Pseudoscience is defined by characteristics including the use of vague, exaggerated or untestable claims, over-reliance on confirmation rather than refutation, lack of openness to testing by other experts, lack of progress, and personalization of issues (see Wikipedia article which is a nice summary). As the article notes, however, it is important to distinguish protoscience with pseudoscience.
At this point in time, 35 years after Jean Ayres wrote Sensory Integration and Learning Disorders, I am willing to consider that this concept of sensory integration theory is no longer 'nascent' and that we have not made appropriate progress toward researching and validating the theory as it is now constituted. For perspective, in 1972 when that book was published, the first hand-held calculator was marketed. Has science not moved forward since that time?
This is not to say that children don't have learning problems, or that they don't have problems in processing sensory information that leads to functional and behavioral problems. What it means is that we need to toss out the proverbial bath water without throwing out the baby when it comes to sensory integration theory.
Why do we need to revisit this? Sensory integration theory does not EXPLAIN why children behave the way they do. Sensory integration theory does not help us PLAN intervention. Sensory integration theory does not PREDICT how behavior will change as a result of intervention.
The most recent textbooks still state that "sensory integration cannot be observed. (Bundy & Murray, 2002, p.3). When I went to school as an undergraduate in the early 1980s I was told that there may be a neurochemical basis to sensory integration difficulties. It was an exciting concept that I was interested to see develop. Twenty five years later after completing my doctorate there is still no definitive cause for the disorder that can be identified.
This matters because there is a credibility gap when an occupational therapist states that a child has SI problems. Therapists have mistakenly believed that if we change the names to SID, or DSI, or SPD that this will somehow change people's minds about the theory. In reality we can call it chicken soup or anything we want - if the theory can't be validated then the name is irrelevant.
Occupational therapists are continuing to dialogue about classification and naming in sensory integration theory. Now we are making distinctions between sensory-seeking vs. sensory avoiding typologies. Or distinctions between active vs. passive coping styles. Ad nauseum. Less attention is paid to neurophysiological underpinnings and basic science research that would actually DO something to elucidate the nature of the disorders.
There is NO LACK of basic science that applies. Why aren't we studying neuronal migration (see Chang, et.al. 2005) or undetected brain damage (see Looney, et.al., 2007)?? Instead we focus on endless classification and naming of some 'unseen process' that we can't define. Then we state that by engaging the child in a series of experiences that they will somehow improve - but here we begin to disagree, with some people advocating passive and prescriptive sensory input and others advocating child-directed approaches. So as a consumer (parent, insurance company, etc) imagine their confusion and hesitancy when they go to some OTs who tell them that their child has some alphabet soup disorder (DSI, SID, SPD, whatever) that can be treated by either active or passive strategies - depending on who you talk to - and that we can't define what is REALLY wrong but we just know that these active or passive strategies will help the brain to function normally.
I wouldn't buy it either.
What I would buy is going to a therapist who tells me that learning disorders often have a physiologic or genetic component and that these causative factors can vary between children. Some of our research points to brain damage, some to neuromigration deficits, and some we just don't understand yet. Either way, the distinct problems that the child has as a result of the organic neurologic problem can be quantified by performance on standardized testing that looks at motor, perceptual, and other cognitive factors that impact occupation. When children have known deficits that we can quantify or measure, we can gather baseline data so that we can measure the effect of our intervention and training attempts. Intervention can involve participation in sensory-based and other activities, but we still need to quantify our progress with a behavioral measure.
After we proceed this way we are left holding data that quantifies the nature of the problem. We have basic science that could be analyzed to determine if there is a 'fit' between the identified organic problem and the clinical signs that are seen. We also have intervention strategies that can be evaluated in terms of progress on standardized measures.
Contrast this to our current state of affairs where all we have is a loosely defined set of behavioral observations, no real science to back it up or validate our intervention strategies, and a pile of people (OTs and non-OTs alike) with products rushing to market music CDs, weighted vests, brushes, and yet another sensory integration book that parents will buy because they are desperate to find answers for the problems that their children are having.
We can do much better.
This is why I now purposefully label sensory integration theory as pseudoscience.
Ayres, A.J. (1972) Sensory integration and learning disorders, Los Angeles: Western Psychological Services.
Bundy, A.C., & Murray, E.A. (2002). Sensory integration: A. Jean Ayres' Theory Revisited in A.C. Bundy, S.J. Lane, & E.A. Murray (2002) Sensory integration: Theory and practice, 2nd ed., Philadelphia: F.A.Davis.
Chang, B.S., Ly, J., Appignani, B., Bodell, A., Apse, K.A., Ravenscroft, R.S., Sheen, V.L., Doherty, M.J., Hackney, D.B., O'Connor, M. et al. (2005) Reading impairment in the neuronal migration disorder of periventricular nodular heterotopia. Neurology, 64, 799–803.
Looney, C.B. et al. (2007). Intracranial hemorrhage in asymptomatic neonates: Prevalence on MR images and relationship to obstetric and neonatal risk factors. Radiology, 242, 535-41.