8. Neuropsychological Assessment Approaches and Diagnostic Procedures

The Halstead-Reitan neuropsychological procedures are the most commonly used batteries available in this country (Nussbaum & Bigler, 1997), and the most well researched neuropsychological battery available. Halstead originally developed a series of tests to measure frontal lobe dysfunction in adults, and Reitan later added new tests and recommended the battery for various types of brain damage in children (Reitan & Wolfson, 2004). The Halstead-Reitan batteries contain measures necessary for understanding the brain-behavior relationship in children and adolescents.

Reitan and Wolfson (1985) indicate that attempts to develop a set of assessment measures resulted in a conceptual model of brain function that is incorporated in the Halstead-Reitan Battery. The battery consists of six categories representing the behavioral correlates of brain function: (1) input measures; (2) tests of concentration, attention, and memory functions; (3) tests of verbal language abilities; (4) tests of visual-spatial, sequential, and manipulatory functions; (5) tests of abstraction, reasoning, concept formation, and logical analysis, and (6) output measures (Reitan & Wolfson, 1985, p. 4).

Reitan and Wolfson (1985) further argue that a neuropsychological battery must have three components: (1) items that measure the full range of psychological functions of the brain; (2) strategies that allow for interpretation of individual brain functions, and (3) valid procedures demonstrated through empirical and clinical evaluation and applications. The neuropsychological batteries for children and adolescents were developed with these components in mind.

Reitan designed two batteries for children, the Halstead-Reitan Neuropsychological Battery for Older Children (HRNB-OC; 9–14 years) and the Reitan-Indiana Test Battery (RINTB; 5–8 years); see Table 8.1. Adolescents 15 years and older are evaluated using the Halstead-Reitan Battery for Adults. Reitan and Wolfson (2004a, 2004b) developed a screening battery for both the HRNB-OC and the RINTB. See Reitan and Wolfson (1992a, 1992b, 2004a, 2004b) for an in-depth description of the HRNB-OC, the RINTB, and the two screening batteries. Table 8.2 lists the various subtests and the abilities associated with each of these measures.

One of the major shortcomings of the CHRNB has been inadequate norms, and insufficient reliability and validity information (Davis, Johnson, & D’Amato, 2005; Strauss, Sherman, & Spreen, 2006). Over the years Reitan developed and expanded his approach for analyzing the CHRNB (9–14 years) and the RINB (5–8 years). Interpretation typically focuses on a Multiple Inferential Approach, including an investigation of Level of Performance, Pathognomonic Signs, Patterns of Performance, and Right-Left Comparisons. Reitan (1986a, 1987) also developed the Neuropsychological Deficit Scale (NDS), a scoring and interpretation model for these batteries, which incorporates multiple factors. The Functional Organization Approach, proposed by Fletcher and Taylor (1984), is less inferential and places neuropsychological measures within a developmental and contextual framework. Each of these approaches will be briefly described and critiqued.

Reitan (1969) and Selz and Reitan (1979b) developed an interpretive system using four levels of inference: Level of Performance, Pathognomonic-Sign, Differential/Pattern of Scores, and Right-Left Differences.

The Neurological Deficit Scale (NDS) incorporates a method for determining the child's Level of Performance, Right-Left Differences, Dysphasia and Related Deficits, and cutoff scores for differentiating brain-damaged from normal youngsters for each battery. The NDS also provides a total score for measuring the overall adequacy of neuropsychological functioning in children. Raw scores are weighted as Perfectly Normal (score = 0), Normal (score = I), Mildly Impaired (score = 2), or Significantly Impaired (score = 3) on the basis of normative comparisons. When using the NDS approach, the examiner takes the following steps: (1) converts raw scores on each test to corresponding weights (0, 1,2, or 3) based on normative tables provided by Reitan (1986a, 1987); (2) calculates right-left difference scores by dividing the score of the nondominant hand by that of the dominant hand, subtracting from 1.00, and converting to weighted scores; (3) makes clinical judgments following Reitan's (1984) guidelines for scoring the Aphasia Screening Test and assigns NDS scores; (4) totals weighted scores across each factor, Level of Performance, Right-Left Differences, and Aphasia Test, and (5) totals the weighted scores on 45 variables for older children and 52 variables for younger children to obtain a Total NDS Score. Reitan provides cutoff scores for brain-damaged versus normal children on the Total NDS score and for each of the factor scores.

