Autism and ASD : Interventions for Reducing Maladaptive Behaviours

 

This review concerns, firstly, assessment and intervention in respect of stereotyped verbal responding.

The following section returns to the theme of video models for fostering appropriate behaviours; followed by a description of the way in which TV viewing could be used as a learning device for individuals with ASD.

Basic but tested strategies for enhancing behaviour, and increasing participation within mainstream class activities, are set out.

The final section highlights some concern about the use of medication among children and young people with ASD (and comorbid conditions).

 

Vocal Stereotypy

 

Stereotypic behaviour is regarded as one of the core characteristics of autism, and is included within the diagnostic criteria.  It is not unique to autism or ASD, but may be observed among individuals diagnosed with other developmental disabilities; and the negative impact upon skill development, notably in the social domain, can be considerable.  

 

This context was the starting point for the work of Ahearn et al (2007) who go on to define stereotypy as repetitive vocal or motor responses for which, it is commonly believed, there is no functional value.  However, it has also been suggested (by researchers adopting a behaviourist approach) that such behaviour could be automatically reinforced by the sensory outcomes.

 

The emphasis in this current study was upon vocal stereotyped behaviour (which has been subject to relatively little investigation) as illustrated in immediate echolalia or in the repetition of vocalisations and/or non speech-like sounds in response to any stimulus or demand.

 

The authors describe how one approach to tackling automatically-reinforced behaviour has involved identifying the specific source of stimulation that maintains the behaviour, and replacing it with some alternative but similar source of stimulation so that the target behaviour can be reduced.  

Evidence is available (see, for example, Piazza et al 2000) that providing access to the sensory stimulation can reduce stereotypy; and even that access to dissimilar forms of sensory stimulation can still have the effect of reducing the stereotyped behaviours.  However, this differential reinforcement is not always effective per se, and the advice is that it needs to be augmented by response-blocking, an extinction process achieved either by modifying the environment or directly obstructing the behaviour.

 

Ahearn et al set out to assess and manage moderate to high levels of vocal stereotypy among children with ASDwhich was thought to interfere with skill acquisition or social acceptance.  Response interruption was the main strategy adopted.

 

The participants were 2 boys and 2 girls with ASD referred by the mental health services as showing vocal stereotypy of a level that was inhibiting their participation in educational activities and elsewhere. 

M. was a 3 year old boy whose stereotypy comprised word approximations and noises.

P. was an 11 year old boy in a residential school who could communicate in order to request items, apply names to objects, answer familiar questions, and avoid tasks; but communication was inconsistent and the stereotypy involved a mix of repeated words and word approximations or noises. 

N. and A. were 7 year old twin girls, also residential pupils.  N. repeated words, word approximations, and noises.  A. hardly used vocal communication, and what little spontaneous attempts there were concerned access to desired items; with the stereotypy involving word approximations and noises.

 

For the purpose of this study, vocal stereotypy was defined as any example of non-contextual or non-functional speech and included singing, babbling, repetitive grunts, squeals, and phrases unrelated to the given situation.

Functional analysis sessions were 5 minutes in duration, with data on stereotypy collected via time sampling.

Initial observations of the patterns of stereotypy indicated that it was not mediated by social contingencies and was, presumably, maintained by the sensory consequences of vocalising. 

 

The effects of response interruption or redirecting were the focus so that, if the child independently vocalised, the teacher offered praise for appropriate language and responded by giving the child the desired object or activity if possible.  If something was requested that was not immediately available, the child was praised for the request and reassured that (s)he might have it soon. 

Incidents of vocal stereotypy were interrupted immediately and redirections to other vocalisations were given.  For example, using a neutral tone of voice, the teacher stated the child’s name while maintaining eye contact and provided a prompt for an utterance that required a response.  Such prompts might be in the form of a social question (eg.  Where do you live ?  What colour is your shirt ?), or a request to repeat a word.  Care was taken to call only upon skills already within the children’s repertoire.  These prompts continued until the child complied with three consecutive correct responses in the absence of stereotypy at which point the teacher delivered social praise for using the appropriate language. 

 

In their summary and discussion, the authors repeated that vocal stereotypy, like other forms of stereotyped behaviours, appears not to be socially mediated but maintained by the sensory consequences.

The results also replicated existing evidence that stereotypical behaviour can be modified by blocking or interrupting its occurrence. In particular, the results here indicated that response interruption alone can produce significant behaviour changes. 

