The purpose of this study was to examine the

Body Part Identification in 1- to 4-Year-Old Children Kathryn MacWhinney, Sharon A. Cermak, Anne Fisher Key Words: body image. child development. pediatrics The purpose ojthis study was to examine the sequence in which body parts are learned and can be identified by very young children. The 101 children tested were divided into Jour age groups: 1-yearaIds, 2-year-olds, 3-year-olds, and 4-year-olds. The children were requested to point to 20 body parts on a doll. Analysis ojthe results indicated significant diiferences in the ability to identify body parts by age and sex. The greatest increase in scores occurred between the ages oj1 and 2 years, with girls achieving a slightly higher score in each age group. The percentage ojsubjects at each age who identified diiferent body parts is presented, indicating the sequence in which body parts are learned. At the time this study was conducted Kathryn MacWhinney was an occupational therapy graduate student in the postprofessional master of science degree program at Boston University. She is now an occupational therapist at North Shore Special Education Consortium, Peabody, Massachusetts. (Mailing address: 75 Orchard Street, Medford. Massachusetts 02155.) Sharon A Cermak, EdD, OTR, FAOTA, is Associate Professor of Occupational Therapy at Boston University and a faculty member of Sensory Integration InternationaL Anne Fisher, ScD, OTR, FAOTA, is Assistant Professor of Occupational Therapy at Boston University and a faculty member of Sensory Integration International. The purpose of this study was to examine the sequence in which body parts are learned and can be identified by very young children. Until recently, minimal information on the development of body part identification in young children has been available although several developmental assessments include items that involve pointing to or naming body parts. One of these assessments was implemented by Gesell (1940), who investigated pointing responses in 18- and 24-month-old infants by comparing scores from his sample (n = 19) of 18-monthold infants to the scores of the 24-month-old infants (n = 50) studied by Muntz (1921) Both groups of infants were asked to point to their hair, eyes, nose, and mouth. However, specific criteria for correct responses were not discussed. Gesell reported that only 5% of the 18-month-old infants he studied were able to point to the four body parts whereas 96% of Muntz's 24-month-old infants were able to point to the four parts. These body parts are similar to items on tests such as the Bayley Scales of Infant Development (Bayley, 1969) and the Revised Stanford Binet Test of Intelligence (Terman & Merrill, 1970), which require a pointing response. The stimulus used on the Bayley is a 3-dimensional doll, whereas a paper doll is used on the Stanford Binet. Age ranges and criteria for receiving credit also vary slightly. On the Bayley, the child receives credit at the 19-month level (age range of 15 to 26 months) if he or she points to three out of the seven parts (hair, mouth, ears, hands, eyes, feet, and nose). On the Stanford Binet, the child receives credit at 24 months of age by pointing to three or more parts (on form 1: hair, mouth, ear, and hands; on form M: hair, eyes, feet, and nose). The naming of body parts has also been studied extensively. Cratty (1970) compiled normative data on the verbal identification of body parts based on a review of the literature; however, the results from the various studies are not presented very clearly. On the basis of this review Cratty observed that at the age of 2 years, children named tummy, legs, feet, arms, and facial parts; that at the age of 3 years they named "planes of the body (ie., front, back, and side) related to objects" (p. 120); and that at the age of 4 years, they named thighs, elbows, and shoulders. Williams (1983) also investigated the age at which children "identify major parts" (p. 302). Although the standard "to identify" was not defined, we felt it described a naming response. Also, the words "major parts" were vaguely defined as "eyes, ears, hands, knees, etc." (p. 302). In Williams' sample of children aged 5 to 9 years en = 250), 55% of the 5 year-olds were able to identify (ie, name) major parts. Williams' conclusion that only slightly over half of the 5-year-olds tested were able to identify body parts seemed in contradiction to other studies on 454 ]uly1987, Volume 41, Number 7

