Los Avances En La Investigación De La Integración Sensorial
La teoría de integración sensorial, que se derivó principalmente de los campos de la medicina, la neurología, y el desarrollo infantil, trata de explicar las relaciones entre los buques de procesos neurológicos y conductas manifiestas (Ayres, 1972b). La teoría de integración sensorial tiene como objetivo aumentar nuestra comprensión de las causas subyacentes de los problemas de conducta, dificultades motoras, problemas de aprendizaje y de algunos niños. Como marco de referencia, la integración sensorial actúa como puente entre la teoría y la practica clínica, proporcionando un conjunto de directrices para la evaluación e intervención. Estas directrices son una herramienta integral para la promoción de la investigación.

La integración sensorial ha sido objeto de una investigación mas que cualquier otro enfoque o marco de referencia dentro del campo de la terapia ocupacional (Miller & Kinnealy, 1993; Parpham y Mailloux, 2001). Durante muchos años, ha sido uno de los marcos mas aplicadas de referencia utilizado por terapeutas ocupacionales pediatricos. A pesar de esta popularidad, la investigación hasta la fecha no se ha establecido claramente el consenso profesional con respecto a la simple pregunta: ¿la intervención efectiva de la integración sensorial?

La integración sensorial ha sido criticado, sobre todo en la literatura educativa, donde la eficacia se ha desestimado por muchos investigadores (por ejemplo, de Hoehn y Baumeister, 1994). La legitimidad científica para este enfoque tiene, por tanto, no ha establecido todavía.
Objeto y Ambito
El propósito de este capítulo es examinar lavalidez de la integración sensorial, tanto como una teoría y un marco de referencia. Para llevar a cabo esta tarea, revisar y sintetizar la literatura mas reciente (a partir de aproximadamente 1986 a 2000) de los métodos basicos y aplicados de la investigación científica. También se discuten las formas en que los profesionales que seleccionan la integración sensorial como un método de intervención pueden aplicarse esta investigación. Llegamos a la conclusión de sugerir direcciones para futuras investigaciones para establecer un consenso profesional, tanto dentro como fuera de la terapia ocupacional con respecto a la validez de la teoría de integración sensorial y su aplicación como un enfoque de intervención.

Investigación Examinar La Validez De La Integración Sensorial
Validación de la teoría de integración sensorial de la integración de la información y el conocimiento de los métodos basicos y aplicados de la investigación científica. La investigación basica trata de responder a preguntas fundamentales sobre la naturaleza de la conducta. Las preguntas de investigación, tales como '¿Qué es la integración sensorial?', '¿Qué es la disfunción de integración sensorial? Y '¿Por qué funciona la integración sensorial intervención?' son ejemplos que caen bajo esta categoría. Ademas, la investigación basica analiza las muchas teorías sobre la cual se deriva la integración sensorial, incluyendo los postulados que lo apoyan.
La investigación aplicada es la investigación científica que directamente los intentos de resolver problemas practicos. Ejemplos de investigación aplicada en el ambito de la integración sensorial son los estudios que examinan la eficacia de laintervención de la integración sensorial, los estudios que tienen como objetivo desarrollar o evaluar los instrumentos de evaluación de la medición de las funciones de integración sensorial, y los informes de casos que describen la intervención y los cambios funcionales que se derivan de la intervención. La investigación es sobre la búsqueda del conocimiento y el desarrollo de formas de pensar, organizar y aplicar lo que sabemos. Sin embargo, los hallazgos científicos y las teorías son siempre considerados como provisionales. Independientemente de la forma elocuente una teoría en particular ha sido probada, siempre se considera tentativa - acercando mas y mas hacia la 'verdad', pero nunca lograr ese fin.
Sensory integration theory and its application as a means of evaluating and intervening with children with certain disabilities have continued to evolve since its inception in the early 1960s. This evolution has occurred in response to research findings over time as well as to the practical demands placed on the professionals who practice within this framework. Intervention based on sensory integration theory is very complex, as are the clients best suited for this approach and the theory on which it is based. Therefore, whether a researcher is interested in conducting basic or applied scientific inquiry, complexity provides a significant challenge.
To establish professional consensus regarding the validation of sensory integration theory, evaluation and support for the basic assumptions on which the theory is based are necessary. A clear conceptualization of what sensory integration is, and what constitutes sensory integration intervention is crucial. In addiction, efficacystudies are necessary to determine whether the intervention is effective, assist in identifying, and identify the functional areas affected by intervention. Finally, efficacy studies must evaluate the intervention in the ways and contexts in which practitioners are applying it in order for the research to be most useful.
We begin by examining research that explores the concept of sensory integration. We emphasize research examining sensory integration intervention as an approach that focuses on remediating underlying etiology or neurologic dysfunction rather than on direct instruction or skill development. We have also included a review of research on central nervous system (CNS) plasticity and hierarchy, two important assumptions supporting sensory processing is discussed. Second, we discuss research examining the concept of sensory integrative dysfunction. Third, we present efficacy studies evaluating the outcomes of sensory integration interventions. Finally, we conclude by discussing challenges related to research in the area of sensory integration and include ideas and priorities for furthering our knowledge in this area.