 

Two possible operant mechanism were considered.

Firstly, the sensory consequences of the vocal stereotypy were dampened by the interruption so that the target behaviour was subject to extinction.

Secondly, response blocking acts as an aversive event stimulated by the child’s behaviour.

Further research is required to assess these effects, although the authors acknowledged that the processes involved may be idiosyncratic across cases; and it was also speculated that the decreased occurrence of the stereotypy could have highlighted the consequences of using appropriate speech as more reinforcing.

 

Given that one could not actually block the child’s utterances, issuing demands that required a vocal response from the child were thought to have the advantage of evoking appropriate behaviour that was incompatible with vocal stereotypy. 

It was also considered possible that the same effect might have been achieved by the use of non-vocal demands, and ongoing research could explore whether the presentation of any demand could reduce the vocal stereotypy (which would be of relevance given that the current procedure would be of limited effectiveness for those children who have very restricted vocal repertoires). 

 

The authors recognise some limitation in this study such as the practical implications for resources necessary for the intensity of intervention, or the short term nature of the work and the absence of data about lasting effects. 

Nevertheless, it was concluded that the adopted procedure produced levels of vocal stereotypy much lower than those observed at baseline among this small sample of children with short term follow up plus anecdotal information suggesting that the procedure can be effective in a natural setting such as a classroom.  

 

Video Models for Enhancing Behaviour    

 

Some direct and practical advice is provided by Banda et al (2007) concerning the effective steps in implementing video modelling (VM) to modify the behaviour of children with ASD. 

 

They begin by citing the quest for strategies which are demonstrably effective and supported by empirical research findings.

VM provides such a strategy, and the authors are able to cite an evidence basis, such as the findings of Ayres and Langone (2005), for its effectiveness in improving skill deficits in the domains of communication, socialisation, scholastic progress, and daily adaptive functioning.

Further, VM intervention may be a preferred option among children with ASD given their commonly reported strengths in the visual channel, and watching a TV screen may be a reinforcing process per se.  Such a procedure is also attractive because of its non-aversive property and simplicity of use (with teachers able to determine the target behaviour and to repeat the showing of the recordings); and evidence exists to show that video models are more likely than “live” models to enhance the maintenance and generalisation of the learned skills.

 

The authors’ review of relevant findings indicates that VM has been effective among individuals with ASD in promoting daily living skills (such as setting a table, shopping, self care, etc); communication skills (such as spontaneous requesting, recognising emotion in speech and facial expression, verbal responses to questions, etc); social skills (such as reciprocal play, spontaneous greeting, sharing, etc); and behaviour (such as reduced aggression); as well as improved performance in the classroom (in spelling, for example).

 

They have highlighted the necessary component steps in setting up a VM interventions to ensure a systematic (and, thus, effective) approach.

1. Identifying and selecting the target behaviour.  The need is for a specific, measurable, and non-fuzzy definition of the target behaviour.  Thus,“ Communication skill ” is not precise enough, and there is a need for a target such as “ Initiating communication by waving or saying hello.”

2. Gaining permission.  Signed consent forms from parents are required plus an undertaking that the video material will be treated as confidential and used only in the implementation of the agreed procedure.

3. Interviewing parents or carers.  The need is to gain an indication of the child’s interest in watching TV and videos (given that such an interest is a prerequisite for this approach but is not universal across the population of children with ASD). 

4. Selecting and training the models.  This involves identifying children whose behaviour would provide a good example of the behaviour to be emulated by the children with ASD.  This could involve capturing instances of positive behaviour in the given domain by the child him- or herself, or selecting other children who can clearly and consistently display the appropriate behaviour in natural settings, albeit, if necessary, after rehearsal using a prepared script.

5. Setting up the equipment.  The need is to ensure that the camera is sited in a position that will provide a clear view of the behaviour modelled, and that the lighting is adequate.  Further, care should be taken that the images do not include visually distracting materials such as posters, brightly coloured toys, etc.

6. Recording the behaviour.  Care is required to ensure that the model is at ease, and, if necessary, recording can be spread over time to ensure a clear presentation of the desired actions which will be presented to the child with ASD.

7. Editing the recording.  The task here is to produce a final version that is not too long, with appropriately paced action, and the behaviours appearing as natural as possible. 

