Please quote me right.
Marilyn Monroe to Maurice Zolowtow.
Introduction
Second language learners use a variety of compensatory strategies such as schemata when trying to complete tasks. Mandler and Johnson（1977）conducted a well- known experiment that explained why a certain group of learners wrote down distortions from the original text in a text reconstruction task. This paper was instrumental in helping Schema Theory enter the Second Language Acquisition（SLA）lexicon. There is no real value in trying to prove Schema Theory but there is value in knowing how Schema Theory can plausibly explain a particular result. However, when a particular set of learners were given a listening recall task, they did not all use or could have used schemata as a compensatory strategy.
This paper presents the results of two research questions. One research question asked “What elements of text do learners first attend to when listening to a text?” The second question asked “What kinds of additions, distortions or subtractions were made by the learners?”
The following details will be presented: a brief description of Schema Theory and how it applies to language learning, an account of the difference between hearing and listening, a description of the students, the procedure of the dictation, some very basic descriptive statistics, an analysis of the results and a plausible explanation of the results.
The final portion of the paper will look at the weaknesses of the study and the pedagogical consequences arising from the study. The individual learner’s schema has a significant
How Schema Theory Can Explain Distortions In Dictation Output
Todd Armstrong
Theory in more detail.
Schema Theory
A schema is a mental structure by which the learner organizes information.
A schema contains the learner’s knowledge on a certain topic and the organized interrelationships among components of that knowledge （Chi, Glaser and Rees 1982）.
Schema theory suggests that the very act of organizing the information makes the material more memorable （Mandler, 2001）. Movement of information from short-term memory to long-term memory involves building schemata, developing them and making them more flexible, and calling on them often for practice and authentic use.
It is helpful to organize information into categories, with larger categories containing more detailed information （Mandler, 2001）. The effort made in such a method unites the learning strategies of organizing and conceptualizing resulting in higher-level chunks of information that are simpler to remember. If the learner labels a chunk （category）
of information meaningfully （for example, words about weather or Japanese expressions for interaction with older, respected people）, this helps to strengthen the retrieval cues for remembering the category in general and the more detailed information inside it. Chunking reduces cognitive load （Pass, Renkl and Sweller, 2004）, thus freeing up working memory for other uses. Organizing information in memorable sequences or strings （Mandler, 2001） can also be valuable. Having briefly summarized Schema Theory, let us move on to understanding the difference between hearing and listening. This distinction will help us to gain a better understanding of the dictation results.
Listening Versus Hearing
Though listening and hearing are generally used interchangeably, a distinction will be made for the purposes of this research. “Hearing and listening involve sound perception, the difference in terms reflects a degree of intention”（Rost, 2002）. A person may have the physical ability to hear but may not be capable of listening. An illustration of this is when a speaker speaks to a person in the speaker’s first language（L1）which is different from
the hearer’s L1. The hearer is able to acknowledge that an acoustic signal is emanating from the speaker’s mouth but that is all. For the hearer, there is no meaning to the signal.
Listening is very much, among other things, the process of interpreting what is heard.
Learners have numerous tasks to carry out while listening to a foreign language.
Here is a list of micro skills presented in O’Malley（1986）:
• Retain chunks of language in short-term memory
• Discriminate between distinctive sounds in the new language
• Recognize stress and rhythm patterns and tone patterns
• Recognize reduced forms of words
• Distinguish word boundaries
• Recognize typical word order patterns
• Recognize vocabulary
• Detect key words, such as those identifying topics and ideas
• Guess meaning from context
• Recognize grammatical word classes
• Recognize syntactic patterns
• Recognize cohesive devices
• Detect sentence constituents, such as subject, verb, object, prepositions, and so on.
As can be seen, the cognitive requirement for listening is quite extensive. Let us look in detail at recognizing individual words, which is what the participants had to do in this study.
Recognizing words in fluent speech is the basis of spoken-language comprehension.
The two main tasks for the listener in word recognition are identification of words and activating knowledge of word meanings. Identification of words is a complex process.