Sensory Integration: A Process Approach
In the educational literature, a process approach focuses or remediating underlying neurologic on mental processes believed to be contributing to a child’s inability to learn or perform a specific skill as opposed to approaches that directly teach skills. In the late 1960s an 1970s, a number of process approaches, including sensory integration, perceptual- motor (Frostig, 1967; Kephart, 1971), and neurodevelopment therapy (Bobath, 1980) were popalla with practitioners intervening with children withneurologic disorders (e.g., learning disabilities and cerebral palsy). Process approaches are based on the belief that the functioning of specific neurologic systems (i.e., those responsible for sensory processing, motor coordination, and sustaining attention was required for adequate cognitive development. After the disordered underlying processes were corrected, academic learning could take place normality. (Hammill, 1993).
Sensory integration con be viewed as a process approach. Through the provision of sensory experiences, within the context of meaningful activity, and the production of adaptive responses, CNS functioning improves. This improvement ultimately leads to better performance in any number of functional areas, including behavior, academics, and motor skill (Fisher Munrray, 1991). However, in the mid-1970s and 1980s, researchers (Goodman & Hammill ; Kavale & Mattson, 1983) conducting efficacy studies evaluating process-oriented inventions in education concluded that these approaches were largely ineffective. This is began to question whether underlying neurologic processes were related to cognitive abilities or academic performance. Eventually, on- task, direct instruction forms of assessment and remediation replaced process approaches. “For the moment, the issue of process training is resolved, and direct instruction has emerged as the model of choice for the remediation of learning disabilities in the United States” (Hammill, 1993, p 303).
Sensory integration was not tested specifically in these studies, and most of these studies were limited to the examination of process approaches with children with learning disabilities. Nonetheless, thedescription of sensory integration simply as a process approach has probably contributed to the negative views of some researchers, particularly in the field of education.
Those who accept effectiveness of intervention based on sensory integration theory believe first that CNS has the capacity for change or remediation. Neuroplasticity is the assumption that the CNS has the capacity for change or is able to modify its structure and function (Lenn, 1991). Second, proponents of sensory integration theory believe in a certain level CNS hierarchy because it is assumed that positive change or adaptation of lower (brainstem level) areas may result in improved higher cortical functions, including any number of functional areas and capabilities. Research related to these two basic assumptions on which sensory integration is based is examined below.
Sensory Integration: A theory Based on Central Nervous System Plasticity and Hierarchical Concepts.
There is an abundance of research that support the concept of neuroplasticity (Bach –y –Rita, 1980; Leen, 1991; Lund, 1978; McEachen & Shaw, 1996; Stein et al., 1974; Stephenson, 1993; Szekely, 1979). Neuroplasticity is evident throughout development; it involves natural changes in the nervous system that occur during maturation. It also is a reactive process that occurs during the recovery form an injury to the CNS (Schaaf, 1994a). Because the majority of individuals who are treated whit sensory integration procedures are children with chronic, developmental conditions (i.e., pervasive developmental disorders, attention deficit hyperactivity disorder, learning disabilities) and because sensory integration theory is not designed toexplain adult-onset deficits (see Chapter 1), research that examines developmental neuroplasticity is most relevant.
Some of the most importante factors that promote developmental neuroplasticity are the inner drive to seek, create, challenge, and master the environment (Aoki & Siekevitz, 1998; Parham & Mailloux, 2001), the just – right challenge are fundamental characteristics of intervention based on sensory integration theory (Ayres, 1979) and may be keys to why it works. Therefore, research that examines such factors provides us with helpful guidelines neural organization through sensory integration intervention.
Neuroplasticity takes place through a number of specific neural mechanisms, such as an increase in myelination and synaptic efficiency and dendritic aborization. Explanations of such mechanisms are beyond the scope of this chapter; however, Schaaf (1994a, 1994b) provided an overview of such mechanisms and how they relate to our understanding of how and why intervention based on sensory integration theory may be effective. Particularly relevant are the ways in which we can enhance neural organization and integration through the types of purposeful activities used during intervention.
Although the idea that a more integrated, more efficient CNS results in improved performance in functional skills seems very logical, this issue has been at the core of many debates. Unlike research that supports CNS plasticity (which is very convincing), the connection between improved sensory processing and functional skills and research examining brain function as a hierarchical process provide useful information in examining this connection.
Case- Smith (1995)explored relationships among sensorimotor components, fine motor skill, and functional performance in 30 preschool children. The sensorimotor components included two measures of sensory processing. One of these, tactile defensiveness, correlated significantly with measures of fine motor skills. However, the same measure showed only a weak relationship to functional social, play, and self-care skills. Case- Smith concluded that practitioners should not assume that functional skills improve when gains are made in underlying sensorimotor skills. Furthermore, other social, cultural, and environmental factors may be just as important for children to learn, perform, and generalize functional skills. She encouraged practitioners to consider such contexts in their interventions with children.