8. Collecting baseline data.  This is important in providing the means of monitoring and measuring progress.  The goal is to produce a clear indication of the child’s level of performance in the targeted skill before the use of the VM.  This might involve, for example, the frequency of occasions when the child does initiate a communication, or the duration of play with a peer or adult.

9. Showing the recording.  Care is needed in selecting a time and place for showing the child the recorded behaviours to ensure that there are no interruptions or distractions, with predetermined points in the tape for pauses to ask questions or to reinforce the demonstrated behaviour.

10. Collecting intervention data.  A pre-prepared chart would be helpful in maintaining a clear record of the child’s performance ... such as the number of communication initiations during the first half hour of the classroom period as the children are arriving and settling in to the routine activities. 

11. Generalising.  The goal here is to promote the use of the learned skills across other settings or with other people so that communications towards a given peer are prompted and reinforced in respect of other members of the group, or in another part of the school environment (playground rather than classroom).

 

The authors summarise by repeating that VM is an evidence-based teaching approach that may help children with ASD to develop skills in any of a range of domains.

The attractiveness of the approach lies in its relative simplicity and its applicability to many children who display similar skill deficits.  Nevertheless, as with any strategy, success may depend upon the experience and efficiency of the persons implementing this VM strategy, with implications for allocating time for preparation, practice, and developing fluency; and the authors recommend a close adherence to the steps provided in order to increase the probability of a successful intervention.

 

On a related theme, it is noted by Janner (2007) that many children or older people with learning and developmental disabilities spend a lot of time watching TV because reading may not be an option for some, or not a preferred activity for others, and because access to alternative leisure possibilities may be restricted.  However, the programmes watched can provide a talking point and something to share with peers or family members.

 

For individuals with autism, it may be that TV has a particular attraction because it can “fit” a number of the characteristic elements of ASD, such as ....

 

In other words, Janner suggests that TV can offer individuals with ASD a source of familiarity, routine, comfort, security, relaxation, virtual relationships, and a sanctuary.

 

The problems of (over)-dependence upon TV and video watching, as summarised by Nally et al (2000) following their survey of parental experiences, include ...

 

On the other hand, the parents were able to see gains from the child’s access to TV, such as the calming effect brought about by watching (familiar?) videos, development of some verbal skills through repeating words, or phrases, or songs learned via the programme or video, channelling the child’s interest in videos to home-made recordings of family activities and events (and, presumably, this could be extended to include access to models of appropriate behaviour and communication), and the potential to use information about viewing preferences as part of a diagnostic process.

 

This survey also described the efforts made by parents to balance the child’s interest in the TV (and the associated benefits) with the needs of the rest of the family.  Such strategies included switching off the TV and wishing it “good night ” before the child goes to bed; having a remote control without a replay button (or somehow preventing a constant replaying of the chosen section of a video), using a recording of the same TV programme as a signal that bed-time is to follow, and providing another TV set to avoid disputes within the family about what is to be screened.

 

Janner also advises that sensory experiences among children with ASD are idiosyncratic so that an optometrist might usefully be consulted to determine if or how the child’s vision affects his or her viewing and what sort of programmes would be most enjoyable and most uncomfortable.

Further, it is recommended that the child has only one (primary) source of sensory input so that even TV watching would better if there are no competing sounds or conversation from elsewhere (or that other games and activities provided for the child are not accompanied by a TV switched on in the same room). 

Where there are special interests, the advice is to seek to build on what is contained in videos.  The example is given of an interest in trains which could be followed up by the creation of scrapbooks, or by finding material about alternative forms of transport, or about the part of the country where a particular locomotive is based.

 

More direct benefit of TV and video watching can be achieved if it is a shared activity and if the action can be paused to allow for pointing out particular behaviours, and discussing the reasons, and what might happen next; and it could be particularly helpful if social interaction can be identified and discussed in this way thus to focus attention upon the meaning of facial expressions, gestures, tone of voice, and socially appropriate responses. 

 

In her conclusion, Janner refers to the ever increasing volume of TV programmes and videos available and to the scope for their use in gathering factual information and social information, while providing the individuals with a sense of control and predictability.  The parental role would involve limiting the choice and establishing a routine for the when and how long of TV and video viewing, but the right balance can create access to valuable experiences. 

