Verification and Clarification of Patterns of Sensory Integrative Dysfunction Zoe Mailloux, Shelley Mulligan, Susanne Smith Roley, Erna Blanche, Sharon Cermak, Gina Geppert Coleman, Stefanie Bodison, Christianne Joy Lane KEY WORDS sensation disorders touch perception vestibular diseases visual perception Zoe Mailloux, MA, OTR/L, FAOTA, is Executive Director of Administration and Research, Pediatric Therapy Network, 1815 West 213th Street, Suite 100, Torrance, CA 90501; email@example.com Shelley Mulligan, PhD, OTR/L, is Chair, Department of Occupational Therapy, University of New Hampshire, Durham. Susanne Smith Roley, MS, OTR/L, FAOTA, is Director of Education and Research; Gina Geppert Coleman, MA, OTR/L, is Executive Director of Practice and Education; and Stefanie Bodison, OTD, OTR/L, is Director of Clinical Education and Professional Development, Pediatric Therapy Network, Torrance, CA. Erna Blanche, PhD, OTR/L, FAOTA, is Associate Professor of Clinical Practice, Division of Occupational Science and Occupational Therapy, and Sharon Cermak, EdD, OTR, FAOTA, is Professor, Division of Occupational Science and Occupational Therapy, Ostrow School of Dentistry University of Southern California, Los Angeles. Christianne Joy Lane, PhD, is Biostatistician, Center for Transdisciplinary Research on Energetics and Cancer, Keck School of Medicine, University of Southern California, Los Angeles.
Building on established relationships between the constructs of sensory integration in typical and special needs populations, in this retrospective study we examined patterns of sensory integrative dysfunction in 273 children ages 4–9 who had received occupational therapy evaluations in two private practice settings. Test results on the Sensory Integration and Praxis Tests, portions of the Sensory Processing Measure representing tactile overresponsiveness, and parent report of attention and activity level were included in the analyses. Exploratory factor analysis identified patterns similar to those found in early studies by Ayres (1965, 1966a, 1966b, 1969, 1972b, 1977, & 1989), namely Visuodyspraxia and Somatodyspraxia, Vestibular and Proprioceptive Bilateral Integration and Sequencing, Tactile and Visual Discrimination, and Tactile Defensiveness and Attention.
Findings reinforce associations between constructs of sensory integration and assist with understanding sensory integration disorders that may affect childhood occupation. Limitations include the potential for subjective interpretation in factor analysis and inability to adjust measures available in charts in a retrospective research. Mailloux, Z., Mulligan, S., Roley, S. S., Blanche, E., Cermak, S., Coleman, G. G., et al. (2011). Verification and clarification of patterns of sensory integrative dysfunction. American Journal of Occupational Therapy, 65, 143–151. doi: 10.5014/ajot.2011.000752 Sensory integration theory was developed by A. Jean Ayres beginning in the late 1950s and early 1960s (Ayres, 1963, 1964). The concepts and body of knowledge subsumed under this framework, now recognized as Ayres Sensory Integration (ASI), are aimed at understanding the sensory and praxis functions that provide the foundation for many aspects of development, social participation, and occupational performance. In addition to the theoretical constructs that provide a road map for understanding normal development, ASI includes a large body of research validating the existence of patterns of sensory integrative dysfunction.
This study was designed to clarify patterns of sensory integration dysfunction to assist with the development and validation of a diagnostic typology for the disorder and to assist therapists with the interpretation of assessment data for intervention planning. Review of the Literature The early formulation of ASI included the development of a set of measures for the systematic investigation of patterns of sensory integrative function and dysfunction. The Sensory Integration and Praxis Tests (SIPT; Ayres 1989), and their earlier version, the Southern California Sensory Integration Tests (Ayres, 1972a), as well as measures in areas such as tactile defensiveness, hyperactivity, distractibility, and language, were included as variables in multiple studies.