Separate tables are available to analyze the neuropsychological test results of older children (Reitan, 1986a) and younger children (Reitan, 1987). Although the NDS approach seems to be an extension of an earlier standardized scoring procedure (“Rules for Neurological Diagnosis;” Selz and Reitan (1979a)), the normative group used to develop the weighted scores reported in the tables is not clearly described in recent test manuals available for the child batteries.

Several other methodologies use and interpret the Halstead-Reitan battery including the normative analysis, the biobehavioral, and the pragmatic approach. These approaches are briefly described.

The normative analysis differs from the “level of performance” approach (+2 Standard Deviations above the mean) by analyzing scores on a continuum rather than cut-offs scores; brain damage is either present or absent. Clinicians use the metanorms to determine a child’s performance within a normative framework (Findeis & Wright, 1995) rather than as dichotomous criteria for determining brain damage versus no brain damage (Nussbaum & Bigler, 1997; Nussbaum & Bunner, 2008).

The Biobehavioral Approach employs a broader, more comprehensive method for interpreting test data and was first proposed by Taylor and Fletcher (1990). This model assumes that neuropsychological functioning occurs with a context, with behavioral as well as neurocognitive, biological and genetic variables interacting and affecting how a disability is manifested (Nussbaum & Bunner, 2008). First, neuropsychological assessment includes information from four major sources: (1) presenting problems; (2) cognitive and psychosocial characteristics; (3) environmental, sociocultural and historical variables (e.g., family and school history), and (4) biological and genetic vulnerabilities (i.e., family history).

Second, the biobehavioral approach assesses other child and environmental factors that impact the child’s basic neurocognitive functions in order to determine the intensity of the disability. These factors may also diminish the disability. Third, it is likely that an uneven profile of performance is associated with disabilities and it is critical to understand the variability to fully appreciate the nature of the disability. Fourth, the clinician must determine the impact of the environment on the neurocognitive functioning of the child. Fifth, neurocognitive deficits influence and ultimately limit the behavioral competencies of the child. These deficits are moderated by family and environmental factors. Finally, neurocognitive deficits must be interpreted within a developmental framework, and are considered correlational and not causal.

The Pragmatic Approach offers a more flexible model (Barron, 2003). Barron suggests a more fluid, non-battery approach for neuropsychological assessment. He advocates a tailored assessment of the child’s strengths and weaknesses that can yield information for targeted interventions. Barron suggests other tests to assess a full spectrum of the child’s assets and deficits.

While there have been relatively few studies on the CHRNB and HINB over the past decade, two studies are noteworthy. First, Vanderslcie-Barr, Lynch, and McGaffrey (2008) found that the screening batteries for the older and younger child produced a high percentage of correct classifications for determining normal or impaired function using archival data from a neuropsychological clinic. The authors found that the screening battery, neuropsychological deficit scale score (SBNDS) for older children had an 85 percent accuracy rate (cut-off scores 16/17), while a cut-off of 22/23 was 100 percent accurate for younger children.

Bello, Allen, and Mayfield (2008) investigated the sensitivity of the Children’s Category Test level 2 (CCT-2) to determine brain dysfunction in children with attention–deficit hyperactivity disorder (ADHD) and children with structural brain damage. Although the Category test was found to be psychometrically sound, it did not differentiate children with brain damage from those with ADHD. The CCT-2 was not particularly sensitive to brain damage; further, both groups performed within the normal range. The authors conclude “we would recommend against the use of the CCT-2, including its factor and subtest scores, for clinical and research applications that aim to draw conclusions regarding the impact of brain injury on abstraction and problem solving abilities” (Bello et al., 2008, p. 338). They recommend using the longer version of the Category test within a comprehensive, larger battery.

