- Selected Topics
- What is Deaf-Blindness
- Definitions of Deaf-Blindness
- Causes of Deaf-Blindness
- National Child Count & Demographics
- Communication Overview
- Early Communication
- Prelinguistic Communication
- Object Communication
- Symbolic Communication
- Sign Language
- Accessing the General Curriculum
- Auditory Training
- Calendar Systems
- Concept Development
- Daily Living Skills
- Environmental Considerations
- Harmonious Interactions
- Lilli Nielsen and Active Learning
- Orientation & Mobility
- Play & Recreation
- Social Interactions
- Tactile Strategies
- Universal Design for Learning
- van Dijk Approach
- Identification & Referral
- Early Intervention
- Assessment Overview
- Assessment Tools and Instruments
- Alternate Assessment
- Program Planning
- IEP Development
- IDEA (Individuals with Disabilities Education Act)
- Assistive Technology
- History of Deaf-Blind Education
- Self Determination
- Person Centered Planning
- Postsecondary Education
- Independent Living
- Customized Employment
- Sex Education
- Adult Services
- Intervener Services
- Support Service Provider
- Personnel Development & Training
- Interpreting for Deaf-Blind Individuals
- Interpreting for Deaf-Blind Individuals - Annotated Bibliography
- Training Resources
- Family Resources
- Personal Narratives - Family Stories
- Personal Narratives
- Art & Writing
- Cochlear Implants
- Functional Hearing
- Functional Vision
- Sensory Integration
- Central Auditory Processing Disorder/Auditory Neuropathy
- CHARGE Syndrome Webcasts and Presentations
- CHARGE Syndrome
- Congenital Rubella Syndrome (CRS)
- Cortical Visual Impairment
- Retinal Degenerative Disease
- Usher Syndrome
- Applications of Technology
- Research to Practice
- Topical Overviews
- Practice Perspectives
- Tools For TA
- Information Packets
- Deaf-Blind Perspectives
- Webinar Recordings
- NCDB eNews
- Archived Webinars
Persons Handicapped by Rubella: Victors and Victims - A Follow-Up Study
Persons Handicapped by Rubella: Victors and Victims - A Follow-Up Study
Jan van Dijk
With Cooperation of Ruth Carlin & Heather Hewitt
SWETS & ZEITLINGER B.V AMSTERDAM/ LISS E
Non exclusive rights to publish this chapter have been granted to DB-LINK by the publishers. The entire publication is available for purchase at
Library of Congress Cataloging-in-Publication Data [applied for] CIP-gegevens Koninklijke Bibliotheek, Den Haag Dijk, Jan van
Persons handicapped by rubella : victors and victims : a follow-up study Jan van Dijk with cooperation of Ruth Carlin & Heather Hewitt, - Arnsterdarn [etc.] : Swets & Zeithnger Met index, U. opg. ISBN 90-265-1128-0 NUGI 725 Trefw.: leerstoorr~ssen : meervoudig gehandicapte kinderen / gedragsstoornissen meervoudig gehandicapte kinderen.
Copyright C 1991 by Swets & Zeidinger B.V., Arnsterclarn/Lisse en J. van Dijk
Cover design and lay out: Rob Molthoff Cover printed in the Netherlands by Casparie ljsselstein Printed in the Netherlands by Offsetdrukkcrij Kanters B.V., Alblasserdarn
All rights reserved. N'o part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or othenvise, aithout the prior twitlen permission of the publisher.
Alle rechten voorbehouden. Niets uit deze uitgaven mag worden verveelvoudigd, opgeslagen in een geautomatiseerd gegevensbestand, ofopenbaar gernaakt, in enige vorrn of op enige wijze, hetzij elektronisch, mechanisch, door Cotokopiedri, opriamen, of op enige andere manier, zonder voorafgaande schriftelijke toesternn-~ing van de uitgever.
ISBN 90 265 1128 0 NUGI 725
In the preceding chapters we have presented empirical and anecdotal information in order to focus attention on the overall development of this group of (multi)-sensorily impaired persons. Although maternal rubella was the etiology common to all the subjects, by the conclusion of the study a wide variance was found to exist in their levels of development. However despite these individual variances some general trends did emerge, with inadequate social skills and social immaturity being the most dominant. These ranged from complete withdrawal from all social contact, observed in a few deaf-blind subjects, to inadequate social interaction with peers and over-dependency on teachers and parents, observed in a number of the "hearing-impaired only" subjects. In the first section of this concluding chapter we will attempt to offer an explanation for these weak social skills and the social immaturity observed in our sample, and propose some appropriate intervention strategies which we believe will facilitate the personal growth of these youngsters. Particular emphasis will be placed on guidelines for educational programs aimed at enhancing social competency. The second section of this chapter will focus on the behavioural problems exhibited by many of the subjects in our sample, and the relationship between aberrant behavioural patterns and impoverished communication skills. The third section will focus on die role of dyspraxia and memory, in the development of an adequate mode of communication, and propose a number of alternative communication modes for dyspractic children. We hope that this chapter will be of particular interest to practitioners in the field, who are faced with the educational and rehabilitative difficulties these (multi-)sensorily impaired persons present. We will illustrate our theoretical model with practical examples wherever possible.
6.1. Towards a theory of social development: attachment
Firstly let's examine the role of attachment behaviour in the acquisition of social skills and social competence. The majority of parents in our sample were aware that they had contracted maternal rubella and were equally aware of the possible adverse effect of rubella on their unborn child. Thus their pregnancy and the first years of their child's development was a time of considerable stress for the parents. Consequently it is not surprising that most of the parents recalled clearly early difficulties associated with nutrition, slow physical development, and periods of hospitalization. The parents of those children who had cataracts, recalled vividly that they had considerable difficulty in establishing rewarding contact with their cii1ld in those early years. A principle contributing factor was that the opacity of' the eye with cataracts which was recognizable almost immediately after birth, and this precluded eye contact between parent and child - a fact which the parents found very distressing indeed. In our 1982 publication we stated that whilst attachment behaviour between mother and child was not fully understood, lack of (eye) contact appeared to significantly affect the child's attachment behaviour and future cognitive and linguistic development (Van Dijk 1982, p, 53), Since we made that statement there has been an overwhelming number of publications elaborating on this theme. Most of the studies have based their findings on data collected from the experimental design developed by Ainsworth and her colleagues and known as the "Strange Situation". We will detail this design later in this chapter. Briefly the technique involves observing the child's behaviour when encountering an unfamiliar adult in an unfamiliar situation, and assessing the degree of security reflected in the child's behaviour. The degree of security is considered to be related to the quality of attachment behaviour the child has developed (Ainsworth, Blehar, Waters and Wall 1978).