younger children that indicated identification of body parts occurred at much younger ages (Bayley, 1969; Cratty, 1970; Gesell, 1940; Muntz, 1921). This apparent contradiction may actually reflect a difference in the learning sequence between naming parts versus a pointing response. Data from studies on pointing versus studies on naming as means to identify body parts have suggested sllch a developmental sequence. Gesell (1940) indicated that there was a developmental progression in the numbers of parts children were able to identifyi while Cratty's (1970) study indicated a sequence with some parts learned before other parts (see Table 1 for a comparison of the studies and assessments discussed). The sequences through which children learn to identify body parts have been studied by examining children's performances on a variety of tasks. Meili Dworetski (as cited by Harris, 1963) who gathered information from more than 100 children through observations, interviews, and studying children's drawings, determined that children initially develop an understanding of their head, trunk, and legs. He hypothesized that the knowledge of head, trunk, and legs forms a basis for the development of further knowledge of the body. Gellert (1975) determined a similar progression from studying 250 elementary school children, aged 5 to 12 years. Each child was asked to construct a person from a selection of 37 body pieces. Gellert found that younger children made more errors and omissions of the upper extremities. He hypothesized that, in normal development, a knowledge of the position of one's trunk, head, and legs is developed prior to a knowledge of the position of one's arms and hands. In older children, drawings of the human figure have also indicated a progression of awareness of the body (Harris, 1963). The head and a vertical orientation of a body were established first. The existence of parts was perceived prior to the knowledge of their spatial relationships to each other, and parts that were important to the child were represented first. Data on the developmental progression of body part awareness (pointing, naming, drawing a person, etc.) are incomplete. None of the studies differentiated between the abilities of boys and girls, nor were they in agreement on the critical ages when particular body parts are learned. Moreover, there was a lack of data on the progressions of types of parts learned (i.e., whether young children learn (a) body parts and then joints [the relationships of parts], or Jearn (b) parts in a cephalic-to-caudal or proximal-to-distal progression). Developmental delays and dysfunctions in body part awareness have been noted in various populations, including persons with physical disabilities (Critchley, 1950), emotional disabilities (McCrae, Summerfield & Rosen, 1982), and neurological impairments (Ayres, 1961). Ayres and MacDonald (1960) are among occupational therapists who have Table 1 Normative Data of Body Part Identification Subjects (n) Ages Gese II (1940) 19 18 momhs Bayley (1969) 360 19 momhs (15-26 momhs range) Muntz (1921) 50 2 years Stanford Binet" 2 years Cratty (1970) from a su rvey of 2, 3. & 4 years available literature Williams 240 5-9 years (1983) Nole. For complete citations, see reference list. 'As discussed by Terman and Merrill, 1970 Task Parts Criteria/Response pointing on self hair, eyes 42%-no parts nose, mouth 21%-1 part 5%-4 parts pointing on hair, mouth credit for 3 parts doll ears, hands feet, nose pointing on self hair, eyes 96% pointed to 4 parts nose, mouth pointing on a hair, mouth credit for 3 or more paper doll ear. hand or parts hair, eyes feet, nose naming 2 years-tummy, feet, legs, arms, face pans 3 years- "planes of body" 4 years-thighs, elbows, lingers, shoulders, thumb "idemifying" "eyes, ears, hands. 55%-"identified knees, etc," maior parts" at 5 years The Americanjournal ojoccupalional Therapy 455

theorized that dysfunction in spatial awareness of body parts hinders the ability to plan movements in everyday living activities, Therefore, the assessment of body part awareness may be an essential component in occupational therapy treatment and should be performed accurately A variety of assessments are presently used by occupational therapists to determine body part awareness, including pointing to and naming body parts, drawing a person, imitating movements, and completing person puzzles (Bayley, 1969; Harris, 1963; Gellert, 1975; Ayres, 1961), However, normative data have been incomplete, Response rates then become difficult to interpret and make the implementation of appropriate occupational therapy programs difficult. Data from a normative sample, indicating a developmental progression and/or critical ages when particular body parts are learned, would assist occupational therapists' efforts to objectively identify developmental delays in abilities to identify body parts. These delays may have a crucial impact on the relationship between body scheme and motor planning, The current study sought to provide addi tional data on the development of the ability to point to body parts in male and female children, ages 1 to 4 years, and to consider the possible progressions of other types of responses. The following hypotheses were tested: (a) Older age groups will point to signifi cantly more body parts than younger age groups and (b) there will be no significant sex effect. An additional guiding question was as follows: Does pointing to body parts show a sequence of types of parts learned (i.e., individual parts, and then spatial relationships of parts [joints], cephalic to caudal, or proximal