Right Hemispheric Dysfunction in Nonverbal Learning Disabilities: Social, Academic, and Adaptive Functioning in Adults and Children
Margaret Semrud-ClikemanDepartment of Educational Psychology University of Georgia
George W. Hynd
Departments of Educational Psychology & Psychology University of Georgia, and Department of Neurology, Medical College of Georgia
This review addresses recent research on social and nonverbal learning disabilities. Involvement of right hemispheric dysfunction in these disabilities has been hypothesized, as studies with adults have suggested that documented right hemisphere damage may lead to deficits in social skills, prosody, spatial orientation, problem solving, and recognition of nonverbal cues. Studies of children purported to evidence nonverbal learning disabilities are reviewed and compared with the results from studies of adults with right hemisphere damage. Specific subtypes of nonverbal learning disabilities are reviewed, including perceptual-organization-output subtype. Aspergers Syndrome, Developmental Gerstmann Syndrome, left hemisyndrome, right hemisphere syndrome, and right parietal lobe syndrome. Finally, implications and future research needs are addressed. The need for a diagnostic nosology and improved and validated intervention techniques is stressed as is early identification of these types of specific nonverbal learning disabilities.In the past literature, the left hemisphere of the brain has often been referred to as the dominant hemisphere, with the right hemisphere called the nondominant or minor hemisphere. This nomenclature was not meant to be prejudicial, but it reflects a historical emphasis on the localization of language (Hynd, 1988). Many studies have been published regarding the linguistic functions and importance of the left hemisphere (Penfield & Roberts, 1949: Ojemann & Whitaker, 1978). More typically, the right hemisphere has seen as important for visual-spatial and lesser functions.
Since the time of Jackson (1915), however, there has been an interest on the right hemispheres function. The work of Milner (1974), Zangqill (1967), and others helped chart functional relations. Following these early investigations, more recent studies have questioned the "nondominance" of the right hemisphere for cogitative functioning (Bear, 1983; Bhatnagar & Rosenthal, Mesulam, Zaidel & Sperry, 1983; Bhatnagar & Andy, 1983; Dwyer & Rinn, 1981; Safer & Leventhal, 1977). Evidence has been reported to show that the hemispheres may function in concert with each other on almost all tasks (Kupfermann, 1985; Mesulam, 1985; Tucker 1981). In other words, the left hemisphere is not just linguistically based and the right hemisphere visual-spatially based; rather, they complement each other.
Goldberg and Costa (1981) have suggested that the two hemispheres have different processing modes that are suited for different aspects and stages of cognition. They proposed that the tight hemisphere has more association, and the left hemisphere processes by specific modality areas, with integration between these modality areas. The left hemisphere is postulated to be superior in analyzing and classifying cognitions into existing schemas. The right hemisphere is most adept at processing novel information and constructing schema that are shared with the left hemisphere for future use (Bever, 1983). Goldberg and Costa (1981) postulated that the reason for these differences in processing abilities is that the left hemisphere has more prominent modality-specific representations and the right hemisphere has more cortical areas devoted to intermodal association. Gur et all, (1980) provided neuroanatomical support for this conceptualization with their finding that the gray to white ratio is greater in the left hemisphere than in the right hemisphere, thus indicating the presence of many nonmyelinated fibers and neuronal masses. On the basis of her review of computed tomography (CT) studies, LeMay (1976) concluded that the right hemisphere is made up of relatively more white matter. Goldgberg and Costa (1981) concluded that the left hemisphere has more interregional connections.
Most of the arborization of dendrites and mylenization occurs following birth (Hynd & Willis, 1988). This dendrites arborization and development of synapses has been found to be interdependent, sequential, and essential for integrated functioning of the central nervous system (Jacobson, 1978). I, in fact, this development is interrupted or delayed and interconnections are lost, the right hemisphere is more likely to be significantly compromised because it has a higher number of interregional connections than is found in the left hemisphere.
Three implications emerge from this conceptualization. The first is that the right hemisphere is more adapted to deal with input from several sensory modalities, and the left hemisphere deals best with a single mode of presentation and processing. Second, the right hemisphere is more able to process complex schematic information. Third, disruptions in perinatal and infant neurological development may have a significantly greater effect on right hemispheric processes. Goldberg and Costa (1981) concluded that neither hemisphere is solely responsible for particular tasks or input. Instead, the hemispheres share complementary involvement in a wide range and variety of cognitive tasks.
These views are similar to the popular view of an analytical holistic dichotomy of right-left hemisphere functions. Information in the left hemisphere is thought to be processed in a step-by-step, analytical fashion, whereas the right hemisphere is believed to be adept at processing in a more global manner (Bradshaw & Nettletonm, 1983; Weinsteinm 1980). The differences are seen as quantitative and not qualitative. Goldberg and Costas (1981) hypothesis fits will with these ideas, as the short fibers between modalities in the left hemisphere would seem to be most adaptable to the analyzing and categorizing of data, and the long myelinated interregional fibers of the right hemisphere would be most adaptable to integrating input from may modalities at once to form a coherent whole of novel and complex schematic stimuli.
The hemisphere not only differ in structure and function, but their development appears to differ in males and females (Witelson, 1976b). The male brain matures at a slower rate than the female brain, and this slower rate appears to have a direct effect on brain specialization (Witelson 1976b) and asymmetry (Kolb & Whitshaw, 1980). That is, the slower the maturation, the more the asymmetry (Waber, 1976). Therefore, the slower-developing male brains should show more asymmetry. Tachistoscopic and dichotic studies have found greater response asymmetry in male brains than in female brains (Kimura, 1969; Lake & Bryden, 1976). Moreover, electrophysiological evidence shows left asymmetry of the auditory evoked response to be more pronounced in males than in females (Harris, 1978). Therefore, it is reasonable to speculate that disruption of right hemispheric function may differentially affect males and females. These factors may related to the frequent finding that makes are at higher risk for many learning and behavioral problems than are females (Hynd & Willis, 1988) and that potentially important interactions may exist among gender, brain morphology, and neuropsychological development (Hynd & Semrud-Clikeman, 1989a, 1989b).
The left hemisphere has been the focus of many studies regarding the extent to which damage will interrupt its normal functioning. Less study has been devoted to the effects of right hemisphere damage on learning and behavioral functioning. Until recently, the major emphasis has been on the reactions of adult patients with trauma or lesions in the left hemisphere.
