ERP Studies of Rearing in Atypical Environments

In this section, we review studies that have utilized ERPs to examine the effects of rearing in atypical emotional environments. The first investigations to be reviewed were carried out by Parker, Nelson, and their colleagues in the Bucharest Early Intervention Project Core Group (Parker & Nelson, 2005a, 2005b). These investigators conducted two studies with two groups of Romanian youngsters; one group was institutionalized and the other had never experienced institutional care. The researchers examined ERP waveforms evoked by facial expressions of emotion. Although not focused on child maltreatment by parents, these investigations have demonstrated the effects of institutional rearing and the associated emotional and physical neglect on brain functioning in young children.

Beginning with early studies by Spitz (1945a, 1945b), ample evidence has accumulated demonstrating the negative consequences of institutional rearing on the development and behavior of young children. While initial studies focused on intellectual development of children reared in institutions, finding drastic and permanent declines in intellectual functioning (Spitz, 1945a, 1945b), more recent work has shown that some aspects of functioning compromised by institutional rearing show recovery (e.g., Tizard, 1977). However, regardless of the degree of recovery of function and the circumstances under which this comes about, the consensus of findings across time and studies has consistently demonstrated the negative impact of institutionalization on all aspects of children's development, including intellectual, physical, behavioral, and socioemotional (for a comprehensive review of this area, see MacLean, 2003).

Researchers have not been able to clearly specify the mechanisms by which institutional upbringing is associated with the observed negative developmental outcomes. The specific factors or characteristics of institutions that are predictive of particular sets of deficits is not known, given the profound deprivation experienced by children across multiple realms (e.g., nutrition, emotional contact) that may each affect multiple areas of functioning and development. The studies by Parker and Nelson (2005a, 2005b) sought to examine the sequelae of institutional rearing on socioemotional development and functioning by utilizing ERPs to index institutionalized children's neural responses to facial expressions of emotion (Parker & Nelson, 2005a) and to familiar versus unfamiliar faces (Parker & Nelson, 2005b).

Parker and Nelson (2005a) worked with institutionalized Romanian infants between the ages of 7 and 32 months, examining the ERPs evoked by fearful, angry, happy, and sad facial expressions in order to ascertain whether these youngsters' deficits in processing emotional and social cues in behavioral interactions might have their origins in difficulties in the recognition of basic human emotions (see, e.g., O'Connor, Brendenkamp, & Rutter, 1999; O'Connor, Rutter, Beckett, Keaveney, & Kreppner, 2000; Rutter et al., 1999). Parker and Nelson (2005a) theorized that early institutional rearing, through its disruption of normal social interactions, may deprive the amygdala and its associated neural networks of vital experiences through which social and emotional stimuli are associated with internal states of pleasure and displeasure, thereby eventuating in deficits in the recognition of facial expressions of emotion. Specifically, these investigators examined four ERP waveforms elicited by these facial stimuli: (1) an early negative component (N170), believed to reflect early perceptual processes; (2) an early positive component (P250), reflecting later perceptual processes that may include recognition of the affective component of a face; (3) the Nc; and (4) the PSW

Parker and Nelson (2005a) found that, compared to Romanian youngsters who had never resided in an institutional environment, children who were institutionalized manifested different patterns of responding in early-latency components of the ERP. Specifically, at both midline and lateral electrode sites, the never-institutionalized infants exhibited greater N170 amplitudes in response to the sad expression; in contrast, the institutionalized youngsters displayed greater N170 amplitudes elicited by the fearful expression. At the midline electrode site, never-institutionalized and institutionalized children manifested greater P250 amplitudes in response to the sad and fearful expressions, respectively.

In addition, Parker and Nelson (2005a) found that there were dramatic group differences in amplitude of all the ERP components when examined across average responses to all of the face stimuli. Relative to the institutionalized groups of children, the never-institutionalized comparison groups of youngsters displayed greater amplitudes of the N170, Nc and PSW ERP components, whereas, relative to the never-institutionalized group of children, the institutionalized group exhibited a larger amplitude of the P250. The group differences in ERP amplitude found in the Parker and Nelson (2005a) investigation provide support for the hypothesis that an early institutional upbringing may contribute to a long-lasting development of cortical hypoactivation, as well as dysfunction in emotion processing and brain function.

