This review is concerned with examining impairments of the developing brain attributable to, or caused by, abuse and neglect, excluding non-accidental injury that causes gross physical injury to the brain. An attempt has been made to review interesting and salient work published in the past decade, selectively citing studies of particular relevance to clinical work.
As is clear, most of the source material for this review emanates from literature not specifically, or indeed in any way, concerned with child maltreatment.
Child abuse and neglect are sometimes alluded to in passing, when mentioned at all. This is not intended as a criticism of the extensive and excellent work being carried out in the fields of neurobiology and developmental psychology. Rather, the challenge is for those of us working in the field of child abuse and neglect to seek out and make connections with these important and clearly relevant sources of knowledge. It is hoped that this review will go some little way towards this endeavour.
Nature and Context of Child Abuse and Neglect
Child abuse and neglect are (wo)man-made phenomena which adversely affect a child's development and sometimes survival, and which should, at least in theory , be preventable. The definition of child abuse and neglect is not predicated on the intention to harm the child. In order for an experience to qualify as child abuse or neglect, it will be considered to be outside the norm of the usual interaction with a child and sufficient to be actually
or potentially harmful to the child. In the U.K., Significant Harm (Adcock & White, 1998) as defined in the Children Act (Department of Health, 1989) has been accepted as the threshold for recognition of child abuse and neglect. Significant Harm relies on evidence of either ill-treatment of the child that has caused or is likely to cause significant harm to the child, and/or impairment of the child's health and development which is attributable to ill-treatment or to the care that the child has or has not received.
The definition of child abuse and neglect includes both qualitative and quantitative aspects. It includes single events, repeated events, or a pattern of interaction that is characteristic of the relationship between the abuser, in this case often a parent or primary carer(s), and the child.
Whereas physical and sexual abuse are in the nature of events, neglect and emotional abuse characterise the relationship between the carers and the child (Glaser & Prior, 1997). It is possible that event-type abuse, which is more likely to be traumatic in nature, leads to different effects on the brain than do chronic emotional neglect and abuse.
Different forms of child abuse and neglect often coexist (e.g. Claussen & Crittenden, 1991; Ney, Fung, &Wickett, 1994). As Manly, Cicchetti,and Bamett (1994) point out, there are many different aspects of abuse and neglect including subtype, severity, frequency, and chronicity. The child's particular characteristics, be they age, gender, temperament, or disability, may make her more vulnerable to particular forms of abuse or neglect. An abuser or a neglectful parent are likely to have been affected adversely by their own past experiences ( e.g. Harmer, Sanderson, & Mertin, 1999). Moreover, the abusing or neglecting person and the child(ren) are often living in a sub-optimal social and, indeed, physical context (T. Boyce et al., 1998). There are factors working in parallel at different levels or systems ( e.g. Bronfenbrenner, 1979) that contribute to the ultimate particular form of child abuse or neglect. The child's own attributes, the child's family context, and aspects of the maltreatment will all contribute to the later outcome for the child.
Effects of Child Abuse and Neglect
Over the last decade, evidence has continued to accumulate about the strong association between childhood maltreatment and social, emotional, behavioural, and cognitive adaptational failure as well as frank psychopathology, both in later childhood and adulthood (e.g. Cicchetti & Toth, 1995; Post, Weiss, & Leverich, 1994).
Many of the papers published in prominent child maltreatment journals such as Child Abuse and Neglect have been concerned with various manifestations, associations, and outcomes of mostly - but not exclusively sexual abuse. The concepts of multi-finality and equi-finality (de Haan, Luciana, Malone, Matheny, & Richards, 1994) are important. Multi-finality recognises that similar, or even seemingly identical, experiences that different individuals undergo can and may well lead to different outcomes. Relatedly, equi-finality describes the observed phenomenon that different developmental trajectories undergone by different individuals may, nevertheless, arrive at a common outcome. For these reasons, it is not possible to predict accurately the outcome of certain experiences. Moreover, some studies reporting associations of abuse with poor later functioning assume causality that may be misplaced (Rind, Tromovitch, & Bauserman, 1998).
Studies of children who have been abused
Examples of outcomes in childhood are included in seminal volumes such as Cicchetti and Carlson (1989) on the (causes and) consequences of child maltreatment, and include specific considerations such as the relationship between maltreatment and later peer relationship problems (Mueller & Silverman, 1989; Rogosch, Cicchetti, & Aber, 1995). The review by Kendell-Tackett, Williams, and Finkelhor (1993) of outcomes following sexual abuse has been helpful in pointing out, first and fortunately, that not all children who have been sexually abused are left with problems, and second that there is no "post child sexual abuse syndrome", the harmful sequelae showing considerable interpersonal differences.