Other researchers have also contributed to the understanding of patterns of sensory integration dysfunction. Lai, Fisher, Magalhaes, and Bundy (1996) proposed that praxis might be a possible underlying construct for both bilateral integration and sequencing deficits as well as for somatodyspraxia. Mulligan (1998) conducted a largescale factor analysis on a sample of >10,000 children and found patterns similar to those found in Ayres’ earlier studies (1965, 1966a, 1966b, 1969, 1972b, 1977, 1989), including problems with bilateral integration and sequencing, somatosensory deficits, visual–perceptual deficits, and dyspraxia. Her study also found that although subtypes emerged from the data, the patterns were highly correlated with one another. Mulligan suggested that they had a common underlying construct and reinforced the concept of integration of multiple sensory inputs from multiple sensory systems as being important for neurological functioning. The factor loadings found in the studies by Mulligan (1998) and Ayres (1965, 1966a, 1966b, 1969, 1972b, 1977, 1989) also provided insight into specific relationships among functions, such as the role of somatosensory perception in relation to motor planning and the relationships between vestibular, postural–ocular, and bilateral functions. On the basis of SIPT scores, Ayres (1989) identified six cluster patterns that emerged from a mixed sample of typically developing children and children with sensory integration problems. Three of the six cluster groupings were thought to represent patterns not specifically associated with sensory integration dysfunction. Two of the groupings were characterized by sensory integration functions in the average range (low-average and highaverage sensory integration and praxis). The third pattern involved language processing (dyspraxia on verbal command) that Ayres (1989, p. 182) suggested might reflect left cerebral hemisphere functions rather than sensory integration dysfunction.
The three clusters characterized by recognizable patterns of sensory integration were Bilateral Integration and Sequencing, Visuo-Somato Dyspraxia, and Generalized Sensory Integration Dysfunction (Ayres, 1989). Mulligan’s (2000) cluster analysis of a large data set of children tested with the SIPT revealed that children fell into one of five cluster groupings: average sensory integration and praxis, moderate sensory integration dysfunction, severe sensory integration dysfunction and dyspraxia, dyspraxia, and low-average bilateral integration and sequencing. The results of this study suggested that in addition to the type of dysfunction, severity of dysfunction was an important consideration.
In a an evidence review of subtypes of problems in sensory integration, Davies and Tucker (2010) identified the need for “comprehensive assessment of sensory function and sensory-based motor performance that includes sensory perception, discrimination, modulation, and praxis in a single study” (p. 399). In this study, we aimed to improve on these methodological challenges by examining data from a large clinical sample in which (1) the test examiners were known to have had extensive trainingin the useof the tests andinASI theory and its application in occupational therapy practice, (2) the background and presenting problems of the children were known, and (3) parent report measures regarding overresponsiveness to sensory experiences and the presence or absence of difficulties with attention were available in addition to SIPT scores.
Another challenge faced by researchers who have conducted studies examining patterns of dysfunction using the SIPT concerned the management of data from the Postrotary Nystagmus (PRN) test. PRN, 1 of the 17 SIPT components, is a measure of vestibular function. PRN is unique in that it is the only SIPT measure in which both a high and a low Z score are indicative of dysfunction (for all other tests, only low scores are indicative of dysfunction). Moreover, a low PRN score is believed to be associated with a type of problem different from a high PRN score. This distribution of high and low scores in children with sensory integrative challenges is suspected to have diminished this measure’s clarity in relation to the other measures (such as bilateral coordination and balance), particularly when mean scores and correlation statistics have been applied.
To verify and clarify patterns of sensory integrative dysfunction, with consideration of previous research findings and challenges regarding variables, we reexamined associations identified in previous studies by including comprehensive measures of sensory integration functions (SIPT) and items from a parent report of responses to sensory experiences and parent reports of problems with attention. In addition, we explored methods for managing the PRN test, which vary from those for other sensory integration tests because both high and low scores are reflective of dysfunction.
Hypotheses We hypothesized that the population under study would demonstrate patterns of sensory integration dysfunction similar to those found in previous studies. The specific hypotheseswere that thefollowing associations and patterns would emerge: • Visual perception and visual praxis would be associated, reflecting a pattern of visuopraxis. • Tactile perception and praxis would be associated, reflecting a pattern of somatopraxis. • Vestibular functions and bilateral motor coordination and sequencing would be associated, reflecting a pattern of vestibular bilateral integration and sequencing. • Tactile perception tests would be associated with each other or with other sensory perception tests. • Sensory overresponsiveness and attention would be associated. Method Research Design Four licensed occupational therapists and research assistants conducted a retrospective chart review to extract data from the charts of children who received occupational therapy evaluations from two pediatric therapy practices in California. The sample was extracted from the files of 425 children who had been tested with the SIPT and the Sensory Processing Measure–Home Form (SPM–Home Form; Parham & Ecker, 2007) or the precursor to this instrument, the Evaluation of Sensory Processing (ESP; Johnson-Ecker & Parham, 2000; Parham & Ecker, 2007). Deidentified scores from these measures then underwent descriptive analyses and factor analysis to examine patterns of sensory integration dysfunction.
