This paper was placed in June this year (2004) on the US Congress Sub-Committee on Human Rights and Wellness research database for future reference on Causes of Autism hearings taking place at Congress in the autumn of 2004.




Celia M. Bibby


B.A. (Hons) Psychology. (University of Kent)


M.Sc. (Eng.) University of Birmingham


Member of The Academy of Experts. UK. (retired)

All text and images © 1998-2004 Celia M. Bibby


- 1 -



This paper unifies scientific work on vision in both human and monkey studies.  Cells dedicated to the memory of faces have been found in the amygdala and the superior temporal sulcus of primates, and any damage to these cells can result in prosopagnosia, an inability to identify faces.  This paper proposes a theory of face cell conditioned response” which hypothesises that there is an underlying mechanism for the avoidance or reduction in eye contact, the core characteristic of autism, and that this mechanism arises from a defensive conditioned response produced by a biochemical life-threatening event.


The network of parallel neural pathways of vision (optic nerve through to visual association cortex) and human immunological response (cytokinetic influence on the hypothalamic function) in respect of biochemical toxicity are shown for the pre, actual and post phases of the toxic event.


A representative case study of an autistic child is used to illustrate the consecutive phases of face cell conditioned response, leading to defensive and self-abusive behaviour.  The second case study is of a previously normal adult experiencing adverse clinical incidents with prescription medicines including encephalitis, memory loss, abnormal eye contact and epilepsy


Neural plasticity, as a function of conditioning, is discussed as a potential reversor of autistic eye contact and behaviour.


It is concluded, the biochemical event, inducing the autistic response, arises from a variety of causes including allergic response to diet, anoxia in labour, genetically determined hypersensitivity responses to pharmaceutical products including prescription medicines and vaccines and environmental toxins.





Eye contact as a developmental process in primates has been recognized by neurophysiologists, developmental psychologists and ethologists for sometime, (Goren, C., et al 1975; Meltzoff and Moor 1977,


- 2 -

Tinbergen and Tinbergen 1983).  Any delay or reduction in duration and direction of infant eye movement would be a developmental factor to be noted by a pediatrician.


Eye contact in conjunction with hearing allows vocalization to become speech by the imitation of mouthing and adult studies (Argyle and Dean, 1965; Exline et al 1965, Exline and Winter 1965, Argyle and Kendon 1967) show its relevance for social interaction with rhythmic patterns between pairs and speech patterns dependent on the rhythm of eye contact.


Autists show dysfunctional eye contact, either avoiding prolonged face to face eye contact or when unavoidable appearing to stare straight ahead looking through others as if they were not there.  Peripheral vision is used to function and some tactile skills learnt.  Face to face eye contact will trigger aggressive behaviour either towards others or self-abuse.  Ethologists (Hutt and Ounsted 1966, 1968) studied gaze patterns of autistic children and concluded that they performed a defensive function preventing hyperarousal.


Temporal lobe studies on monkeys revealed that damage can lead to alteration in behaviour and failure to react appropriately to faces i.e the Kluver-Bucy syndrome (Kluver and Bucy, 1939).  The discovery by researchers (Sanghera et al 1979, Rolls 1981 (b)) of cells in the amygdala of monkeys dedicated only to recognizing faces led other researchers (Meadows, 1974, Desimone and Gross 1979) to link this finding with prosopagnosia – an inability to identify individual faces.  Perrett (Perrett et al 1981) experimented on rhesus monkeys showing that in these normal monkeys there were cells in the superior temporal sulcus which also responded only to faces not to other visual stimuli. Similar work was also being carried out (Pigarev et al 1979, Rosenfeld and Van Heoesen, 1979, Bruce et al, 1981).


      The only primate, besides the macaque, in which the effects of temporal lobe damage has been studied is man, and Zeki (Zeki, 1993) acknowledges that the area of the human cerebral cortex critical for the recognition of familiar faces lies in close proximity to the area V4 on the fusiform gyrus near to the superior temporal sulcus where the macaque face cells are located.



- 3 -

These primate studies involving face cells provided a scientific key to dysfunctional autistic eye contact and, when considered in conjunction with the work on the possible causation of autism in the literature, the following hypothesis was formulated.



 “Face cell conditioned response following biochemical trauma is the basis of the characteristic symptom of dysfunctional eye contact associated with autism”

 Face cells are part of the survival mechanism (fight/flight response) for primates, enabling them to discriminate life- giving stimuli from life threatening stimuli and, as such, are species specific.