The purpose of this review is to elucidate the role of the right hemisphere with respect to social-emotional development and deficits on social perception, and its contribution to specific learning disabilities, especially, perhaps, in arithmetic. It is felt that if the right hemisphere were more adept at "reading" complex and novel situations, as the earlier conceptualization suggests, then deficits in the right hemisphere might significantly affect a persons perception of his or her environment.
Specifically, we discuss the possible relation of right hemisphere deficits to nonverbal learning disabilities in children. An attempt is made to show in what manner right hemisphere deficits may interfere with child-development and contribute to learning disabilities in the area of social perception and arithmetic. This area of study is particularly relevant as the National Institutes of Health Interagency Report to Congress recommended that significant deficits in the attainment of age-appropriated social skills be reorganized as a specific learning disability (Wyngaarden, 1987). It should be pointed out that there is little documentation of right hemisphere damage in children with this disability. Hence, we compare and contrast behavioral data with what is known about adults with documented right brain damage (RBD). Caution must be stressed, as data from adults are not directly comparable to that from children: Brain injury may have different consequences for development at various ages (Hynd & Willis, 1988). However, as Denkla (1973) suggested, the utilization of analogies between childhood and adult syndromes may be both conceptually useful and productive in deriving clinical classification schemes.
Nonverbal Learning Disabilities
Early descriptions of nonverbal learning disabilities were presented by Johnson and Myklebust (1971). They defined this disorder as characteristic of children who are unable to comprehend the significance of many aspects of the environment; who cannot pretend and anticipate; and who fail to learn and appreciate the implications of actions such as gestures, facial expressions, caresses, and other elements of attitude. In Johnson and Myklebusts view, the nonverbal learning disabled child is unable to acquire the ability to determine the significance of basic nonverbal aspects of daily living, even though their verbal intelligence is at or above the average level. The nonverbal disability impairs perception and imagery, and therefore constitutes a more fundamental distortion of the total perceptual experience. Johnson and Myklebust termed this finding a social perception disability and found that the social quotient correlated significantly with the diagnostic findings of neurologists.
Myklebust (1975) further clarified these ideas with his definition of social imperception. Social perception, or social imperception, is defined as the "childs ability or lack of ability to understand his social environment, especially in terms of his own behavior" (p. 86). Myklebust found that the child with this type of disability is often immature and unable to make many of the routine judgements needed to succeed in everyday life. He felt that the difficulty is not due to the perceptual problems, per se, but rather with memory and imagery, with a basic deficiency in storage rather than in recall.
This idea is consistent with the findings of Borod, Koff, and Caron (1983) and Ross and Mesulam (1979) that the defect on right-brain-damage adult patients was not in perception of faces but rather in the identification of expression. It seems reasonable to speculate that a person needs an intact image of an expression in order to facilitate processing by the left hemisphere. Thus, in right brain damage a deficiency may be present in the ability to match the image of an expression with a current impression.
The following problems are believed to be present concomitantly with nonverbal learning disabilities; disturbed social relationships; poor self-help skills; difficulty learning right from left; difficulty learning to tell time, to read maps, or to follow directions; disturbances in math, and deficits in learning the meaning of the actions of others (Myklebust, 1975). Moreover, these childrens poor social perception is believed to limit their inner experience, which consequently has a delimiting effect on reasoning and adaptive behavior (Myklebust, 1975). Presumably, nonverbal disorders may be more handicapping than verbal learning disabilities, as verbal deficits have little effect on nonverbal experience but nonverbal deficits may make major contributions to the misreading of verbalizations. Various researchers have found support for the relation between nonverbal learning disabilities and deficits in social perception of self and others (Gaddes, 1985): interpretation of emotion and visual-spatial tasks (Wiig & Harris, 1974); and deficits in differentiating and interpreting facial expressions with a defective body image (Badian, 1983).
Development of Social Perception
Before discussing in more detail the social perception problems of learning disabled children, it is important to determine through use of a developmental framework what social skills are necessary for success. Krudek and rile (1982), in a study in the development of social competence skills, found that the ability to project oneself into anothers place becomes more important with age. They investigated social skills and their development in third through eighth graders. It was found that popular children were most adept at communicating effectively, knowing how to initiate a conversation, and being aware of the other persons affective state and empathizing with then and, most important, having the ability to match their social skills to the demands of a particular situation. In other words, popular children are able to succeed in all areas that the nonverbal learning disabled child shows exceptional difficulty. It has also been found that the social skills most important for development between the ages of 9-12 were the ability to conceptualize alternative scenarios, to anticipate, and to use cause-and-effect reasoning. To complete these tasks successfully, children must be able to evaluate a social situation and then make judgements based on their perceptions (Bruno 1981).
The accuracy of emotional recognition and emotional labeling has been demonstrated to increase in direct relation to age. Normally developing children will recognize emotions far sooner than they can apply appropriate labels to them. The finding of similar growth curves between cultures confirms that emotional recognition is strongly related to age and development and probably has a biological basis. In addition, studies of preschool children have found that emotional recognition and emotional labeling are positively and significantly correlated with indices of intelligence and perceptual-motor skills in both normal preschoolers and disadvantaged children. Emotional recognition skills have also been found to be a significant predictor of academic achievement in the areas of reading and arithmetic (Izard, 1977).
It is, therefore, not unreasonable to suggest that children who are unable to acquire these skills because of difficulty in evaluating facial expressions, gestures, or prosody would be at high risk for the development of significant learning difficulties in, at the very least, the social-emotional arena of competence. One might also assume that this disability may be present early in development but may not necessarily be recognized until the child begins school and moves into the age range in middle childhood where peers and peer relationships become more crucial. However, it also seems reasonable to propose that problems are most likely to be present from birth and, although of unknown etiology, unfold as the child develops.
Because the right hemisphere develops faster than the left from 18-24 months of age, this faster growth may reflect greater involvement of the right hemisphere in infancy. Much of the prelinguistic childs learning consists of the perception of visual-spatial relations, patterns, environmental sounds, and rhythms. The emotional experience of the infant develops through sounds, images and pictures that constitutes much of an infants early learning experience, and are disproportionately stored or processed in the right hemisphere during the formative stages of brain ontogeny (Ley & Bryden, 1981). Therefore, congenital dysfunction or arrest of the right hemisphere during development could be seen as very deleterious to the childs early development. As Boll (1974) suggested, the "general development rate and time of onset and chronicity of the disorder, while relevant for both adults and children, appears to play a far more crucial role in the latter group" (p. 92).