Sensory integration largely focuses on the remediation of brainstem functions as a way to improve functional skills (Ayres, 1972b). This hierarchical view understands the CNS as having vertically arranged levels that are interdependent yet reflect a trend of ascending control and specialization. This hierarchical approach led Ayres (1972b) to believe that the more primitive or subcortical systems such as the tactile, the vestibular, and the proprioceptive systems provide the foundation for the development of higher-order cortical functions such as academic ability, complex motor skills, and the development of social skills. This relationship is, however, controversial because the indirect correspondence between the stimulus (i.e., improved brainstem functioning) and performance (e.g., motor skill, academic achievement) is, for the most part, unobservable. More importantly, over the past10 to 20 years, a more popular view of brain function, one that views the brain as a more integrated, holistic system, has emerged (Cohen & Reed, 1996).
Sensory integration theory has evolved to incorporate this view, and today it is consistent with a more holistic view, and today it is consistent with a more holistic view of brain function. Fisher and Murray (1991) described sensory integration theory as being based on a systems view, rather than viewing sensory integration problems are believed to result from a number of interrelated systems that are not functioning optimally. Each part of the system performs a different role. Some are control centers; some fine-tune the instructions; some carry feedback messages. The extent of give and take from each part supports a holistic rather than a strictly hierarchical view.
The amount of literature that describes interventions that apply sensory integration techniques in conjunction with other approaches has increased in recent years. This literature suggests that sensory integrative dysfunction is only one of many problems that contribute to a child’s difficulty in performing or learning tasks effectively. Today, intervention based on sensory integration theory is rarely provided isolation; it is more often combined with other approaches, including the teaching of specific skills (Case- Smith, 1997) and promoting positive play behaviors (see Chapter 15). Intervention activities providing enhanced tactile and vestibular sensations are often used in conjunction with activities that tap higher cortical systems such as those involved in forming the idea about how to perform motor tasks and with practice of specificfunctional skills in context (see Chapter 3).
A more holistic intervention approach does however, dispute the emphasis that sensory integration has on the remediation of primitive sensory systems or the importance of healthy sensory integration for performing functional skills. Rather, in has placed this emphasis in context of a more holistic and interactive view of brain function and question our ability to separate subcortical from higher-cortical functions. Subsequently, the use of other approaches in combination with intervention based on sensory integration theory is encouraged (see also Chapter 1).
Sensory Integration: A Component of Sensory Processing
There has been some confusion related to the use of terminology associated with sensory integration. The inconsistency in the use of this terminology has made interpretation rather difficult. The term sensory integration has been used to describe a nervous process (occurring at the cellular or nervous system level), a behavioral process (observable behaviors that result from these processes), and a frame of reference useful for assessment and intervention. Thus, researchers must differentiating clearly between what they observe in children and what they infer happening within the nervous system. Miller and Lane (2000) addressed the meant to establish a consensus for the use of various terms particularly those associated with neurophysiological processes, including peripheral sensory processes and central sensory processes (see also Chapter 4).
Sensory processing is one term that has because popular, and become it is often used interchanges ably whit sensory integration, the differentiation between these two termsrequires clarification. Both terms represent theoretical frameworks for explaining the same types of functional deficits and behaviors observed in children. However, sensory processing is a more global, encompassing construct than is sensory integration. Sensory processing is the way in which the central and peripheral nervous systems mange incoming sensory information from tactile, vestibular, proprioceptive, visual, auditory, olfactory, and gustatory sensory systems. According to Miller and Lane (2000), the “reception, modulation, integration, and organization of sensory stimuli, including the behavioral responses to the sensory input are all components of sensory processing” (p. .
A conceptual model of sensory processing proposed by Dunn (1997) described the relationships among a number of neurobiological factors, including sensory registration (i.e., how one regulates incoming sensory information), and habituation and sensitization (i.e., whether the CNS reacts to or ignores incoming sensory information; see Chapter 7). In addition, Dunn attempted to link sensory processing with the ways in ways individuals perform daily life activities. Consistent with this more global use of the term sensory processing, DeGangi (1991) reported that the sensory processing problems of ported that the sensory processing problems of post-institutionalized children often lead to problems with sensory integration. Therefore, sensory integration should be viewed as only one component of sensory processing. Researchers must make explicit their use of various terms, and consumers of research must attend to the ways in which various terms are used.