(Note) Since it is unethical to deliberately attempt to produce an autistic state in any individual two case studies have been used as the subjects for this paper.


The two case studies will be dealt with separately.  CASE STUDY A is that of an autistic child with

typical autistic dysfunctional eye contact.  CASE STUDY B is that of a previously normal adult who, after two life threatening adverse clinical incidents in 10 days caused by prescriptive medicines, suffered encephalitis with dysfunctional eye contact.


In CASE STUDY A the effects of eye contact on behavior will be shown, followed by an explanation of the neural network which underlies such behavior.  In CASE STUDY B the evidence of biochemical trauma leading to dysfunctional eye contact will be given, with figures illustrating the neural network involved.


THERAPY - CASE STUDY B describes the methods used to assist the subject and the results obtained. Its relevance to neural plasticity is covered in DISCUSSION – CELLULAR CONDITIONING

THERAPY – as applied in cases similar to STUDY A are covered in DISCUSSION – HOLDING



- 4 -


Description – visual

The subject was an eight year old boy resident at the Human Development Research Unit, Park Hospital for Children, Oxford, England.  He was chosen as a representative autistic child, for this study.  His behavior was filmed by researchers Hutt and Hutt (1968) as part of their work “Stereotypies their relation to arousal”.  The photographs Figs. A – F are film stills from that work.  In Fig. A the subject was seated in a room with other children and with a nurse who was also seated opposite the subject.


The nurse was instructed to put her arms out in a welcoming gesture to the subject. Fig B.  The subject’s response was filmed.

Results – visual

In photo-stills. C – F the subject is showing increasing fear, pushing himself to the back of his chair, mouth open, biting his thumb, grimacing in a defensive manner and finally, unable to flee, reverting to self-abuse biting the back of his hand

Brain structures relevant to autism. Fig. 1.

(a)     limbic system ( amygdala and hippocampus responsible for emotional response in primates) involved with selection of items for memory storage, this selection is affected by emotional motivational and drive factors, hunger, thirst, pleasure, pain avoidance.

(b)     the hypothalamus which receives chemical signals cytokines, from the immune cells.  These protein

molecules target specific cells, permeating the blood/brain barrier, or sending signals along nerve routes and those from the periphery nerves signal evasive and self protective behaviour  to allow time for healing

processes to occur. (Sternberg and Gold 1997).  The hypothalamus reacts to external stress by inducing release of cortisol from the adrenal glands needed for the fight/flight response or by modulating inflammatory response to internal stress, bacterial infection or chemical toxicity.

(c)  face cells in the area V4 and amygdala in the human brain.






Fig. A – F Autistic child’s response to nurse’s face



 A          D



 B           E



 C           F

- 7 -

Neural networks


Fig. 2(a) Central connections of visual neuronal network. Normal subject


When the normal subject has a critical level of eye contact, light input is relayed through the two optic nerves to the midbrain nuclei, the suprachiasmatic nuclei (hypothalamus) and to the lateral geniculate nuclei (thalamus) and dorsolateral area of the amygdala.  Each of the lateral geniculate nuclei projects to the primary visual cortex and from there to visual association cortex (prestriate cortex) where the face recognition cells in man are located, in the area approximate to V4 on the fusiform gyrus (Zeki 1993) and a non-aggressive response to the human face occurs.


Fig. 2(b) Neuronal pathways that are proposed for unconditioned stimulus


The visual and other associated sensory stimuli are held in the hippocampus gyrus before being consolidated in the neocortex as an engram.  The visual signals received in the amygdala triggers output to the hippocampus and neocortex.  The hippocampal gyrus has been indentified as the site of recent gross memory ‘trace’ that interacts with other systems to elicit information in order areas of the brain (Halgren 1978).  The hypothalamus influences the anterior pituitary to produce a fight/flight response.


Fig. 2 (c) Neuronal pathways activated by face cells previously conditioned by toxic event


This memory trace or engram with its emotional content is the key to the behaviour of

autists when forced to submit to face to face eye contact  . The unconditioned stimulus, (the toxic event) may

have occurred only once, but the conditioned response remains as a survival mechanism (in the same way as the memory of being burnt by fire for the first time).