In a study of 1-year old children, Ainsworth (1979) found that the way infants express themselves toward the mother will also affect the way in which they interact with their environment and organize their experiences into a meaningful and predictable framework. This organization, in turn, provides continuity in cognitive and emotional development. If a child with a dysfunctional right hemisphere has difficulty in retaining an image of his or her mothers face and thus does not respond appropriately, it is not difficult to believe that the mother-child relationship may be significantly altered. It is possible that bonding will not occur, and the childs emerging view of the world is sure to be altered.
The human face is probably the primary vehicle for early development and exercise of emotions in a safe and secure environment. Accordingly, facial expression plays a critical role in the development of social responsiveness. Vision and other stimulation such as vocalization and facial expression directed toward the infant may contribute more to determining the amount of social responsiveness than does meeting the infants physiological needs (Izard, 1977). Furthermore, the perception of the mothers face seems related to the development of attachment behaviors and social adequacy. The infant who has difficulty in processing and retaining visual-spatial and auditory stimuli and in prediction of temporal events may well experience deficits in the nuances of human form and sound. Deficits in the recognition of facial expressions will delay attachment which will in turn delay behavior that promotes exploration cognitive reorganization, and separation (Kaslow & Cooper, 1978). Moreover, social learning is implicated in facial expressions as only certain expressions appear on the adults face with any frequency in front of the infant. The child learns from these displays how and when to display affects (Mayo & LaFrance, 1979). It is not unreasonable to speculate as to the major developmental effect of an infants intuition to decipher facial expressions even at an early age. During the first year of lie, children normally communicate through one nonverbal channel at a time. The integration of these channels emerges as part of the cognitive developmental process. In addition, children learn how to integrate their nonverbal communication with others. Thus a childs appreciation of distance and gaze as cues to attraction and liking become well established in the preschool years (Mayo & LaFrance, 1979). Horton (1976) agreed as to the importance of childrens ability to orient themselves to the mothers and to "recognize" her. He believed that personality disorder is "related to or caused by in part, a disturbance in the right hemisphere" (p.783). Because babies perceive emotional stimuli before verbal stimuli, engrams for emotional voices are more strongly imprinted in the right hemisphere (Carmon & Nachson, 1973). It may well be that the diffuseness of the communicating fibers of the right hemisphere are more suited to the processing of nonverbal material and, with early development, take initial precedence in the infants processing of environmental stimuli (Campbell & Whitaker, 1986: Hynd & Willis, 1988).
Learning Disabilities and Social Perception
The learning disabled childs comprehension of nonverbal communications has been studied only in the past 10-15 years. Bryan (1974) summarized five studies and concluded that the learning disabled child was more egocentric and less attuned to the affective states of others. Children with learning disabilities have also been found to be significantly less empathic, thus suggesting that learning disabled childrens perceptual problems prevent them from developing the appropriate experiences to make appropriate judgements regarding the recognition of emotion to context (Bachara, 1976). Significant differences in the interpretation of affective states between learning disabled adolescents and a normal control group have also been found (Axelrod, 1982: Bruinicks 1978: Bruno, 1981; Bryan, 1977; Gerber & Zinkgraft, 1982; Pearl and Cosden, 1982; Wiig & Harris, 1974)> In contrast, a number of studies have not found significant differences between learning disabled and normal individuals in social perception and interaction skills (Bryan, Donahue, & Pearl, 1981; Bryan & Wheeler, 1972; Bryan Wheeler, Felcan & Henel, 1976; Connolly, 1969; Maheady, Maitland & Sainato, 1984; Richy & McKinney, 1987).
Such inconsistency is not surprising, perhaps, as these studies vary widely in measures used, age of subjects, criteria for learning disabilities, and number of subjects. Little attempt was made in most of these studies to control for attentional or verbal variables. The learning disabled children were also seen as constituting one group. It is noteworthy that none of these studies reported how many of the children had speech and language disabilities, how many were diagnosed as also having Attention Deficit Disorder (ADD), or how many had similar family histories. Denckla (1979) provided convincing evidence that the presence of various codiagnoses, including types of language disorder for ADD with hyperactivity, is meaningfully related to subtypes of childhood learning disabilities. More recent studies have provided further support for the idea that learning disabilities frequently co-occur with other psychiatric disorders, particularly ADD without hyperactivity (Hynd et al, in press). Thus the lack of this information in these reports represents a significant and shared problem in methodology.
There are several difficulties with the studies comparing children with and without learning disabilities. From the 14 representative studies reviewed, a number of conclusions seem warranted. First, each study had differing criteria for defining learning disabilities. Not only is this variance in definition a problem, but in 11 of 14 studies the criteria for selection were not specified. IQ and achievement scores were not fully reported, and groups were vaguely defined. One of 3 studies that reported IQ and achievement, 1 study (Wiig & Harris, 1974) used different criteria for their learning disabled group depending on the school attended. Second, few of the studies evaluated speech and language skills, and the three that did (Bachara, 1976; Bruno, 1982; Wiig & Haris, 1974) did not report appropriate scores. Language difficulties are known to influence scores on tests used to diagnose learning disabilities (Hynd & Semrud-Clikeman, 1989b). Third, no study conducted emotional assessment directly. One must rely on the authors assurance that these children were free of "primary emotional disturbance." Sociometrics, an area one would thins would be particularly relevant to assess, was ignored in 11 of 14 studies, with teacher ratings used in 1 additional study. Surprisingly, both studies (Bruinicks, 1978; Siperstein, Bop, & Bak, 1978) that used sociometrics found nonsignificant correlations between their measures of social perception and sociometrics. No study supplied any developmental history: some included children with neurological soft signs and some excluded these groups. Moreover, another major oversight was that not one study assessed attention or hyperactivity . Because attention to the task would appear to be a major confounding variable, this omission in these studies must be viewed as a serious oversight.