Historically, studies utilizing the HRNB-OC and HINB have focused on the ability of these batteries either to distinguish between children with brain damage and those with learning disabilities, or to elucidate the profiles achieved by differing disorders (e.g., conduct disorder, psychiatric disorders). The longer Category test is the best discriminator for children with learning disabilities. Results of studies attempting to distinguish children with learning disabilities from those with brain damage and typically developing children suggest that children with LD have normal motor development with consistent weaknesses on the Category test (Shurtleff et al., 1988). Moreover, a relationship between reading and/or math difficulty and the Category test has been found (Shurtleff et al., 1988; Strom, Gray, Dean, & Fischer, 1987).

Intelligence scores show moderate correlations with the HRNB-OC (Shurtleff et al., 1988). In a review of studies that attempted to differentiate learning-disabled from brain-damaged children using the HRNB-OC, Hynd (1992) suggested that differential performance on intelligence tests may account for much of the ability of the HRNB-OC to discriminate between the two groups. On the other hand, Strom et al. (1987) found that the HRNB-OC provided unique data that were not redundant with data from the WISC-R. Because the issue of the overlap between the WISC-R and the HRNB-OC has not been resolved, it is important that the clinician recognize the overlap between the two measures and take intelligence into consideration when interpreting results.

In addition to the caution as to the influence of intelligence on performance on this battery, children with psychiatric disorders have also performed poorly on the HRNB-OC (Tramontana, Hooper, & Nardillo, 1988). The HRNB-OC's ability to distinguish children with psychiatric disorders from children with brain damage is not clear from existing research. This finding is consistent with the adult Reitan Battery, which also is not diagnostically specific for brain damage versus psychiatric disorder (Hynd & Semrud-Clikeman, 1992). The length and expense of the battery for general use with clients is another concern. The average amount of time to administer the battery ranges from six to 12 hours. Reitan (1986b) suggested that although reducing the length of the battery or developing a screening protocol would have value, the information necessary to answer referral questions makes the development of a screening protocol problematic. Alth-ough Reitan demonstrated a remarkable hit rate for his ability to determine brain damage (Selz & Reitan, 1979a, 1979b), there has not been sufficient documentation of the battery's ability to localize dysfunction or predict recovery from brain injury (Hynd, 1992). Furthermore, the HRNB-OC does not have adequate norms, and lacks detailed information on the validity and reliability of the battery.

Finally, the HRNB-OC requires intensive training for administration and interpretation of results, which also can be problematic for its use in general clinical or school environments. It may be more appropriate for general clinicians to use other measures to screen for possible neuropsychological involvement and to refer clients to a trained neuropsychologist for a full evaluation if areas of concern are identified. “In terms of future directions for the HRNB-OC and the RINB, if the HRNB-OC is to remain relevant and employed as a battery, then it should be updated with broader and more indepth measures, particularly in areas such as memory and attention” (Nussbaum & Bunner, 2008, p. 264).

Luria (1980) described brain activity in terms of functional that incorporated elements of localization and equipotential theories. Localization theorists argued that specific brain regions were responsible for discrete brain functions with visual functions in the occipital lobe, auditory functions in the temporal lobe, and so on (Kolb & Whishaw, 2003). Equipotential theorists pointed out that complex human behaviors are controlled by functional CNS regions in such a way that when one portion is damaged, another adjacent or analogous region can assume its function (Kolb & Whishaw, 2003).

Luria's theory was different from other hypotheses at the time because he made four major assumptions:

Research supports aspects of each of these theories in various degrees because functions appear localized to some extent; however, a particular behavior may be impaired because of damage to a number of different brain areas. Kolb and Whishaw (2003) suggest that the important questions center on how damage to a particular site can affect specific behaviors or performance. Luria's functional systems approach conceptualizes brain function as follows. Luria (1980) discussed three functional units as: (1) the arousal unit; (2) the sensory receptive and integrative unit, and (3) the planning, organizational unit (see Table 8.4). The nature of each functional unit is briefly described.