to distal);> Methodology Subjects Of the 101 normal (nondisabled) subjects tested, approximately 24 were drawn from each of the following four age groups: 1-year-olds (10 to 14 months), 2-year-olds (22 to 26 months), 3-year-olds (34 to 38 months), and 4-year-olds (46 to 50 months). There were 12 boys and 12 girls in each of the groups of l and 2-year-olds. The group of 3-year-olds consisted of 10 boys and 14 girls, and the group of 4-year-olds consisted of 13 boys and 16 girls. An additional 10 children (six 2-year-old boys, one 3-year-old boy, two 3-year-old girls and one 4-year-old boy) would not participate in the test. Two of the children started the test, but would not complete it. Subjects were recruited from nursery schools, well-baby clinics, and day-care centers from the Boston, Massachusetts, and Rochester, New York, areas. Children belonged to predominantly white, middle-class families. Only Table 2 Percentage of Subjects Correctly Identifying Body Parts (Parents' Scores in Parentheses) l year-olds 2-year-olds 3-year-olds 4-year olds Body Part (n = 24) (n = 24) (n = 24) (n = 29) Ankle a 25 28 Arm 71 (75) 96 93 Back a 71 96 97 Chin a 67 75 90 Ears 8 (21) 96 100 100 Elbow a 21 71 69 Eyes 25 (50) 100 100 97 Finger 0(4) 92 (96) 100 100 Foot a (8) 96 96 93 Hair 4 (21) 96 (100) 100 100 Hand 4(17) 96 (100) 96 100 Knee 0(4) 46 83 90 Leg o (4) 58 (63) 92 90 MOlHh 0(13) 92 (96) 100 100 Neck 0 42 96 97 Nose a (50) 96 100 100 Shoulders 0 21 71 76 Toes 8 (29) 100 100 90 Tummy 4 (21) 92 (96) 100 100 Wrist a 4 17 28 those children who showed no observable physical disabilities were included in the study. Instruments A 19 in Baby Tunene doll l was used for this study, and each child was asked to point to 20 body parts (see Table 2) (Barval, 1985). Body parts were selected to depict appropriate categories, such as cephalocaudal, proximal-distal, and parts versus joints. Procedure After the child entered the testing room and the parent Signed the informed consent form, the child was presented with a doll to hold or to play with, As the child played with the doll, the examiner asked the child to point to the specified body parts. The easiest method of evaluating the children's awareness of body parts seemed to be through a painting response, which enabled very young children to be included in the sample and decreased the effects of verbal abilities. Because research has shown that children asked to point to parts on themselves often became selfconscious (Terman & Merrill, 1970), a doll was used. Questions such as "Where's the doll's hair?" or "Show me the doll's eyes" were asked. Each of the 20 body parts was randomly assigned a number and all were grouped into 15 ordered sequences by random selection. The order of body parts to be identified varied for each child. Therefore, each boy and girl in each age group was assigned a randomly selected se 1 The doll is manufactured by Barval Toys, Inc., 8801 Northeast 23rd Street, Miami, Florida 33137 456 july 1987, Volume 41, Number 7

quence of parts to identify from the possible 15 orders. One point was recorded for each correct painting response. Correct identification was recorded when the child placed his or her hand or finger on the specified body part or painted clearly towards the body part. When responses were questionable, encouragement was provided to attain a clear response. Yes or no questions such as "Is this the doll's hair/" were not used. Additional information was obtained concerning the child's ability to identify body parts from parental reports when available If the child did not point to specified body parts, the parent or caretaker was asked which items the caretaker felt the child knew. This information was not used in statistical analyses but was used in descriptive analysis to further examine types and progressions of parts learned. A parent or caretaker was present for the testing of all the 1-,2-, and 3-year-olds The majority of 4-year-olds were tested in nursery schools; however, 34% (10) of those children were tested in the presence of their caretaker. Results Statistical analyses were performed on accuracy scores, the total scores for correct painting on the doll (see Table 3 for means and standard deviations) Analysis of variance was performed for accuracy scores using a 4 X 2 (Age X Sex) design. Age was significant, F(3,93) = 418.24, P <.0001, and sex was significant, FO,93) = 6.82, P <.Oll The Age X Sex interaction was not significant, F(3,93) = 1.3, P =.28. Post hoc tests were performed on accuracy scores to examine the age effect. The Newman-Keuls test (Bruning & Kintz, 1977) revealed a Significant difference in scores between the ages of 1 and 2 years, (p <.01), and between the ages of 2 and 3 years, (p <.01). There was no significant clilference between the ages of 3 and 4 years. Table 2 depicts the percentage of children in each age group who were able to correctly identify specific body parts. The additional information ob- Table 3 Accuracy Scores of Body Part Identification by Age and Sex Age Boys