DbI Review

Repetitive Behaviours In Children With Sensory Impairments And Multiple Disabilities
Heather Murdoch

Many children with sensory impairments and multiple disabilities show repetitive or stereotyped behaviours - for example, rocking, head shaking, mouthing objects or repeating sounds. Those working with the children may be unsure how best to respond, particularly when programmes emphasise communication but the child's behaviours appear self-directed. Heather Murdoch recently completed her PhD thesis at the University of Birmingham on this topic, looking at the roles which repetitive behaviours may play in children's development and the ways in which educators interpret and respond to them. Repetitive behaviours are very common in children with multiple disabilities and sensory impairments, and in some cases dominate the child's choice of activities. Expert advice on how to respond has differed over time: from using behavioural techniques to reduce the behaviours, to finding activities which the child finds more interesting (often a difficult task), to addressing the development of communication skills in the belief that repetitive behaviours will reduce as the child's communication improves, to imitating and extending the behaviours themselves as a means of developing interaction. Current wisdom suggests that each of these approaches has a place, and that the 'best' approach will vary from child to child and from behaviour to behaviour. In a sense, however, this leaves educators no better off, unless the expertise and resources needed to evaluate which approach is best are readily available. Children with sensory impairments have long been known to show functionally equivalent behaviours - in other words, achieving a recognised end by unusual means. Blind children, for example, often become still when listening, instead of looking towards the source of sound; children with poor peripheral vision may swing their heads from side to side when walking in order to scan the environment. Many of these behaviours are highly repetitive. The concept of functionally equivalent behaviour has been extended by Anne Nafstad and Inger Rodbroe in their book Co-Creating Communication: they suggest that developmentally significant sequences of activities, not just individual behaviours, may be replaced by functionally equivalent actions. This suggests a new interpretation of repetitive behaviours: as an essential part of a child's development. This idea is generally accepted for nondisabled children - all babies show repetitive behaviours, peaking at around 6-12 months of age, and Piaget identifies actions performed repeatedly on the child's own body or on objects as part of the sensorimotor stages of development (primary and secondary circular reactions respectively). For children whose development is delayed for any reason, these stages (unsurprisingly) start later and last longer. For children with sensory impairments, in addition, functionally equivalent actions may make it harder to interpret behaviour. This is not to say, by any means, that the 'old' identification of repetitive behaviours as stereotyped actions reinforcing the child's isolation is always wrong - just that many repetitive behaviours, especially in children with sensory impairments and multiple disabilities, are more complex, and potentially of more value in development, than we used to think. My research was motivated by an interest in repetitive behaviours from a developmental perspective, and by the lack of guidance currently available to educators on this topic. The aim of the study was to gain a clearer picture of how children use repetitive behaviours in educational settings, and how educators perceive and respond to the behaviours. Three aspects were identified for research:

  • relationships between the performance of repetitive behaviours and the characteristics of sensory impairments and multiple disabilities;
  • the development of repetitive behaviours over time in children with sensory impairments and multiple disabilities;
  • the ways in which adults working with these children perceive and respond to repetitive behaviours.