Another problem with these studies is the variation and dissimilarity between measures of social inference and actual, socially appropriate behavior. Although four studies used observation of children in their natural setting (Bryan, 1974; Bryan & Wheeler, 1972; Bryan et al, 1976; Richy & McKinney, 1978), none of these studies assessed awareness on the childrens part of their own behavior and their view of their interpersonal relationships. This oversight is serious, as the studies purported to investigate the learning disabled childs social skills and social perception, yet included no assessment of related cognitions.
It is not surprising that divergent results were found, considering such methodological variability. However, these studies do represent an initial effort in an area of research that had previously not been empirically addressed.
Arithmetic Learning Disabilities
Johnson and Myklebust (1971) originally tied dyscalculia to deficiencies in visual-spatial organization and nonverbal integration. They found that children with this disability were unable to quickly distinguish differences in shapes, sizes, amounts, and lengths. Their subjects were also not able to examine groups of objects and tell which contained the greater amount. Studies of their case histories revealed nonverbal difficulties early in life. The parents reported that their children rarely played with puzzles, blocks, or constructions-type toys. Many of then showed extraordinary auditory skills and spoke early in development. Word reading was appropriate, but reading comprehension (especially beginning in fourth through fifth grade when inferential thinking is required) was deficient. They also found that most of these children had disturbances in body image and their human figure drawings lacked detail and organization.
It is likely that some children are born with a deficiency in the ability to discriminate and manipulate spatial and numerical relations (Landsdown, 1978). Although arithmetic skills have traditionally been associated with left hemisphere dysfunctioning, subjects with right hemisphere dysfunction also show difficulties with basic arithmetic operations. For children, the acquisition of number concepts may be based on exploration of the spatial and physical attributes of objects, thus suggesting that neural mechanisms in both hemispheres contribute to math skills. Therefore, arithmetic skills may be substantially hindered by early right hemisphere dysfunction (Weintraub & Mesulam, 1983).
Arithmetic underachievers have been found to have differing evoked potentials in only the right hemisphere. Moreover, these right hemisphere evoked potentials differed significantly from the findings with normal subjects (John, Karmel & Corning, 1977). Further studies have concluded that arithmetic may well be a task shared by the hemispheres for calculation and numerical judgements about the relative magnitude of two digits (Dimond & Beaumont, 1972: Katz, 1980: Murphy, Darwin & Murphy, 1977). In their study with right-brain-damaged, left-brain-damaged, and control subjects, Querishi and Dimond (1979) found that more deterioration of calculation occurred when damage existed in the right hemisphere.
Subcortical processes may additionally be implicated in arithmetic processing. For example, Qjemann (1974) found that electrostimulation of the right and left thalamus affects arithmetic ability differently. Left thalamic stimulation accelerates the rate of counting backwards and increases calculation errors with no increase of latency of number identification. Right thalamic stimulation slowed the rate of counting and increased the latency of number identification and calculation errors. Ojemann (1974) concluded that the left thalamic stimulation evokes an alerting response and acceleration of higher cognitive functions, and right thalamic stimulation affects number reading and arithmetic calculations. The thalamus may well serve as a gating mechanism, in that, when stimulated, the left thalamus not only accelerates mental arithmetic processes but also serves as an alerting system for the input of these stimuli. Stimulation of the right thalamus may open the perceptual and processing gate for number reading and computations of numbers. On the basis of this research, Ojemann (1974) concluded that abstract reasoning skills draw on the whole brain and specific subcortical mechanisms so that dysfunction anywhere is likely to increase mental rigidity and to reduce adaptivity.
Luria (1980) also emphasized the relation between arithmetic operations and spatial imagery and concepts. He suggested that spatial acalculia results from right hemisphere lesions and that acalculia itself implies bilateral dysfunction. Furthermore, Gaddes (1985), in an overview of arithmetic difficulties, suggested that a medical posterior right hemisphere dysfunction contributes to difficulty in spatial perception and imagery and results in inferior achievement in arithmetic, geometry, map drawing, graphic arts and all types of mechanical and constructional skills. These conclusions regarding the interrelations among acalculia, social intelligence, and right hemispheric dysfunction are supported by postmortem studies of patients with Turners Syndrome (Brun & Skold, 1968: Reske-Nielsen, Christensen, & Nielson, 1982), as well as by neuropsychological studies of patients with Turners Syndrome (Hynd & Willis, 1988).
Therefore, it may be concluded that arithmetic skill, at least the early precursors of arithmetic, can be significantly impaired by right hemisphere dysfunction. It is not unreasonable to suggest that nonverbal social-emotional problems and arithmetic difficulties may be related to right hemisphere dysfunction, as both involve the manipulation of spatial and visuoperceptual processes. From her review of literature on social adjustment, Badian (1983) concluded that there is a positive correlation between social skills and arithmetic. High achievers in arithmetic were found to be well adjusted and sociable, and low achievers were found to have severe emotional problems. Moreover, children with deficient arithmetic skills have been found to experience difficulty in learning appropriate social generalizations (Kirby & Asman, 1984). These findings suggest that research may productively be spent on more carefully characterizing the neuropsychological profiles of children with these specific learning disabilities.
Research on Subtypes of Nonverbal Learning Disabilities
Studies by Badian (1983), Denckla (1978), Nagy and Szatmari (1986), Rourke and Associates (Strange & Rourke, 1983: Rourke & Finlayson, 1978). Weinberg and MacLean (1986), Weiner (1980), and Wing (1981) have all examined the relations among visual-perceptual skills, social skills, motor development, and arithmetic. Because there is considerable variability in how these learning disabled children are conceptualized, a discussion of each of the classification schemes listed in Table 1 is needed.