In Luria's theory (1980) the arousal system is the first unit and comprises the reticular activating system (RAS), the midbrain, the medulla, the thalamus, and the hypothalamus. Visual, auditory, and tactile stimulation comes through this unit to higher cortical regions. The structures work together in concert in Unit I to regulate energy level and to maintain cortical tone. This unit raises or lowers cortical arousal depending on internal needs. When cortical tone is too low, the brain loses its ability to discriminate between stimuli. Another function of this unit is to filter out irrelevant stimuli. The RAS prevents the cortex from being flooded by unimportant stimuli that could interfere with cortical functioning. If the RAS filters out too much stimulation, sensory deprivation and hallucinations may be present as the cortex attempts to generate its own activity to keep itself aroused. Severe injury to Unit I can result in marked deterioration of wakefulness, with loss of consciousness and possible death. Less severe injury can result in disorganization of memory, distractibility, attentional problems and insomnia. If Units II and III are functional, then in later development or in adulthood these units can assume the functions of Unit I and monitor hyperactive and/or impulsive behavior. Methylphenidate has also been found to activate Unit I and thereby decrease behaviors of impulsivity and poor attention.

Unit II is considered the sensory system and consists of the parietal, temporal, and occipital lobes; its major function is sensory reception and integration. Therefore, the areas of Unit II correspond to their sensory modality (temporal for auditory, parietal for sensory tactile, and occipital for vision). Unit II has been hypothesized to be guided by three functional laws: (1) hierarchical structures of cortical zones do not remain the same during ontogenesis; (2) hierarchical zones decrease in their specificity of function with development, and (3) progressive lateralization of function within hierarchical zones in-creases with development (Luria, 1980). This hierarchy is further divided into three zones: primary, secondary, and tertiary. The primary zones are responsible for sorting and recording sensory information. The secondary zones organize the sensory information and code it for later retrieval. The tertiary zones combine data from various sources in order to lay the basis for organized behavior.

Unit III is responsible for output and planning. It is located in the frontal lobes which are further demarcated into three hierarchical zones. The primary zone, in the motor strip of the frontal lobe, is concerned with simple motor output. The secondary zone, in the primary premotor regions, is involved in sequencing motor activity and speech production. The tertiary zone located in the orbitofrontal region of the frontal lobe (the prefrontal region) is the last region to myelinate and develop. Development continues until the third decade of life. The tertiary zone of Unit III is primarily involved with planning, organization, and evaluation of behavior, functions similar to the executive functions. Damage to this area has been linked to problems in delaying gratification, controlling impulses, learning from past mistakes in behavior, and focusing attention. In many cases damage to this zone can be difficult to distinguish from psychiatric and behavior problems. When dysfunction occurs in Unit I, later development of Unit III can compensate or modulate levels of arousal. Moreover, Unit III can activate other parts of the brain and has rich connections to all regions of the brain.

Luria’s conceptual framework is based on the theory that certain skills are acquired at different rates depending on the neurodevelopmental stage of the child (Golden, 1981). Further, specific problem solving strategies, behaviors, and skills are dependent on biochemical as well as physiological maturation, including myelination and the growth of cells, dendritic networks, and interconnecting neuronal pathways. Although physiological development is related to psychological maturation, this relationship can be altered by adverse environmental events. Table 8.5 outlines the five major developmental stages described by Golden (1981).

Injury during any one of these stages is thought to produce various deficits depending upon the site and severity of injury. Golden (1981) suggests that damage to the developing brain during Stage 1 is likely to produce deficits in arousal and that, when severe damage ensues, death or mental retardation may result. Damage after 12 months of age is less likely to produce attentional deficits, although physiological hyperactivity is associated with damage prior to 12 months. Paralysis, deafness, blindness, or tactile deficits may result from unilateral injury to the primary sensory areas during Stage 2 development. In some instances, sensory or motor functions may be transferred to the opposite hemisphere if damage occurs during this stage. Although damage after this developmental stage is likely to produce more serious deficits, there are still compensatory factors that play a role in recovery of function. Golden cautions, however, that bilateral damage is more serious, producing deafness, blindness, and/or paralysis, where compensation is less likely. During Stage 3 development, the two hemispheres begin to show differentiation of function in terms of verbal and nonverbal abilities (Golden, 1981). Unilateral damage is likely to result in loss of language functions if injury is sustained in the left hemisphere once verbal skills are present, at about the age of two years. Damage prior to age two may result in transfer of language to the right hemisphere, whereas damage after age two begins to mimic recovery of functions similar to what is seen in adults (Golden, 1981). However, Golden (1981) suggests that plasticity (i.e., transfer of function) is less likely when injuries are diffuse in nature, or in cases of mild injury. Thus, small injuries early in development can have more deleterious effects than larger injuries later in life. Recovery of function will be explored in more detail in later chapters.