 

Increasing Engagement in Inclusive Classrooms  

 

The starting point for the advice offered by Goodman and Williams (2007) was the recognition of the increasing numbers of children with ASD who are included in mainstream classes for most or all of the school day, reflecting not only the implementation of inclusive policies but also the apparently growing incidence of ASD diagnosed among young children. 

 

The implication concerns determining the provisions and routines that will be most helpful for the children’s social and scholastic development, and the means of averting or reducing maladaptive behaviours that inhibit successful performance in the classroom.  Such behaviour might include perseveration in a given and small number of actions, self-stimulatory behaviours, and impaired social and communication skills ... with the likely outcome of limitations in relationships or in participation in a range of shared activities, and a maintenance of the narrow range of interests. 

 

Learning will be impaired as a result of these behavioural deficits or anomalies, with the further risk that some of the ASD-characteristic behaviour or “style” may be interpreted as a form of non-compliance or even defiance. 

 

However, evidence exists for the positive ratio for many children with ASD of benefits over problems associated with inclusive practice, notably in terms of social development (see, for example, Schreibman 2005) and learning to function with a range of other people in relatively complex group settings; and, while evidence is somewhat inconsistent, it appears that inclusion can be associated with academic gains, especially among children with the higher measured ability.

 

These authors go on to acknowledge the existence of research findings by which to inform practice, but note that strategies devised and tested in well-controlled clinical settings may be difficult to implement in the classroom. The implication is to highlight those strategies which not only have a research base, but have been field tested for their practicality. 

It is this kind of intervention for young children in the early stages of schooling upon which these current authors focus.

 

Their first theme is auditory engagement which students typically demonstrate through their following of verbal instructions and responses to questions from teachers or peers.  However, the communication weaknesses characteristic of children with ASD may include poor comprehension and inefficient auditory processing such that verbal information and pointers are missed ... and the risk is that the children will withdraw from verbal instruction. 

 

The advice is that, in such situations, the children will benefit from the use of some cue for auditory focusing, such as a bell or rhythmic clapping, so that the attention of the whole group, not just the children with ASD, is gained.  Further, evidence exists that music can bring about heightened attention among such children, with the implication, therefore, of helping children to switch activities, or to listen for cues about what to do next, if the message is delivered through a simple song.

 

This advice is based on the converging findings that transitions from one task to another in the classroom can be the occasion for maladaptive behaviour given the problem in shifting attention common among children with ASD.  A simple song is considered likely to attract their attention and to cue the required action.  The children can learn which of a small number of songs is associated with a given transition, thus highlighting what to do and where to go.

 

Reference is made to the importance of shared activities like circle time or story time, which require a higher degree of listening and which may prove challenging to children with ASD thus leading to disengagement; and the advice is to maintain attention by the use of brief songs interspersed with verbal cues when the children show signs of drifting off task. 

 

Visual engagement is typically shown via making eye contact with the teacher or partner and via focusing attention upon the materials presented and described. The children with ASD, however, may tend to follow movements of people in the class or generally to attend to some less salient stimulus.

 

Visual schedules are described as very helpful in clearly setting out what has been done and what is the next thing to do.  Evidence exists that their availability reduces the delay between the completion of some task and the transition to the next activity.  This is held to be a relatively simple aid to produce given that classroom activities are readily separable into component parts that can be represented by photographs, or symbols, or written words.  As each task is completed, so the visual cue is removed, revealing the cue for the next activity.

 

Similarly, different parts of the classroom can be labelled to reflect the different activities to be completed; and specific tasks can be aided by the use of pictures.

For example, the use of construction toys or building blocks can be facilitated by providing photographs of a completed object or pattern. 

 

The target is a reduction in stereotyped activities via ensuring that the child knows what to do and how to do it, and that the child can increase independence in directing attention to the given activity with fewer instances of prompting needed to maintain on-task behaviour.

 

It is emphasised that many activities in a mainstream class involve the children sitting in a group while listening to the teacher and/or looking at some materials set out in front of them.   This may prove challenging for the child with ASD who may have some difficulty in focusing upon the relevant stimulus in a visually-complex environment.

 

The advice, as set out by Marks et al (2003), is to provide tangible support material.  This could include bulletin boards to display information ... not simply one main board at the front of the class, but small replicas setting out the information for the child to hold or to place on his or her table.