While this study was designed to assess neural responses to facial expressions of emotion and not memory per se, the results may have some bearing on memory processes. For example, some investigators have recently discussed the close relation between visual attention and recognition memory in infants (e.g., Reynolds & Richards, 2005; Richards, 2003). Richards (2003) argues that attention serves to enhance memory processes, and demonstrated that Nc was of greater amplitude in infants when they were indeed attending to a stimulus (as indexed by heart rate deceleration). More importantly, this increased Nc, directly tied to attentional processes, was associated with a higher probability of the occurrence of late slow waves in infant ERPs, associated with updating of memory

Although the proposition that greater attentiveness to a stimulus is associated with enhancement of processes involved in updating memory for that stimulus may be self-evident, these co-occurring processes demonstrate the facilitative effect of attention on infant recognition memory. Thus, given that the institutionalized group of children displayed lower amplitudes of the Nc and PSW ERP components compared to the never-institutionalized group, it could be inferred that this general hypo-activation may reflect some general neurofunctional deficit in attention and memory processes. Also, early deficits in attentional processes revealed by ERPs (namely the Nc) may mark a vulnerability for these high-risk children to develop later deficits in memory functioning. Parker and Nelson (2005b) utilized the same sample of Romanian youngsters to examine neural processes involved in recognizing familiar and novel people, by recording ERPs elicited by images of caregivers' and strangers' faces. As in their other study (Parker & Nelson, 2005a), prominent amplitude differences between the institutionalized and never-institutionalized children were obtained in all of the ERP components examined (i.e., N170, P250, Nc, and PSW). With the exception of the P250 component, the amplitude of the other wave forms studied were larger in the never-institutionalized group, with the group of institutionalized children exhibiting a larger-amplitude P250.

Moreover, consistent with prior studies utilizing a similar paradigm (e.g., Carver et al., 2003; Dawson, Carver, Meltzoff, Panagiotides, McPartland, & Webb, 2002; de Haan & Nelson, 1997, 1999; Nelson & Collins, 1991), the amplitude of the Nc was greater in response to strangers' faces than to those of familiar caregivers for both the institutionalized and never institutionalized groups of children. In contrast, the institutionalized groups of youngsters displayed a greater PSW in response to the caregiver's face relative to the face of the stranger, suggesting that the faces of primary caregivers in institutionalized environments may not be fully encoded by the children assigned to their care.

The findings of deficits in recognition memory of faces of primary caregivers in this sample may have long-term consequences on several levels. Most basically, the findings may reflect general deficits (or at least delays) in neural processes related to recognition memory regardless of the type of stimuli or context. In addition, from a human-attachment perspective, these findings may have negative implications concerning the development of a secure attachment, which has as a prerequisite the ability to develop an internal working model of the attachment figure (Bowlby, 1980, 1988). The development of such a model almost certainly requires the ability to encode and easily remember the face (or some salient aspect) of the primary caregiver (Sroufe, 1996). The pattern of ERP evidence from Parker and Nelson (2005b) suggests that the ability to fully encode the face of the primary caregiver is compromised in the institutionalized children.

The next set of studies to be reviewed has examined ERP waveforms in children who have experienced child abuse and neglect. The bulk of this work has been conducted by Pollak and his colleagues, who have examined ERPs in response to various classes of emotion stimuli (Pollak et al., 1997; Pollak et al., 2001). This research has focused on elucidating the potential mechanisms through which the chronic stress experienced by children who have been maltreated could eventuate in problems in the processing of emotion (Pollak, Cicchetti, & Klorman, 1998), focusing on the construct of human attachment.

Attachment systems have been theorized to be constructed to permit flexible responses to environmental circumstances, to influence and be influenced by emotion-regulation abilities, and to function through internal working models that children hold of themselves and of their relationships with others (Bowlby, 1969/1982; Bretherton, 1990; Cassidy, 1994; Sroufe, 1996). In their ERP investigations, Pollak and colleagues have strived to ascertain whether the activation of these mental representations, which could be characterized as involving memory processes (i.e., retrieval), may be reflected through physiological activity as well as behavior. Pollak and colleagues reasoned that the P3b (P300) ERP component, which is believed to reflect neural processes involved in the updating of representations in working memory, may be useful in illuminating the cognitive processes that accompany the encoding of salient emotional stimuli.