This technique is suitable for assessing attachment behaviour in children between 12 and 24 months of age. Most non handicapped children within this age range have learnt that their mother is the most important and stable provider of security and comfort. This "Strange Situation" technique does not however explain how this bonding process develops, It merely demonstrates whether it exists or does not exist. Limited research with multi-sensory impaired infants indicate that impairment of hearing or vision, renders this population very vulnerable in the development of adequate attachment behaviour (Fraiberg 1975, Burlingham 1975, Greenberg and Marvin 1979). Psycho-analytic theory has also contributed significantly in explaining attachment (bonding) behaviour. Initially the proposal was that infant-mother bonding stemmed from the child's need for nourishment, and as the mother was the main provider of this nourishment, this resulted in the infant developing a desire to interact with her, even in situations where no nourishment was provided (Freud 1965). Harlow's research refuted this psycho analytic theory by demonstrating that monkeys reared with wire surrogate mothers, which provided nourishment via an attached feeding bottle, rejected this provider in favour of the cloth surrogate who had never provided nourishment, but who had provided tactile comfort. Furthermore Harlow demonstrated that this tactile comfort provided sufficient security for the young monkeys to engage in further exploratory behaviour including social exploratory behaviour (Harlow and Harlow 1962).
Later research associated Harlow's findings with studies of children raised in emotionally deprived environments such as orphanages, hospitals, and similar such institutions. These studies indicated that despite the fact that the children's physical needs were met, they failed to thrive physiologically and psychologically. This failure was attributed to inadequate social interaction being provided because of constant staff changes.
John Bowlby's research laid the basis for the now widely-accepted and well-researched theory of early social and emotional bonding in the physiological and psychological growth of young children. Central to this theory is the importance of contact comfort and appropriate care and stimulation in the early years from a limited number of people. Bowlby observed that the infant spontaneously sought this "contact comfort" from the mother or caregiver, and if responded to appropriately, used this base to explore changes in the environment providing these were regular and of moderate intensity. Intense or irregular stimuli resulted in withdrawal behaviour (Schnierla 1959).
This approach-withdrawal behaviour can be readily observed in neonates. Gentle responses will elicit extensor responses, such as opening hands when gently touched, while intense stimuli will elicit flexor responses such as closing hands or moving away. These behaviours are the beginnings of internal experiences and primitive consciousness. As early as two weeks of age this early level of non verbal communication emerges via a differential crying response between global crying and general discomfort crying. At this stage however the behaviour is not consciously internal but stems from a complicated neurological system - the Limbic System (Lamandella 1976). Adequate communication depends entirely on whether people in the child's environment can correctly interpret the state of the neonate from his crying behaviour. It can be assumed that in the very early stage of life, the first response of the neonate is to differentiate between those stimuli which create comfort within him, and those which create discomfort. His reactions to these are known as "affective" and "aversive", behaviour(Salzen 1979). The limbic system is also responsible for facial expressions and vocalization. As a result of environmental responses to these expressions, the infant becomes involved in a network of social relationships, so that by four weeks of age the foundations are laid for attachment behaviour towards his mother or principal caregiver. In developing this behaviour, the infant pays particular attention to the human face, even if the face is at some distance.
Attention to vocal cues emerges later in the child's development. As this facial stimulation to vocal cues becomes a regular part of the child's social interaction, the model of the mother's face and voice becomes familiar to the child, thus when a strange face and/or voice appears, the child experiences a mismatch between the face and/or voice, and his expectations. This mismatch evokes an anxiety response from the child, and for a time the child will experience fear of strangers when confronted with such a mismatch (Salzen 1979).
Thus it can be seen that vision, hearing and contact play a vital role in attachment behaviour. However, as Harlow's more recent research demonstrates, all three senses are essential components of this process. Visual or auditory stimuli without contact stimuli are insufficient for the development of adequate social interactive skills (Harlow 1978).
The role of memory is frequently overlooked in studying attachment behaviour. As instrumental learning takes place through the reactions of the environment on the child's expression of his internal state (comfort vs discomfort), or through the child's responses to environmental stimuli, situations become more familiar, oriental reactions diminish and habitual responses begin to emerge. The child also learns to expect certain responses from others, towards his reactions. Thus at this early age we see the child beginning to learn to respond to other people's emotions - to be soothed by pleasant tones, to be upset by angry faces. Operant learning and memory, functions of the hippocampus, play an important role in the child's basic understanding of the emotive response of other people towards himself.
In the development of the early interaction between mother and child, distance senses, particularly vision, play an important role in this process. Selma Fraiberg drew special attention to the role of vision in her observations of the interaction between mothers and their blind child, as well as utilizing her own personal experiences with blind infants as a further source of reference. In describing her own reactions she states "when the eyes do not meet ours in acknowledgment of our presence, it feels curiously like a rebuff' (Fraiberg 1975).
As described earlier, in the first few -months of life, the infant demonstrates a strong assertive reaction to the mother's changing facial reactions. Up until five months of age, many faces and voices, will elicit an indiscriminate smile from the sighted infant. At about five months of age, this indiscriminate smiling response becomes reserved for familiar faces only. In blind infants however this response develops differently. The blind infant Mules only when he hears a familiar voice, or feels a familiar touch (Van Opstal 1982). As Collins and Schaffer (1978) pointed out, in normal sighted children visual co-orientation in mother/child interaction can be demonstrated (Collins et al 1975). However in the mother/blind child interaction this co-orientation is hampered. Not only is the child's ability to match the mother's face impeded by his disability, but the mother experiences difficulty reading her infant's face because of the infants relatively undifferentiated facial expressions - a difficulty which understandably arouses considerable anxiety in the mother (Egan 1979) particularly on those occasions when the infant is distressed and the mother is unable to readily determine the reason for the distress. It has been observed that if the blind infant produces a gaze or a smiling response, even if there may are few other signs of responsiveness, this was sufficient to elicit a strong response from the mother. This observation lead to Rowland's remark that the blind infant's early behaviour affected the mother's behaviour more strongly (although unintentionally and unsystematically) than the mother's behaviour affected the infant (Rowland 1984). This early smiling response however should not be confused with the beginning of social activity, such as the stretching out of arms to indicate a desire to be picked up. This type of social activity does not begin to emerge until the end of the first year of life.