Golden (1981) suggests that learning during the first five years of life is primarily unimodal in nature, with little cross-modal, integrative processing. Early reading during this stage is characterized by rote strategies involving memorization of individual letters, words, or letter sounds. The visual symbol is meaningful only in its relationship to spoken language. Cross-modal learning is possible during Stage 4 when tertiary association regions of the sensory cortices are developing. Injury to these association regions can result in significant learning impairments, such as mental or cognitive deficits or learning disabilities. The type of deficit depends on the location and severity of the injury, and even small insults can affect the integration of one or more sensory modalities (Golden, 1981). Golden (1981) suggests that injuries to tertiary regions are not always evident until Stage 4 development. Injury in one stage may not produce observable deficits until a later stage because the brain regions subserving specific psychological and behavioral functions are not mature. For example, a child sustaining injury to tertiary regions at the age of two may appear normal at age three, but may show serious learning deficits at age 10 (Golden, 1981). Golden further indicates that predicting future deficits is complicated by these neurodevelopmental factors, and that neuropsychologists must consider these issues when injury is sustained early in life. Finally, according to Golden (1981), Stage 5 involves the development of the prefrontal regions of the brain, which begins during adolescence. According to this neurodevelopmental theory, deficits resulting from injury to frontal regions may not begin to emerge until 12–15 years of age or later. Others have argued that frontal lobe development may occur at earlier stages than suggested by Golden (1981). For example, Becker, Isaac, and Hynd (1987) and Passler, Isaac, and Hynd (1985) describe a progression of frontal lobe development beginning at age six. In these studies it was found that some tasks thought to be mediated by the frontal lobes begin at six years of age (e.g., flexibility during verbal conflict tasks), continue to emerge at age eight (e.g., inhibition of motor responses), and still are incomplete at age 12 (e.g., verbal proactive inhibition). Neurodevelopmental stages are of primary importance in child neuropsychology, and further research is needed to more clearly map these stages of brain development. Although the question of frontal lobe development will be of continued interest to researchers and clinicians in the next decade, the relationship between brain development and psychological and behavioral function has a strong empirical base.

Although the Luria-Nebraska Children’s Battery Revised (LNCB-R) was designed to assess brain functioning based on Luria’s model, the test has not been adequately researched. Attempts to standardize and validate the LNCB-R have been slow and “the lack of current research findings present a major concern” (Leark, 2003, p. 155). “Several contemporary neuropsychological assessment tools are available to assess skills similar to those tapped by the Luria-Nebraska domains and were designed solely for children” (Hale & Fiorello, 2004, p. 137). There are several newer batteries that were developed using Luria’s conceptualizations of brain functions. The NEPSY and the Cognitive Assessment System (CAS) will be briefly reviewed next.

The Attention and Executive Function Scale measures inhibition of learned and automatic responses, monitoring and self-regulation, vigilance, selective and sustained attention, nonverbal problem solving, planning and organizing complex responses, and figural fluency.

The Language Scale measures major phonological processing, repetition of nonsense words, identification of body parts, verbal semantic fluency, rhythmic oral sequences, and the comprehension of oral instructions.

The Memory and Learning Scale measures immediate memory for sentences; narrative memory and free recall, cued recall and recognition recall; recall with interference, and immediate and delayed memory for designs, faces and lists.

The Sensorimotor domain measures hand movements, repetitive finger movements, and use of a pencil with speed and precision.

The Social Perception domain measures facial affect recognition, comprehension of others’ perspectives, and intentions and beliefs.

The Visuospatial domain measures line orientation, copying geometric figures, three-dimensional designs, mental rotation of objects, whole-part relations, and schematic map reading.