In their first investigation, Pollak and colleagues (1997) looked at a group of school-aged children, comparing the ERPs of maltreated children to those of nonmaltreated children of comparable socioeconomic background and cognitive maturity. Across a variety of experimental conditions, different facial expressions of emotion (i.e., happy, angry, neutral) were utilized as stimuli, and the probability of occurrence (i.e., rare or frequent) and task relevance (i.e., target or nontarget) were manipulated. Children were instructed to respond to either an angry or a happy face, both of which appeared less frequently than the nontarget neutral face. The amplitude of the P3b ERP waveform of the nonmaltreated children was equivalent in both the happy and angry target conditions. In contrast, the P3b of the maltreated children was larger in the angry than in the happy target conditions. This pattern of results was interpreted by these investigators to reflect the neural correlates of a process whereby angry and happy targets activated affective representations differently for maltreated versus nonmaltreated children. Such activation may involve a process whereby experience with negative affect, likely to predominate in a maltreated child's environment, is remembered and becomes a predominant affective representation. Subsequent viewing of an angry facial expression may elicit memory processes, represented by the P3b seen in the Pollak et al. (1997) study.

In another study, conducted to determine the specificity of the relation between the ERP response of maltreated children and the nature of the eliciting stimuli, Pollak et al. (2001) examined and compared the ERP response of maltreated and nonmaltreated children to prototypic happy, angry, and fearful facial expressions. As was the case in their prior study, it was discovered that nonmaltreated children exhibited equivalent P3b amplitude in response to all of the target facial expressions of affect (e.g., anger, happiness, fear). However, the amplitude of the P3b waveform of the maltreated children exceeded that of the nonmaltreated children only in response to the angry target (Pollak et al., 2001). These investigators suggested that these findings demonstrated that there was specificity in maltreated children's differential processing of emotional information, and that maltreated children are uniquely sensitive to detecting anger over other facial expressions of emotion.

In ascribing psychological meaning to these psychophysiological results, Pollak et al. (1997; Pollak et al., 2001) conjectured that attention to negative versus positive affect activates mental representations in these children that are distinct from the patterns of activation observed in non-maltreated children. Emotion systems have been postulated to function as associative networks wherein input that matches significant mental representations activates memory systems (Lang, 1994). P3b amplitude, in this context, may mark the match of facial stimuli with more complex emotional memories.

The results of the Pollak et al. (1997; Pollak et al., 2001) studies suggest that that the nature of experiences encountered by maltreated children during their lives has caused particular emotional stimuli to become more salient, based in part upon the stored mental representations (i.e., memories) that have been associated with the stimulus over time. Consequently, the prior experience of maltreated children is thought to be reflected in these children's psychophysiological response.

In a subsequent study, Cicchetti and Curtis (2005) conducted an investigation of maltreated toddlers' ERP responses to emotion stimuli. These investigators focused on neurofunctional processes in toddlerhood because it is a period when critical aspects of synaptogenesis, driven in large part by experience in the environment as well as continued my-elination, are occurring (Cicchetti & Curtis, 2006; Huttenlocher, 2002; Thompson & Nelson, 2001). Moreover, development of higher cognitive functions primarily mediated by the prefrontal cortex, such as speech production and language, as well as advances in self-development and representational/symbolic processes, are taking place during the toddlerhood period (e.g., Cicchetti, 1990b; Thompson & Nelson, 2001).

Cicchetti and Curtis (2005) examined the association of maltreatment during the first year of life on the neural correlates of processing facial expressions of emotion at 30 months of age. ERPs evoked while maltreated and nonmaltreated children passively viewed standardized pictures of female models posing angry, happy, and neutral facial expressions were examined. Four ERP waveforms were derived in the study, including early perceptual waveforms (N150, P260), Nc, and a PSW

As expected, the maltreated and nonmaltreated toddlers did not show latency or amplitude differences on the early perceptual negative component (N150). However, the maltreated toddlers displayed greater P260 amplitude at frontal leads compared to the nonmaltreated toddlers in response to viewing angry facial expressions. In addition, the scalp topography of the P260 differed between the two groups, as illustrated in Figure 3.3 (also see color insert). Specifically, within the maltreated group, the P260 showed equivalent maximal amplitude at frontal sites for all three emotion conditions (i.e., angry, happy, and neutral). However, for the nonmaltreated toddlers, P260 showed maximal amplitude at more central scalp sites while viewing angry faces but was maximal at frontal sites in response to happy and neutral faces.

