ABSTRACT

Assessing the cognitive skills of young children with physical disabilities can be difficult since most standardized assessment batteries for children under three years of age require specific motor responses to evaluate cognitive skills and are not flexible in administration procedures to accommodate for physical limitations. Further, many existing assessment batteries either do not provide scores for separate cognitive domains, or do not contain enough items to adequately assess children across the entire age range from 18 to 36 months. In order to assess cognitive skills across several domains in children with physical disabilities, an assessment battery with flexible administration materials and procedures was developed. A preliminary battery of 83 items with 148 scorable responses across five cognitive domains (object permanence, cause/effect, problem-solving, spatial relations and symbolic play) was developed. The battery was administered to 26 children with severe physical disabilities between the ages of 20 and 36 months. Rasch analysis was applied to the data to eliminate non-discriminatory, redundant and/or misfit items. This resulted in a final battery with 35 items, all with high internal consistency within their respective cognitive domains. Inter-rater reliability, intra-rater reliability and test-retest reliability were all high. Clinical applications of the battery are discussed.

INTRODUCTION

Educators and clinicians who work with young children have long recognized the usefulness of cognitive assessment instruments in identifying a child’s developmental strengths and weaknesses, developing educational plans and objectives, establishing early intervention therapies, and measuring progress in cognitive domains. In addition, rehabilitation professionals who work with very young children with physical limitations need to be able to assess cognitive skills for a variety of rehabilitation objectives. There are a number of tools available to assess cognitive skills in infants and young children (e.g., Bayley Scales of Infant Development - II,1 Battelle Developmental Inventory,2 Developmental Activities Screening Inventory - II,3 Carolina Curriculum for Handicapped Infants,4 Uzgiris and Hunt Infant Psychological Developmental Scale - Dunst Revision5). However, difficulties can sometimes exist for the clinician who wishes to assess individual cognitive skill domains in young children with physical limitations. For example, traditional norm-referenced evaluation tools often indicate only a global cognitive developmental age (i.e., scores reflect an average of all cognitive skill areas). One such battery, the Bayley Scales of Infant Development - II1, combines language and other cognitive skills such as object permanence, cause/effect and spatial relations into a single global score, and scores for individual cognitive skills are not readily attainable. Summary scores can sometimes be misleading because children with the same global developmental score may have very different scores across the contributing developmental domains, resulting in widely varied abilities across different learning situations.

Further, the evaluation of cognitive skills in a young child with physical disabilities can be challenging since many assessment batteries infer cognitive skills from sensory and motoric responses to tasks presented. In fact, most standardized assessment batteries for children under three years of age require specific motor responses to evaluate cognitive skills and are not flexible in administration procedures, regardless of the child's physical abilities or interest. When a young child with a physical disability is unable to demonstrate an age-appropriate task response, it is often difficult to determine whether the child has not developed these cognitive skills or has developed the skills but is unable to perform the task because of physical limitations.

For clinicians and researchers who use this information to make clinical decisions or observe clinical findings, this dilemma can pose a difficult challenge. For example, the current researchers proposed that certain cognitive skills may influence a young child's ability to operate a powered wheelchair. Currently however, because of the difficulty in obtaining an objective measure of these cognitive skills in young children with disabilities, the decision to place a child in powered mobility is typically based on clinicians' experience and/or trial and error. Better knowledge of a child's level of development in these skills could lead clinicians to a more timely placement of young children in powered mobility devices and to improved intervention plans.

After examining existing assessment batteries, it was found that no single instrument could adequately assess cognitive skills in children with physical disabilities over the entire age range between birth and age three, that would provide scores for separate cognitive domains, and would be flexible in materials and administration procedures to accommodate physical limitations. In order to evaluate cognitive skills in these children, there was a need for an assessment instrument that could assess these skills without penalizing children for motoric limitations.

Thus, the purposes of the current research were as follows: 1) to develop a cognitive assessment tool with flexible task materials and administration procedures to evaluate cognitive skills across various domains in young children with physical disabilities, and 2) to determine which of these cognitive skills can predict powered wheelchair driving ability. A program to teach and evaluate powered mobility skills has been described previously,6 and the results of the comparison between cognitive skills and powered mobility skills are described elsewhere.7 The current article describes the development of the cognitive assessment battery.

METHODS

Selection of Cognitive Domains

'Piagetian-based' assessment instruments8-10 often assess critical concepts assumed to be the foundations for later learning. A Piagetian-based framework of sequential stage development within distinctly different skill domains allows the clinician to determine a child's developmental strengths and weaknesses across various cognitive domains, such as object permanence, problem solving, receptive language and seriation. Despite criticisms of Piaget’s cognitive developmental theory (e.g., neglect of the role of social or cultural factors,11,12 disregard for postadolescent development,13 underestimate of the competence of children14 ), this is nevertheless a well researched and validated theory of infant/toddler cognitive development.15-17

Cognitive development of children from birth to 36 months of age is encompassed by Piaget’s Sensorimotor and early Preoperational stages. Development in the Sensorimotor stage (birth to 24 months) involves integrating motor and sensory movements that eventually lead to representational or symbolic abilities. During the early phase of the Preoperational Stage (2 to 4 years), children learn to use internal cognitive images to organize thoughts logically and become active problem solvers. At each stage of development, a typical child demonstrates specific behavioral attributes across cognitive skill domains (e.g., operational causality, means-ends relations, object permanence, seriation). Normally developing children progress through each domain in a hierarchical manner but may achieve different stages of development simultaneously across domains.