Korkman (1999) provides a nice review of validity studies with the NESPY. Specifically the NEPY appears to have validity for differentiating subtypes of learning difficulties, discriminating ADHD from other learning disabilities, and identifying deficits in infants with prenatal alcohol exposure. Further, Korkman reports that children with brain damage had deficits on the NEPSY, but did not have lateralizing effects. This is consistent with findings that children show functional reorganization of the brain with early brain damage, and that children tend to have more diffuse rather than focal brain dysfunction.

The NEPSY can also be used to identify neurodevelopmental deficits in a number of clinical populations. Mikkola et al. (2005) found that extremely low birth weight (ELBW) infants had decreased performance on measures of the NEPSY (i.e., Attention, Language, Sensorimotor, Visuospatial, and Verbal Memory) compared to controls. Shum, Neulinger, O’Callaghan, and Mohay, (2008) also reported poor performance on verbal memory and attention on the NESPY and low scores on the Trail Making B test. These problems were associated with parent and teacher ratings of attentional difficulties. Young children at risk for ADHD were also found to have deficits on NEPSY measures of executive tasks [e.g., Attention, Fluency (Perner, Kain, & Barchfeld, 2002)

The Planning scale measures mental processes for determining, selecting, applying, and evaluating problems. Performance on this scale is dependent on retrieval of knowledge and impulse control, and is reflective of prefrontal lobe functions.

The Attention scale measures selective focus on stimuli, while inhibiting other responses. Selective attention, focused cognitive activity, resistance to distractions, orienting to tasks, and vigilance reflect reticular activating system functions.

The Simultaneous scale measures the ability to perceive and integrate parts into a whole or group. The parietooccipital regions are primarily involved with this mental process.

The Successive scale measures the ability to integrate stimuli in a sequential, serial order.

The CAS appears to have been well conceived and researched. While initial factor analyses support the four-factor solution of the PASS model (Naglieri & Das, 1999), others have not found the same results (see Strauss et al., 2006 for a review).Research has shown that the CAS has utility for measuring cognitive processing deficits in clinical groups including children with reading difficulties (Joseph, McCachran, & Naglieri, 2003), children with written expressive disorders (Johnson, Bardos, & Tayebi, 2003), children with ADHD (Naglieri & Das, 2005), and children with moderate to severe TBI (Gutentag, Naglieri, & Yeates, 1998).The following results have been reported: (1) Children with ADHD had lower scores on the Planning scale of the CAS compared to children without ADHD; (2) the Planning scale best discriminates children with and without written expressive disorders, and (3) children with TBI show low scores on the Planning and Attentions scales.

The CAS is correlated with achievement measures, other intelligence and neuropsychological tests, including the NEPSY (Naglieri, DeLauder, Goldstein, & Schwebach, 2005; Naglieri & Bornstein, 2003). Specifically, low scores on the Successive Processing scale is related to poor reading scores, while low Planning scores are related to decreased performance on calculation, dictation and writing scores on the WJ-R (Naglieri & Bornstein, 2003)

The Spreen-Benton Aphasia Tests or the Neurosensory Center Comprehensive Examination for Aphasia (NCCEA) comprises 20 subtests measuring language functions, and four subtests measuring visual and tactile skills (Spreen & Benton, 1969). Spreen and Benton (1977) describe the revised NCCEA tests in detail and list the following tests for the language domain: Visual Naming, Description of Use, Tactile Naming (right hand), Tactile Naming (left hand), Sentence Repetition, Repetition of Digits, Reversal of Digits, Word Fluency, Sentence Construction, Identification by Name, Identification by Sentence (Token Test), Oral Reading (Names), Oral Reading (Sentences), Reading Names for Meaning, Reading Sentences for Meaning, Visual-Graphic Naming, Writing Names, Writing to Dictation, Writing from Copy, and Articulation. The NCCEA items cover a range of language functions and were selected to be sensitive to aphasic symptoms, but not to mild intellectual impairment.