 

In respect of story time designed to build listening and language skills, one aid to improve focus on the part of the children with ASD is the provision of their own copies of the story (or sets of appropriate illustrations to cover the main events or concepts in the story) by which more closely to follow the reading from the teacher.

 

A further weakness characteristic of ASD is perseveration ... a weakness in disengaging attention from one activity and shifting the focus to something else ... or a limitation in the capacity to filter information thus to attend to what is relevant and ignore what is not relevant.

 

The advice, again, is to highlight the significant information in some way ... perhaps by producing a bulletin board where topic cards are attached, and which are handed to the child when a given topic is introduced. 

 

Social engagement involves active participation in classroom activities, but children with ASD commonly tend to spend time in self-stimulatory activities while avoiding other and more social activities. 

Various strategies are suggested, such as requiring responses whereby teachers promote engagement in the shared activity by asking questions at regular intervals and facilitating responses by prompts and social reinforcement ... largely an exaggerated form of standard practice.  As the child begins to respond to interactions initiated by peers, the probability of communication and shared attention is enhanced, and the number of prompts or the leading nature of questions (ie questions designed to elicit the repetition of a piece of information just given) can be reduced.

 

The encouragement of making choices can also promote engagement.  For example, children with ASD are more likely to engage in free play when they make the choice of toys.  Similarly, given some degree of choice over activities to be completed as part of the teacher-assigned task is associated with greater engagement and reduced disruptiveness of behaviour. 

 

During structured activities, choice can still be given in respect of the materials to be used, crayon or pencil or marker, paper or white board, etc, or of the order in which tasks set out on the visual schedule are to be completed.

 

Similarly, peer interaction might best be fostered if it is the child with ASD who selects an activity and then one has the peers join in ... especially if those peers prompt the child with ASD by means of simple questions to act as cues for the next step of some task.

 

Physical engagement is reflected in the child’s maintenance of appropriate body posture and correct use of materials for the task in hand.  The child with ASD may tend, instead, to engage in repetitive motor movements, and such stereotyped actions may be all the more likely during sessions which involve sitting and listening so that attention is not maintained. 

 

One preventive measure involves giving the child something to hold, such as a soft ball, while giving frequent positive feedback for the child’s adopting this alternative to some idiosyncratic and repetitive action. 

 

Another possibility is giving the child specific opportunities to move about, thus to break up the periods of sitting still and listening.  This has been found to maintain active engagement in the group activity and to reduce the tendency to withdraw. Having the child act as the teacher’s helper in leaving his or her seat to carry out simple tasks is one way of providing the legitimate respite. 

 

Further, the maximising of opportunities for imitation is valuable as a promoter of play and communication and interaction.  Certain reading activities involve an opportunity for actions on the part of the children, or they can be encouraged to mime elements of a story (with prompting and reinforcement of the children with ASD to copy the actions of other children).  The establishment of willingness and ability to copy behaviour may then be utilised by the teacher to encourage the imitation of pro-social and communicative behaviours.  

 

The authors summarise by describing the strategies listed as being intended to be unobtrusive; and the progress observed in the children with ASD as a result of the standard and specific provisions available will offer pointers about the overall viability of the inclusion of the children in question. 

It is further pointed out that many or all of the children in the class could benefit from the kind of interventions described in respect of their participation in group activities. 

 

 

However, the authors do introduce some caveats, such as the need to avoid any assumption that the child with ASD will learn incidentally and to accept that the strategies must be explicitly taught or modelled.  It is also the case that the children with ASD are a heterogeneous group and there will be much individual variation in existing skill levels and in the time and effort required to establish their independent use of the strategies introduced, and consolidation and generalisation will require continuing attention and guided practice. 

 

The authors stress that the strategies described in this study are not a finite list, and that good practice tends to evolve as a result of observation and experience ... however, it is likely that the use of a range of strategies rather than single strategies will be significantly effective.

 

(One might also highlight the implication that the provision of individually-planned and repeated strategies, plus the production of specific materials, will involve a significant amount of adult time. Accordingly, the successful inclusion of children with ASD in a mainstream class is likely to require increased time and supervision from a teaching assistant supporting the class-teacher whose responsibilities cover the full range of needs among the children in the class.)

 

Medication and ASD

 

This final section refers to the somewhat worrying survey finding that a large proportion of children diagnosed with autism or ASD in the USA are treated with at least one form of psychiatric medication.