Girls I year-olds 2-year-olds 3 year-olds 4 year-olds 1\1 1\1 M M 058 067 12.83 3.27 16.10 179 1646 127 042 0.67 1.3.67 2.90 17.86 1.66 1806 J6J tained from parental reports is provided in parentheses next to the accuracy scores. Discussion Age and Sex Effects The findings of this study support Gesell (1940) and others who have indicated that there is a developmental progression in the number of parts that children are able to identify (Bayley, 1969; Terman & Merrill, 1970; Cratty, 1970). Older children pointed to signifi cantly more body parts than younger children. It is interesting to note that there was a large increase in the number of parts identified between 1- and 2-yearaIds, and the 2- and 3-year-olds, with 3- and 4-yearaIds achieving similar scores. This suggests that there is a critical time period between the ages of 1 and 3 years during which children learn to identify body parts. In this age span, children encounter increased experiences and interactions with the environment. They are more mobile and may become more aware of their own body parts through new sensory and motor experiences. Attention span, language abilities, and abilities to follow directions also improve, all of which would improve the ability to perform on a task such as pointing to requested parts on a doll. The majority of 1-year-old children were unable to point to specified body parts. No body parts were reported by 50% of the children at 1 year although, according to parental reports, half of the children knew eyes and nose On the other hand, a substantial increase occurred in the number of reported responses recorded for the 2-year-old children's sample; 11 body parts or more were identified by 75% or more of the children (parent scores). The two older age groups were able to identify all but one to two parts_ This study coincides with the information from previous studies regarding some of the parts learned first by young children (hair, eyes, nose, mouth, hands, feet, ears) and has similarities with items on the Bayley Scales of Infant Development and the Revised Stanford Binet Test of Intelligence. However, the new information suggests that 2-year-old children are able to identify a much greater number of body parts than has previously been reported by Gesell (1940), Bayley (1969), and Cratty (970) Male/female differences have not previously been reported. However, this study found a significant overall sex effect, with girls' scores slightly higher than the boys' scores. A variety of factors may have influenced this male/female difference Girls may typically have more experience playing with dolls ancl may be more familiar with dolls' body parts_ In this sample, girls were not as reluctant to participate in the test as were boys (70% of the children who The American Journal of Occupational Therapy 457

chose not to participate were boys). Other differences in language abilities, attending, attending to detail, and following directions may have intluencecl the results. A replication of this study with a larger sample of children using both a doll and a male-specific referent would assist in further delineating the presence of male/female differences in knowledge of body parts. Testing Method The designed testing situation was found to be useful with older age groups; however, it was found to be inadequate in determining the abilities of 1-year-olds and some of the 2-year-olds. Many of the very young children did not appear to understand what was asked of them; they were fearful or shy of the examiner or unable to attend for the duration of the time required to complete the task. Many of the very young children were unable to complete the task but showed some understanding of it; for example, they would look at a body part or touch their own body part. Some of the young children's abilities seemed to diminish toward the end of the test, suggesting the duration was too long for a young child's attention span. Although accuracy scores indicated the child's actual performance on the task, they may not have been indicative of his or her actual knowledge. Therefore, in order to learn more about the typical parts that very young children know, parental reports were also obtained. Information obtained in this manner relies heavily on the reliability of parental reports. If one uses parental reports, the scores increase, particularly in the younger age groups. Parents of 1- and 2-year-olds reported that the child "knew" an average of one body part more than the child demonstrated knowing. There was no change in older children's scores. Progressions and Types ofparts Learned Table 2, which lists specific body parts identified by each age group, indicates that there was no apparent sequence or progression of types of parts learned in terms of proximal to distal, or cephalic to caudal. However, body parts were correctly identified more frequently than joints, suggesting that joints were harder parts to learn_ In the sample of 2-year-olds, none of the joints were correctly identified by 50% or more of the