediator of individual stressresponse baseline and reactivity; nonverbal signals of safety or threat frommembers of one’s ‘‘clan’’ modulate one’s stress response. The bottom lineis that healthy relational interactions with safe and familiar individuals canbuffer and heal trauma-related problems, while the ongoing process of ‘‘tri-balism’’—creating an ‘‘us’’ and ‘‘them’’—is a powerful but destructive aspectof the human condition, exacerbating trauma in individuals, families, andcommunities attempting to heal.

The clinical implications of this understanding of the power of relationalhealth are, again, profound. As one would predict, research suggests thatsocial connectedness is a protective factor against many forms of child mal-treatment, including physical abuse, neglect, and nonorganic failure to thrive,as well as a means of promoting prosocial behavior (Belsky et al., 2005;Caliso & Milner, 1992; Egeland, Jacobvitz, & Sroufe, 1988; Rak & Patterson,1996; Travis & Combs-Orme, 2007; Chan, 1994; Coohey, 1996; Guadinet al., 1993; Hashima & Amato, 1994; Pascoe & Earp, 1984; Altemeier,O’Connor, Sherrod, & Vietze, 1985; Benoit, Zeanah, & Barton, 1989; Crnic,Greenberg, Robinson, & Ragozin, 1984; Gorman, Leifer, & Grossman, 1993).The number, quality, and stability of relational interactions matter to thechild. Removing children from abusive homes also may remove them fromtheir familiar and safe social network in school, church, and community.And worse, the presence of new and unfamiliar individuals can actually acti-vate the already sensitized stress-response systems in these children, makingthem more symptomatic and less capable of benefiting from our efforts tocomfort and heal. Our well-intended interventions often result in relationalimpermanence for the child: foster home to foster home, new schools,new case workers, new therapists as if these are interchangeable parts. Theyare not. Even ‘‘best-practice’’ therapeutic work is ineffective in an environ-ment of relational instability and chronic transition.

TRANSLATIONAL NEUROSCIENCE: THE NEUROSEQUENTIALMODEL OF THERAPEUTICS

Over the last 20 years, we have been adapting our clinical practice to incor-porate emerging findings from neuroscience. This has resulted in a shift froma traditional medical model approach to a more developmentally sensitive,neurobiology-guided practice. The results are promising (see Perry, 2006;Barfield et al., 2009). A brief overview follows.

The neurosequential model of therapeutics (NMT) is not a specifictherapeutic technique or intervention; it is an approach to clinical workthat is informed by neuroscience (Perry, 2006). It is, in short, an effort in

248 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

translational neuroscience that has been evolving over the last 15 years. TheNMT process structures assessment and identification of primary problemsand strengths, and it sequences the application of interventions (educational,enrichment, and therapeutic) in a way that reflects the child’s specific devel-opmental needs in a variety of key domains and is sensitive the to coreprinciples of neurodevelopment—some of which have been articulatedabove. There are three central elements of the model: a developmentalhistory, a current assessment of functioning, and a set of recommendationsfor intervention and enrichment that arise from the process.

NMT Developmental History

The brain organizes as a reflection of experiences both good and bad. Tounderstand an individual, therefore, one needs to know his or her history.The NMT assessment is focused on the developmental history of the child.The NMT core assessment reviews the timing, nature, and severity of devel-opmental challenges; these are scored, resulting in an estimate of develop-mental ‘‘load.’’ This also allows an estimate of which neural networks andfunctions would plausibly be impacted by the child’s developmental insultsor history of trauma (Perry, 2001, 2006). For example, intrauterine insultssuch as alcohol use or perinatal caregiving disruptions (such as an impaired,inattentive primary caregiver) will predictably alter the norepinephrine, ser-otonin, and dopamine systems of the brainstem and diencephalon that arerapidly organizing during these times in life. These early life disruptions, inturn, will result in a cascade of regulatory functions impacting a wide distri-bution of other brain areas and functions that these important neural systemsinnervate (for more, see Perry, 2008).

A second important element of the NMT core assessment is a review of therelational history of the child during development. As discussed above, rela-tional milieu can be protective and confer some capacity to buffer the impactof trauma, while relational instability and multiple transitions can exacerbatedevelopmental insults. This NMT relational health history provides importantinsights into attachment and related resiliency or vulnerability factors thatmay have impacted the functional development of the child (see Figure 1).