- 11 -


Description – clinical

The subject was an adult female with a dominant left side.  She was being treated by her clinician for pleurisy

Action – clinical

She was prescribed amoxycillin 500 mg x 3 for 2 days and then 250 mg x 3 for 4 days. She was also taking 75 mg of aspirin daily.  On the third day the subject suffered anaphylaxis, and her clinician instructed her not to take penicillin again. After 10 days the pleurisy returned and she was prescribed ciprofloxacillin 500 mg x 2 for 1 day and then 250 mg x 2 for 5 days.  After 4 days she suffered anaphylaxis again and was instructed by her clinician not to take the 4 – fluoroquinolone ciprofloxacillin antibiotic again.  She was also subsequently instructed not to take aspirin again.

Results clinical

The subject suffered many of the ADR’s listed in the literature by the drug manufacturers, initially, tremor, gastrointestinal upsets and skin photosensitivity, visual disturbance. Dizziness, haemorrhagic bullae.  The subject also began to react adversely to flashing lights; she suffered disorientation and incontinence (Ciprofloxacin-induced complex partial status epilepticus manifesting as an acute confusional state, Isaacson et al 1993).

Within six months the subject was experiencing delayed retro-amnesia covering 6 years and was diagnosed as suffering from encephalitis with both eyesight and hearing deficits. 20% loss of visual acuity and nerve damage to left ear.  Visual field tests Fig. 2(d) showed absolute defects in lower right quadrant and relative defects in upper left quadrant confirming lesions in the visual pathways.  Electromyography and nerve conduction tests in the upper limbs showed atrophy arising from central nervous system damage.

Dysfunctional eye contact following adverse clinical incident

The subject had visual agnosia for 24 months following the reaction.  Her husband also reported that during this period she would not face him in bed but always faced away. The subject herself reported feeling anxious and tense when having any face to face conversation with friends and family and her husband noted that she rarely looked directly at them.  Family photographs taken during the first 2 years showed a marked change in her eye contact and direction, as she never looked directly at the camera.  A stranger also complained that she stared straight past him when having a conversation.


- 13 -

Results CASE STUDY A and CASE STUDY B – neural network

See Figs. 2(b) and Fig. 2 (c)



Description - clinical

The subject’s medical history regarding prescription medicines and vaccines had been assessed by a consultant pharmacologist at a London Hospital with the conclusion that the subject was extremely hypersensitive to prescription medicines and vaccines and had been all her life.


It was decided to encourage the subject to use what memory remained and to rediscover those aspects of her life which she could still remember or which slowly came back to her.  For example she did not remember driving, but after being driven around by her husband, her memory for driving returned.  She was able to manoevre a car very successfully without any rehearsal but could not continue because of the epilepsy.  She

was encouraged to attend social gatherings, business functions, to travel, to shop, speak foreign languages, and swim, all activities which she could do previously.  Some activities she remembered well, swimming and foreign languages (being left handed her language skills had not been lost).

The subject was slowly reintroduced back into her normal activities, although it was limited by the

photosensitive epilepsy.


Results –

It was noted however that the subject was aware of having had a long and full life prior to the adverse clinical incident (she was a middle-aged business woman) and when she regained any part of her memory that had been lost she suffered severe emotional pain and became either angry or very depressed.  She explained that it was

like going over a dark void which connected her present reality with the memories of her past, a past when she was not damaged as she was now.



- 14 -

As she was encouraged to re-experience old skills her eye contact improved until, some four years after the incident, she had a passport photograph taken without a flash camera, which revealed that her eye contact and facial expression was now normal. See DISCUSSION – CELLULAR CONDITIONING and Fig.3.





Conditioning at cellular level has been proposed before, (Hebb 1949, Bliss and Lomo 1973, Kandel 1979, Buzsaki 1989, Kandel and Hawkins 1992), where the hippocampal memory function is implicated and the timing interval and the release of the neurotransmitters serotonin and NMDA are also intrinsic to the conditioned reflex.


Plasticity of cells as a function of learning and memory was supported by the long term potentiation exhibited by hippocampal dentate granule cells following high frequency stimulation (as demonstrated by Bliss and Lomo) and the neurotransmitter NMDA (N-methyl-D-aspartate) has been shown to have temporal properties which facilitate learning. An additional factor of sharp wave activity associated with LTP (long term potentiation) has also been identified in humans suffering epilepsy. Therefore, there are three factors: LTP in cells, NMDA as a neurotransmitter and sharp wave activity, which appear together in the hippocampus when stimulation occurs under laboratory conditions


Buszaki has stated that in the natural state three corresponding conditions must be met to reflect the experimental induction of LTP. There should be strong electrical activity, with a bursting pattern, coactivation of cells and behavioural changes. It is hypothesised that this is the neural code, which relates to the autistic experience, the bursting pattern of sharp waves in the hippocampus, the synchrony of cellular activity across a variety of cells, including face cells in the amygdala and visual association cortex, and consequential behaviour, which is defensive or aggressive.       