Instruments The variables measuring sensory integration and praxis functions included in the study are summarized in Table 1 and described here. Sensory Integration and Praxis Tests. The SIPT is a battery of 17 tests measuring visual form and space perception, somatosensory functions (tactile and proprioceptive processing), vestibular processing (including balance), bilateral coordination, sequencing, and praxis. The test is standardized on children ages 4 yr, 0 mo, to 8 yr, 11 mo. The validity and reliability data provided in the test manual (Ayres, 1989) are strong. The major Z scores for all SIPT measures described in the test manual were included in the analysis, with the exception of the Manual Form Perception test, for which both Part 1 and Part 2 scores were entered separately. Because Part 1 involves matching tactile stimuli with visual stimuli and Part 2 involves tactile processing simultaneously with two hands, the two scores for this test were kept separate.
The PRN test has created challenges in previous studies because both low (£–1) and high (³1) Z scores represent dysfunction; combining high and low PRN scores results in mean scores for the group that appear average, so the research team entered this variable in several ways, including initially entering high and low scores as separate variables. However, initial review of the data revealed that only about 8% (23 of 273) of the children in the sample demonstrated prolonged (³1.0 standard deviation) PRN. Because the canceling effect was not present in this sample, the major Z scores were entered in the same manner as the other SIPT scores, thus resulting in a total of 18 variables for the SIPT. Sensory Processing Measure–Home Form. The SPM– Home Form is a parent questionnaire standardized on children in Grades K–6. Test–retest reliability was strong (r 5 .94; Parham & Ecker, 2007). Internal consistency of items yielded a coefficients >.80, and data support the SPM–Home Form’s construct and content validity (Parham & Ecker, 2007). Items considered for analysis were those that targeted areas of overresponsiveness to tactile, auditory, and movement experiences; vestibular and proprioceptive seeking; and ocular and postural control. Because many of the children included in the study had been administered an earlier research version of the SPM, the ESP, only items in these areas on both forms were included. After preliminary analyses, we decided that the six items related to tactile defensiveness would be most useful because relatively few items were included in the other areas and tactile defensiveness was a pattern that had emerged in previous studies (Ayres, 1965, 1966a, 1966b, 1969, 1972b). Thus, the following items (listed by their accompanying SPM or ESP item number) were included: • SPM 30/ESP T4 (overresponsive): Does your child pull away from being touched lightly? • SPM 32/ESP T7 (overresponsive): Does your child react negatively to the feel of new clothes? • SPM 33/ESP T16 (overresponsive): Does your child prefer to touch rather than be touched? • SPM 34/ESP T27 (overresponsive): Does it bother your child to have his/her finger or toe nails cut? • SPM 35/ESP T36 (overresponsive): Does it bother your child to have his/her face touched? • SPM 36/ESP T12 (overresponsive): Does your child avoid getting his/her hands in finger paint, paste, sand, clay, mud, glue, or other messy things? The items were scored on a scale ranging from 1 to 4, with a lower score indicating more difficulty, to match the Table 1.
The mean score for the six ESP and SPM variables was calculated and included in the analysis. Behavior Rating of Attention. Because we were interested in revisiting the existence of patterns related to overresponsiveness to sensory experiences, we included a behavioral measure of attention–inattention and distractibility. Descriptive data from the intake form in the charts provided a means of capturing the presence of problems and parental concerns related to the child’s ability to attend to tasks, activity level, and impulsiveness. A scale ranging from 0 to 2 was used to rate attention problems provided by parents (2 5 definite evidence, significant problem; 1 5 some evidence, mild or potential problem; 0 5 no evidence of problems with attention). Interrater reliability was established on this measure using 20 cases across three raters, who achieved 100% agreement on the ratings after refinement of definitions. Data Analysis After exploratory, descriptive data analysis, we conducted factor analysis to examine patterns of sensory integration dysfunction.