NMT Functional Status and Brain ‘‘Mapping’’

The second component of the NMT process is a review of current functioningthat allows us to make estimates of which neural systems and brain areas areinvolved in the various neuropsychiatric symptoms. An interdisciplinary staff-ing is typically the method for this functional review. This process helps inthe development of a working functional brain map for the individual (seeFigure 2). This visual representation gives a quick impression of develop-mental status in various domains of functioning: A 10-year-old child, for

Examining Child Maltreatment Through a Neurodevelopmental Lens 249

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

example, may have the speech and language capability of an 8-year-old, thesocial skills of a 5-year-old, and the self-regulation skills of a 2-year old. Thisvisual ‘‘map’’ is very helpful when talking about trauma, brain development,and the rationale of various recommendations with educators, mental healthstaff, caregivers, and clients. It is also very useful to help track progress;improvement, as shown in changes in the shadings of various brain areas,is quick to see in the comparison of today’s brain map with one from6 months ago and is a powerful reinforcement for tired parents and hard-working frontline staff who feel their efforts are for naught.

This review requires a working knowledge of neural organization andfunctioning. In order to ‘‘localize’’ a set of functions to any set of brain net-works or regions, the senior clinician leading the interdisciplinary NMT staff-ing must know child development, clinical traumatology, and developmental

FIGURE 1 Relationship between developmental insults (trauma and neglect) and functionalorganization of the brain. Using the NMT developmental history measure (higher scores indicatemore developmental insult such as trauma and neglect) and the NMT functional brain mappingscores (higher scores indicate positive functioning), a linear relationship is seen between num-ber and intensity of developmental insults and the compromise in normal development andfunctioning of the brain. It is of interest to note that individuals who fall below the line tendto have more profound relational poverty (e.g., multiple placements, disengaged or unhealthyprimary caregiving) during development, and those above the line have relatively more protec-tive relational health (e.g., extended family, few placement disruptions, more stable family rela-tionships). The individuals at upper left (NMT FS! approximately 90–100) are healthycomparison children. The outlying individual at lower left is a child with autism, healthy caregiv-ing, and minimal adverse experiences during development. (Bruce D. Perry # 2008)

250 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

neurosciences. At present, this is the major impediment in exporting the NMTapproach: It requires a senior clinician to lead the process with a uniquecombination of clinical and preclinical skills.

NMT Interventions

The third major element of the NMT process is providing specific recommenda-tions. The NMT ‘‘mapping’’ process helps determine a unique sequence ofdevelopmentally appropriate interventions and enrichments that can help thechild reapproximate a more normal developmental trajectory. As outlined inbrief below, these recommendations are made with various principles of neu-robiology in mind; while many deficits may be present, the sequence in whichthese are addressed is important. The more the therapeutic process can repli-cate the normal sequential process of development, the more effective the

FIGURE 2 NMT functional brain ‘‘map’’: 6-year-old traumatized and neglected child vs. com-parison child (normal development). This map is generated from an interdisciplinary staffingprocess examining the functional status of various brain-mediated functions. Each rectangle inthe diagram indicates a brain function. Each rectangle is shaded to indicate functional status(see key above). Brain functions (e.g., regulation of heart rate: Brainstem; speech and lan-guage: CTX; attunement: Limbic) are ‘‘localized’’ to a brain region mediating the specific func-tion (this oversimplification attempts to assign function to the brain region that is the finalcommon mediator of the function with the knowledge that almost all brain functions are influ-enced and mediated by complex, trans-regional neural networks). This approximation allowsa useful estimate of the developmental=functional status of the child’s key functions, estab-lishes the ‘‘strengths and vulnerabilities’’ of the child, and determines the starting point andnature of enrichment or therapeutic activities most likely to meet the child’s specific needs.

Examining Child Maltreatment Through a Neurodevelopmental Lens 251

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

interventions are (see Perry, 2006). Simply stated, the idea is to start with thelowest (in the brain) undeveloped=abnormally functioning set of problemsand move sequentially up the brain as improvements are seen. This mayinvolve initially focusing on a poorly organized brainstem=diencephalon andthe related self-regulation, attention, arousal, and impulsivity by using anyvariety of patterned, repetitive somatosensory activities (which provide thesebrain areas with the patterned neural activation necessary for reorganization)such as music, movement, yoga (breathing), and drumming or therapeuticmassage. Once there is improvement in self-regulation, the therapeutic workcan move to more relational-related problems (limbic) using more traditionalplay or arts therapies; ultimately, once fundamental dyadic relational skillshave improved, the therapeutic techniques can be more verbal and insightoriented (cortical) using any variety of cognitive-behavioral or psychodynamicapproach. Further, the recommendations and enrichments are not limited to theconventional limits of ‘‘mental health’’ symptoms; issues in speech, learning,motor functioning, and social functioning are all addressed as part of a compre-hensive, more holistic approach to the child and her or his family.