- 16 -



The lack of modelling on factors such as attention and arousal, emotions, such as fear, hope, joy, and the role of moods, desires etc. has been pointed out. (Churchland and Sejnowski 1992) They state that such factors are often considered "fluff" compared to the hardware (neural networks in the case of computational scientists, physiology & neurobiology to those dealing with in vitro and in vivo data)



A re-evaluation in the context of dysfunctional eye contact


Holding therapy arises from an American Psychiatrist Dr. M. Welch who, in the early 1980's, acting on the theory  (Tinbergen and Tinbergen 1984) that a breakdown in bonding between mother and child was the cause of autism, introduced "holding therapy". This was not normal cuddling but holding the child firmly so that face to face eye contact was inevitable. The autistic child invariably resisted but Dr. Welch believed that if the eye contact could be maintained long enough the child would cease struggling, bonding would begin and an normal emotional response with the mother would result. Although a degree of success was claimed for this method, other researchers, (Aarons and Gittens, 1992) cited lack of scientific studies to support the therapy and claims of success, and it was therefore not given any scientific credence.



This therapy in its limited application to small children was correct, in that it proposed that the bonding between mother and child had broken down but the causal factor for the breakdown was not correctly identified. At this time, a biochemical factor which could affect the child's eye contact with the mother was not known. However the therapy which involved forced eye contact with the mother, although arousing a good deal of emotional pain and anger in the child did, in fact, bring about a new emotional response.


This process of reversal of mood in the presence of constant visual stimulus has been researched (Brown 1970, 1974). Experiments showed that subjects could alter their emotional response to colours using

- 17 -

biofeedback EEG monitors and this involved the replacement of the first memory by the second, which is based on learning and experience. Memory traces (engrams) are stored in the brain and can be positive (producing adaptive behaviour) or negative (producing instability, frequent triggering & homeostatic failures and maladaptive learning patterns, (Stoyva ,1976).

The regaining of normal eye contact by CASE STUDY B indicates that in adults the re-learning process can also occur where there is appropriate support and the subject can be motivated to work past an emotional pain barrier.



A common pathway?

A biochemical theory of autism has become increasingly hypothesised by researchers into autism over the last few years and the case for a biochemical cause clearly set out (Baron-Cohen and Bolton, 1993). They record those medical conditions, including epilepsy and other neurological causes, in which autism is an associated factor as well as links with other conditions, genetic, metabolic, viral infections and congenital syndromes. They acknowledge, however, that there are cases of autism where these factors are not to be found. It is also understood now that autism can be "acquired" and is not only a condition evident from birth. Therefore, if it can be "acquired" in previously healthy, normally developed children and adults, there must be some definitive causal factor. Baron-Cohen and Bolton conclude that there is some common pathway proceeding from brain damage arising from the conditions previously listed and also from other causes not yet identified


Dietary factor

The dietary factor in autism has been investigated by the School of Biochemistry - University of Birmingham - and (Waring and O'Reilly, 1993) has highlighted the link between autistic behaviour and an inability to process certain foods. Autistic children compared with controls were less effective at sulphation (detoxification of chemicals), in this case a single paracetamol tablet. Foods which trigger off autistic episodes are high in the neurotransmitters serotonin, dopamine and tyramine. Serotonin excess in the body leads to bufotenin being produced and bufotenin has psychedelic properties. Serotonin is also the only chemical found to be extremely high in 30 - 50% of autistic children, therefore this biochemical factor may play a key part in

- 18 -

autism (this may well link up with conditioning at cellular level with release of serotonin as proposed by Hebb and Kendal page 10) and it has also been linked with auto-immune disorders (Warren and Singh 1996).



Medicines & Vaccines and genetic hypersensitivity

Studies (Tsai, 1987, Smalley et al, 1988, Folstein and Rutter, 1988, LeCouter et al, 1989) show that genetic factors are important in the likelihood of developing the condition. As the work on allergy-induced autism increases, and with the present data held regarding known enzyme deficiency conditions and the polymorphic cytochrome isoforms which includes CYP2D6 and CYP3A3/3A4, (Ketter et al, 1995) which affect individual responses to a variety of pharmaceutical drugs, this link between the inability to process certain foods/chemicals which is genetically determined is gaining ground in scientific minds. It has been suggested that an autosomal inheritance pattern could be relevant in the raised levels of sulphite, thiosulphile and taurine present in autistic children. (Murch et al, 1993).  Researchers in the U.K. and Ireland have been studying the possible connection between autism and Crohn’s disease and the MMR vaccine currently used in the U.K. and in many other countries (Wakefield et al 1998 and Sheils, O. 2002).