The study was divided into three sections, one for each aspect, and the findings in each case were interpreted from a developmental perspective, focusing on children's changing abilities to interact with the physical and social environment. Because of the dependence of children with sensory impairments and multiple disabilities on mediation (usually from adults) to access and interpret the environment, the social dimension of learning was emphasised throughout the study. The first part of the study investigated possible relationships between the characteristics of sensory impairments and multiple disabilities and children's performance of repetitive behaviours. Past studies have found that sensory impairments (especially visual impairments) increase the likelihood of repetitive behaviours, with deafblindness and multiple disabilities associated with their persistence over time. Most studies of children with single sensory impairments, however, have not considered those with multiple disabilities, and studies of children with deafblindness have largely focused on those with congenital rubella syndrome - now barely found among the school-age population in the UK and many other countries. In the current study, specialist teachers working with children with sensory impairments and multiple disabilities were asked to complete questionnaires about children who showed repetitive behaviours. They were given a list of 23 behaviours (see below) and asked to record how often, for how long and in which circumstances each behaviour occurred; they could also add other behaviours. The list of behaviours was compiled from previous studies and in consultation with a group of experienced specialist teachers. Only non-injurious behaviours were considered (with the exception of eye-poking), because, if a behaviour causes injury, a developmental approach may need to be overridden by interventions to safeguard the individual. In addition, those completing questionnaires were asked for information about the child's disabilities and their effects on interaction with others and with the physical environment. 170 completed questionnaires were returned, of which 153 could be analysed. Very few responses (less than 10%) concerned children with hearing impairments and multiple disabilities, but full vision. In line with the findings of previous studies, this suggests that children with visual and dual sensory impairments are more likely to show repetitive behaviours than children with hearing impairments, whether or not the latter have additional disabilities. Of the children with hearing or dual sensory losses who were reported to show repetitive behaviours, however, those whose hearing levels fluctuated tended to show more repetitive behaviour than those whose hearing losses were stable. This occurred whether the fluctuation was innate (for example, due to conductive or mixed hearing loss) or environmental (for example, due to wearing hearing aids intermittently). Fluctuating visual impairments, and the intermittent wearing of spectacles, had no effect on rates of behaviour. These findings may be explained by the different roles of sight and hearing in our interaction with the environment. Sight is a 'foreground' sense, allowing us to choose the space and level of detail viewed, whilst hearing (operational even in sleep) maintains a 'background' role in alerting us to change. Studies of infants with hearing or visual impairments suggest that visual loss has a greater impact on early interaction, with people or objects, than hearing impairment. Children with visual impairments are less easily able to initiate, control or respond to interactions. Hearing, in contrast, alerts us to change, allowing us to predict and act to control events around us. Fluctuating hearing levels may affect the ability to identify changes, and hence the ability to predict and control. Predictability and control may be the common factors relating global visual impairment and fluctuating hearing impairment: for children unable to predict events, or to access feedback from their own or others' actions, the immediacy and accessibility of feedback from repetitive behaviour, and the ability to control its occurrence and results, may explain its increased performance. Degree of sensory impairment did not affect rates of behaviour in the current study, except in that fewer different behaviours were recorded for children with profound hearing losses. This may be because fluctuations in hearing level cause less variation in environmental information for those with profound hearing losses than for those with milder impairments. The explanation above fits the results of the current study, but requires further investigation before it can be accepted - in particular, of the confounding effects of dual sensory impairment and the range of strategies used by people with deafblindness to identify changes in the environment (cutaneous perception of changes in air pressure, for example). Factors affecting the performance of specific behaviours were also examined. Children with mild-moderate visual impairments, for example, were more likely to wave than those with more profound losses; those with severe visual impairments were more likely to light gaze. Children whose hearing fluctuated were more likely to mouth objects, rock or suck their thumb or fingers. These and the other distinctions found can plausibly be explained in terms of available sensory feedback, or as responses to limited predictability and control of events. Children who self-injured, were aggressive or showed socially unacceptable behaviours were reported to show more repetitive behaviour. Although, in the current study, only non-injurious repetitive behaviours were considered, stereotypy and self-injury are very frequently performed together, and there is an urgent need for more research looking at these two types of behaviour. The second part of the study investigated changes in repetitive behaviours over time. Three case studies of pre-school children attending a specialist Early Years Centre were carried out over the course of a year. Two of the children were deafblind and the third had a visual impairment and multiple disabilities. Monthly visits were made to the Centre: on each visit, the children were observed and videotaped during their normal activities, each child's keyworker was interviewed and school records were used to give information on developmental progress or regression. The lists of behaviours developed for the first part of the study were used to gain an initial picture of each child's behaviour; then individualised lists were drawn up and amended after each visit. The keyworkers estimated the frequency, duration and circumstances of each behaviour each month and these were discussed, together with any changes in the use or interpretation of the behaviours and any significant events in the child's life (for example, illness). After nine visits, the information was collated and examined. Changes in the performance of behaviours were graphed and evaluated in the light of contemporaneous changes in sensory function, developmental stage, health and other circumstances. Progressive changes in the frequency and duration of some behaviours were clearly identifiable for all three children, as were changes in the use of behaviours. One child, for example, often explored his keyworker's face and hair with his fingers. During the course of the study, he began to use speech to communicate, and this change was accompanied by more focused