RESEARCH EXMINING SENSORY INTEGRATIVEDYSFUNCTION
Sensory integrative dysfunction is a diagnostic label that is relatively unpopular outside the profession of occupational therapy (Missiuna & Polatajko, 1995), and it is not included in the Diagnostic and Statisyical Manual of Mental Disorders (DSM-IV; American Psychiatric Association, 1994.). Rather, sensory integrative dysfunction is often associated with the underlying sensory processing problems characteristic of any number of diagnostic conditions included in the DSM-IV. The most common DSM-IV diagnostic groups that coexist with sensory integrative dysfunction are pervasive developmental disorders (e.g., autism) (Kientz & Dunn, 1997; McIntosh et al., 1999), attention deficit-hyperactivity disorders (Mulligan, 1996), learning disorders (Ayres, 1979), fragile X syndrome (Miler et al., 1999), and developmental coordination disorder (Missiuna & Polatajko, 1995). Although not all children in these diagnostic groups have evidence of symptoms associated with it. Based on research to date, it also is not correct to assume causative relationships between sensory integrative dysfunction and these other conditions when the do coexist. The same underlying neurologic deficit may well cause both conditions.
Sensory Integrative Dysfunction and Other Diagnostic Labels
The issue of classification of children has been debated over the years, and the confusion of “labeling” has led to problems both in identifying appropriate children for specific interventions and in selecting subjects for research purposes. A review of four terms: sensory integrative dysfunction, clumsy child syndrome, developmental dyspraxia, and developmental coordination disorder byMissiuna & Polatajko (1995) concluded that these four terms should not be used interchangeably and that clear definitions and characteristic feactures of each need to identified.
Typically, children are labeled as having sensory integrative dysfunction when they do poorly on tests specifically designed to measure sensory integration (Missiuna & Polatajko, 1995). Such tests, such as the Sensory Integration and Praxis Test ( SIPT; Ayres, 1989), measure the processing of basic sensory systems (i.e., tactile, propioceptive, vestibular) as well as visual-perceptual and motor planning functions. Specific types of sensory processing deficits are beginning to be identified in children with different diagnostic classifications. This research provides valuable information about the nature of dysfunction characteristic of certain types of children. For example, Mulligan (1996) indentified specific patterns of sensory integrative dysfunction in children with attention disorders defensiveness specific to children and adults with developmental disabilities. Kientz and Dunn (1997) described the sensory processing of children with autism, and Demaio- Feldman (1994) identified somatosensory processing deficiencies of low-birth-weight infants at school age.
The Zero to Three Diagnostic Classification Task Force included a diagnostic classification called multisystem developmental disorders in the Diagnostic Classification of Mental Health and Developmental Disordesr of Infancy and Early Childhood (DC 0-3) (as cited by Wieder, 1996). These multisystem disorders are consistent with what has been described as characteristics of sensory integrative dysfunction because they view avariety of functional problems as being secondary processing deficits rather than as primary deficits. The different types of multisystem disorders, however, place more emphasis on sensory modulation (a component of sensory processing) than on sensory integration. For example, the three types of multisystem developmental disorders are a hypersensitive type (type 1), a hypersensitive type (type II), and a motorically disorganized and impulsive type (type III). Research by Miller and colleagues (personal communication, March 15, 2000) is exploring the possibility of sensory modulation disorders as a separate diagnostic entity. This research group has indentified and demonstrated types of sensory modulation problems in children with autism and fragile X sindorme, using both traditional measures and neurphysiological measures such as electrodermal responses (McIntosh et al., 1999; Miller et al., 1999)
Types of Sensory Integrative Disorders
Ayres (1972s) believed that many different types of sensory integrative disorders existed and that each was associated with dysfunction in a particular neural substrate. She then developed a typology of sensory integrative dysfunction based on a series of multivariate analyses (Ayres, 1989; see also Chapter 1 and Appendix B). Sensory integrative dysfunction, therefore, has been conceptualized as multidimensional. The identification of different types of sensory integrative dysfunction assists with the understanding of underlying etiology and, more practically, is useful for developing specific intervention for the discrete patterns identified.
Categorical systems for conceptualizing sensory integrative dysfunction, largely based on Ayres’factor analytic work in the 1960s and 1970s, are described by Parham and Mailloux (2001), Fisher and Munrray (1991), and Ayres (1989) (see also Chapter 1 and Appendix B). Although there is not perfect consensus on the best way to categorize the patterns of dysfunction, there are recurring themes, and much commonality expressed by all authors.
The exploratory analytic studies on which these patters of dysfunction were based, however, must be interpreted with caution and can be criticized appropriately for limitations in design (Cummins, 1991; Hoehn & Baumeister, 1994; Parham & Mailloux, 2001). Because Ayres was constantly exploring new ideas, she used a different battery of tests in each study. Therefore, none of the studies was a true replication of a preceding one. Furthermore, her samples were heterogeneous and consistently small in number relative to the number of test scores that were analyzed. Terminology used to describe the factors that emerged in these studies was also inconsistent; therefore, comparing results from these studies and drawing conclusions based on their combined contributions are difficult. Nonetheless, the practical implications of these patterns have been very important for our understanding of the nature of the sensory integrative dysfunction seen in children and in interpreting the results of children on the SIPT (Ayres, 1989).