In view of the data supporting a biochemical theory for autism it is not unreasonable to suggest, as stated at the beginning of the paper, that biochemical trauma (from a variety of sources) produces a conditioned response in the subject so traumatised. The traumatic event, being recorded by the brain at a subcognitive level, is laid down in the memory system of the neocortex as an engram (or series of engrams relating to a variety of senses, smell, touch, hearing, taste and, in the case of the visual system, the face cells in area V4 and amygdala).

Such engrams are triggered by the conditioned stimulus (the face to face eye contact) and produce defensive behaviour. Where eye contact can be kept to a minimum the defensive behaviour need not be elicited, but minimising eye contact leads to social and developmental deficits. It has, however, been modified by "holding therapy" in small children or kept within tolerable limits by specialist schools who aim to reduce anxiety levels and to keep to a very structured curriculum.


- 19 -

The role of memory as a feedback tool has not been recognised previously in this condition but it is an essential factor, demonstrated by the account in CASE STUDY B. Memory is necessary for normal functioning and also for learning and survival however it has been more often related to cognitive levels of brain functioning but the cellular and subcognitive levels are where the answers to autism are found and not only in relation to eye contact but other aspects of autistic behaviour which are beyond the scope of this paper.




Aarons, M., and Gittens, T. The Handbook of Autism. Routledge. 1992

Argyle, M., and Dean, J. Eye Contact, Distance and Affiliation. Sociamtry.1965: 289 -304

Argyle, M. and Kendon, A. Advances in Exp. Soc. Psychology. Academic Press New York: 1967 Vol. 3.

Baron-Cohen, S. and Bolton, P. "Autism The Facts" Oxford University Press 1993.

Bliss, T.V. and Lomo, T. Journal of Physiology (London) 1973 232: 331 -356

Brown, B. Awareness of EEG-subjective activity relationships detected within a closed feedback system. Psychophysiology 1970: 451-464:

Brown, B."New Mind, New Body" Harper & Row. 1974.

Bruce, C.J., Desimone, R., Gross, C.G. Visual Properties of Neurons in a polysensory area in superior temporal sulcus of the macaque. Journal of Neurophysiol. 1981 46: 369 - 384.

Buszaki, G., Ponomoreff, G., Bayardo. F., Ruiz, R. and Gage, F.H.  Neuronal Activity in the sub-cortically denervated hippocampus: a chronic model for epilepsy. Neuroscience. 1989 28: 527-38

Churchland, P.S., and Sejnowski, T.J. The Computational Brain. The M.I.T. Press Bradford Books 1992: Chapter 5

Desimone, R. and Gross, C.G. Visual areas in the temporal cortex of the macaque. Brain Research 1979 178: 393-380.

Exline, R.V., Gray D., and Shuette, D. Visual Behaviour in a dyad as affected by interview content and sex of respondent. Journal of Personality and Social Psychology 1965 1: 201-9.

Exline, R.V. and Winters, L.C. 'Affective relations and mutual glances in dyads' in Affect, Cognition and Personality  Tomkins, S., and Izard, C. (eds.) Springer, New York  1965

Folstein, S. and Rutter, M., Journal of Autism Dev. Disord. 1988: 18: 3-30

Goren, G., Starty, M. and Wu, P. Discrimination of face-like stimuli by newborn infants. Pediatrics 1977 56: 544-549.


- 20 -

Halgren, E., Babb, T.L. and Crandall, P.H. Activity of human hippocampal formation and amygdala neurons during memory testing. J. Electroenceph. clin. Neurophysiol. 1978  45: 585-601.

Harlow, H.F., and Suomi, S.J.  Nature of Love - Simplified. American Psychologist 1970 25: 161-68

Hebb, D.O. Organisation of Behaviour Wiley New York 1949

Hutt, C., and Ounsted, C. The biological significance of gaze aversion: with special reference to childhood autism. Behavioural science. 1966 11: 346-56

Hutt, C., and Ounsted, C. Gaze Aversion and its significance in childhood autism. Behaviour studies in Psychiatry Oxford,  Hutt, S.J. & Hutt C, (eds). Pergamon 1968

Hutt, C., and Hutt, S.J. Stereotypies and their Relation to Arousal: A study of Autistic Children, part of "Stereotypy, arousal and autism" Human Development 1968; 11: 277-86

Kandel, E.R. Small Systems of Neurons. Scientific American September 1979.