Since, in humans, neglect often coexists with physical or sexual abuse, there have been few studies that have examined specifically the effects of emotional neglect on development. The longitudinal study by Egeland and Erickson ( 1987) included a proportion of children whose mothers were neglectful and/or emotionally unavailable, although there was also overlap with physical abuse. Follow-up findings show the emotionally neglected children to be socially withdrawn, inattentive, and cognitively underachieving in their elementary-school years (Erickson & Egeland, 1996).
A study of 25 children aged between 23 and 50 months (mean = 35 months) who were being raised in a Romanian orphanage were found to be seriously delayed in their cognitive and social functioning. This study used recognised and nontraditional tests and, interestingly, showed that the findings were not related to the Apgar scores at birth, age of entry into the orphanage, or length of time of residence in the orphanage, the mean here being 26 months (Kaler & Freeman, 1994). The cognitive developmental progress at the age of 4 years of Romanian children raised in orphanages and later adopted into families in the U .K. has been shown to be predicted by the age of entry into the U.K. Being adopted before the age of 6 months was protective of later development (Rutter et al., 1998).
There must be some debate about the extent to which maternal depression can be appropriately linked with child abuse and neglect. The later inclusion of some findings about the neurobiological consequences for the development of infants and children who have been cared for by mothers suffering from depression, is justified by the fact that these studies shed light on the effects of emotional deprivation and insensitive-intrusive care on the children's development.
In a recent review, Weinberg and Tronick ( 1998) summarise findings about the effects of maternal depression on infant cognitive, behavioural, and emotional functioning, showing compromise of development in all three domains, effects that last beyond the mother's resumption of normal interaction with the Infant.
Studies of adults abused in childhood
Notable among the evidence about adult sequelae of childhood maltreatment is the study by Mullen, Martin, Anderson, Romans, and Herbison (1996) of a community sample of women from New Zealand. The authors found that a history of any form of serious childhood abuse was associated with prior risk factors in the family, which were, in themselves, associated with negative outcomes in adult life. However, sexual, physical, or emotional abuse showed an independent association with adult mental
health problems, found on the PSE (Present State Examination) (Wing, Nixon, Mann, & Leff, 1977) to be in the depressive and anxiety areas. To a limited extent, different forms of childhood abuse were found to be associated with different adult psychopathology as well as personal and social problems.
Thus, adult eating disorders were more commonly found following reported childhood emotional and sexual abuse. A history of sexual and emotional abuse by a male caregiver was associated with sexual difficulties in adulthood. Emotional abuse by the mother was associated with worse self-esteem and physical abuse with marital breakdown. The authors caution against an exclusive emphasis on sexual abuse.
In psychiatric patients, associations have been shown between a history of childhood physical and sexual abuse and adult psychopathology (e.g. Swett, Surrey, & Cohen, 1990). A higher incidence of post-traumatic stress disorder (PTSD) in Vietnam veterans has been shown to be associated with a history of childhood physical abuse (Bremner, Southwick, Johnson, Yehuda, & Charney, 1993). Similarly, Yehuda, Kahana, Schmeidler, et al. (1995) have shown that early trauma predisposes adults to suffer PTSD following traumatic event experiences in adulthood.
Childhood abuse as a risk factor for parenting
A history of childhood maltreatment in a parent's own past is now recognised as one important risk factor in the abuse of children (e.g. Widom, 1989). This is, however, not an inevitable outcome (Langeland & Dijkstra, 1995). Egeland and Susman-Stillman (1996) have found some significant differences between mothers who had been abused and went on to abuse their own children, and those who did not.
High scores on the Dissociative Experience Scale (Bernstein & Putnam, 1986), idealised and inconsistent accounts of their childhoods, as well as attempts at psychological "escape" including early abuse of drugs and alcohol and attempts at self-harm, distinguished the 14 mothers who went on to abuse their own children from the 10 mothers who had been able to break the cycle of abuse. These latter mothers were able to talk about their past abuse in a coherent and integrated fashion, which enabled then to reflect on their style of parenting their own children.
There are clear similarities here with findings from adults' attachment status on the Adult Attachment Interview (George, Kaplan, & Main, 1985), in which security of attachment is positively related to the coherence of the adult's account of their own childhood, rather than to the absence of maltreatment during childhood (Main, 1990).
Further work on the relationship between childhood abuse and later sensitivity to a child's emotions shows that, in contrast to mothers who had an uneventful childhood, mothers with a self-reported history of physical abuse responded to a video of a smiling, but not a crying, infant with physiological arousal, as measured by increased skin conductance (Casanova, Domanic, McCanne, & Milner, 1994). This is a paradoxical finding that requires further exploration but which indicates an insensitivity and lack of attunement to infants' emotional cues in mothers who had a history of childhood physical abuse.
Despite the methodological weaknesses in some of the research on the effects of child abuse and neglect, there can be little doubt of its potential for adverse effects on the developing child. Questions remain, however, about neurobiological mechanisms by which abuse and neglect become linked with behavioural and emotional impairment.
The remainder of this review will explore ways in which the developing brain responds to the external stimuli of the child's environment, in particular child abuse and neglect.