Mulligan (1998) attempted to validate a five factor model of sensory integrative dysfunction based on the SIPT scores of large, heterogeneous group of children. The model tested was consistent with current views of patterns of sensory integrative dysfunction and included a bilateral integration and sequencingpattern, a somatosensory pattern, and a postural-ocular movement pattern. Although the results supported the hypothesize model as reasonable, a number of weaknesses were indentified with it that supported further analyses of alternative models. One of the most important findings was the very string relationship among all of a higher-order general factors, which initially as termed generalized practic dysfunction. (Editors` note: Mulligan later agree that her higher-order factor may reflect a general inefficiency of CNS functioning, particularly in the areas or systems measured by the SIPT. It might be more appropriate to label this higher-order general factor general sensory integrative dysfunction. [Mulligan, personal communication, June 1, 1999].).
IN view of this research, rather than purport the existence of separate patterns of dysfunction (related to dysfunction in particular neural substrates), as previous studies have done, Mulligan suggested that specific patterns of dysfunction (based on deficient SIPT scores) should be viewed as extensions of generalized sensory integrative dysfunction (see Chapter 1 for more specifics on Mulligan’s factor-analytic study)
The idea of a general factor emphasizes the complexity of our CNS and supports the systems or holistic view of the CNS discussed earlier in this chapter. In examining construct validity of the SIPT, Lai et al. (1996) provided evidence that praxis is a unidimensional construct and that both bilateral integration and sequencing and somatopraxis were a part of this unidimensional construct. Although their study did not examine all patterns of dysfunction believed to comprise sensory integrative dysfunction,these result indicated a shift in thinking regarding the multidimensionality of sensory integrative dysfunction.
Mulligan (1998) also found that the SIPT was not sufficient to detect problems related to postural functioning, supporting Fisher and Bundy (1991), who discussed the importance of using other clinical observations such as examining equilibrium reactions and antigravity postures to determine postural problems (see also Chapter 7). Somatopraxis also did not emerge as a separate pattern (Mulligan, 1998) as it had in previous studies (Ayres, 1966, 1971, 1977; Ayres et al., 1987). A relationship between sonmatosensory processing and praxis was evident; however, unlike previous models, which identified somatodyspraxia as a separate pattern of dysfunction, the relationship between these two areas was explained by the presence of the general factor (Mulligan, 1998). In view of this new information, Mulligan suggested that therapists be cautions when identifying a child as fitting one of the five specific patterns of dysfunction (based on SIPT scores) previously indentified.
In summary, a multitude of diagnostic labels is used to describe children with sensory integrative dysfunction, and care must be taken not to use various terms interchangeably. When conducting research, coexisting diagnoses of study participants need to be identified and reported so that the effects of intervention for different types of dysfunction can be examined and considered in the interpretation of results. Although it appears that the multidimensional view of sensory integrative dysfunction is shifting to a more unified model, a consensus on the best way to conceptualize it has not beenreached. Further understanding of how sensory integrative dysfunction relates with other models of sensory processing is necessary.
RESEARCH EVALUATING INTERVENTION BASED ON SENSORY INTEGRATION THEORY
In applied research, the definition of intervention based on sensory integration theory, as an independent variable, has been inconsistent. This is problematic when one attempts of intervention. Rosenthal and Rosnow (1984) defined and independent variable as an observable or measurable event manipulated by a researcher to determine whether there is any effect on another event (i.e., the dependent variable). Research based on a traditional definition of sensory integrative- based intervention often requires that the independent variable or intervention be administered to every individual in precisely the same manner. Clearly, intervention based on sensory integration theory defies a simple definition.
Integration based on sensory integration theory is complex because the way in which it is administered depends on the individual needs of a client (Ottenbacher, 1991). In addition, the application of intervention as it was conceptualized classically must be distinguished from more current applications that involve the use of indirect, consultative models, and the use sensory integration principles and activities in combination with those from other frames of reference. Because this distinction is so important, research evaluating each type, “classic” and “nontraditional”, will be addressed separately.
Studies of Classic Sensory Integration Intervention.
The classic form of sensory integration intervention tends to be practiced more often in private, medically orientedclinics than in educational environments. Such intervention is highly specialized, and it has been recommended that only therapists who have received advanced training administer it. Kimball (1988, 1999) described several characteristics of sensory integration intervention. These can be summarized as follows.
* The goal of the intervention is to facilitate appropriate physical and emotional adaptive responses by improving CNS processing rather than to teach specific skills.
* Intervention activities are individualized and at the upper levels of the client`s capacity.
* Intervention is administered by a vigilant practitioner who provides constant feedback.
* Intervention involves purposeful activities that are client directed and result in an adaptive response.
* Intervention activities provide enhanced proprioceptive, vestibular, and tactile sensation.
Although these characteristics provide guidelines for defining the intervention, they are very broad, and there is room for variability within each of them. However, interventions that involve the application of sensory stimulation only, structured group interventions, combined approaches, use of sensory integration in consultative models, or intervention that applies sensory integration principles during the practice of specific functional skills are not considered classic sensory integration intervention. Although all of these approaches are very appropriate for some children, in a research context, care must be taken to differentiate them from classic sensory integration intervention.
