The action of audio-psycho-phonology on stammering

Experimental study by Dr P. E. Van Jaarsveld*, psychologist, Department of Psychology of the* University of Potchefstroom (South Africa). Parallel English-French translation document published by the Association Française d’Audio-Psycho-Phonologie (Amiens). Probable date: c. 1975-1980.

The action of audio-psycho-phonology on stammering. Experimental study on 30 young adult stammerers — the first rigorous protocol conducted in South Africa on the efficacy of the Electronic Ear. All measured parameters show a significant improvement at the 1% threshold.

1. Introduction

A survey of the current literature on stammering indicates that, despite the precision and depth of research, experimental investigations have contributed proportionally little to giving a clear, concise and generally acceptable schema of stammering. The contradictory and conflicting results of the many empirical investigations carefully controlled with a view to an aetiology of stammering have left many speech therapists and psychologists — interested in an effective therapeutic technique for the stammering problem — in a state of confusion and frustration. It is not rare to hear an expert in pathology say: “We do not know where stammering comes from”.

Without contesting this confusion, two aspects are constantly stressed by researchers:

the listening function as the initiator of the act of speech;
the reciprocal interaction between mother and child in the acquisition of speech.

These two factors also form the central nucleus around which Tomatis’s theories and his APP therapeutic techniques have developed.

Tomatis defines stammering in the context of a developmental problem. He regards stammering as a fixation at the bisyllabic phase of speech development — as a chronic continuation of the natural stammering of the small child, such that his language appears always to be a “song to the mother” (Tomatis, 1973). This implies an inability to overcome the cleavage between the two poles constituted by the mother and the father — a cleavage which, in a normal development, ought to be introjected into the child’s psyche as a functional unity.

In the end, an unconscious monologue (as if the child remained attached to the mother alone) dominates the child’s processes of communication and reinforces the stammering syndrome in such a way that all the characteristics symbolised by the maternal pole — the left, the past, the primary somatic tendencies, egocentrism — develop in priority over those symbolised by the father: the right, the future, the psychic and spiritual tendencies, altruism.

Secondly, Tomatis’s hypothesis (1957a), empirically demonstrated, says that the listening function — seen as the initiator of the act of speech — not only perceives the pressure of sound waves but also analyses stimulations within the limits of various dimensions: volume, pitch, timbre, rhythm. In the human being, the auditory input is not accepted at its face value alone; the ear is not a passive receiver but analyses and transforms the input by means of subtle internal variances in a mental process.

Characteristically, the dynamic organisation of the stammerer’s listening is revealed by the listening test as a relative hypo-function in the speech zone of the directing ear (Tomatis, 1954a).

Audio-Psycho-Phonology has been defined as the study of the various incidences existing between man’s “auditory posture” and his psychological and phonatory attitudes. The Electronic Ear has been described as a device permitting — for its therapeutic purpose — the restoration of self-control in the chain of audio-psycho-phonological speech, a chain that may have been dislocated by the traumas mentioned earlier.

The aim of this study, carried out at the Department of Psychology of the University of Potchefstroom, was to examine the above hypotheses through a theoretical and experimental study.

2. The parent-child relation as a determinant of stammering

2.1. The development of speech

The development of speech in the child is presented by Tomatis as the dynamic parent-child interaction through the phonic phase towards the syllabic and linguistic phases (Tomatis, 1972, 6, pp. 56-66):

The first period (phonic) — audio-vocal exercise with special refinement in the mother-child feedback.
The second period (syllabic) — described as natural stammering, because this bisyllabic character of speech is still directed primarily towards the mother.
The third period — establishes a bridge with social language and requires communication with the father, the “other” and the social environment.

The critical role of the mother-child relation in speech development has been demonstrated by numerous studies — for instance those conducted on children raised in orphanages (Bowlby, 1951). Baker (1955) regards “reciprocal identification” between speech partners as the central mechanism of all speech relations. A deprivation of the mother during the period of reciprocal identification — when the child is still immature and unprepared — may leave him in a state of intense and acute frustration.