 

Gerhard (2007) completed a study of the prescription patterns followed by physicians to highlight the initial therapeutic response on the part of medical clinicians to children presenting with these conditions. 

His presentation of findings included a reference to the variable degrees of impairment associated with autism and ASD, and to the marked increase in the numbers of children so diagnosed over the last decade, albeit with no definite explanations of aetiology.

Intervention may include behavioural, educational, and pharmacological elements, and Gerhard focused upon the latter given some uncertainty about the level of use of medication. 

 

Data from the years 2002 to 2005 indicated that the average age of children referred to physicians was 9·7 years among cases of autism and 11·1 for children with Asperger syndrome.  The presence of comorbid psychiatric conditions was 36% for autism and 44·4% for Asperger syndrome.  Disruptive behaviour was involved in around 30% of the cases ... perhaps reflecting the far greater numbers of boys referred compared to girls.

Around 80% of the children diagnosed with ASD were treated with at least one drug ... 30% with antipsychotic drugs, 40% antidepressants, 40% stimulants, and 30% some other form of drug such as mood stabilisers or anticonvulsants.  Some of the children were on several forms of medication. 

The trend was for more psychoactive drugs to be used for Asperger syndrome than for autism (89% compared to 64%) with a particular disparity in respect of the use of psycho-stimulants. 

 

Gerhard recommended research to explore whether this drug prescription pattern reflects the clinical efficacy of the treatment, or a response to parents who are requesting some “aggressive” form of intervention.  His concern is that, as with many conditions among children, there is little known about how the drugs work and how they might best be used; and the implication is for ongoing studies by which to produce more information to guide practice. 

 

(The present writer –MJC – wonders whether this kind of prescription pattern would be replicated in a similar study in the UK.  There is the further speculation that psycho-stimulant drugs, presumably prescribed to manage impulsive and off-task behaviours, are more commonly used among children with Asperger syndrome than children with autism because there is a greater probability that the majority of the former group will be included within mainstream classes where behavioural issues are highly salient given their potential impact upon their own progress and that of the  majority of typically-developing classmates .  It is also noted that the average age of referral to the medical practitioners was 9+ for the children with autism and 11+ for children with Asperger syndrome, and it is likely that, in most cases, the diagnosis will have been made much earlier and various forms of intervention put in place.  It may be possible, therefore, that the referrals were made because [ as Gerhard has pondered ] of a quest for medical [ ie pharmacological ] treatment to add to an existing intervention package in the light of ongoing difficulties of behaviour, perhaps linked to comorbid difficulties as much as to the autism or ASD per se.)

 

                                        *          *          *          *          *         *

 

M.J.Connor                                                                                           December 2007

 

               

 

 

 

REFERENCES 

 

Ahearn W., Clark K., and MacDonald R.  2007   Assessing and treating vocal stereotypy in children with autism.   Journal of Applied Behaviour Analysis  40  263-275

 

Ayres K. and Langone J.  2005   Intervention and instruction with video for students with autism.   Education and Training in Developmental Disabilities  40  183-196  

 

Banda D., Matuzny R., and Turkan S.  2007   Video modelling strategies to enhance appropriate behaviours in children with ASD.   Teaching Exceptional Children  39(6)  47-52

 

Gerhard T.  2007   Psychiatric drugs and ASD.   Presentation to the National Conference of the American Academy of Pediatrics.   San Francisco : October 28th 2007

 

Goodman G. and Williams C.  2007   Interventions for increasing the academic engagement of students with ASD in inclusive classrooms.   Teaching Exceptional Children  39(6)  53-61 

 

Janner M.  2007   Supporting people with autism through TV and film.   Good Autism Practice  8(1)  31-34

 

Marks S., Shaw-Hegwer J., Schrader C. et al   2003   Instructional management tips for teachers of students with ASD.   Teaching Exceptional Children  35  50-55

 

Piazza C., Adelinis J., Hanley G., Goh H., and Delia M.  2000   An evaluation of the effects of matched stimuli on behaviours maintained by automatic reinforcement.   Journal of Applied Behaviour Analysis  33  13-27

 

Schreibman L.  2005   The Science and Fiction of Autism.   Cambridge, MA : Harvard University Press        

 

 

This article is reproduced by kind permission of the author.

© Mike Connor 2007.

Back to NAS Surrey Branch Welcome Page