We calculated descriptive statistics, including frequencies, means, and standard deviations, for all variables, and estimated bivariate Pearson correlation coefficients between all variables (Table 2). Factor analysis was performed using SPSS 16.0 (SPSS Inc., Chicago). Exploratory factor analysis using principal components extraction with varimax rotation was performed using the variables listed in Table 1 to assist with determining the patterns of dysfunction or domains of sensory integration dysfunction being measured. Results The sample included 193 boys (70.7%) and 80 girls (29.3%) ranging in age from 4 to 9. Most of the children (78.1%) had no diagnosis on file and were referred for evaluation because of occupational performance problems suspected to be the result of underlying sensory integration dysfunction. Diagnoses or conditions that were reported included sensory integration disorder (n 5 29; 10.6%), attention deficit disorders (n 5 20; 7.3%), and speech–language or learning disorders (n 5 17; 6.2%). The results of the factor analysis yielded six factors with eigenvalues >1. However, because two factors included only one variable, we felt that a four-factor solution best fit the data, accounting for 48% of the variance.
The four-factor solution with accompanying factor loadings is shown in Figure 1. All factors in this study were significantly related to one another (Table 3). The first factor, named Visuodyspraxia and Somatodyspraxia, is characterized by high loadings on the tests of visual perception and visuopraxis functions, including Table 2. Pearson Correlations for Variables Included in Factor Analysis Variables 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 1. ATTN — 2. BMC .08 — 3. CPR .06 .28 — 4. DC .08 .34 .51 — 5. FG .08 2.02 .13 .10 — 6. FI .14 .21 .29 .33 .16 — 7. GRA .20 .37 .25 .35 .18 .35 — 8. KIN .07 .25 .31 .36 .07 .30 .37 — 9. LTS .10 .09 .25 .22 .10 .34 .30 .21 — 10. MAC .10 .32 .22 .46 .07 .31 .40 .33 .26 — 11. MFP 1 .07 .27 .48 .42 .14 .31 .28 .26 .16 .25 — 12. MFP 2 .05 .23 .29 .26 .02 .30 .25 .23 .16 .25 .39 — 13. OPR 2.04 .38 .15 .21 2.07 .17 .36 .17 .15 .23 .14 .15 — 14. PPR .03 .31 .29 .42 .07 .35 .25 .26 .05 .30 .33 .14 .37 — 15. PRN .04 .07 .02 .07 .04 .09 .17 .17 .00 .19 .08 .15 .10 .09 — 16. PRVC .19 .36 .36 .39 .12 .29 .34 .34 .08 .24 .32 .33 .22 .40 .14 — 17. SPR .14 .52 .38 .49 .16 .38 .48 .41 .24 .39 .36 .31 .33 .37 .19 .44 — 18. SV .21 .18 .34 .48 .16 .32 .35 .24 .17 .35 .33 .27 .14 .34 .08 .30 .41 — 19. SWB .03 .35 .29 .39 .01 .30 .36 .28 .23 .52 .21 .25 .32 .35 .22 .33 .41 .32 — 20. TD .04 2.05 2.03 .04 .00 2.07 .02 .03 2.03 2.07 2.02 .10 2.05 2.05 2.06 .03 .04 .06 .00 — Note. ATTN 5 Attention Behavioral Rating; BMC 5 Bilateral Motor Coordination; CPR 5 Constructional Praxis; DC 5 Design Copying; FG 5 Figure Ground; FI 5 Finger Identification; GRA 5 Graphesthesia; KIN 5 Kinesthesia; LTS 5 Localization of Tactile Stimuli; MAC 5 Motor Accuracy; MFP 1 and MFP 2 5 Manual Form Perception, Parts 1 and 2; OPR 5 Oral Praxis; PPR 5 Postural Praxis; PRN 5 Postrotary Nystagmus; PRVC 5 Praxis on Verbal Command; SPR 5 Sequencing Praxis; SV 5 Space Visualization; SWB 5 Standing and Walking Balance; TD 5 Tactile Defensiveness.