2.2. Wyatt and the “developmental crisis theory”

Wyatt (1969) applies this view in her book Theory of Stammering — Developmental Crisis. She found that stammerers, in general, experience an intense fear of being separated from their mother, more often than non-stammering groups. Stammerers also seem to have a far greater need for the presence of their mother.

The stammering child seems to be the victim of a perpetual problem: the fear of being separated gives rise to acute feelings of rage and hostility directed against the mother; and the feelings of aggressiveness then give the stammerer a fear of separation that is integrated by the child as a means the mother would use to take revenge. Wyatt concluded that the dislocation of the mother-child relation during a critical period of the child’s development can lead to forced repetitions and to a fixation at this period.

2.3. Verticality, laterality and speech

Tomatis’s very interesting paper on “Verticality, laterality and speech” (Tomatis, 1963) is well known. A few points:

The rich symbolism of the polarities left/right, mother/father, etc., presented by Tomatis is confirmed in the writings of several researchers: Ornstein (1972), Domhoff (1969), Fagan and Shepard (1971).
Man is distinguished from all other living creatures by his vertical bodily posture. By his verticality, man is raised in such a way that his sensations become truly human perceptions. This, however, requires the maintenance of balance between the vertical force and the gravity of the earth.
In the same way as his feet need a firm support on Mother Earth for his vertical bodily posture, man’s independence and final maturity also develop from the security of a firm relation with the mother, characterised by an empathic bond and a freedom of growth towards the other pole constituted by the father, the social environment, the spiritual degrees and the higher developments.

At the psycho-social level, the achievement of an effective balance between these poles — left/right, mother/father, soma/psyche, past/future — appears to be a necessity inherent in human development, the realisation of which is today the foremost goal of psychotherapies.

2.4. The filtered maternal voice

Tomatis’s research to establish this balance led him to suppose a prenatal rhythmico-acoustic engrammation (Tomatis, 1972b, p. 21) and to use the filtered maternal voice. The latter appears to be a unique and original APP technique, not found in any other therapy. Several authors confirm Tomatis’s theory according to which the ear would be functional very early during the prenatal period: Ormerod (1960), Johnsson, Wederberg and Westin (1964), Grosser and Ortmann (1966), Clauser (1971). Salk’s experiments at the City Hospital, Elmhorst NY (1960, 1961, 1962), constitute a well-documented indication of prenatal rhythmico-acoustic engrammation.

The use of the filtered maternal voice — according to the Tomatis technique — aims at eliminating anxiety and establishing feelings of security (cf. Sarkissoff, 1973). Clinical observations confirm that the child recognises his mother’s voice very early and that anxious, insecure or aggressive children manifest positive reactions when listening to the filtered maternal voice.

3. The listening function as a determinant of stammering

Language may be described as a cybernetic system. Phonation is controlled by audition. The sensory input is individually selected, digested, stored and transformed by inner forces. The output (speech behaviour) is continuously monitored by the feedback process and sent back through a central regulating mechanism (cf. Wiener, 1954).

At the psycho-social level, Mead’s model is applicable: man is always testing reality in a selective manner. Perception is organised in terms of hypotheses which, after manipulation, lead to meaning — with the result that behaviour is paralysed or facilitated respectively by negative or positive feedback. Wyatt’s developmental crisis model applies this principle to the problem of stammering.

At the physico-acoustic level

The following deductions have been made from the research mentioned in the specialised literature:

In the stammerer’s speech-listening feedback circuit, a delay that perturbs the act of speech is most probable (Fairbanks, 1955; Stromsta, 1958; Berry, 1969; Cherry and Sayers, 1956).
This perturbation is probably active in the feedback systems of all the modalities serving language (auditory, visual, tactile-kinaesthetic, proprioceptive). But these modalities are secondary to the ear — in the archaeo-, palaeo-, and neo-stages of development as described by Tomatis (1974). Traumas may act as precursors and precipitators.
In the structure of audition feedback, a relation seems to exist between low frequencies and stammering (Cherry and Sayers, 1956; Stromsta, 1958; Ham and Steer, 1967; Bachrach, 1964).