Patterned, repetitive activities shape the brain in patterned ways, whilechaotic experiences create chaotic dysfunctional organization. Therapeuticactivities, then, are most effective when implemented with focused repetitiontargeting the neural systems one wishes to modify. One cannot change aneural system without activating it; one cannot learn how to write bywatching a DVD on how to write—one has to hold the pencil, make the move-ments, and practice and master the skill. The NMT assessment and functionalmapping allow targeted therapeutic efforts in the neural systems that mediatethe child’s specific symptom array. When symptoms related to the persisting‘‘fear’’ response (common in maltreated children) are addressed, therefore,remembering that these first arise in the brainstem and then move throughthe brain up to the cortex, the first step in therapeutic work is brainstemregulation. The child may also have a host of cortically mediated symptomssuch as self-esteem problems, guilt, and shame. The most effective interven-tion process would be to first address and improve self-regulation, anxiety,and impulsivity before these cognitive problems become the focus of therapy.

A key component of the NMT recommendations relates to the child’scurrent relational milieu. A primary finding of our clinical work (and manyother researchers; see above) is that the relational environment of the childis the major mediator of therapeutic experiences. Children with relational sta-bility and multiple positive, healthy adults invested in their lives improve;children with multiple transitions, chaotic and unpredictable family relation-ships, and relational poverty do not improve even when provided with thebest ‘‘evidence-based’’ therapies. A simple metric, the NMT relational healthindex, scores the number and quality of relational supports capable ofproviding the safe, nurturing, and attuned environment in which therecommended therapeutic, educational, and enrichment activities are to be

252 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

provided. In many cases, these children’s caregivers or parents have similardevelopmental traumas, loss, or neglect; we can generate a similar ‘‘map’’ forthe key members of the child’s relational network with the goal of identify-ing the strengths and vulnerabilities of the adults who will be involved inhelping the child. Recommendations for co-therapeutic activities whereparent and child can engage and receive mutually beneficial services arecommon. In some cases, the specific interventions required to help the childare obvious, but the relational environment is so chaotic, and so relationallyimpoverished or impermanent (e.g., foster care), that the recommendedinterventions are impossible and the ultimate outcomes poor.

FUTURE DIRECTIONS

Awareness of key principles of neuroscience and neurodevelopment canimprove practice, programs, and policy in child maltreatment. A key chal-lenge is translating the emerging concepts into practical improvements inour clinical systems and in our therapeutic approach. A first and very impor-tant step is increasing capacity. Not enough parents, teachers, therapists,judges, or physicians know enough about child development or the basicsof brain organization and function. Simply increasing awareness of the keyprinciples of development and brain function would, over time, lead to inno-vations and improved outcomes; oddly enough, even though neurodevelop-mental principles impact all child-related disciplines, we rarely teach the coreconcepts and facts of neurodevelopment to our trainees in education, socialwork, medicine, law, pediatrics, psychology, and psychiatry.

An additional step is to continue to develop and study the impact ofinterventions that begin to incorporate some of the plausible clinical implica-tions of these principles (e.g., massage, yoga, EMDR, music and movement).While funding for research in ‘‘alternative’’ interventions is difficult to obtain,federal and philanthropic funders should be educated about the neurobiolo-gical plausibility of some seemingly ‘‘fringe’’ interventions and encouraged tofund clinical trials; if these interventions are proven to be effective, theycould be included in conventional mental health reimbursement models.

While in its ‘‘infancy,’’ we believe that the NMT, as well as other neuro-biologically informed, developmentally sensitive clinical approaches, offersmuch promise. We continue to learn and remain hopeful that this approachwill help us better understand and heal maltreated children.

REFERENCES

Altemeier, W., O’Connor, S. M., Sherrod, K. B., & Vietze, P. M. (1985). Prospectivestudy of antecedents for nonorganic failure to thrive. Journal of Pediatrics,106, 360–365.

Examining Child Maltreatment Through a Neurodevelopmental Lens 253

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

Anda, R. F., Felitti, R. F., Walker, J., Whitfield, C., Bremner, D. J., Perry, B. D., et al.(2006). The enduring effects of childhood abuse and related experiences: Aconvergence of evidence from neurobiology and epidemiology. EuropeanArchives of Psychiatric and Clinical Neuroscience, 256, 174–186.

Barfield, S., Gaskill, R., Dobson, C., & Perry, B. D. (2009). Implementing the neuro-sequential model of therapeutics (NMT) with filial therapy in a therapeutic pre-school setting: Implications for work with children with serious emotionaldisturbance. Manuscript submitted for publication.

Belsky, J., Jaffee, S. R., Sligo, J., Woodward, L., & Silva, P. A. (2005). Intergenerationaltransmission of warm-sensitive-stimulating parenting: A prospective study ofmothers and fathers of 3 year olds. Child Development, 76, 384–396.

Benoit, D., Zeanah, C. H., & Barton, M. (1989). Maternal attachment disturbancesand failure to thrive. Infant Mental Health Journal, 10, 185–202.

Bremner, J. D., & Vermetten, E. (2001). Stress and development: Behavioral and bio-logical consequences. Development and Psychopathology, 13, 473–489.