Whilst watching people's face and eves during social interaction, is a natural human tendency, it was Fraiberg who first drew attention to the importance of minor hand movements in understanding the blind infants emotional intentions. For example if the child's hands dwell longer on one toy than on another, this could well indicate a preference for that toy over another. The education of mothers and caregivers in observing these hand and other body signals, and interpreting them correctly, is a crucial part of early intervention programs for blind infants (Fraiberg 1975). At the commencement of the second year of life, a social dialogue through the medium of touch commences. Rowland demonstrated that this dialogue lacked synchronicity when mothers were unable to detect the right time to pause, and this made mutual conversation patterns almost impossible (Rowland 1984). In the sighted child this dialogue is initiated not only by touch, but also by vision. For example take a situation where the child can not only feel something interesting but he can also see that it is interesting. In such a situation face to face interaction between mother and child decreases, and both mother and child pay increasing attention to the object itself, which they are both observing and discussing. This coordinated joint engagement supports the notion that communication forms are imbedded and supported by a social context (Bakeman and Adamson 1984). The sighted child acquires knowledge of people, particularly his mother, and a knowledge of objects, by coordinating these social and object realms. By sharing traditional play situations such as "peek-a-boo", "hide and seek", and other such games, the sighted child, learns that persons and objects can be out of sight but still exist. Thus it is through a social context that the sighted child establishes object permanency, which in turn enables him to search for his mother when she has left the room. Blind children however, experience a three month delay in acquiring object permanency - the reason being that deprived of vision, they have to rely on the later development of auditory cues, which develop at about 12 months of age, to establish object permanency (Fraiberg 1973). Thus blind children can only become aware of the absence or presence of their mother through the sounds she makes, and it is only when the mother has acquired a really meaningful value for the child, that he can tolerate her absence and her silence does not arouse anxiety. This explains why the blind child may remain very silent and still for long periods of time. He is using silence to listen intensely to any sounds that his mother may make, and so signal to him her presence. Unfortunately parents frequently misinterpret these periods of silence and inactivity. They fall to recognize the highly adaptive nature of the silence and regard it as very passive behaviour. This arouses their anxiety and they try anxiously to engage the child in some activity. Naturally this disrupts the adaptive listening activity of the child and understandably this often elicits an aggressive withdrawal response in the child (Van Opstal 1982).
Burlingham maintains that because the blind child feels so dependant on others, he is afraid to show his cathartic -emotions, particularly when he is frightened (Burlingham 1975). Thus for example, when placed in a cr8che with sighted children, the blind child may well become very quiet (which for him is adaptive behaviour), using hearing to try to locate his mother or caregiver. These observations support Freedman's assumption of the importance of vision in the early mother-child interactions (Freedman 1964). Fraiberg also reports that, "in the general blind population, there is a significant number of blind children, otherwise intact, who show gross impairment in their human object relationships" (Fraiberg 1979 p. 231). The effect of this impairment may be long term. Blind adolescents frequently require additional training in social skills, and consequently a number of training programs have been developed to teach blind persons effective high level social skills such as social conversation, assertiveness and offering and receiving personal assistance (Van Hasselt, Hersen, Kazdin, Simon and Mastantuono 1983).
6.2. The "Strange Situation"
Interestingly enough, although there is ample evidence that the behaviour of a child in the "Strange Situation", is indicative of his feelings of anxiety, there are no available studies pertaining to attachment behaviour in blind infants using this technique. The "Strange Situation" procedure is a type of eighteen minute miniature drama where, on two occasions, the mother leaves the child in an unfamiliar room with a stranger. The behaviour the child exhibits, when the mother returns, is said to be indicative of the quality of the mother-child relationship. If the child greets the mother, than returns quietly to his toys, showing no maternal avoidance behaviour, then the child is said to be demonstrating a secure mother-child relationship. Studies in the United states classify 70% of children as falling in this B-normative category, i.e. the secure attachment group of children who are able to adapt to new situations, maintain contact with the attachment figure, and have a secure base from which to explore their environment.
A second group comprising about 20% of children in the United States has been categorized as "avoidant attached"(A-group). These children when placed in the "Strange Situation", ignore their mother's return to the room and demonstrate little or no contact behaviour with her when picked up. By actively avoiding the mother the child is said to be demonstrating emotional instability, and failing to use the adult as a secure base for exploring the environment.
A third group (C-group) comprising 1O% of the population, behave extremely passively when placed in the "Strange Situation" They engage in little or no exploratory behaviour and on the mother's return to the room, engage in aggressive interaction with her, frequently biting or hitting her. A more detailed description of these categories can be found in Lamb et al. (1985 p. 37).
Greenberg and Marvin used the "Strange Situation" behaviour to assess attachment behaviour in deaf infants. No significant difference in the distribution of the three categories was found from that found in hearing children. However it should be noted that the deaf infants assessed were between 2-4 years of age, and this were slightly older than the age group for which the "Strange Situation" was designed (Greenberg and Marvin, 1979.)
Greenberg, Calderon and Kusche have used a method with a similar goal as the "Strange Situation", that is to assess the mother's ability to maintain conversation, or social interaction with her young deaf child (3-3 year) on a topic e.g. making a puzzle together. The study revealed that those parents of deaf children, who had not received an appropriate guidance program, tended to be directive and rather controlling in their interaction with their child (Greenberg et al. 1984). This finding is partially supported by Henggeler and Cooper who found a less satisfactory reciprocity in the parent-child interaction in mothers who have not received appropriate intervention programs compared to those who had (Henggeler et al 1983)). The latter were found to be more in synchrony in their interaction with their child's ongoing activity and therefore more effective in their responses to their child's behaviour (Greenberg et al 1984 p. 614). It was suggested that this synchronic type of maternal behaviour was a significant contributor to the advocated communication skills demonstrated by the children in the intervention group. In general it appears that mothers of disabled infants have difficulties in synchronizing turn-taking. This altered social interaction pattern may be the beginning of social deficits later in life (Rogers, 1988). A further study of 268 deaf adolescents and adults demonstrated that 17% of profoundly deaf individuals could be retrospectively classified as A-category. These adolescents and adults manifested relatively withdrawn behaviour from their attachment figure, exhibited ambivalent behaviour, poor relationships, and displayed lack of ego strength and over ego control. This sample of 268 included 22 deaf/blind adolescents of whom eleven could be classified as A-individuals. Of the total sample 59% were categorized as B-individuals of whom only 23% belonged to the deaf/blind group. Twenty-four percent was categorized as C-group, meaning that they had interpersonal conflicts, sought strong proximity with the attachment figure, displayed lack of emotional contact, and apparent lack of ego strength. Twenty-seven percent of the deaf-blind were classified as belonging to this group (Broesterhuizen in press). This research is a convincing demonstration of the way in which social emotional behaviour at an older age is related to the quality of attachment formed during infancy. This concept has recently received sustained attention in the literature (Burbach, Kashani and Rosenberg 1989; Kobak and Sceery 1988; Bretherton 1985; Main, Kaplan and Cassidy 1985).