In the 1970s and 1980s, many studies were conducted specifically evaluating the effectiveness of sensory integrationintervention, with mixed results. Many detailed reviews of this literature are available (Cermak & Henderson, 1989; Hoehn & Baumeister, 1994; Mulligan, 2001; Polatajko et al., 1992, Vargas & Camilli, 1999). These early studies demonstrated that sensory integration therapy improved performance in motor, language, and academic areas (Ayres, 1072a, 1972c; Magrun et al., 1981; White, 1979). Ottenbacher (1982) performed a meta-analysis of eight intervention effectiveness studies, which provided support for sensory integration in the remediation of motor, academic, and language functions, whit the most improvements noted in the motor area. Specifically in relation to children whit learning disabilities, he reported that “the average learning disabled students receiving sensory integration therapy performed better than 75.2 percent of the learning disabled subjects not receiving therapy” (Ottenbacher, 1982, p.576).
There have not been many group studies in the past 10 years evaluating the outcomes of sensory integrations interventions in comparison with other approaches such as tutoring (Wilson et al., 1992) and perceptual- motor training (Humphries et al., 1992). The more recent studies, however, improved on previous studies in terms of methodological rigor. These studies concluded that sensory integration therapy is not any more effective than these traditional approaches. Polatajko et al. (1992) reviewed seven two- and three- group experimental studies conducted from 1879 to 1992 using sensory integration intervention with samples of children with learning disabilities. They concluded that the results of previous studies do not indicate that sensory integrationintervention improves the academic performance of children with learning disabilities more than placebo does. With respect to sensory or motor performance, the results were inconsistent and indicated overall that sensory integration intervention may produce minimal positive effects. However, the generalization of these studies is limited because study subjects included children with learning disabilities only and because the nature and extent of the subjects’ sensory integration dysfunction were often not known. No negative effects sensory integration interventions were ever reported.
Wilson Kaplan (1994) conducted follow up study of the children who had received either tutoring or sensory integration therapy in their 1992 study. The concluded that the children who received sensory integration intervention performed better on tests of gross motor function than the children who received the tutoring intervention. No significant differences were found on tests that measured reading skills, fine motor skills, visual motor skills, or behavioral factors. Allen and Donald (1995) conducted a pilot study to determine the effect of occupational therapy intervention (based of sensory integration theory) on five children wed documented sensory integrative dysfunction. There case studies revealed that four of the five children receiving the therapy improved in the area of motor performance- the one participant who did not show motor gains was older than the other children (is age 11 years compared with ages 5 to 8 years).
In a meta-analysis of 32 experimental group studies (16 comparing sensory integration intervention with other interventions and 16 comparing sensory integrationintervention with no intervention), Vargas and Camilli (1999) expressed there main conclusions. First, when compared with a control situation, the effects of intervention based on sensory integration theory showed positive results in earlier studies but no differences in later studies. Second, effect sizes for measures of motor performance and cognition such as IQ and academic achievement were greater than those for measures of behavior, language, and sensory and perceptual ability. Third, overall intervention based on sensory integration theory was as effective as various alternative intervention approaches.
One way of minimizing or controlling the variation in complex intervention approach when evaluating it is to reduce the intervention to a small number of standardized and strictly controlled activities. The advantages of doing so are that in allows the researcher to examine the effectiveness of specific components of the intervention while making it easier to operationally define the independent variable. The main disadvantage in doing so is probably not very representative of the actual intervention as it is typically practiced, nor of the construct itself.
Reductionist definitions of sensory integration intervention are common in the literature (Jarus & Gol, 1995; Ottenbacher et al., 1981; Ottenbacher, 1991; Well & Smith, 1983). Altough not specifically defined as sensory integration therapy, De-Gangi et al. (1993) compared a therapist-directed sensory motor intervention approach. They concluded that the children-directed approach, which is more characteristic of sensory integration, was better than the other approach for the development of fine motorskills but less effective for the improvement of gross motor skills, sensory integration functions and functional skills. One must take care in interpreting this study because it measures a certain component of sensory integration intervention (i.e., child directed versus therapist directed) and not the intervention itself.
Tickle-Degnen and Coster (1995) examined another component of intervention based on sensory integration theory in the context of intervention sessions. With the use of videotape, they examined a social element, the nature of therapist-child interactions during sensory integration intervention. This social element is one of the most important facets of sensory integration intervention; Ayres emphasized that achieving positive results depends on the involvement of a skilled practitioners who can match both input and adaptive demands to the child`s current needs and capacities. Kaplan et al. (1993) also suggested that the bond between a therapist and child may account for some of the positive perceptions regarding intervention outcomes.
In summary, efficacy research of sensory integration in its “purest” form provides conflicting results. Overall, the results provide little supporting evidence that the intervention is helpful in the remediation of learning or academic difficulties of children. There is some support, however, that it improves sensory processing; improves some behavioral problems in children; and, most clearly, improves gross motor performance in children with learning, behavioral, or motor problems. Results also indicate that intervention based on sensory integration theory is most effective with children who have been specificallyidentified with sensory integration problems and that the intervention appears most effective with younger children.