Many previous studies demonstrated a bilateral integration and sequencing pattern, associating bilateral integration with vestibular and proprioceptive functions. This pattern was characterized by high loadings on Oral Praxis, Standing and Walking Balance, Postrotary Nystagmus, Bilateral Motor Coordination, Graphesthesia, Motor Accuracy, and Sequencing Praxis. Of the tests loading on this factor, Oral Praxis, Standing and Walking Balance, Bilateral Motor Coordination, Graphesthesia, and Sequencing Praxis all loaded on the Bilateral and Sequencing factor reported in the SIPT manual (Ayres, 1989). All of these tests were noted to involve bilateral and sequential actions. In this study, the addition of Motor Accuracy to this group is hypothesized to reflect the vestibular and bilateral functions needed for the coordination of eye, head, and hand movements; postural adjustments; and crossing of body midline needed in this test.
In the 1989 data from the SIPT manual, these independent patterns were notably stronger (i.e., first-order factors) in the samples that had large numbers of typically developing children. A cluster analysis of the same data revealed a Visuodyspraxia and Somatodyspraxia cluster group that included more children with learning disabilities or sensory integration issues (n 5 22) than children from the normative group (n 513). It is possible that visuo- and somato-based praxis functions overlap more in samples of children with identified problems, as was the case with this study’s sample. The second factor finding that is especially important for both therapists and researchers is the demonstrated link between the PRN test and the low scores on tests of vestibular and bilateral functions.
Although the relationship of hyporesponsiveness to vestibular input, evidenced by low PRN scores, has been hypothesized, this study marks the first time this relationship has been supported by empirical data. The small percentage of cases with prolonged nystagmus (8.4% [n 5 23] vs. 35.9% [n 5 98] with shortened-duration nystagmus) in this sample avoided the averaging effect that occurs when both high and low PRN scores are prevalent in a sample. Thus, these findings reinforce a long-held view of the significance of low PRN scores and validate Ayres’ (1989) theoretical assumption that bilateral integration problems and other signs of vestibular inefficiencies are associated with a shortened duration of postrotary nystagmus.
The added factor loading of the Motor Accuracy Test (which requires vestibular- and proprioception-related coordination of head, eye, and hand movements and crossing the body midline) and a moderate loading on Kinesthesia also reinforce the underlying commonality of vestibular and proprioceptive functions in this group. Although the hypothesis related to the association of low PRN with bilateral and sequencing measures was supported, an association between the measures of directionality, crossing midline, and laterality with this pattern was not found. The subscores used to measure these functions may not have had adequate biometric properties, or these measures may not reflect functions specific to this pattern.
Finally, the hypothesized association between attention problems and heightened sensitivity to touch rather than to tactile discrimination and practic functions was supported by empirical data. At this time, fully exploring the hypothesis that other forms of sensory overresponsiveness (e.g., auditory defensiveness and gravitational insecurity) are associated with problems in attention was not possible because the number of items representing those variables was smaller and less specific to overresponsiveness than the items for tactile defensiveness. Continued clarification of the relationships between generalized sensory overresponsiveness and specific sensory responsiveness is still needed. Efforts to clarify underlying physiological functions associated with over- and underresponsiveness will also provide insight into the impact of these problems on behavior, learning, and participation (Reynolds & Lane, 2009). The significant relationship between the factors in this study and Ayres’ (1989) studies reinforces the longheld idea that the patterns revealed in these studies show specific associations among sensory, motor, and praxis; these associations reflect that sensory systems function in a synergistic, rather than isolated, way.
Identifying and understanding underlying problems that affect health, performance, and participation are important first steps for practice and research in occupational therapy. The results of this study confirm the presence of sensory integration patterns of dysfunction in children commonly referred for an occupational therapy evaluation.
Knowledge of these patterns allows therapists to understand the nature of difficulties interfering with performance, contributes to more precise intervention planning, and provides a means by which to determine the interventions’ effectiveness. Using standardized assessments, such as the SIPT, in conjunction with caregiver reports will enhance participant selection in efficacy studies of occupational therapy applying a sensory integrative intervention approach. Limitations This study’s limitations include the subjectivity of factoranalytic studies, the subjectivity of the tools used to measure attention and tactile overresponsiveness, and the degree of representation in the sample. First, as with all factor analysis, the interpretation of the factors has a level of subjectivity (e.g., the decision to use the four-factor solution and the naming of the factors). However, given the breadth of previous studies on which to base decisions and the many researchers who collaborated throughout the interpretation process, subjectivity was minimized. Second, the methods for measuring attention and tactile overresponsiveness in this study do not have established reliability and validity data; however, other means for assessing these variables were not available in this retrospective sample. Finally, data were retrieved from only two private practices and may therefore have limitations in generalizability. However, the similarity with previous findings and hypothesized constructs reduces this potential concern.