6.3. What determines attachment behaviour
In view of this data, the question must be asked what is it that determines exactly the formation of a secure bond between mother and child? Secondly are there any typical maternal characteristics which explain why any particular child falls into one of the ABC categories? We believe these are significant factors which can be identified. There appears to be a correlation between the mother's pedagogical style, her personality, and the type of category to which the child is classified as belonging. The Baltimore Longitudinal Study undertaken by Ainsworth provides considerable information on this correlation. Ainsworth studied the relationship of 26 infants and their mothers during the first year of life, for four hours every three weeks. These infants at age 51 weeks were then tested with the technique of "Strange Situation". The findings indicate that the most important underlying dimensions explaining individual differences in mother-child reactions were: "mother's sensitivity", and "appropriate responsiveness to infant signaling". The mothers of infants classified as "B-children" were assessed as being more sensitive, accessible, co-operative and caring, than the mothers of infants classified as A or C (Ainsworth 1983). She hypothesized that the mother's personal characteristics enabled her to pick up her infant's cues and react appropriately to them. This in turn had a significant effect on the child in a sense that it provided feelings of security and facilitated the development of the notion within the child that in moments of distress, the mother would be available "to be in contact". This maternal behaviour contrasted with that shown by children classified as A-category who demonstrated avoidance -attachment bonds. The mothers of these children appeared to be rather ambivalent towards their infants, demonstrating moments of non-acceptance when the child's behaviour interfered with their life style, and becoming easily frustrated when the child cried at inappropriate times (Van 1jzendoorn 1986).
The mothers of children classified as C-category on the other hand were described as clumsy in their interaction with their child. Sometimes they responded immediately to a minor cue, and at other times the child had to persist with crying for a long time before the mother picked him up. Inconsistency in response to their children's needs, appeared to account for the strong proximity and contact seeking and maintaining behaviour, in these infants, who cried frequently. Later research has shown that children, in this category had mothers who appeared to deal with their child's needs in a relatively superficial way, whereas category A children received such intensive stimulation from their mothers, that they tended to engage in avoidance behaviour (Belsky, Rovine, Taylor 1984). It should however be noted that this classification is a "global pattern", obtained from U.S. studies. However in a cross, cultural study involving 8 different countries and 2000 "Strange Situation" classifications, it became evident that B category is modal in all countries, whereas A and C categories differ among countries (Van Ijzendoorn and Kroonenberg 1988).
The reason we have detailed these theories at length is because we believe they may well lead to an alternative educational approach to that based on social learning theory which currently dominates educational programs for handicapped children. For example learning theorists would argue that prompt response to an infant's crying will increase the crying behaviour, whereas those using attachment theory would argue that crying behaviour will decrease when prompt appropriate parental response is given. There have been some studies with handicapped children which use the "Strange Situation" procedure to assess the quality of maternal child relationship. Studies with premature and low birth weight babies without additional disabilities indicated that the ABC distribution of these babies followed the "global" U.S. pattern (Rode, Chang, Fisch and Sroufe 1981; Holmes, Ruble, Kowalski and Lavesen 1984). However it should be noted that all subjects were from middle class families, with access to social supports.
However in cases where the additional disability of respiratory distress syndrome was present, as well as low birth weight and prematurity, the attachment distribution differed from the "norms". Plunkett, Meisels, Stiefel, Pasick and Roloffs (1984) study revealed that 41% of these children were in the C category. Gordon and Jameson's study on 12 children diagnosed as "failure to thrive" syndrome (as many rubella children do), is also pertinent as half of this group were categorized as "non B" status (Gordon and Jameson 1979).
Brooks-Gunn and Lewis used a different approach to the "Strange Situation" procedure to assess the mother-child interactive patterns of handicapped children. Using proximal - distal behaviour they assessed three types of handicapped children within the age range 3-36 months. The sample comprised Downs Syndrome children (N=56), developmentally delayed children (N=21) and cerebral palsy children (N=34). The aim was to ascertain whether or not the infant's behavioural characteristics had an influence on maternal behaviour. The general finding was that the mother's sensitivity to her child's behaviour correlated more highly with her child's mental age, than either the child's chronological age or type of handicapping condition (Brooks-Gunn et al 1984). Thus different aspects of maternal interaction appear to be controlled by different features of infant behaviour. The relationship between classification of the infant, and the mother's personality remains obscure, although different maternal temperamental characteristics have been reported between mothers of A infants and mothers of B and C infants (Lerner, Palermo, Spiro and Nesselroade, 1982).
6.4. Attachment behaviour in rubella children
By applying these research findings on attachment behaviour, to our sample of (multi) handicapped rubella persons, we have constructed the following general profile. Most of the mothers of the multi handicapped group in our sample were aware of the possibility of giving birth to a handicapped child as a result of contracting maternal rubella in pregnancy. Following this stressful pregnancy most of the babies were born full term. Most exhibited management problems in early infancy attributed to low birth weight and failure to thrive. It can be assumed that in many of the babies the rubella virus was still present after birth so that these mothers were dealing with sick children (Van DiJk 1982, Coll, Dumoulin and Souriau 1983).