Nonverbal Perceptual-Organization-Output Disabled Classification
Rourke has advanced the idea that central processing deficiencies can lead to both social-emotional disturbances and learning disabilities (Rourke & Fisk, 1981). Studies by Petrauskas and Rourke (1979), Weiner (1980), and Porter and Rourke (1985) have reported the finding of four separate subtypes of learning disabled children based on social-emotional characteristics. Rourke and Finlayson (1978) identified what were termed as nonverbal perceptual-organization-output disabled (NPOOD) children whose reading and spelling performances were average or above and whose arithmetic skills were relatively weaker, with impairments found in visuospatial skills. This group evidenced strong auditory perceptual skills. Rourke and Finlayson concluded that NPOOD children have a dysfunctional right hemisphere, and children with reading and spelling difficulties have a relatively dysfunctional left hemisphere. Strange and Rourke (1983 & 1985) found that the arithmetic errors evidenced by NPOOD children on the Wide Range Achievement Test arithmetic test were in spatial organization, misreading of visual detail, procedural errors, failure to shift operations, visual-motor skills, and judgement and reasoning. They were also found to experience difficulty in adapting previously learned operations to slightly different problems. Concern was expressed regarding these childrens inability to adapt to a new situation. This deficit may well have great import for the social skills of these children. Although no empirical data are offered, Strange and Rourke hypothesized that the majority of these children have experienced these general impairments since birth. Additional findings on the NPOOD children were bilateral tactile-perceptual impairment (more pronounced with left), bilateral psychomotor impairment (more pronounced with left hand), and poorly developed visual-perceptual organizational abilities. Rourke (1982) suggested that conceptual development is dependent on the right hemisphere and generally is established during the sensorimotor stage. With a dysfunctional right hemisphere these skills may show significant deficits. Infants attachment and exploration, which Ainsworth (1979) and other see as crucial for the childs well-being, may be significantly impaired. Utilizing the premise that central processing deficits predispose a child to social emotional successes or difficulties, Ozols and Rourke (1985) focused on the social functioning of the learning disabled child. They suggested that the NPOOD child has difficulty attending to and interpreting nonverbal social cues such as gestures and facial expressions. Children with auditory-perceptual problems have also been found to perform more poorly on tasks requiring verbal labeling and explanations, whereas children with poor visual-spatial and arithmetic skills (NPOOD) experienced difficulty on the tasks requiring matching of facial expressions or gestures with verbal content (Ozols & Rourke, 1985). As a consequence, these children may stop exploring the world because of their inability to profit from nonverbal experiences.
It has also been found that NPOOD children have difficulty in processing and integrating new situations or tasks and in understanding facial expressions; their social relations are routinized and stereotypical; and their speech is flat and monotonous, as is their affect (Rourke, 1982). In addition, these children appear to experience significant difficulty in concept formation and problem solving and this encounter great difficulty in benefiting from past experience. Rourke (1982) concluded that these children have a significantly dysfunctional right hemisphere with a well-developed left hemisphere that has adapted to the difficulties by overprocessing information verbally and analytically. In Rourkes study, 50% of the children with these types of difficulties were female a distinct difference from the usual 4:1 incidence in favor of males. Exploratory studies suggest that males develop this same specialization until at least adolescence (Witelson, 1976a). It may well be that this slower maturation of the right hemisphere in females predisposes them to higher risk for nonverbal learning disabilities. Given the usually earlier-developed verbal skills of females, it is not unreasonable to hypothesize that these same verbal skills are used to compensate for these weaknesses.
Further support for Strang and Rourkes (1985) syndrome of the NPOOD child was found when the social behavior of children with differing math skills was compared. The child who was a good reader but who did poorly in math computation was frequently inattentive, disorganized, avoided responsibility, and had poorer social behavior than children with reading problems (Badian & Ghubilikian, 1983).
Strang and Rourke (1985) carried the idea of the NPOOD childs dysfunctional social-emotional functioning one step farther. They assessed these children on the Personality Inventory for Children and compared the resulting profiles with children having linguistically based learning disabilities. The NPOOD children obtained a profile suggestive of psychopathology. Elevated scores were found on psychosis, social skills, anxiety, withdrawal, and depression. This profile distinguished the NPOOD child from the other children, who scored within the average range. Other characteristics found were an inability to adapt to change, avoidance of novel situations (preference for routine); clumsiness: lack of friends; a good vocabulary, language full of jargon and used to maintain contact with the listener but not to communicate: and a focus on what is said, not the context in which it is said. Again, 50% of the children in this group were female, possibly indicating a gender-related manifestation of nonverbal learning disabilities.
One may wonder what the results of the studies discussed in the previous section may have been if the learning disabled groups had been carefully defined as to subtype of learning disability. Information as to developmental history and direct assessment of social skills is notably absent in almost all of the studies, and few studies included significant numbers of females. Because Rourke (1982) and others have noted that this subgroup is not usually identified until late adolescence and adulthood, it is noteworthy that few studies of this age group were found.
Follow-us studies of small samples of adults diagnosed a NPOOD children found that they continued to show similar difficulties as adults (Del Dotto, Rourke, McFadden & Fisk, 1987; Rourke, Young, Strang, & Russell, 1985). In the Rourke et al. (1985) study, the adults difficulties were not identified until late in their school career, if at all. A recurring theme was social inadequacy. Four of the eight subjects had been treated in an inpatient psychiatric unit for depression. At least two of these were thought to be suicidal. Of the four not under psychiatric care, three were found to be prone to depression. None of the patients had been able to work at a job commensurate with their education.
Brumback and Staton (1982) suggested that depression, learning disabilities, and ADD are all associated with right hemisphere dysfunction. They proposed that ADD may be the result of an anatomically based right hemispheric dysfunction with cortical or subcortical in nature, whereas depression may be a physiological disturbance. A further clinical report Brumback, Staton, and Wilson (1984) found that when children with learning disabilities and depression were treated with tricyclic antidepressants, they showed significant improvements on right hemisphere tasks. Two of their treated children showed a "striking" disappearance of sensorimotor neurologic abnormalities on the left side during antidepressant medication therapy. Several additional studies of right hemisphere dysfunction have also offered evidence as to the role of the right hemisphere in depression (Kinsbourne & Bemporad, 1984; Ross, 1984; Weintraub & Mesulam, 1983; Wexler, 1980).
The findings of depression in several of these subjects with right hemispheric difficulties is consistent with the hypothesis that the right hemisphere plays a mediating role in the expression, both affectively and motorically, of sadness (Ladavas, Niciletti, Umilta & Rizzolatti, 1984). This sadness response may well be mediated through the right hemisphere premotor area, which " modulates and controls the more primitive responses mediated by the hypothalamus and other visceral centers" (Ladavas et al., 1984, p. 483). Moreover, electroencephalographic studies have found right frontal activation when confronted with negative affect, whereas the converse was found with positive affect (Davidson, Schwartz, Saron, Bennett, & Galeman, 1979; Tucker, Stenslie, Roth & Shearer, 1981). The EEG data from patients with pathological depression have found increased activity in these same tight frontal regions.