children. Three- and 4-year-olds showed an increase in knowledge of joints, although they were still unable to identify all the joints. Conclusion Tools to evaluate body scheme functioning are varied. Drawing a person, pointing to and naming body parts, imitating movements, and completing person puzzles are a few of the methods that can be used. This study provides baseline data on the ability of young children to identify body parts. This information is important because body scheme has been implicated as an important construct in the child's development of motoric and perceptual abilities. Motor skills, spatial awareness, learning abilities, and social!emotional growth are areas of development that are of crucial concern to occupational therapists. These areas of development have all been implicated as functions that result from the development of body scheme (Ayres, 1961; Arnheim & Sinclair, 1979; Cleveland & Fischer, 1957; Head, 1911; MacDonald, 1960; Schilder, 1964). Ayres and others (Arnheim & Sinclair, 1979; Cratty, 1970) theorize that movement of body parts is the groundwork for the development of spatial awareness and the understanding of spatial concepts_ Kephart (1971) states that "we use our own bodies as this point of reference. Objects about us are referred to our body and oriented in space with reference to it" (p. 93). Feeding, dressing, using a pencil, or sitting at a table, all rely on the knowledge of where one's body is in space. The evaluation method that employs pointing to body parts on a doll can be a useful tool in assessing a child's knowledge. However, consideration must be given to very young children who are not easily persuaded to complete the entire task. Longer durations of time in additional sessions and additional methods of assessing the child's abilities may be useful. For 1-year-olds the task could be pared down to 12 items, omitting those items that none of the children could identify and that none of the parents reported as known by the child. Parental reports may also provide valuable information that cannot be obtained in a traditional testing situation and that may help to establish emerging knowledge, or knowledge that has not yet been generalized to different situations. Therapeutic intervention can then be initiated at the child's perceived level of emerging knowledge and proceed in the appropriate sequence. Facial parts seemed to be learned first, with detailed parts, or joints learned last. This finding is not consistent with studies that show that the head, trunk, and legs are learned first (Harris, 1963). Further research to supplement this investigation would be beneficial. Detailed research focused on the emerging of abilities of children between the ages of 1 and 3 years would provide more detailed information on the progression of parts learned and help to establish the age at which the ability to identify body parts is realized. References Ayres, A.]. (1961) The development of the body scheme in children. American journal of occupational Therapy, 15(3),99-128 458 july 1987, Volume 41, Number 7

Arnheim, D., & Sinclair, W. (1979). The clumsy child. Sr. Louis: Mosby. Bayley, N. (1969). Bayley Scales of Infant Development. New York: Psychological Corporation. Bruning, J. L. & Kintz, B. L. (1977). Computational handbook ofstatistics. Glenview, IL: Scott, Foresman & Co. Cleveland, S., & Fischer, S. (1957). Prediction of small group behavior from a body image schema. Human Relations, 10, 223-233 Cratty, B. (1970). Perceptual and motor development in infants and children. New York: Macmillan. Critchley, M. (1950). The body image in neurology. Lancet, 1, 335-340. Gellert, E. (1975) Children's constructions of their self images. Perceptual and Motor Skills, 40,307-324. Gesell, A. (1940). Thefirstfive years oflife. New York: Harper & Row. Harris, D, B. (1963). Children's drawings as measures of intellectual maturity. New York: Harcourt, Brace, and World, Head, H. (1911). SensoiY disturbances from cerebral lesions, Brain, 34, 187-188. Kephart, N. (1971). The slow learner in the classroom. Columbus, OH: Charles E. Merrill, MacDonald, J. C. (1960) An investigation of body scheme in adults with cerebral vascular accidenrs. American}ournal of Occupational Therapy, 14(2),75-79. McCrae, c., Summerfield, A., & Rosen, B. (1982). Body image: A selective review of existing measurement techniques. British}ournal ofmedical Psychology, 55, 225-233. Munrz, L. (1921), A study of individual differences in two-year old children. Unpublished master's thesis, Yale University, New Haven, CT. Schilder, P (1964). The body image. In L. Bender (Ed.), Contributions to developmental neuropsycbiatry (pp. 42-60). New York: International Universities Press, Terman, L., & Merrill, M. (1970) Measuring intelligence: A guide to tbe administration of tbe New Revised Stanford Binet Tests ofintelligence, New York: Macmillan, Williams, H. (1983). Perceptual and motor development. Englewood Cliffs, New Jersey: Prentice Hall. The American Journal of Occupational Therapy 459