The development of eye contact, so critical in the feeding situation was virtually impossible in the visually impaired group because of the presence of cataracts. Consequently the highly sophisticated signal relating system between infant and mother was absent from the time of birth and this precluded the adequate development of mother/child synchrony. This partially explains the behavioural differences we noted throughout the study, between those rubella children with cataracts and those without cataracts. It should be noted that at the time of the birth of these children, it was not common medical practice to remove cataracts until the second half of the first year of life or even later. Thus not only was the early stage of mother/child face to face interaction hampered, but the subsequent stage of mutual looking at the same object was also hampered - a stage where the mother appears to take intensive notice of the focus of the baby's eyes towards a novel or attention worthy feature or an object, and respond to this by looking at the same object (Collins and Schaffer 1975) As the child's visual disability precluded the development of this mutual regulating system between mother and infant, and the infant's additional hearing impairment precluded auditory cues playing the normal compensatory role present in "blind only" infants, the deaf-blind infant's ability to seek comfort and security, could only be developed through direct tactual contact.
As explained earlier, in non disabled children distance senses play a vital role in the development of orientative behaviour. The child both sees and hears the adult approaching him to seek direct contact, and so learns to anticipate to be touched. With a deaf-blind child this anticipation is impossible, unless specific procedures are developed e.g. using a plywood floor, which made it possible for the deaf-blind baby or infant to anticipate his mothers approach. Otherwise the mother's contact approach will only arouse confusion in the child and he will not learn to associate contact approach behaviour with comfort and security. This explains why many deaf-blind children do not recognize their mother as a special person who can provide security until much later in life, if they learn it at all.
The seeking of security is a biologically imprinted behavioural pattern in virtually all primates. When confused, the primate clings to any figure or object which may provide a feeling of comfort and security. In seeking security one can observe behaviours such as clinging to the object, jumping to be pulled up, self orality, i.e. biting one's own lips and fingers, and motor stereotypic behaviours. This behaviour is also seen in deafblind children. A striking analogy can be observed in monkeys reared in total isolation and deprived of adequate sources of contact stimulation (Salzen 1979). This proximity seeking behaviour (hanging on to persons) inter-mixed with aggressive resistant behaviour, is present as reported in our deaf-blind sample. It is our theory that these persons lacked a secure base in infancy and so never developed adaptive exploratory behaviour. Other deaf-blind persons in the sub-sample were extremely passive and socially withdrawn. All these behaviours viz. social withdrawal, passivity, stereotyped behaviour and aggressiveness, so commonly seen in low functioning deaf-blind persons, can be partially explained by attachment theory, as these behaviours are typical of the C-category. Unlike the higher functioning group of deaf-blind or "hearing-impaired only" group, the low functioning deaf-blind group in our sample, did not possess compensatory functions such as high intelligence, easy temperament, or a stable environment.
A follow up study of severely emotionally deprived infants on the island of Corfu demonstrated that these factors play an important role in determining whether therapeutic socialization techniques can be successfully used with this population. (Schneider-Rosen, Braunwald, Carlsen and Cicchetti 1985).
The extent to which a lack of basic security influences intellectual development remains an intriguing question. If the "guided touch" approach had been used in infancy with our low functioning deaf blind group, would they still have developed the behaviours described? Experiments with a school program using the guided touch approach, or co-active movements, indicates that low functioning deaf blind persons level of functioning can be significantly increased by using this approach (Appel 1980, Nielsen 1987, Van Dijk 1990).
Many of the "hearing-impaired only" rubella infants in our sample exhibited behaviours in their early years which indicated they were experiencing stress at that point in time, as they had no visual impairment, and thus they were able to seek direct eye contact with their mother, and engage in shared joint visual attention. Detailed examination of the data revealed that many strongly sought proximity or avoided proximity by back arching from the mother. In general however the body language was easier for their mothers to read, than the behaviour of the deaf-blind children. This synchrony behaviour in the "hearing-impaired only" sample was facilitated when the child began to develop gestures according to age appropriate norms. The emergence of natural gestures was a crucial factor in determining whether the mother continued to seek contact, or withdraw from the child.
This emergence of pre-verbal communication (pointing and natural gesture) is critical in patterns of mother/child interaction. In our sample it could be that the period of strong proximity seeking and resistance, indicating an insecure base, shifted to a more secure base, when the child started developing signs and gestures. Factors which could have facilitated this shift were family stability and social class (Lamb et al 1985). It was noted earlier that the family status of our sub-sample of "hearing-impaired only" group was found to be relatively stable over the years, and no family was categorized as being in the lower socio-economic class. This stability is very important, because it may indicate the partner's support of the mother. In virtually all interviews which were carried out in the homes of the "hearing- impaired only" subjects, both parents were present. This indicates at least a genuine interest of both parents in the well-being of their son or daughter.
Recent research on families of autistic children revealed that maternal burn-out correlates highly with partners providing some help with the care of the child (Milgram and Atzil,1988).
In our sample of "deaf only group" and the higher functioning deaf-blind group, both had received family and community support in managing the 91 early problems they had experienced with their disabled child. It is felt that this is an important contributing factor to the mother's perseverance and henceforth to the child's positive development. However these same maternal and family characteristics which may have facilitated these changes, could also in some instances counteract the child's development of autonomy. As reported earlier, mothers of deaf children have a tendency to develop a controlling style of mother-child interaction. This tendency possibly makes them less likely to facilitate the child's increasing need to independently explore his environment and develop autonomy. In our sample there was a striking evidence of social immaturity in virtually all the higher functioning deaf-blind persons, and in a high proportion of the "deaf only" group. Many were still seeking proximity with the attachment figure (the mother) and were reticent with peers, and lacked ego control. An over controlling environment can lead the child to believe that his life is totally controlled by significant people e.g. his parents, teachers and peers, rather than any ability he may have to control events in his own life (Chau 1978).
Persons who have developed an external locus of control tend not to think about their future because they perceive it as being controlled by some external source (Chau and Keogh, 1973). This lack of any future realistic goals is evident in several of our case studies detailed in chapter 3. The literature supports the view that a lack of internal locus of control in pre-school children, is related to early attachment behaviour (cf Van Lieshout) and that this may affect social competence in later life, because of an inability to independently solve social problems with peers, colleagues, or partners (Parker and Asher 1987).