As Cutler（1997）notes, in listening to continuous speech, there is no direct auditory equivalent to the white spaces when reading continuous text because there are no reliable cues marking every word boundary. In listening, individual word recognition is a very problematic and highly cognitive, demanding process. Furthermore, if a listener does recognize a word but has limited or no knowledge of the word’s meaning, the entire process of word recognition can be subverted, and the listener must resort to compensatory How Schema Theory Can Explain Distortions In Dictation Output
words in speech, whether through faulty identification of word boundaries or inadequate knowledge of word meanings, is the major source of confusion in language comprehension, particularly second-language comprehension.
The Participants
The participants in this study were all Japanese learners who had been learning English for approximately six years. Their ages ranged from 18 to 19 years old. There were a total of 27 participants, 21 men and six women, in this study. They can all be classified as false beginners which means that they have had six years of English language instruction in the grammar/translation method with no communicative element whatsoever in either written or verbal form. The English class would be conducted 95% in Japanese with very few listening tasks carried out in a typical class. The aim of the class is to prepare students for a multiple choice exam that would mainly test their grammatical knowledge of English.
The Script
Choosing the text was quite difficult due to the low proficiency of the learners.
The text is accessible to the learners but has parts that are beyond the learner’s ability to fully comprehend. The script lasts for about one minute, which is long enough so that the participants are not able to memorize it but short enough to allow them to remember the main idea and some key facts. In other words, just enough information to stretch their short-term memory load. The 12-word script was taken from Talcott and Tullis（2007）
which can be found in Appendix 1.
The Procedure
Nation（2016）outlines the basic procedure for a dictation and it is this method which was used in the study. The listeners heard the text three times. The text was first read aloud in its entirety so that the listeners could attempt to reach a general level of comprehension. The second reading was delivered in chunks of seven to eight words at a time with brief pauses in between each chunk to allow the listeners enough time to write
down what they thought they had heard. In the third reading, the text was read aloud again in its entirety to allow the listeners to add or make corrections to what they had written during the second reading.
The Analysis of the Notes
Rost（2002）defines Implication Scaling as a technique for “ordering instances of an observed behavior in order to assess the order in which behaviors are acquired or exhibited by a group of subjects”（277）. In other words, the scale is used to measure the relative frequency of complete versus incomplete responses from subjects in any study. In order to measure how much information the participants were able to write down, we turn to Kim（1995）who derived one such scale as outlined below:
Phase 1 Pre-key words: The listener cannot identify key words that bear phonetic prominence in speech（e.g. reporting “milk” or “meal” for “mail”）.
Phase 2 Key words: The listener identifies phonetically prominent words and forms associative relationships between them to aid understanding（e.g. hearing “mail”, “machine”, and “stamps”）.
Phase 3 Phrases: The listener encodes not only key words but also less prominent surrounding elements that form a small grammatical unit（e.g. hearing “mail”, “put through the machine” and “canceling stamps”）.
Phase 4 Clauses: The listener encodes not only grammatical relationships between lexical words, but identifies semantic relationships between arguments and predicates in a clause（e.g. hearing “the mail is collected”, “taken to the post office”, “it is put through a machine”）.
Phase 5 Clause plus: The listener encodes not only almost all clauses in the input but also the relationships between them （e.g. hearing “the mail goes through several steps before it is delivered”）.
The Data
Let us briefly refer back to the first research question.
１. What speech elements are learners able to attend to when they first
How Schema Theory Can Explain Distortions In Dictation Output
From the notes that the participants wrote down and their reconstructed texts, and with reference to the implication scale developed by Kim, all the participants wrote down proper nouns. The average amount of words written down was 12 words. There were distortions from the original text.