tactile exploration of the keyworker's mouth. Another child held objects at arm's length and tapped them; at the start of the study, he never looked at the objects when tapping. During the study, his visual function improved, and he began to show more visual curiosity; at the same time, he began fleetingly to glance at the objects he tapped, and by the end of the study he looked at tapped objects approximately 50% of the time. As he had cortical visual impairment, it is plausible that moving objects were easier for him to see than stationary ones, but further investigation would be needed to test whether this was the case. Each child showed some behaviours which varied little in frequency, duration or circumstances throughout the study. In addition, one child (whose health was very variable) showed some behaviours which varied over time with no apparent pattern. The first child's consistently-performed behaviours were used in interaction with his keyworker, often for turn-taking games (for example, bouncing or jumping, tickling arm or leg, tapping surfaces in rhythm). Most of the second child's consistent behaviours were schemes with objects (for example, mouthing objects, dropping objects and looking after them), interpreted as self-occupation or interactive behaviours. These two groups of behaviours appeared to be developmentally adaptive behaviours used for communication or exploration - functionally equivalent to strategies used by nondisabled children, and in this sense compensating for the restrictions imposed on environmental interaction by the children's disabilities. In contrast, the variable behaviours and a few of the consistent behaviours of the second child appeared to give sensory feedback without involving another person (for example, light gazing, rubbing head with body swaying, rocking), whilst for the third child, behaviours showing consistent frequency and duration generally involved non-auditory, non-visual behaviours using his own body (for example, rubbing ear, tooth grinding, thumb sucking). These were interpreted as providing predictability amid the inconsistency caused by his fluctuating visual and auditory function and epilepsy. These groups of behaviours showed many of the characteristics of stereotypy - self-directedness, persistence, unresponsiveness to changes in the external environment and the generation of sensory feedback. Although most models of stereotypy suggest development only in the direction of more intensely stereotyped behaviour, the evaluation of behavioural change in the current study included change in the opposite direction, with apparently stereotyped behaviours becoming used in communication or exploration as children's abilities to interact with the environment improved (with, for example, improved sensory function enabling a child to scan the environment visually). The third part of the study investigated staff perceptions and interpretations of repetitive behaviour. Two groups of specialist teachers (eight in the first group, five in the second) were interviewed and asked how they understood repetitive behaviours, why they thought some children showed these behaviours, how they felt about the behaviours and how they would respond. The interviews were taped, transcribed and analysed to identify sections addressing similar concepts, and themes which linked different sections. Five themes were identified. The first concerned repetitive behaviours as a source of similarity or difference between teacher and pupil: some of the interviewees saw repetitive behaviour as something that they did, providing common ground with the children with whom they worked, whilst others saw repetitive behaviour as tied to disability and emphasising the difference between themselves and their pupils. The second theme was that of obstruction: repetitive behaviour was felt by some teachers to interfere with the child's learning, with their own teaching and/or with teacher-child interaction. Some saw interference with learning and with teaching as virtually indistinguishable, whilst others felt that the child should lead the teacher in teaching, and that the idea of repetitive behaviours interfering with this process was therefore inappropriate. The third theme concerned teachers' understanding of pupils, with the identification of potential functions of repetitive behaviour. The functions identified fitted with the accepted theories of deafblindness and its implications for development, and often also with accepted models of stereotypy. The fourth theme concerned decisions about action from the teacher - judgements about whether and how to intervene - and the fifth theme identified repetitive behaviours as powerful, with some interviewees speaking of them almost as if they had a life of their own. Some of the themes (the first and fifth) dealt with emotional responses to the behaviours, whilst others (the third and fourth) were cognitively-based. This part of the study was a very small-scale, preliminary enquiry, but the results (especially the range of responses) suggest that further research is warranted, and would assist those developing guidance for teachers in this area: the ways in which teachers currently perceive repetitive behaviours are an essential consideration. Overall, the study findings suggested that the repetitive behaviours of children with sensory impairments and multiple disabilities comprise at least three categories: self-directed, stereotypical behaviours; delayed, developmentally normal behaviours, and functionally equivalent behaviours compensating for the effects of sensory impairments. Although these categories could each be identified in the literature, they had not been brought together before, and empirical evidence for their existence was sketchy at best. Neither had the idea of movement between categories previously been explored in detail. All the study participants wanted to change children's behaviours in some way, either directly or by appropriating existing behaviours for new (usually communicative) functions. The purpose of education, and the raison d'etre of teachers, can be understood as the engendering of change in children's understanding and behaviour. For this reason, there is a need to make explicit the factors affecting children's performance of repetitive behaviours in educational settings, and teachers' responses to the behaviours. The use of a developmental framework in studying this topic seemed appropriate, and potentially the most helpful approach for educators. Further work is needed on the use of repetitive behaviours in developmental, interactive teaching approaches, and it is hoped that the information from this study is useful both to those working directly with the children, and to those engaged in research and training. Further information about the study is available from Heather Murdoch, who can be contacted c/o the Editor, DbI Review, Sense, 11-13 Clifton Terrace, London N4 3SR, UK.

Repetitive behaviours listed in the survey:

  • mouthing objects
  • eye poking
  • light gazing
  • sniffing or smelling
  • screaming
  • crying or whimpering
  • tooth grinding
  • repetitive vocalisations
  • head movements
  • hair twisting or pulling
  • arm waving
  • clapping
  • hand flapping
  • hand, thumb or finger sucking
  • strobing light with hand
  • complex hand or finger movements
  • throwing objects
  • repetitive manipulation of objects
  • repetitive rubbing or wiping movements
  • body rocking
  • bouncing or jumping
  • twirling
  • foot kicking