These results bring substantial support to Tomatis’s therapeutic approach: an education of the ear through high-frequency stimulations as a therapeutic aid to stammering.

The treatment of audition by Electronic Ear implies that the muscles of the middle ear can be conditioned to a state of adaptation to the sonic environment. On the basis of various studies (Kostelijk, 1950; Wersall, 1958; Reger, 1960; Moller, 1972), it has been established that through the contractions of the eardrum muscles, the physical properties of the conduction system of the middle ear are changed: a reduction of sensitivity to low frequencies and an increase in tones above 1,000 Hz give rise to an ascending audiometric curve — such as Tomatis had described years before these results as the graphic representation of a well-adapted ear.

4. Experimental investigation

4.1. Aim

The principal aim of the study was to apply the technique of Audio-Psycho-Phonology to a group of stammerers, and to evaluate the result by various measures.

4.2. Subjects

The investigation was conducted on a group of 30 young adult stammerers (22 men and 8 women), with:

mean age: 21 years;
mean IQ: 112.

All received auditory training with the help of the Electronic Ear.

4.3. Tests and procedure

The following tests were administered before the start of auditory training and again after its completion:

Lanyon Stuttering Severity Scale (1967) — indication of severity.
Analysis of speech and reading samples by the Johnson method — estimation of the number of dysfluencies (categories: interjections, word repetitions, phrase repetitions, revisions, incomplete sentences, broken words, prolonged sounds).
Johnson measure of speech and oral reading rate.
Iowa Scale of Attitudes towards Stuttering — evaluation of the subject’s feelings about his own speech.
Audiometric examination — effect of training on auditory acuity.
Spectral analysis of the reading voice — effect on the physico-acoustic aspects of phonation.

The statistical design used the experimental group as its own control (paired measurements). To evaluate the difference of means, the t test for correlated samples was applied.

4.4. Results

4.4.1. Severity of symptoms (Lanyon scale)

A mean value of 71.50% before re-education (severe stammering problem) was brought down to 41.76% after auditory training (mild stammering problem). Difference significant at the 1% threshold (t = 10.57; p < 0.01).

4.4.2. Measure of dysfluencies

Total dysfluencies index per 100 words:

In reading: 16.35 before → 6.20 after (difference 10.15; t = 5.84; p < 0.01);
In spontaneous speech: 26.55 before → 11.90 after (difference 14.65; t = 6.5; p < 0.01).

Highly significant differences between the mean before and after training, in oral reading as in spontaneous speech.

4.4.3. Rate of speech and reading (words/minute)

Reading: 99.1 → 117.5 words/min (gain 18.5; t = 4.13; p < 0.01);
Speech: 88.2 → 111.5 words/min (gain 23.3; t = 5.98; p < 0.01).

Rate significantly faster after auditory training, in both modalities. This gain — given that the severity of stammering is in part a function of the inhibited behaviour represented by a slowed rate — is interpreted as a progress in speech.

4.4.4. Attitude towards stammering (Iowa scale)

Mean score 2.46 before → 1.65 after (difference 0.81; t = 7.158; p < 0.01). The subjects reflect a significantly healthier attitude to the stammering problem after auditory re-education.

4.4.5. Listening test

Before re-education, the audiometric curves (air and bone conduction, right and left ear) reveal a relative listening loss in the passband of the speech zone — particularly marked in the right ear (the directing ear), as Tomatis had predicted.

After re-education: significant gain in auditory acuity in all frequency zones (low, middle, high), at the 1% threshold, in both ears, in air as in bone conduction. Particularly marked in the middle zone (language zone).

4.4.6. Spectral analysis

Evaluation made on 24 subjects, by tape loop (duration 20 s), one-third octave band, Brüel & Kjær recorder (measurement error < 0.5 dB). Results on the mean spectral energy in dB:

Zone 200-500 Hz: 84.56 → 91.09 (+6.53; t = 4.98; p < 0.01);
Zone 630-2,000 Hz (language zone): 72.46 → 84.10 (+11.64; t = 6.68; p < 0.01);
Zone 2,500-8,000 Hz: 53.57 → 65.48 (+11.91; t = 8.64; p < 0.01);
Total energy: 100 → 107.91 (+7.91; t = 3.99; p < 0.01).