The main distortion was confusing home address and email address. Schema Theory can provide a plausible explanation for this result. Not one of the participants had a magazine subscription nor did they have any experience in dealing with customer service representatives over the telephone. However, given that all of these participants have smartphones, the address that fits their schema is therefore an email address, resulting in six students writing down “mail address” as opposed to five writing “home address”. The other participants did not write anything at all in relation to address. The other distortion was names like “Tan” and “Tim” being written down as opposed to “Tanya”.
Of the 27 participants, only two were able to write down five sentences that resembled the original text. So, using Kim’s scale, two participants could be classified as having phase 3 proficiency while all the others demonstrated phase 1 proficiency.
Weaknesses of the Study
The obvious weakness of this study is the sample size. Case studies, by their very nature, have a small data size is and it is somewhat difficult to extrapolate the results accurately as regards the wider population. However, since the methodology and the theories that were used in this study have been widely accepted within the field of SLA, the results should not be totally discounted.
Another weakness of the study is that there were only two distortions. Given, the low proficiency of the learners and the relatively short length of the text, few distortions should be expected. It could be argued that a longer text would likely have created richer data. The problem with using a longer text, as outlined in Nation（2016）, is that a longer text could be too taxing, especially for lower proficiency learners, which could result in even fewer words being recorded. However, any distortions from the original text can be plausibly explained by Schema Theory.
Consequences for Teaching
The teacher should be aware of what kind of schemata the listeners of the text possess. Eliciting schemata through questioning of the listeners prior to the dictation can create conditions that enable more words to be written down. If the listeners do not have many schemata in regards to the script, the teacher could provide some by paraphrasing the information in the text in order to give the listeners a better chance of increasing the amount of words they write.
Conclusion
This paper outlined a small case study which explored what a particular group of learners attend to when listening to a text. In this case study, most were only able to write down proper nouns and only two were able to write down any semblance of sentences.
One distortion was observed when some of the learners wrote “email address” as opposed to “home address” which was in the text while email address was not mentioned. Schema Theory can provide a plausible explanation regarding the result as most of these learners associate “address” with email address and not home address. Therefore, teachers should be aware of the schemata of their students when preparing the content of a dictation for their learners to perform. Teachers should use scaffolding and elicitation techniques to improve the accuracy and quantity of their learners’output.
Works Cited
Chi, M. T., et al. “Expertise in Problem Solving.” Advances in the Psychology of Human Intelligence, edited by R.S Sternberg, Erlbaum, vol. 1, 1982, p. 1–75.
Cutler, A. “The Comparative Perspective On Spoken Language Processing.” Speech Communication, vol. 21, 1997, p. 3–15.
Kim, Hae-Young. “Intake from Speech Stream: Speech Elements That Learners Attend to.” Attention and Awareness in Foreign Language Learning, edited by R Schmidt, University of Hawaii Press, 1995.
Mandler, Jean, and Nancy Johnson. “Remembrance of Things Parsed: Story Structure and How Schema Theory Can Explain Distortions In Dictation Output
Mandler, G. “Remembering.” Oxford Guide to the Mind, edited by G. Underwood, Oxford University Press, vol. 1, 2001, p. 30–32.
Nation, Paul. How Vocabulary Is Learned. Oxford University Press, 2016.
O’Malley, J., and A. Charmot. Learning Strategies in Second Language Acquisition.
Cambridge University Press, 1990.
Paas, F., et al. “Cognitive Load Theory: Instructional Implications of the Interaction between Information Structures and Cognitive Architecture.” Instructional Science, 32, 2004, p.
1–8.
Rost, Michael. Teaching and Researching Listening. Pearson Education, 2002.
Talcott, Charles, and Graham Tullis. Target Score Second Edition. Cambridge University Press, 2007.
Appendix 1 The Script
Hello. This is Tonya Jackson in the subscriptions department of Business News publications.
You recently subscribed to our magazine but the postal service has returned your first three issues of Business News. They’ve indicated that they are unable to deliver because the address listed is incomplete. Could you please call me back to confirm your home mailing address? The toll-free number is: 1-800-825-9838. Please accept our apologies.
（2017年11月30日受理、2017年12月11日採択）