The birth of a handicapped child who is difficult to manage, subjects a family to considerable stress. Consequently it is understandable that in some instances, attachment behaviour is vulnerable and the child may lack a fully secure base from which to explore his environment. This in turn may well affect the child's social competence. Further study is required to explore fully these complex interrelationships. Instruments need to be developed to assess early mother-child interaction patterns' in handicapped children and the effect of early intervention programs on the mother-child interaction. Finally there needs to be a theoretical structure developed on which such intervention programs can be based and which hopefully will preclude the development of behavioural problems described in the previous chapters. In studying the effect of early intervention programs with different types of handicapped children, Rogers reaches the conclusion that gentle support of parents' spontaneous interactions with their handicapped child seems to be more effective than emphasizing situations in which the mother has to carry out teaching tasks or therapy. In these situations the mother needs to be more directive, and her interaction is thus often negatively loaded (Rogers, 1988, p. 313). We have developed a theory in which mother - (handicapped) child synchrony is crucial.) We will describe this approach which we have tried to put into practice for many years.
6.3. Appetite versus Aversion
Fox, in her impressive paper on the effects of the combined handicaps of vision and hearing on a child's development, describes the experiences of such a child as "Imprecise, unpredictable, and un-filtered". The child's life is dominated by chaos, and he does not know which stimulus to orientate to (Fox 1985). This is certainly an apt description of the multi-sensorily impaired child's world, if the child lives in an unstructured environment where the child is subjected to indiscriminate touch contact and
presentation of objects, which raises his anxiety level and leads to stereotyped motor behaviour. In this context we again draw the reader's attention to the fundamental act of human communication: the approach withdrawal system which is based on gentle and regular stimuli, versus strong irregular stimuli (Carmichael 1970, Schnierla 1959, 1963). Gentle contact leads to orientation movements such as exploring the person who is providing the contact, whilst unpredictable contact will lead to withdrawal or acceptance of the contact in a passive non exploratory way. With multi-sensory impaired infants the reaction of the child to a particular kind of contact is quite unpredictable. Assessment procedures which attempt to determine what specific type of contact stimuli will stimulate approach behaviour, and what specific type of contact behaviour stimulates withdrawal behaviour, have to be carried out before an intervention program can start. We have used the following approach with deaf-blind children. We use a variety of different touch contacts; we carefully touch the child on different places on his body, we observe the child's reaction after each touch contact, and note whether the child pays particular attention to any one type of contact. At this level of very fundamental interaction there are numerous behavioural signs which communicate information about the affective state of the child (Lamandella 1977). For example one may observe an increase in the child's attention when blowing gently on the child's mouth. We term this an appetite reaction to oral facial stimuli. Or the child's attention may be caught by spraying him gently with lukewarm water. We use a great variety of "approach stimuli", on different parts of the child's body and attempt to map out areas on the child's body. By using specific stimuli we may be able to elicit orientative reactions. Having selected the stimuli and responsive area of the body we hope that constant repetition may elicit signal behaviour (Sokolov 1960).
This carefully differentiated procedure should not be confused with the massage approach where the child is overwhelmed with stimuli. Rather we aim to create an environment which enhances the development of orientative behaviour and communication (Van DiJk 1986; Meshcheryakov 1979; Nielsen 1987).
With constant repetition some children will eventually become used to the stimuli and it's novelty disappears. The child becomes habituated to the stimuli. This observation is important to note for it indicates memory function in the child as he has to remember the stimuli in order for habituation to occur (Miller, Sinnott, Short and Haines 1976, Goossens 1990). In such children the orientative behaviour can only again be elicited by presenting a slightly different stimuli. The child may then respond to this slightly different stimuli with surprise because his expectations are mismatched. Eliciting such reactions should be part of an intensive intervention program because they may enhance intellectual performance (Kimmel 1981). The child, by changing his level of attention by indicating surprise, signals that he is aware of the mismatch. (Waters, McDonald and Koresko 1977) In turn by responding to it, by giving an appropriately appetite response, the child starts to feel that he is able to control the events happening in his environment and the beginning of an internal locus of control starts to emerge (Rotter 1966). Intervention programs based on such careful assessment and centered around activities and objects which elicit appetite behaviour, can only be established when such activities are conducted on a regular daily time schedule, and carried out by the same person, preferably the mother (Riksen-Walraven, 1977). As appetite responsiveness is the key concept in this approach, it is essential that the person conducting this daily program is carefully supervised, whilst carrying out the program (Seys & Duker 1986, Pogers, 1988).
6.6. Development of attachment
The literature supports the view that attachment is established when the mother or caregiver provides positive appropriate responses to the child which in turn enables the child to gradually learn what he may expect from this attachment figure, and to predict his or her behaviour. This pattern develops feelings of security within the child.
In sensonily deprived children, the signals the child emits may be difficult for the mother or caregiver to recognize and so make it impossible for them to respond sensitively and appropriately. In order to assist mothers and caregivers to accurately interpret the child's signals we have developed the technique of co-active movement and responsiveness (Van DiJk 1965, 1966, 1986, Hammer 1988, Writer 1987, Siegel-Causey and Downing 1987).
Mother/caretaker and child move or act together in close physical contact. In doing so the caregiver becomes aware of topographically minute behaviour (Siegel-Causey et al 1987, p. 33) and thus is more likely to be able to respond appropriately. We emphasize here that responding appropriately does not necessarily mean responding affirmatively. In the development of a person's predictability, it is crucial to learn when to expect "no" for an answer. Regular repetition of activities e.g. child and caregiver moving co-actively along an obstacle course. Once the activity has been signaled by always presenting the child with same object to be used during the activity, this will serve to introduce order in the child's life, and so help to overcome some of the memory difficulties so prevalent in these children.
We have recently introduced a similar technique which enables the child to successfully identify his principal care-giver. The caregiver always wears a particular piece of clothing, such as a scarf, or earring, during their one to one activity with the child. When commencing these activities she always brings the child's attention to that specific piece of clothing, thus enabling the child to identify her. These techniques have been elaborated on in Van DiJk (1986) and Rowland and Schweigert (1988).
We have found this technique to have a positive effect on the bonding behaviour of deaf/blind individuals and their parents (Broesterhuizen,1990).
There is today an urgent need to focus on specific techniques for the following
- for guiding parents in their early interactions with the child
- for recruiting appropriate intervention staff
- for ways of providing immediate effective feedback to the child for demonstrating appropriate responsive behaviour for the development of a valid and reliable assessment instruments for measuring the effectiveness of the above strategies.
6.7. Enhancement of social skills in deaf and deaf-blind persons
Throughout this study we have constantly referred to the physical and emotional problems which complicate the management of the handicapped child in the first years of life. These may well lead to a strong closed family system where strong bonds exist between parents and the handicapped child. At the time of adolescence when these youngsters may request more autonomy and personal space for themselves, this strong closed family system may well come under stress, and differences between individual family members may become evident. These problems may be overcome in those families where there is strong mutual support and the family is able to adjust to the new situation. However in families where the attachment bonds are so strong that the adolescent may continue to feel dependent on his parents, the child will not seek increased autonomy, personal space, or participation in decision making.