The gain in spectral energy after re-education is notable throughout the range of the spectrum but particularly evident in the middle frequencies — that is, the zone of language.

4.5. Conclusions

The severity of stammering, measured by the participants’ attitude towards their language problem, was significantly less important after auditory education.
The participants made clear progress as to the number of dysfluencies of speech, in the spoken and read sample. Conclusion: auditory education is the most important factor responsible for the change.
The significantly faster rate of speech and oral reading, recorded after the programme, is attributed to a progress in speech.
The significantly lower score on the “Iowa Scale of Attitude towards Stammering” indicates a better attitude and considerable tolerance towards stammering.
Stammerers, in general, show a listening loss in the passband of speech — particularly important in the language zone of the right ear, as Tomatis had predicted.
Better acuity scores were recorded after education, in the low, middle and high zones — particularly important in the middle frequencies corresponding to the language zone.
Audio-vocal conditioning appears to be the only factor responsible for the improvement in acuity scores and the diminution of dysfluencies.
The spectral analysis carried out on the oral reading samples also supports Tomatis’s observations: a significant gain of energy and vocal output by the group after audio-vocal education, particularly in the passband of the language zone.

General conclusion

Auditory education — or audio-vocal training with the help of the Electronic Ear, as defined by Tomatis — may lead to:

an improvement in listening behaviour;
a corresponding modification of the physico-acoustic and temporo-rhythmic characteristics of speech, giving greater ease of elocution and a healthier behaviour in the face of stammering.

Bibliographical references (selection)

Ammons R. & Johnson W. (1944), The construction and application of a test of attitude toward stuttering, Journal of Speech Disorders, 9, 39-49.
Baker S. J. (1955), The theory of silence, Journal of Genetic Psychology, 53, 145-67.
Bowlby J. (1951), Maternal care and mental health, Geneva, WHO.
Cherry C. & Sayers B. M. (1956), Experiments upon the total inhibition of stammering by external control, Journal of Psychosomatic Research, 1, 233-246.
Fairbanks G. (1954), Systematic research in experimental phonetics: a theory of the speech mechanism as a servosystem, Journal of Speech and Hearing Disorders, 19, 133-39.
Johnson W., Darley F. L. & Spriestersbach D. C. (1963), Diagnostic methods in speech pathology, New York, Harper & Row.
Lanyon R. I. (1967), The measurement of stuttering severity, Journal of Speech and Hearing Research, 10, 836.
Salk L. (1962), Mother’s heart beat as an imprinting stimulus, Transactions of the New York Academy of Sciences, 2, 24, 753-763.
Tomatis A. (1954), Recherches sur la pathogénie du bégaiement, Journal Français d’Oto-Rhino-Laryngologie, III, 4, 384.
Tomatis A. (1956), Relations entre l’audition et la phonation, Annales des Télécommunications, no. 7-8, Cahiers d’Acoustique.
Tomatis A. (1957), Audiométrie objective, Journal Français d’Oto-Rhino-Laryngologie, 3, 379.
Tomatis A. (1963), L’oreille et le langage, Le rayon de la science, Paris.
Tomatis A. (1969), Dyslexia, Univ. of Ottawa Press.
Tomatis A. (1972), Education and Dyslexia, Paris.
Tomatis A. (1972), La libération d’Œdipe, ESF, Paris.
Tomatis A. (1974), Vers l’écoute humaine, ESF, Paris.
Wiener N. (1948), Cybernetics, New York, John Wiley.
Wyatt G. L. (1969), Language learning and communication disorders in children, New York, The Free Press.

— Dr P. E. Van Jaarsveld, psychologist, University of Potchefstroom (South Africa). Parallel English-French translation document, published by the Association Française d’Audio-Psycho-Phonologie, 2bis rue Charles Dubois, 80000 Amiens.