The Effect of Interactive Visual Feedback on the Improvement of English Intonation of Japanese EFL Learners

Evelyn Abberton (University College London)

Masaki Taniguchi (Chikushi Jogakuen University)

0. This presentation is based on research dedicated to helping to improve the teaching and learning of English intonation (prosody) for Japanese EFL learners. It attempts to evaluate the effectiveness of the use of real time interactive visual feedback on the learners’ approximation of their fundamental frequency contours to those of native speakers. It also attempts to investigate characteristic features of Japanese EFL learners’ English intonation and how their Japanese accents are affecting their English intonation.

1. We conducted three kinds of experiments.

1.1 In the first experiment, we examined how Japanese learners of English can improve their English intonation not only by listening to and repeating after a model speaker but also by listening to and watching a model speaker’s fundamental frequency (henceforth Fx) patterns displayed on a computer screen. Three intermediate Japanese learners of English read a prepared dialogue three times. First they read it on their own without any help. Second they listened to and immediately repeated after a model speaker, a trained phonetician, sentence by sentence. Third they listened to the model speaker, watched her/his Fx contours on a computer screen in real time, and repeated after her. Throughout the experiment, the Laryngograph Processor (by Laryngograph, Ltd) was used to display and analyse the Fx contours.

1.2 The Laryngograph Processor operates with a personal computer using a suite of speech processing programs, PCLX, which analyses laryngeal excitation waveforms, Lx, and speech fundamental frequency. One of its characteristic features is that it can display the Fx contours of the speaker in real time on the screen of the computer attached to it. It can also display two persons’ Fx contours on the same computer screen, for instance, the model speaker’s at the top half and the learner’s at the bottom.

1.3 We must note here that Fx and pitch are not the same. The traces on the screen are Fx contours, which are a physical correlate of pitch. Pitch is an auditory perceptual category which depends on fundamental frequency variation: we perceive a rising pitch, for example, when the fundamental frequency values change from lower to higher.

1.4 What we are aiming at in this study is to examine how we can more effectively help the Japanese learner to produce the same shaped contours as the model speaker in the appropriate places in an utterance: the location and type of the nuclear tone. When we try to copy a visual fundamental frequency contour, we are not trying to change the absolute fundamental frequency values, but only to imitate the overall pattern and not aiming to produce the exact fundamental frequency values; this is called "normalisation" and is at the heart of speech perception.

1.5 The overall result is that it is much more effective to give the Japanese learner an audio cue than nothing or just the tonetic marks, and still more effective to give both audio and visual cues. An important point that must be mentioned about the above experiment is that one assumption can be made that as the learner read the same dialogue three times, on two of which she repeated after the model speaker, she could have improved even without the visual cue. However, the extent to which these three subjects improved their pitch patterns with the audio-visual cues was so great that it may be concluded that in the teaching of English intonation to Japanese learners it is more effective to give the learner both the audio and visual cues than just the audio cue.

2. In the second experiment, four kinds of material were used: the well-known passage, The North Wind and the Sun without tone marks, a group of short phrases with six kinds of nuclear tone marks, a dialogue without tone marks, and a dialogue with tone marks. They were all presented in written form and shown to the learners in orthography, not in phonetic transcription. The tone marks used were from the O’Connor and Arnold system of 1973.

2.1 Twelve learners were used as informants and were divided into two groups, a control and a test group, with six members in each. The learners were all female Japanese college EFL students who attended the Summer Course in English Phonetics (SCEP) at the Department of Phonetics and Linguistics, University College London. Stereo DAT recordings were made of their speech and simultaneous larynx output.

2.2 During the SCEP, both groups attended all the required lectures and practical lessons, including the ones on intonation. The difference between the two groups was that the control group did not use Laryngograph-based feedback at all, while the test group used it regularly in their intonation lessons.

2.3 This experiment enabled us to reaffirm our confidence in the effectiveness of of the interactive visual feedback provided by the Laryngograph Processor in helping Japanese EFL learners improve their English intonation. We saw that there was a striking difference in improvement between the two groups. We also found that the use of tone marks helped the learners a great deal, but an important finding was that if no tone marks were provided, it was extremely difficult for the learners to improve without any interactive visual feedback.