Kobak and Sceery investigated the way parent-child relationships influence adjustment during the first year of college life. They found that those students who overcame the anxiety of living away from home, had a working model of their parents as loving and available when needed, whereas those students who were experiencing a high level of anxiety and a low level of social competence, had a working model of their parents as pre-occupied with attachment relationships. These adolescents were very dependant on others and exhibited emotionally clinging behaviour in an attempt to reduce their anxiety (Kobak et al 1988).
Whilst parent-child management which has occurred in the early years cannot be reversed, intervention strategies aimed at facilitating independence and enhancing social skills, can be introduced at adolescence. Greenberg and his associates have developed a curriculum for hearing impaired youngsters known as PATHS (Providing Alternative Thinking Strategies), (Greenberg, Kusch6, Gustafson and Calderon 1985). The program is based partially on ideas derived from cognitive and social learning theory and psycho-analysis. The youngsters are placed in a circle and exchange affective themes with their peers - such as personal experiences of being discriminated against, being falsely accused of an action, etc. Appropriate vocabulary is taught, and problem solving strategies discussed. The technique of "turtling" is a vital part of this approach. The authors of the program were aware that one of the reasons why deaf youngsters have social problems [and this is particularly true for rubella handicapped persons], is because of their impulsivity. The aim of playing "turtle" is to encourage these youngsters to withdraw for a moment and contemplate their reactions and responses, before acting. The influence of cognitive behavioural therapy is clearly present in this approach (Meichenbaum 1979, Mahoney and Arnkof 1978, Alberto and Troutman 1986). This extensive social program requires specific teacher training and careful supervision of the ongoing program. It is not designed for use with multi-handicapped children. The effect of the program was evaluated by comparing the social behaviour of matched groups with and without exposure to the PATH curriculum (Kusche 1984).
Another approach aimed at provoking social skills in hearing impaired youngsters was proposed by Schloss. This program is based on the learning principles of modeling, behavioural rehearsal instruction, contingent reinforcement, and practicing in a natural setting. The rules of a card game are utilized. Each card has a consumer related problem written on it, e.g. obtaining attention in a restaurant, responding to suggestive selling etc. Points are obtained for giving appropriate responses to these real life situations (Schloss 1982).
On our first author's initiative a social skills program for language delayed deaf youngsters (7-20 years) has been devised and implemented at Instituut voor Doven, Sint-Michielsgestel (The Netherlands). Elements of the above two programs have been incorporated into this program, together with the inclusion of pro-social problems, which are presented on videotapes. The deaf students observe the way their hearing peers solve their social problems. These reactions are then discussed and rehearsed through the technique of role playing. The effectiveness of this program is currently being assessed (Rasing and Duker). It should be noted that successful interaction with significant persons in the environment is not only dependent on social skills or social cognition, (although there may be considered pre-requisites) but most importantly participating as a fully accepted member in the group. For successful social interaction, being accepted as a participating member of the group appears to be more important than knowing how to behave socially (Sroufe and Rutter 1982, Broesterhuizen, Van Lieshout and Riksen-Walraven, in press). This is clearly illustrated in the second case study in chapter II, where the youngster's bowling skills when discovered by the group, led to him becoming a more attractive group member, and other people then started interacting more frequently and more positively with him.
The "buddy" system as practiced with case A is interesting from the perspective of social learning theory. Contact experience with peers in adolescence has proved to be effective in reducing aberrant behaviour. With deaf pre-school cl-:ii1dren it has been demonstrated that communication between deaf and hearing peers improved when the deaf child had a steady partner who lived nearby (Lederberg, Ryen and Robbins 1986). To merely mainstream a multi-handicapped sensorily impaired child, without explicit involvement of the non handicapped peers, and their teachers, will only lead to the rejection of the handicapped student as Gresham has so convincingly portrayed (Gresham 1984).
6.8. Dyspraxia, Communication and learned Helplessness
The data presented in Chapter 4 on the relationship between dyspraxia and communication skills, demonstrated clearly the way in which the cognitive sensory motor function influences the communication competency of the (low functioning) deaf-blind child. The inability to carry out simple self care skills has a devastating influence on the child's feeling of competency. The following behaviour can be understood when one comprehends the all pervasive nature of this dysfunction. Personal observation: a teacher presents the child with two triangles, one of which she has neatly coloured in. She requests the deaf-blind child to colour the second triangle in and presents the child with the crayon. The child examines the crayon, then gives it back to his teacher, and proceeds to bang his head on the table and then attempts to withdraw from the situation. Clearly this episode has added to the child's long history of failure. Manipulation of teacher or caregivers hands indicating they want to merely observe, while the caregiver carries out the actual task is common in these children. Some become very effective manipulators, particularly if their manipulative efforts are not restricted by the caretakers and/or teachers, and help too willingly offered. When this occurs the significant adult becomes increasingly active in the child's life, and the deaf-blind child increasingly passive. Within the context of the theory of locus of control, the individual clearly perceives himself as being unable to influence events in his environment, and feels controlled by his events. A situation of learned helplessness exists. To encourage the child to start to take control and to become active, does not make sense, if he is given tasks which are not within his eupractic ability. To facilitate the learning of sensory motor skills in dyspractic children, we advocate teaching these skills co-actively. This entails the teacher carefully guiding the child's hands as he completes the task (Van DiJk 1965a, 1965b, 1986, Writer 1987). A similar approach has recently been employed in teaching self dressing skills to a severely retarded deaf-blind girl aged eleven years. The therapy commenced with total hand over hand manual guidance in putting on a garment. Gradually this guidance was completely faded out. Reinforcement therapy was combined with this procedure (Sisson, Kilwein and Van Hasselt 1988). Sometimes this fading out procedure is difficult to maintain because the subject may unexpectedly grasp the hand of the trainer and so gain full assistance again. We experienced this when trying to teach a deaf-blind child to walk independently without hand holding. To overcome this an automatic prompting device was developed which gives the subject a tap when he stops walking. It was found that the time intervals of independent walking increased with training and the results were maintained when the prompting instrument was withdrawn (Lancioni, Van Dijk, Driessen and Manders 1988). New prompting devices for staying on task, which enable adults to remain some distance from the child are currently being devised.