Studies of Nontraditional Sensory Integration Intervention
Although the classic approach is still used with some children, intervention based on sensory integration theory is more often modified or used in conjunction with other approaches. For example, practitioners working with children in educational settings have modified the way in which they use sensory integration as a frame of reference to meet the requirements of special education law and to accommodate the specific needs of students. Additionally, they have tried to join the move of educators from process approaches toward more direct instruction models and toward inclusive education (i.e., educating students with special needs in regular classroom environments and within the context of regular classroom activities). For example, the used of consultative services has increased over the past 10 to 15 years (Dunn, 1988; Kemmis & Dunn, 1996) and occupational therapists are applying sensory integration principles to help students to be successful with classroom activities already a part of the typical curriculum (Mulligan, 1997).
There is osme empirical evidence supporting the efficacy of consultative models (Davis & Gavin, 1994; Dunn, 1990; Kemmis & Dunn, 1996), which often include a component of sensory integration theory. For example, Case- Smith (1997) found that one of the most important factors in successful interventions was the practitioner`s ability to reframe a child`s classroom behaviors using sensory integration theory. Sensory integration was reported to be the primary frameof reference for five or the 13 children examined in her study.
In addition to the increased use consultation, practitioners often combine sensory integration interventions with other approaches, including the provision of adapted equipment, gross and fine motor skill training, self-care skill training, and parent and teacher support, as part of child`s total intervention program. Few studies however, have examined the effectiveness of sensory integration with other approaches. This is not surprising. As with consultation, it is very difficult to insolate the contributions of the specific components of combined approaches to the ultimate outcomes of the intervention.
A promising study by Case-Smith et al 1998) demonstrated the effectiveness of school-based occupational therapy services on fine motor skills and other functional outcomes of preschool children with developmental delays. In this study, 44 children with developmental delays and 20 children without delays were pretested with a number of outcome measures at the beginning of the school year and posttested at the end of the school year. The intervention approaches varied depending on the individual needs of each preschooler. The use of activities characteristic of intervention based on sensory integration theory was very common (40.4% of sessions emphasized motor planning and tactile activities; 31.7% emphasized vestibular and propioceptive activities) as was the use of visuomotor and manipulations activities (81% of sessions). The results of this study supported the use of sensory integration techniques in combination with other approaches for improving fine motor skills, and functional skills ofpreschoolers with developmental delays.
Operationally defining sensory integration as an independent variable has not become any easier in the past 10 to 15 years. Rather, with the increased use of sensory integration in consultative models and in combination with other approaches, it may very well be more difficult to isolate and define than it was previously. We do, however, have guidelines that can be used (Kimball, 1988, 1999) to help researchers define sensory integration intervention in its classic form (see also Appendix A). Examining components of intervention is also a technique that has been used to further our understanding of the effects of certain qualities of the intervention. Finally, the use of experimental designs that do not attempt to control the intervention, as in the study design used by Case-Smith et al. (1998) and supported by Bower and McLellan (1994), appear to be a useful way to evaluate the effects of complex, multifaceted, and individualized interventions.
DIRECTIONS FOR FURTHER RESEARCH
A number of scholars in the field (e.g., Cermak & Henderson, 1989; Kaplan et al., 1993; Mulligan, 1997; Ottenbacher, 1991; Tickle-Degnen, 1988) have provided direction for future research related to sensory integration theory and intervention. Research questions range from the basic premises on which the theory and hypothesized dysfunction are base to the efficacy of intervention as it is practiced in its varied forms.
For professional consensus to be reached regarding the usefulness of intervention based on sensory integration theory, it must be better defined and conceptualized. Basic research that explores the theoretical bases of sensory integrationincluding the underlying mechanisms on which it is based, is needed (Tickle-Degnen, 1998). Such research is useful for determining how and why intervention works rather than if it is effective. Further understanding of the ways in which sensory integration relates to other information processing models, such as those described in the education and psychology literature (See Swanson, 1987) and other sensory processing models (e.g., Dunn. 1997) is also necessary. Such work may someday allow researchers and theorists to converge on a unified paradigm for the study of individuals with sensory integrative dysfunction.
Second, we must better understand the individuals for whom intervention is useful and be better able to describe participants in research samples. Specific tests of sensory integrative dysfunction a recommended to identify individuals indentified as having sensory integrative dysfunction should be used to evaluate intervention based on sensory integration theory. Research examining specific patterns of sensory integrative dysfunction may also be helpful in identifying intervention best suited for certain individuals. Furthermore, coexisting conditions, including other diagnoses and characteristics of participants in studies, should always be reported so the influence of multiple variables can be considered in the interpretation of results.