A Piagetian based developmental approach has been found to be appropriate for children with a diverse range of disabilities (e.g., mild to profound developmental delays, cerebral palsy, limb deficiencies, and visual or hearing impairments). For example, in a study of 21 children ages 17-44 months who were physically affected by thalidomide,18 all children 18 months and older had successfully achieved Piaget’s highest level object permanence; only the youngest two who were 17 months had not yet reached this level. In reviews of research on this topic,5,19 it was found that children with diverse disabilities typically acquire behaviors in the same sequential stage development as able-bodied children, but may differ in the rate at which skills are attained and in the final level of mastery. Other research has also found that the developmental sequence for symbolic play was similar among children with Down syndrome and normally developing children.20

Twelve Piagetian-based developmental skill domains representing the sensory-motor and preconceptual stages were considered for inclusion in the current assessment battery (see Table 1). It was believed that some of these areas would be less likely to relate to wheelchair operation skills (e.g., numbers, vocal imitation). Since it was not practical to evaluate children with disabilities across all twelve domains due to the length of time to conduct such an evaluation and the limited patience and attention span of children in this age range, it was determined that a subset of domains would be selected for inclusion. To establish content validity and determine the most appropriate cognitive domains to be included, ten experts in the field of early child development, occupational therapy, speech pathology and powered mobility were asked to rate the twelve developmental skill domains in terms of their presumed relevance to powered mobility skills. Based on these ratings, a subset of five domains (cause/effect, object permanence, problem-solving, spatial relations, symbolic play) was selected. A brief description of these cognitive skill domains is presented in Table 2.

Item Selection

To identify items in these five domains for inclusion in the battery, existing cognitive assessment batteries were reviewed. It was determined that no single existing assessment battery contained an adequate number of items across the age span of 18 to 36 months (range where children might be able to operate a powered wheelchair) and could be modified for use with children with physical disabilities. Thus, items from various tests were used to create a continuum in each domain that would assess this entire age range. The preliminary assessment battery incorporated 73 tasks from several standardized developmental batteries5,10,21-23 along the five cognitive domains identified. This battery was then pilot tested with three children, and necessary revisions in item content were made. For example, items in the spatial relations domain that required physical locomotion (e.g., ‘Child makes a simple detour to obtain a desired object’) were replaced by items at the same age range that evaluated spatial relations without emphasizing physical movement, and items were added to better assess the upper end of the symbolic play scale (i.e., play with mini toys). The final battery included 83 tasks evaluating cause/effect, object permanence, problem solving, spatial relations and symbolic play. Items in the cause/effect scale assess development through 21 months and items in the object permanence scale ‘ceiling’ at 23 months, the ages at which these skills are believed to be fully developed and integrated into more complex domains. The other three scales include items assessing development through 36+ months. Several tasks had multiple scoring levels depending on the nature of the child’s response, for a total of 148 scorable responses across the entire battery. Examples of items are included in Table 3.

Modification of Item Administration

Tests with standard procedures for administration and response often require modification for use by people with functional impairments.24 Since the battery was being developed for use with children with functional limitations, it was essential that the response requirements not penalize children with motor disabilities. Therefore, the assessment battery materials were modifiable where appropriate to accommodate various physical limitations. Modifications of materials, positioning or environment were made to allow children to lift, move, or manipulate the objects, however, care was taken to insure that the basic task premise always remained the same. For example, one task assessing spatial relations required the child to stack large plastic rings on a long plastic dowel. During assessment of children who were too weak to lift the plastic rings or who had limited range of motion, small, lightweight styrofoam rings and a shorter dowel were substituted. During a problem solving task that required children to dangle a long chain to insert it into a small opening, shorter chains were used for children who were unable to lift their arms above chest height. For a child with almost no upper extremity movement, the three-screen object permanence task was conducted with one-inch squares of construction paper, vertically positioned on a small shelf instead of napkins placed on the table. If a child physically attempted the critical component of a task and demonstrated the intent of the task, the test administrator was permitted to complete the movement to reward the child and limit frustration. For example, during a task assessing problem solving and tool use, if a child picked up a stick and attempted (but was physically unable) to obtain an out-of-reach object using the stick, the test administrator used the stick to pull the object to the child.

Modifications were also made in the children’s physical positioning during assessment to maximize upper extremity range of motion. One child was evaluated in a gravity-reducing position, lying on his side, while another child was assessed in a standing position, leaning on the assessment table.

PARTICIPANTS

Participants included 26 children (20 male, 6 female) between the ages of 20 months and 36 months, with various physical disabilities. The average age at the time of cognitive assessment was 28.3 months. Children were recruited from clinics at a large rehabilitation medical center in Los Angeles and from the California Children Services, an agency providing school-based physical and occupational therapy services. Table 4 presents the participants’ diagnoses. Only children with diagnoses not typically associated with severe cognitive and/or sensory-motor impairments were included (e.g., muscle disease, spinal cord injury, arthrogryposis). This served to minimize the potential influence of sensorimotor integration problems which may occur with conditions with central nervous system involvement (e.g., cerebral palsy, acquired brain injury). Diagnoses classified as 'other' include burn/amputee, polio, spina bifida, osteogenesis imperfecta and femur hyplasis syndrome. Children’s hand function varied from completely functional to a minimal or non-functional grasp. For example, children with muscle disease had a weak grasp and limited active reach. Children with arthrogryposis who had no functional grasp compensated by moving their arms and hands via trunk movement or picked up objects with their teeth. All children were positioned to maximize the functional use of their arms (e.g., sitting, standing, lying down) prior to administration of the assessment battery. All participants were unable to ambulate at the time of participation in the study.