Future Research Future research should continue to identify and enhance our understanding of the underlying neurodevelopmental problems that affect health, performance, and participation. Measures for children younger and older than the range included in the SIPT normative sample are needed to determine whether these patterns of dysfunction hold true for these age groups and are also needed to conduct outcome studies.
The development of perceptual–motor abilities: A theoretical basis for treatment of dysfunction (Eleanor Clarke Slagle Lecture). American Journal of Occupational Therapy, 17, 221–225. Ayres, A. J. (1964). Tactile functions: Their relation to hyperactive and perceptual motor behavior. American Journal of Occupational Therapy, 18, 6–11. Ayres, A. J. (1965). Patterns of perceptual–motor dysfunction in children: A factor analytic study. Perceptual and Motor Skills, 20, 335–368. Ayres, A. J. (1966a). Interrelations among perceptual–motor abilities in a group of normal children. American Journal of Occupational Therapy, 20, 288–292. Ayres, A. J. (1966b). Interrelationships among perceptual– motor functions in children. American Journal of Occupational Therapy, 20, 68–71. Ayres, A. J. (1969). Deficits in sensory integration in educationally handicapped children. Journal of Learning Disabilities, 2, 160–168. doi: 10.1177/002221946900200307 Ayres, A. J. (1971). Characteristics of types of sensory integrative dysfunction.
American Journal of Occupational Therapy, 25, 329–334. Ayres, A. J. (1972a). Southern California Sensory Integration Tests. Los Angeles: Western Psychological Services. Ayres, A. J. (1972b). Types of sensory integrative dysfunction among disabled learners. American Journal of Occupational Therapy, 26, 13–18. Ayres, A. J. (1977). Cluster analyses of measures of sensory integration. American Journal of Occupational Therapy, 31, 362–366. Ayres, A. J. (1989). Sensory Integration and Praxis Tests manual. Los Angeles: Western Psychological Services. Brett-Green, B. A., Miller, L. J., Schoen, S. M., & Nielson, D. M. (2010). An exploratory event-related potential study of multisensory integration in sensory over-responsive children. Brain Research, 1321, 67–77. doi: 10.1016/j.brainres. 2010.01.043 Davies, P. L., & Tucker, R. (2010). Evidence review to investigate the support for subtypes of children with difficulty processing and integrating sensory information. American Journal of Occupational Therapy, 64, 391–402. doi: 10.5014/ajot.2010.09070 Dunn, W. W. (1997).
The impact of sensory processing abilities on the daily lives of young children and families: A conceptual model. Infants and Young Children, 9, 23–25. doi: 10.1097/00001163-199704000-00005 Dunn, W. W. (1999). Sensory Profile: User’s manual. San Antonio, TX: Psychological Corporation. Johnson-Ecker, C. L., & Parham, L. D. (2000). The evaluation of sensory processing: A validity study using contrasting groups. American Journal of Occupational Therapy, 54, 494–503. Lai, J.-S., Fisher, A. G., Magalhaes, L. C., & Bundy, A. C. (1996). Construct validity of the Sensory Integration and Praxis Tests. OTJR: Occupation, Participation and Health, 16, 75–97. McIntosh, D. N., Miller, L. J., Shyu, V., & Hagerman, R. J. (1999). Sensory-modulation disruption, electrodermal disruption and functional behaviors. Developmental Medicine and Child Neurology, 41, 608–615. doi: 10.1017/ S0012162299001267 Miller-Kuhanek, H., Glennon, T., & Henry, D. (2007). Sensory Processing Measure—Main Classroom Form. Los Angeles: Western Psychological Services. Mulligan, S. (1996). An analysis of score patterns of children with attention disorders on the Sensory Integration and Praxis Tests. American Journal of Occupational Therapy, 50, 647–654. Mulligan, S. (1998).