3 In the third experiment, fifty first-year college students of English, having had six years of English learning but with no knowledge of phonetics, were asked to read a dialogue written in orthography without any tone marks. They were allowed to read it silently for five minutes prior to the experiment. The aim of this experiment was to examine the characteristic features of Japanese students’ English intonation.

3.1 In this experiment, we found that Japanese EFL learners tended generally to be narrow in pitch range and it was often difficult to discern what pitch pattern was being employed. This seemed to be an influence of a general characteristic of Japanese intonation.

3.2 We also found that characteristics of Japanese were transferred to the intonation of their English. For instance, in words and compounds like "passport" and "summer course," even though the primary stress is placed on the first syllable in each of them, their counterparts (loan words) in Japanese have the following patterns:

3.2.1 "passport": [pasupooto] The pitch accent goes as follows: LHHLL (L=Low, H=High). (The vowel in the second mora may be devoiced.) ([u] can be either rounded or unrounded in Japanese .)

3.2.2 "summer course": [samaakoosu] The pitch accent goes as follows: LHHHLL. ((The vowel in the last mora may be devoiced.)

3.2.3 Thus many of the subjects failed to produce the appropriate accentuation and gave primary stress to a later syllable, placing the nucleus on the second syllable of "passport," and on the word "course" in "summer course."

3.3 The third characteristic of Japanese students’ English intonation is that old information is not necessarily deaccented. For instance, "here" in "I’m here for a month" in response to "How long are you here for?" was not deaccented by many of the subjects. Neither was "sightseeing" in "Well, partly for sightseeing" in response to "For sightseeing?" Nor was "young" in "I don’t look young" in response to "You look so young." Since the same word is used in each of these dialogues, it is advisable to deaccent it when it is used the second time at a point very close to its first occurrence.

3.4 Many Japanese learners failed to deaccent in what might seem to be fairly obvious cases of old information, let alone where it was not that obvious. For instance, "purpose of visit" in "What’s the main purpose of your visit?" in response to "Well, partly for sightseeing" was not deaccented by many of the subjects. Since "sightseeing" is a purpose of visit, it is advisable to deaccent "purpose of your visit" in "What’s the main purpose of your visit?"

4 Our overall results were that Japanese EFL learners’ English intonation exhibited some characteristic features of Japanese intonation, but that they were more effectively cured with the help of the real time interactive visual feedback function of the Laryngograph Processor than without it.

Acknowledgements

Sincere thanks are due to Professor John C. Wells, head of the Department of Phonetics and Linguistics, University College London, and many of his colleagues for providing me with an invaluable opportunity to work there as an academic visitor from April 1996 to March 1997. I was also deeply grateful for the kind guidance and collaboration of Dr Evelyn Abberton, Senior Lecturer at the same department. My sincere thanks are also due to Professor Jack Windsor Lewis who has given me generous advice on my research throughout my stay in Britain, and to Dr Judith Broadbent and Ms Josette Lesser who were the practical lesson tutors at the SCEP for their kind cooperation and advice. The enthusiastic cooperation of my students was of course greatly appreciated as well. Masaki Taniguchi

References

Abberton, E. and Fourcin, A. J. (1975a) "Visual feedback and the acquisition of intonation" in Lenneberg and Lenneberg (eds) Foundations of Language Development: A Multi- disciplinary Approach. London: Academic Press.

--------- (1975b) American English Intonation: Audio-Visual Practice Materials, textbook and tapes. London: Linguaphone.

Abberton, E., Fourcin, A. J. and Miller, D. (1994) "Using pc-based interactive visual displays for teaching and learning intonation" Language Centre 5:5-6.

Allen, W. S. (1954) Living English Speech. London: Longman.

Bradford, B. (1988) Intonation in Context. Cambridge: CUP.

Taniguchi, M. (1997a) "A Pilot Study for Research on Improving the Teaching of English Intonation to Japanese Learners of English" Journal of Chikushi Jogakuen University. No. 9 (151-169).

---------- (1997b) "A Study of the Effect of Interactive Visual Feedback on Improving the Intonation of Japanese EFL Learners" in M. Tsuzuki (ed) English Phonetics.