A recent case report by Ternes, Woody, and Livingston (1987) of a child with right hemisphere deficit is illustrative. This 11-year-old boy evidenced arithmetic and spelling deficits, social problems, withdrawal, lethargy, and frequent right temporal spikes on his electroenecephalogram. Performance Scale IQ on the Wechsler Intelligence Scale for Children-Revised was average (96), and his Verbal Scale IQ was below average (85). Reading skills were at grade level. Following treatment with carbamazepine, his Performance Scale IQ increased to 114 and his Verbal Scale IQ remained the same. Affective difficulties and interpersonal problems showed significant improvement at a 1 year follow up.
The possible role of the right hemisphere dysfunction in depression, ADD with or without hyperactivity, and concurrent arithmetic/nonverbal/social-emotional learning disability needs to be further explored (Hynd et al., in press). This area of investigation has been neglected in light of all the research efforts focused on dyslexia (Hynd & Semrud-Clikeman, 1989a). This inattention has been unfortunate for the children and adults who have not received adequate assistance and may, in fact, have been incorrectly diagnosed with psychiatric labels. Given the possible co-occurrence of depression and right hemisphere learning disabilities, we now discuss another psychiatric syndrome known variously as Aspergers Syndrome, autistic psychopathy, or developmental schizotypal personality disorder in relation to learning disabilities.
There are many similarities between NPOOD children and children who are diagnosed as having a constellation of symptoms, and developmental schizotypal personality disorder. This syndrome was first identified by Han Asperger in 1944. He described 200 cases of children who experienced difficulty with nonverbal and motivational aspects of communication; had limited facial expressions, gestures, and vocal intonation; and maintained poor eye contact. These children were characterized by a need for routine and sameness; early speech and good grammar, language used for its own sake, not as a means of interpersonal communication; narrow, repetitive play; poor peer relations (they are often the brunt of cruel jokes); circumscribed interests, and delayed visual-spatial and motor skills (Wing, 1976, 1981, 1985). Moreover, these children showed a significant lack of humor and the gender ratio was 4:1 in favor of boys. In addition, some have found that children with Aspergers Syndrome have good math and science skills (Wing, 1982; Wolff & Chick, 1980; Wolff and Barlow, 1979), whereas others have found significant difficulties in arithmetic (Baron, 1987; Isaev & Kuipers, 1962).
Although Aspergers Syndrome is often not recognized until age 3, parents, in retrospect, report that the child lacked active involvement in social interaction from early infancy. There is the perception by some that Aspergers Syndrome is at the end of the infantile autism continuum (Burgoine & Wing, 1983; Gillberg, 1985; Wing, 1985), whereas others feel it is a distinct syndrome (Asperger, 1979; Van Krevelan, 1971; Wolff & Barlow, 1979). Similarities among Asperger' Syndrome, developmental schizotypal personality disorders (SPD), and Pervasive Developmental Disorder (PDD) have been noted by Nagy and Szatmari (1986). They conclude that Aspergers Syndrome should be conceptualized as separate from PDD but very similar to SPD, which may be a more appropriate term. There is consistency in that many of the children with these traits have experienced prenatal and postnatal trauma and have strong family history of genetic predispositions for developmental Aspergers Syndrome.
Well-controlled studies must be conducted to determine the existence of Aspergers Syndrome independent of Kanner, (1943) definition of infantile autism. In addition, neuropsychological studies must further verify the uniqueness of Aspergers Syndrome as a specific type of learning disability. There may well be overlap between these two syndromes. In this vein, Shey and Mesibov (1985) recognized the overlap between learning disabilities and higher-level autism, and suggested that study their concurrent overlap will assist in understanding both conditions more fully and provide interventions appropriate to both conditions.
Developmental Gerstmann Syndrome
Another possible subtype of relevance is the Developmental Gerstmann Syndrome (DGS). Although we recognize that occurrence of this syndrome is indeed very controversial, we have included it in this review because some investigators have noted similarities to this syndrome in children with right hemispheric learning disabilities, and because recent more carefully documented cases of DGS have appeared in literature (PeBenito, 1987; PeBenito, Fisch & Fisch, 1988).
Developmental Gerstmann Syndrome consists of four elements: dyscalculia, dysgraphia, right-left confusion and agnosia (Kinsbourne, 1968). Rourke and Strang (1978) suggested that their NPOOD learning disabled subgroups are "analogous to that seen in the Gerstmann Syndrome", (p. 65). These similarities can be seen in Table 1.
What is of relevance to the present discussion is that DGS adults is usually attributed to left parietal dysfunction. There is some evidence that in children with DGS the child may be more related to never having developed the needed neurocognitive skills, whereas in adults, trauma accounts for the loss of skills. In studies of adults with DGS, Benson and Denckla (1969) and Poeck and Orgass (1966) found that the aphasia symptoms interfered with calculation abilities, no the lack of computational skills. It is likely that DGS is a separate entity and that in this latter case the localization is unknown. Benson and Geschwind (1970) suggested that constructional apraxia can accompany DGS, and children with DGS usually show a strong difference between verbal IQ and performance IQ, with verbal being the highest. Benton (1979), in an attempt to distinguish the neuroantomical basis for symptoms of DGS, suggested that deficient verbal mediation is the basis for failure in arithmetic for left-hemisphere-damaged patients, and that visuospatial deficits may account for associated deficits for right-hemisphere-damaged patients.
Traditionally, the underlying mechanism for DGS has been thought to be a defect in sequential processing (Kinsbourne & Warrington, 1963). However, as PeBenito et al. (1988) pointed out, such a deficit may explain poor performance on spelling tasks, finger agnosia, and perhaps even some components of the arithmetic disability, but it would not necessarily explain the often profound deficits on tasks assessing constructional apraxia or spatial orientation. Furthermore, the arithmetic deficits in these cases, including other reported cases (e.g., Benson & Geschwind, 1970; PeBenito, 1987), often reflect visuospatial problems. Thus although some suggest that in adults DGS reflects left parietal deficits (Strub & Geschwind, 1974), others have considered these symptoms in children as reflective of bilateral parietal disturbance (Weinberg & McLean, 1986) or right parietal dysfunction (Rourke & Strang, 1978).