Caliso, J. A., & Milner, J. S. (1992). Childhood history of abuse and child abusescreening. Child Abuse and Neglect, 16, 647–659.

Chan, Y. (1994). Parenting stress and social support of mothers who physically abusetheir children in Hong Kong. Child Abuse and Neglect, 18, 261–269.

Chiao, J. Y., Iidaka, T., Gordon, H. L., Nogawa, J., Bar, M., Aminoff, E., et al. (2008).Cultural specificity in amygdala response to fear faces. Journal of CognitiveNeuroscience, 20, 2167–2174.

Coohey, C. (1996). Child maltreatment: Testing the social isolation hypothesis. ChildAbuse and Neglect, 20, 241–251.

Crnic, K., Greenberg, N., Robinson, N., & Ragozin, A. (1984). Maternal stress andsocial supports: Effects on the mother-infant relationship from birth to eigheenmonths. American Journal of Orhospychiatry, 52, 550–557.

Egeland, B., Jacobvitz, D., & Sroufe, L. A. (1988). Breaking the cycle of abuse. ChildDevelopment, 59, 1080–1088.

Fitzpatrick, K. M., & Boldizar, J. P. (1993). The prevalence and consequences ofexposure to violence among African American youth. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 32, 424–430.

Fowler, J. H., & Christakis, N. A. (2008). Dynamic spread of happiness in a largesocial network: Longitudinal analysis over 20 years in the Framingham HeartStudy. British Medical Journal, 337, a2338.

Gaudin, J., Polansky, N., Kilpatrick, N., & Shilton, P. (1993). Loneliness, depression,stress, and social support in neglectful homes. American Journal of Orthopsy-chiatry, 63, 597–605.

Gorman, J., Leifer, M., & Grossman, G. (1993). Nonorganic failure to thrive: Maternalhistory and current maternal functioning. Journal of Clinical Child Psychology,22, 327–337.

Graham-Berman, S. A., & Levendosky, A. A. (1998). Traumatic stress symptoms inchildren of battered women. Journal of Interpersonal Violence, 13, 111–128.

Hashima, P., & Amato, P. (1994). Poverty, social support, and parental behavior.Child Development, 65, 394–403.

Malinosky-Rummell, R., & Hansen, D. J. (1993). Long-term consequences of child-hood physical abuse. Psychological Bulletin, 114, 68–79.

254 B. D. Perry

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009

Margolin, G., & Gordis, E. B. (2000). The effect of family and community violence onchildren. Annual Review of Psychology, 51, 445–479.

Pascoe, J., & Earp, J. (1984). The effect of mothers’ social support and life changes onthe stimulation of their children in their home. American Journal of PublicHealth, 74, 358–360.

Perry, B. D. (2001). The neuroarcheology of childhood maltreatment: The neuro-developmental costs of adverse childhood events. In K. Franey, R. Geffner, &R. Falconer (Eds.), The cost of maltreatment: Who pays? We all do (pp. 15–37).San Diego, CA: Family Violence and Sexual Assault Institute.

Perry, B. D. (2002). Childhood experience and the expression of genetic potential:What childhood neglect tells us about nature and nurture. Brain and Mind,3, 79–100.

Perry, B. D. (2006). The neurosequential model of therapeutics: Applying principlesof neuroscience to clinical work with traumatized and maltreated children. InN. Boyd Webb (Ed.), Working with traumatized youth in child welfare(pp. 27–52). New York: Guilford Press.

Perry, B. D. (2008). Child maltreatment: The role of abuse and neglect in develop-mental psychopathology. In T. P. Beauchaine & S. P. Hinshaw (Eds.), Textbookof child and adolescent psychopathology (pp. 93–128). New York: Wiley.

Perry, B. D., & Pollard, R. (1998). Homeostasis, stress, trauma, and adaptation: Aneurodevelopmental view of childhood trauma. Child and Adolescent Psychia-tric Clinics of North America, 7, 33–51.

Rak, C. F., & Patterson, L. E. (1996). Promoting resilience in at-risk children. Journalof Counselng & Development, 74, 368–373.

Travis, W. J., & Combs-Orme, T. (2007). Resilient parenting: Overcoming poorparental bonding. Social Work Research, 31, 135–149.

Bruce D. Perry, M.D., Ph.D. is the Senior Fellow of The ChildTrauma Academy,a not-for-profit organization based in Houston, Texas that promotes innovations inservice, research, and education in child maltreatment and childhood trauma(www.ChildTrauma.org). Dr. Perry serves as Adjunct Professor, Department of Psychia-try and Behavioral Sciences, Feinberg School of Medicine, Northwestern University,Chicago, IL.

Examining Child Maltreatment Through a Neurodevelopmental Lens 255

Downloaded By: [Perry, Bruce] At: 05:07 2 July 2009