At the Instituut voor Doven, the workplace of the first author, specific task rooms have been designed aimed at enabling deaf-blind individuals to complete simple tasks with minimum prompting from adults. This is achieved by ensuring the objects are self reinforcing. For example, every time a coloured disc is manipulated it produces bright patterns of colour. In such an environment, with self reinforcing objects, we have found that independent play has increased by 80% in some deaf-blind persons (Van Diik, Carlin, Janssen 1989).
In assessing communication skills the eupractic ability of the child is frequently overlooked. The interest in the field of education in communication systems for persons unable to speak is overwhelming (see for critical survey: Kiernan 1987). Fristoe and Lloyd's survey which focused on the selection of appropriate forms of communication for individuals report that the actual selection is based more on the individual's existing familiarity with a specific system, than a knowledge of available options (Fristoe and Lloyd 1977).
In general there are two types of communication systems - the graphic system e.g. rebus (Clark 1981, Bliss 1963), Makaton (Walker 1987) ("aided" systems) and the "unaided" techniques of sign language and finger spelling (Prinz 1987). Different sign language systems combined with aided forms of communication, are widely used with multi-handicapped individuals, including the deaf-blind (Wolf, Delk and Schein 1982)). It seems obvious to us that the selection of any non-verbal system should consider the eupractic ability of the potential user. Our findings confirm those of Shane and Wilbur who purport that signs are often the only forms of communication that are offered to persons with mild to severe motor impairment (Shane et al 1980, p. 29). We believe that not only does the motor component play a vital role in proficient signing, but so does the individual's total eupractic ability, which contains cognitive elements such as motor memory which is essential to ensure retention of signs. Wetherby and Prutting held similar beliefs. In referring to nonverbal autistic children they reported that few research studies assessed the sensory motor functioning of the children in relation to the communication and linguistic development (Wetherby et al 1984, p. 364).
When assessing the motor components of signs, the sign structures can be divided according to body height at which the sign has to be produced, e.g. chest, chin, or mouth level. Thus to distinguish between the ASL sign APPLE" and "ONION", the sign has to be placed at the lower chest and upper chest respectively.
A second differentiation is whether or not the hand actually touches the body. A third is the actual hand shape which is used in almost all signing systems eg. a fist hand shape, or one using extension of the index finger etc. (Stokoe, Casterline and Cronenberg, 1965). Movements in signs are also essential in differentiating structured elements of the sign. For example do the hands/arms rotate, or make circular movements, or follow linear directions etc.
These various motor components play an essential role in even basic vocabulary. We have observed that even in competent deaf-blind individuals, expressive communication can be seriously affected by dyspraxia.
In studying eupractic ability Van Uden's distinction between motor acts with or without "material" is very significant (Van Uden 1983). This means that the development of a signing system for dyspractic individuals should initially be a "body touching" system. If the individual is unable to locate the place on his body to be touched, the wearing of a communication apron with different colours and/or textures located on specific parts of his body will greatly assist the development of accurate signing (Van Dijk 1990). In general signs used with persons with dual sensory and motor disabilities, should only be used if they are within their eupractic competency, otherwise they may well withdraw from attempting to initiate communication. When this occurs teachers and caregivers often react by bombarding the person with signs in a desperate attempt to elicit some initiation of communication, whereas a more effective teaching strategy would have been to discard the signing system and select a communication code that is within the person's eupractic ability.
Finally any signing system requires a good sequential memory for receptive and expressive communication. There is a high correlation between sequential memory, rhythmic ability and level of language functioning (Van Uden 1983).
Many multi-handicapped individuals have poor sequential memory and rhythmic ability. This supports our theory that simultaneous communication codes such as the use of objects of reference (Visser 1988), tangible symbols (Rowland and Schweigert 1988) schematical pictures (Lancioni 1983)and the written word combined with pictures (iconic graphic system) are more effective communication codes. (See for further details on this approach Writer 1987.)
6.9. Behavioural problems
When reviewing the behavioural problems of our sample of rubella handicapped youngsters (see Chapter 4) it is apparent that even after many years of intensive education and behavioural therapy, many aberrant behavioural patterns still remain. This is particularly true of the low functioning deaf-blind group. As revealed previously many are overtly aggressive, damaging property and engaging in self mutilative and self-stimulatory behaviour.
This population does not appear to be very sensitive to it's environment. Educators have tried to stop these maladaptive behaviours because they interfere with productive learning and contravene human dignity. Furthermore, such behaviours create considerable tension in parents and caretakers, and frequently lead to the youngster being placed in a very restrictive environment.
Sisson raises the question on the efficacy of extinguishing these aberrant behaviours when they may be a natural reaction to severe frustration to anxiety aroused by an inability to communicate basic needs (Sisson 1988a, p. 14). We are of the opinion that it is essential to differentiate between behaviours which are detrimental to the individual and persons in his environment (e.g. aggression) and stereotyped behaviours. The first behaviour should be extinguished as soon as possible as significant people in the person's environment Will want to avoid him, and this leads to even further social isolation, which will aggravate the problem. Sisson examined the occurrence of undesirable behaviours such as hand mouthing, which interfered with "on task" behaviour, self injury and self stimulation in severely retarded deaf-blind individuals. She found that in order to reduce these aberrant behaviours, physical restraint or firmness had to be applied in combination with positive re-enforcement. In our experience application of aversive stimuli sometimes had a greater positive effect than appropriate positive re-enforcement.
Whilst Sisson used highly sophisticated behavioural techniques she failed to counter the physical and social environment of her clients, and so detect any discriminative stimuli in the environment which may have triggered off some of these aberrant behaviours. Recently it has been' demonstrated that deaf-blind subjects' aggressive behaviour eg. scratching, throwing objects at others, or self-injurious behaviour such as hand-biting, face-slapping or head-banging can be extinguished by teaching them to use tangible symbols or signs in order to express their needs. When a task was presented, thev could request a break by showing a typical object or by means of a sign "adult attention" (Durand 1986).
Apparently if persons in the environment place heavy demands on the subject, or give too little or too much attention, or fall to understand his wants or needs, and the subject is unable to communicate his feelings about this situation, then aberrant behaviour may well arise. This behaviour may manifest itself by either inappropriate attention-seeking behaviour, through aggressive and self-injurious behaviour, or by