Thus, whether DGS can be considered to reflect left or right hemispheric dysfunction or bilateral disturbance in children is unclear, as the pattern of symptoms could implicate many different dysfunctional systems. Be that as it may, of all the syndromes reviewed here, there is the least support for the inclusion of this syndrome among those that may be associated with right hemispheric learning disabilities.
Left Hemisyndrome Classification
Denckla (1978) suggested an additional way of classifying children with deficits in arithmetic, visual-spatial, and social perception skills. She labeled these children as having left hemisyndrome, which reflects tight hemisphere dysfunction. These children present with at least three motor system markers asymmetrically indicating right hemisphere involvements. These motor markers include disorders in reflexes, weakness, muscle tone, gait, tremors, uncoordination, nystagmus, fixed or complex squints (manifest strabusmus), and dysarthia. They also have a neuropsychological profile that documents early mild delays in speech and reading, which later develop adequately with the exception of verbal reasoning; deficits in inferential thinking; higher-order language and sociolinguistics; math disability; geographical disability in processing gestures and vocal expression. In addition, these children will experience difficulty in the perception of subtle gradations of facial expression and vocal intonation. Although these children seek social interaction, they attempt to gain approval in ways that almost ensure disapproval (for example, giving unsolicited bear hugs, chattering on without noticing anothers disapproving facial expression). These children are often verbal and are overly dependent on verbal rules and regulations. Denckla (1978) localized the dysfunction in the tight hemisphere. From Table 1 it can be seen that there are numerous similarities between the NPOOD syndrome. Dencklas (1978) left hemisyndrome, and Aspergers Syndrome would appear to be the most severe disorder on this possible continuum.
Right Hemisphere Syndrome
Voeller (1986) profiled 15 children who show striking similarity to the NPOOD and left hemisyndrome classifications. Using the results from neuropsychological test batteries, CT scans, and EEGs, she found indications of right hemispheric deficits in all these children. In addition, these children displayed significant difficulty in displaying appropriate affect and an inability to decipher the emotional states of others. As a group, these children showed left-sided neurological findings, higher verbal than performance skills, and higher reading than arithmetic abilities. In addition, there was a high percentage of children with ADD (both with and without hyperactivity). This particular finding was not mentioned by Denckla (1978). Although Rourke and Strang (1978) did not note ADD without hyperactivity, their descriptions of lethargy, monotonous voice, and attentional problems are similar to what might be expected (Hynd et al., in press, Lahey, Schaughency, Hynd, Carlson, & Nieves, 1987).
Voeller and Heilman (1988) provided evidence that children with a diagnosed right hemisphere dysfunction show evidence of motor impersistence. Using the length of time various motor movements could be sustained as a dependent measure, they found that children diagnosed as having a right hemisphere syndrome also net the diagnostic criteria for Attention Deficit Hyperactivity Disorder (ADHD: American Psychiatric Association, 1978) and evidenced deficits in motor persistence as compared with control children. Because motor impersistence is believed to be related to right hemisphere damage in adults (Kertesz, Nicholson, Cancelliere, Kassa & Black, 1985). Voeller and Heilman reasoned that ADHD may also be related to a right hemisphere syndrome in children.
Right Parietal Lobe Classification
Another system of classification was developed by Weinberg and McLean (1986). They identified two different types of learning disabilities in children with arithmetic and social emotional difficulties. Type R1, is also called developmental right parietal lobe syndrome. It is marked by hyperprosody and hyperelemental emotionality, and is characterized by the misperception of social situations with overemotional gestures word usage, and tone of speech. Tasks that are repetitive are difficult for children with Type R1. They often start tasks they do not finish. Reading is slow and labored but intact. Arithmetic problematic and writing is messy and overly wordy. They have a good vocabulary and use extremely animated and emotional speech. Prosody is poor and often inappropriate to the situation.
Type R2 is also called developmental right parietal lobe syndrome. It is characterized by hyperprosody and hypoelemental emotionality, and is identical to R1 with the exception that these children show flattened affect. Both of these syndromes are thought to be caused by a disturbance of mechanisms located in the right parietal lobe.
Unfortunately, no data or clinical samples are provided with these classifications. The rationale for assignment of these particular categories and their localizations are also not provided. Weinberg and McLean (1986) localized all of these difficulties to mechanisms solely in the right parietal lobe. This localization is inconsistent with other studies cited earlier that show that neurocognitive processes associated with arithmetic skills are probable bilateral, with input from subcortical as well as cortical mechanisms. Moreover, Ross (1981) demonstrated that prosody not only is a right hemisphere activity but also differs in type depending on anterior/posterior right hemisphere: Anterior regions contribute to the expression of prosody and posterior regions contribute to the recognition of prosody. In addition, circumscribed portions of the temporal and frontal lobe are most likely involved. Therefore, it would appear that Weinberg and McLean (1986) have grossly oversimplified both the syndromes and the locus of the observed deficits. It appears that these skills are most likely influenced not only by bilateral processes but also differentially by anterior/posterior portions of the right hemisphere (Kupfermann, 1985).
It would appear that these different methods of classification of nonverbal learning disabilities are similar in more ways than they are dissimilar. Most of the schemes suggest difficulties in arithmetic rather than reading. In addition, most of the classifications indicate gross-motor and visual-motor delay. On tests of language skills, these children show difficulties in abstract thinking and concept formation, although superficial language is intact. Social skills are also notes to be deficient in all of the methods of classification.
Aspergers Syndrome and PDD seem to be more severe and handicapping types of disorder. They may well represent the more involved end of a continuum of psychopathology. All of these classification schemes should be viewed as exploratory at this point. Replication and refinement of these diagnostic schemas is needed, particularly with regard to studies that link documented deviations in brain morphology (due to trauma or neuodevelopmental processes) to disordered neurobehavioral processes.
As was stated in the beginning of this article, caution must be exercised in extrapolating from research with adults to neuodevelopmental disorders found in children. As suggested by Denckla (1973) , however, it is, nevertheless, instructional to compare behavioral observations of right-brain-damage patients with children purported to have right hemisphere dysfunction.