Third, psychometrically sound tests of sensory integration are necessary for indentifying appropriate candidates for intervention as well as for measuring the effects of intervention, supporting research in the area of test development. Mulligan (1998) suggested that a new, shorter test emphasizing the identification of praxisproblems and the underlying sensory integration functions that may be contributing to the praxis problems be considered. In addiction, further tests of vestibular function and tests that measure sensory modulation need to be developed because it is believed that these are important aspects of sensory integration that have not yet been captured adequately within a standardized tool (Mulligan, 1998) Mulligan also reported that a tool that provides an overall score of sensory integrative dysfunction would be helpful to identify clearly whether dysfunction exists and what the overall severity of the dysfunction is.
The specific goals of intervention based on sensory integration theory and expectations for areas of gain differ for various individuals. Therefore, the dependent measures selected should be consistent with these expectations and be sensitive enough to detect changes. Cermak and Henderson (1989) reported that research is greatly hampered by the lack of good measures of intervention effectiveness (see also Chapter 1). Ways to document both the short-and long-term effects of intervention are needed. In addiction to measures evaluating motor, language, academic and behavioral outcomes, measures of the effects of intervention on occupational performance (e.g., play) and performance components (e.g., attention, organization, and affect) are need (Cermak & Henderson, 1989). Based on their research examining parental hopes for therapy outcomes, Cohn et al. (2000) discussed the importance of including both child-focused measures (i.e., self-regulation, perceived competence, and social participation) and parent outcomes (i.e., validation of their child’s problems andcompetency in applying strategies to assist their children) when measuring the effectiveness of intervention.
Fourth, the complexity and individualized nature of intervention based on sensory integration must be carefully considered when evaluating the effectiveness of the intervention. For traditional experimental designs, sensory integration intervention must be better defined and controlled. Kimball’s (1988, 1999) description is useful for identifying the basic and necessary components and characteristics of classic intervention. Specific questions researchers can ask themselves when operationally defining interventions have been outlined by Miller and Kinnealey (1993). For example, researchers must determine how much of the child’s responses. Protocols that describe the interventions need to develop and reviewed by experts in the field, and procedural reliability checks should be implemented to ensure the consistency and accuracy of the application of the protocols. (See also Appendix A for the STEP-SI protocol developed by Miller and her colleagues).
Reductionistic definitions of intervention based on sensory integration theory, as note earlier, can also be used to minimize intervention variation and allow the researcher to insolate and examine the effectiveness of specific components of this multifaceted approach. However, car must be taken when generalizing the results of these studies because the sum of the effects of individual components of an intervention do not necessarily add up to the total effects of the intervention.
If interventions are based on a consultative model or are used in combination with other approaches, then all approaches need to beindentified, descried, and accounted for in the study. It is, however, not always necessary to control the intervention in order to evaluate it. Case-Smith et al. (1998) demonstrated how researchers are able to evaluate therapy services being individualized and provided in natural contexts. Miller and Kinnealey (1993) cited single subject experimental designs as well as case studies and other qualitative approaches as being particularly relevant when studying the effects of sensory integrative interventions. Rather than masking individual differences and attempting to produce homogeneous groups, these approaches explore the effects of individual differences, allow for individualized intervention, and are useful in determining for whom intervention is most helpful. Regardless of the research method, however, it important for investigators to carefully describe the delivery of the intervention and monitor its implementation throughout the research process.
SUMARY AND CONCLUSIONS
Sensory integration as a frame of reference has provide a useful framework for assisting occupational therapy practitioners, parents, and teachers in understanding children’s behavior in ways that make sense. Despite contradictory empirical evidence regarding the effectiveness of this approach over the years, it has remained very popular. We must not lose sight of the fact that sensory integration has a great deal in common with more global and accepted frames of reference such as developmental approaches and information processing. This commonality with other approaches enhances its credibility and the comfort level of the practitioners who choose to use it. Sensory integration as a frame ofreference takes a unique perspective in organizing largely accepted view regarding the way in which the CNS works, what we know about child development, and how one learns and process sensation.
Developing consensus regarding the validity of sensory integration will require a creative synthesis of past and present empirical efforts. Such efforts require cooperation of multitude of researchers from various backgrounds and expertise in diverse research methodologies. The science of sensory integration is still in its infancy, and no single research approach has emerged as the methodology of choice in establishing empirical consensus. The absence of a unifying research paradigm is a function of the highly complex subject that sensory integration is. The exploration and application a multitude of research approaches should be encouraged and is viewed as a positive development.
As concerned, caring professionals, we must believe in the therapeutic merit of what we do. At the same time, we must acknowledge the limitations of the scientific evidence that support the outcomes of the interventions that we strive to achieve with our clients. There is scientific evidence to support the use of intervention based on sensory integration theory for children with sensory processing disorders, and there is scientific evidence to dispute its use. Careful monitoring of the progress of our clients is, therefore, crucial because we must remain accountable. Finally, we must all take some level of responsibility for furthering our knowledge base and our understanding of sensory integration. As a part of this process, we must value research as an integral part of practice.