This test requires the child to match faces with three different conditions: identical view orientation, matching front view with three-quarter views, and front view with lighting differences. The first six items require a match of only one pose with six selections. The final 16 items require the child to match three selections to the sample. This test is sensitive to language comprehension difficulties as well as to visual-spatial processing problems.

The Cancellation Task requires the child to select a target visually and repetitively, as quickly as possible. One task that may be used (D2 task) requires the child to cross out all the Ds with two marks above them. There are 15 lines of Ds, and the child is asked to cross out the Ds in each line for 20 seconds and then to switch to the next line. Lower scores may indicate problems in visual scanning, inhibition problems, and inattention.

Clients with difficulties in sequencing and inattention do poorly on this task compared to those individuals without these problems (Spreen & Strauss, 1991; Sohlberg & Mateer, 1989). The Symbol Search task from the WISC-III and the Visual Matching and Cross-Out Tasks from the Woodcock-Johnson are additional tasks that require quick visual scanning and attention to task, and which may be utilized if this area is of concern. There are several more measures which may be utilized to more fully evaluate various aspects of functioning. The astute clinician will seek out these measures in order to determine their appropriateness for various children or adolescents.

In summary, the Boston Process Approach begins with a sampling of behaviors and then fine tunes the evaluation depending on the initial findings. The strength of the Boston Process Approach is also its weakness; namely, the ability to determine the client's areas of strength and deficit through qualitative data. Qualitative information has improved the prediction of brain damage found on radiological evidence to more than 90 percent (Milberg et al., 1986). Heaton, Grant, Anthony, and Lehman (1981) also found that qualitative data gathered by clinicians using the Reitan battery also showed significant improvement over quantitative scales.

The weakness of the Boston Process Approach lies within the examiner. To avail him- or herself of this approach, the clinician not only must have a wide array of measures in his or her knowledge base, but also sufficient experience in which to apply behavioral observations to brain-behavior relationships. It is also imperative that the clinician have a good database of a “normal” child's performance at various ages.

Although there is a beginning database for the use of the Boston Process Approach with adults (Lezak, 1994; Milberg et al., 1986), data on its efficacy with children are limited. The astute clinician will recognize that best practice will always dictate careful observation of how the child solves the tasks presented to him or her. Although the Boston Approach may be intuitively appealing, further research is needed to determine the benefit of this approach with children.

D. Delis, Kaplan, and Kramer (2001) designed the Delis-Kaplan Executive Function System (D-KEFS) for individuals between the ages of eight to 89 years. The D-KEFS is comprised of nine individually administered tests, and the “battery offers one of the first psychometrically sound, nationally normed set of tests designed exclusively for the assessment of verbal and nonverbal executive functions in children, adolescents and adults” (Shunk, Davis, & Dean, 2006, p. 275). While the D-FEFS contains new tests, other tests have been modified. The battery has an empirical basis and reflects the principles and procedures from the extensive body of literature on executive functions (EF).

The test is organized into four domains: Concept Formation, Flexibility, Fluency and Productivity, and Planning. The D-KEFS contains improved versions of older tests previously described, including the following (Shunk et al., 2006): (1) D-KEFS Trail Making Test measures visual scanning and motor speed; (2) D-KEFS Word Context measures abstract thinking and deductive reasoning; (3) D-KEFS Sorting Tests measures verbal and nonverbal concept formation; (4) D-KEFS Twenty Questions measures object naming and recognition, visual attention and perception; (5) D-KEFS Tower Test measures planning and problem solving, learning, inhibition of impulsivity, and maintenance of instructional sets; (6) D-KEFS Color-Word Interference test is a version of the Stroop test, and measures inhibition and attention; (7) D-KEFS Verbal Fluency Test measures verbal fluency; (8) D-KEFS Design Fluency Test measures verbal fluency in the spatial domain, and(9) D-KEFS Proverb Test measures verbal abstraction.

The technical adequacy of the D-FEFS is an improvement over earlier EF measures particularly with newer, expanded norms and modified measures (Shunk et al., 2006). Initial research with the D-FEFS shows promise with special populations including ADHD (Wodka et al., 2008), and individuals with autism and Asperger’s disorders. However, additional reliability and validity research for children is needed (Barron, 2003).