Several issues for discussion emerge from this review. For example, the role of the thalamus as a "gate-keeping" mechanism in learning and attention must be further explored. In a separate review, Hynd and Semrud-Clikeman (1989a) outlined how thalamic dysfunction may impede the allocation of hemispheric attentional resources possibly implicated in developmental dyslexia. It would not be unreasonable to suspect that deficits in sensory processing, particularly in right hemispheric learning disabilities, may involve similar subcortical structures, perhaps, in this case, those thalamic bodies related to visual-spatial input and the expression of emotion (Kelly, 1985).
Many of the children in these studies were found to be somewhat lethargic and apathetic. One might well ask if these children are underaroused. In her initial study, however, Voeller (1986) found several children with ADD without hyperactivity, and all of her subjects in the second study had ADHD (Voeller & Heilman, 1988). Because some evidence suggests that learning disabilities are more frequently found in children with ADD without hyperactivity (Hynd et al., in press; Lehey et al., 1987), one must wonder what the incidence of subtypes of ADD might be in subclassifications of children with right hemispheric learning disabilities. Because research articulating the nature of children with or without such codiagnoses may lead to a better understanding of etiology and direct attention toward more productive models of brain-behavior relations in these children, the incidence of codiagnoses (e.g., learning disabilities and ADD with or without hyperactivity) is an important issue (Wyngaarden, 1987).
Are these children a different subtype from Strang and Rourkes (1985) population? From the information reported in these studies, it is difficult at this time to know if these are separate disorders. Furthermore, it cannot readily be seen from the available research whether it is possible to have nonverbal social learning difficulties without deficits in attention and arithmetic, visual-spatial, or motor delays. Research in all of these areas would shed light on the existence and interactive nature of these difficulties. Similarly, evidence from children with documented right hemispheric damage is needed to identify behavioral correlates of this disorder in a developmental context.
It is of interest that learning problems in arithmetic occur in many of the reports on children with developmental learning disabilities presumably of right hemispheric origin (e.g., Badian & Ghubilikian, 1983; Johnson & Mykelebust, 1971; Strang & Rourke , 1983; Weintraub & Mesulam, 1983). Considering the very complex and interactive cognitive and perceptual processes involved in number processing and calculation (McCloskey, Caramazza & Basili, 1985), it may well be that only very specific deficits in arithmetic are manifested in children with learning disabilities of possible right hemispheric origin.
Hecaen, Angelergues, and Houllier (1961) provided an early conceptualization for the disorders of calculation that included those referred to as (a) alectic acalculia, (b) spatial acalculia, and (c) anarithmetria. This classification schema was derived from neuropathological findings in patients with left and right hemispheric damage. Consistent with Strang and Rourkes (1983) findings that their NPOOD children had deficits o the WRAT arithmetic test involving spatial orientation, misreading of visual detail, procedural errors, and a failure to shift operations, Hecaen et al., (1961) attributed similar errors to those patients with documented right hemisphere damage.
Additional, and more recent, support for this notion is provided by Daltman, Hartje, Bussing, and Strum (1982), who examined disorders of calculation of Brocas and Wernickes aphasics, right-hemispheric-damaged patients and controls. Brocas aphasics were found to perform the poorest on most tasks, possibly because of the disrupted verbal-production system (McClosky et al., 1985) that characterizes Brocas aphasics with calculation disturbance (Benson & Denckla, 19869). However, both the Wernickes aphasics and the right-brain-damaged patients did poorest on the tasks involving visuospatial processes. This dinging is not only consistent with the positron emission tomography finding that the homological right hemisphere for Wernickes region is specialized for visual pattern processing (Kushner et al., 1988). More detailed studies of the very specific errors made by these right-hemispheric learning disabled children, using the conceptual and methodological approaches provided by Daltman et al (1982) and McClosky et al., (1985), may provide additional insight. Of great interest is the possibility that the deficits evidenced by these children are related more to visuoperceptual processing deficits, as one might suspect in retrorolandic, inferior parietal lesions in the right hemisphere, or are more reflective of the lexical and syntactic-processing components that may be associated with left perisylvian lesions. Certainly, one can conclude that the arithmetic deficits reported in these children have not been carefully examined in this context.
It should also be noted that some have postulated a relation between some forms of depression and right hemisphere involvement (Weintraub & Mesulam, 1983). Some patients have responded favorably to antidepressants. A related area, and one of great concern, is the overlap of some psychiatric conditions with right hemispheric learning disabilities. Weintraub and Mesulam (1983) found a pronounced overlap with depression and "diagnosed" schizophrenia. Clinicians may need to be suspicious of diagnoses in which the patient has several of the signs identified as possible indicative of right hemispheric involvement. Treatment goals may differ radically depending on a psychiatric or social-educational diagnosis.
One might also ask, is there a continuum of this disorder with PDD and schizoid personality disorder-Aspergers Syndrome at the severe end of this continuum? Or are these nosological classifications separate neuropsychiatric entities? One cannot make a definitive statement on the basis of existing research. Although children with Aspergers Syndrome of PDD may be conceptually similar to children with nonverbal learning disabilities, these syndromes appear to be distinct from the syndromes outlined by Rourke, Denckla, and Voeller in terms of severity, possible genetic etiology, incidence, and clinical manifestation.
An additional issue that has been very poorly addressed is the possible interaction of early deviations in neurological development, gender, and right hemispheric difficulties. Although Rourke (1982) found an increased incidence of nonverbal learning disabilities in females, Aspergers Syndrome shows the classic gender ratio for learning disabilities in favor of males. The other classifications do not directly address this issue. It would appear that gender is clinically and theoretically relevant, particularly given the differential effects of brain dysfunction between the sexes (Harris, 1978). It would be important, for example, to document carefully any prenatal effects on the emergence of various symptomatology associated with right-hemisphere learning disabilities, particularly as they relate to gender. On the basis or work by Witelsen (1976b) and others (e.g., Kimura, 1969), one might expect that the impact of trauma perinatally may produce different clinical manifestations in males and females.
Finally, successful treatment depends on accurate and reliable diagnosis, underscoring the need for a comprehensive nosology in which various right hemispheric learning disability syndromes are delineated. It will be important, however, not to develop a taxonomy of symptoms while neglecting the urgent question as to what assessments and interventions are most appropriate with regard to these specific developmental learning disabilities.