Kandel, E.R., and Hawkins, R.D. The biological basis of learning and individuality Scientific American September Special Issue 1992.

Ketter, A., Flockhart, D.A., Post, M., Denicoff, K., Pazzaglia, P.J., Marangell, L.B., George, M.S., Calaban, A.M. The emerging role of cytochrome P450 in psychopharmacology. J. of Clinical Psychopharmacology 1995 15: No. 6 (NIMH)

Kluver, H., and Bucy, C. Preliminary analysis of functions of the temporal lobes in monkeys. Arch. Neurol. Psychiatry 1939 42: 979-1000.

LeCouter, A. et al. Abs. In World Congress in Psychiatry. Genet. Cambridge U.K. 1989

Meadows, J.C. The anatomical basis of Prosopagnosia. J. of Neurol. Neurosurg. Psychiatry  1974 37: 489-501

Meltzoff, A.N., and Moore, M.K. Imitation of facial and manual gestures by human neonates. Science 1977:  198: 75-78

Murch, S.H., Macdonald, T.T., Walker-Smith, J.A., Levin, M., Lionetti, P., Klein, N.J. Disruption of sulphated glycosaminoglycans in intestinal inflammation Lancet 1993; 341: 711-41.

Perrett, I., Rolls, E.T., and Cann, W. Temporal lobe cells of the monkey with visual responses selective for faces. Neuroset. Lett. (Suppl. 3), S358 1979.

Pigarev, I.N., Rizzolatti, G., and Scandalora, C. Neurones responding to visual stimuli in the frontal lobes of macaque monkeys. Neurosci. Lett. 1979 12: 207-212.

Rolls, E.T. Responses of amygdaloid neurons in the primate.  In Ben Ari Y (Ed) The Amygdaloid Complex Elsevier Amstdm. 1981(b) 383-393


- 21 -

Rosenfeld, S.A., and Van Hoesen, G.W. Face Recognition in the rhesus monkey. Neuropsychologia 1979. 17: 503-509

Sanghera, M.K., Rolls, E.T., and Roper-Hall, A. Visual Responses of neurons in the dorsolateral amygdala of the alert monkey. Exp. Neurol. 1979 63: 610-626

Sheils, O., Smyth,P., Martin, C., O’Leary, J.J. Development of an ‘allelic discrimination’ type assay to differentiate between the strain origins of measles virus detected in intestinal tissue of children with ileocolonic lymphonodular hyperplasia and concomitant developmental disorder. The Journal of Pathology Vol. 198 suppl. Page 5A. September 2002.

Smalley, S.L. et al. Arch. Gen. Psychiatry 1988: 45: 953 - 961.

Sternberg, E.M., and Gold, P.W. The Mind-Body Interaction in Disease. Scientific American Special Issue (NIMH)  1997.

Stoyva, J. Why should physical relaxation be clinically useful? Some data and models. In biofeedback and behaviour A NATO symposium Munich. 1976

Tinbergen N., and Tinbergen, A.E. "Autistic children" George Allen & Unwin 1983

Tinbergen, N., and Tinbergen, A.E. "Autistic children new hope for a cure" George Allen & Unwin 1984.

Tsai, L.Y. Neurobiological Issues in Autism Plenum New York (eds. Schopler and Mesibov) 180-189.1987

Wakefield, A.J., Murch, S.H., Anthony, A., Linnell, J., Casson, D.M., Malik, M. Berelowitz, M., Dhillon, A.P., Thomson, M.A. Harvey, P., Valentine, A., Davies, S.E., Walker-Smith, J.A. Ileal-lymphoid-nodular hyperplasia, non-specific colitis, and pervasive developmental disorder in children. The Lancet 1998 351: 637-641.

Waring R.H., and O'Reilly B.A. Enzyme and sulphur oxidation deficiencies in autistic children with known food/chemical intolerances. J. of Orthomolecular Medicine. 1993 8: No. 4

Warren, R.P., and Singh V.K. Elevated Serotonin Levels in Autism: Association with the major Histocompatibility Complex. Neuropsychobiology 1996; 34: 72-75.

Zeki, S.M. A Vision of the Brain. Blakewell Scientific Publications, 1993