Similar in morphology to the P250 waveform seen in the Parker and Nelson (2005a, 2005b) studies, the P260 waveform may represent a later perceptual process that includes recognition of the affective component of a face. Thus, the findings from the P260 may indicate a lack of early perceptual differentiation amongst the three facial expressions of emotion in the maltreated children, while, based on the differential scalp topography

Angry Happy Neutral

Angry Happy Neutral

figure 3.3. Topography of the P260 event-related potential (ERP) component across three emotion conditions for maltreated and nonmaltreated groups. Each map is constructed based on the latency of the P260 peak at Fz (mid-line frontal scalp region, where P260 was maximal) from the grand mean of that group and condition. Because of amplitude variations between groups and across emotions, the scales were adjusted for each condition separately in order to best illustrate the P260 component. Source: Cicchetti & Curtis, 2005.

figure 3.3. Topography of the P260 event-related potential (ERP) component across three emotion conditions for maltreated and nonmaltreated groups. Each map is constructed based on the latency of the P260 peak at Fz (mid-line frontal scalp region, where P260 was maximal) from the grand mean of that group and condition. Because of amplitude variations between groups and across emotions, the scales were adjusted for each condition separately in order to best illustrate the P260 component. Source: Cicchetti & Curtis, 2005.

of the P260, the nonmaltreated toddlers processed the angry facial expression "differently" than the happy or neutral facial expressions. Given its probable role in the early detection of affect, these findings could also indicate that the angry facial expression was more "novel" than the other expression for the nonmaltreated children.

A similar pattern of findings emerged for the Nc component, illustrated in Figure 3.4 (also see color insert). For this waveform, the maltreated toddlers showed greater Nc amplitude at the central midline scalp site while viewing angry faces, whereas no regional variations in Nc amplitude within the nonmaltreated comparison toddlers emerged. However, the nonmaltreated toddlers exhibited greater Nc amplitude at lateral parietal and temporal electrode sites while viewing pictures of happy faces compared to angry and neutral faces.

As reviewed earlier in the chapter, the Nc waveform is believed to represent the neurofunctional processes associated with the automatic allocation of attentional resources, theorized to be linked with an orienting

figure 3.4. Topography of the Nc event-related potential (ERP) component across three emotion conditions for the maltreated and nonmaltreated groups. Each map is constructed based on the latency of the negative central (Nc) peak at Cz (midline central scalp region, where Nc was maximal) from the grand average of that group and condition. Because of amplitude variations between groups and across emotions, the scales were adjusted for each condition separately in order to best illustrate the Nc component. Source: Cicchetti & Curtis, 2005.

figure 3.4. Topography of the Nc event-related potential (ERP) component across three emotion conditions for the maltreated and nonmaltreated groups. Each map is constructed based on the latency of the negative central (Nc) peak at Cz (midline central scalp region, where Nc was maximal) from the grand average of that group and condition. Because of amplitude variations between groups and across emotions, the scales were adjusted for each condition separately in order to best illustrate the Nc component. Source: Cicchetti & Curtis, 2005.

process (Nelson, 1994; Richards, 2003). In light of Richards' (2003) finding of an association between Nc amplitude and an objective measure of attention (i.e., greater Nc amplitude occurred during trials with greater attention), it is perhaps reasonable to interpret differences in Nc amplitude across stimulus categories as representing differences in the degree (or extent) of attention to those stimuli. Thus, maltreated toddlers in the Cicchetti and Curtis (2005) study may have been automatically allocating more attention to the angry faces relative to the happy and neutral faces, consistent with results from the Pollak et al. (1997; Pollak et al., 2001) studies of school aged maltreated children.

The relation of Nc to memory processes is less clear; the Nc has been shown to reflect some aspects of recognition independent of stimulus probability, with greater Nc amplitude observed in response to familiar stimuli in infants and young children (de Haan, Johnson, & Halit, 2003). Therefore, one inference that could be made is that angry facial expressions may be more familiar to maltreated children than happy or neutral facial expressions, as consistent with the findings from de Haan et al. (2003), where greater Nc amplitude was seen in response to familiar stimuli. Likewise, the increased Nc amplitude elicited by happy facial expressions in the nonmaltreated toddlers may be interpreted as reflecting increased attention to more familiar happy facial affect.

Finally, the nonmaltreated group in the Cicchetti and Curtis (2005) study showed a greater PSW area score in the right hemisphere in response to viewing angry facial expressions in comparison to the maltreated group. The PSW, as discussed in the overview of normative studies of ERP waveforms, is believed to represent updating of working memory for stimuli that have not been fully encoded. Thus it could be inferred from this finding that, for the nonmaltreated children, angry facial affect may be a less familiar stimulus and, as such, may require neural resources for updating in working memory.

Parenting Teens Special Report

Parenting Teens Special Report

Top Parenting Teenagers Tips. Everyone warns us about the terrible twos, but a toddler does not match the strife caused once children hit the terrible teens. Your precious children change from idolizing your every move to leaving you in the dust.

Get My Free Ebook

Post a comment