Commentary: I'm only trying to help: a role for interventions in teaching listening.

INTRODUCTION

In my work as an author and teacher trainer, I have the opportunity to travel around the world and talk to teachers in a variety of settings. Though I meet teachers with a range of backgrounds and a wide disparity of resources, I find that a few common themes come up whenever I talk with teachers about language teaching and technology. One of the familiar refrains is that most of us claim to lack the technological resources we feel we need to teach effectively. There's always something new on the horizon that we feel we just have to have. Another recurring theme is the lament that most of our students just don't seem to take advantage of the extra learning opportunities we present them anyway! Teachers want to help, but often feel underappreciated for their efforts.

Personally, I have relished the ongoing advances in technology over the course of my teaching career. I started out as a secondary school teacher in Togo, West Africa with chalk--sometimes yellow or pink!--and a blackboard as my only teaching technology. When teachers express a sense of being overwhelmed by new technology, I sometimes talk about my own beginnings and also remind them of a few of Donald Norman's principles of human-centered design. According to Norman (2004), for any new technology to be effective, it must be intuitively helpful and elegantly efficient. In the case of language teaching, this means the technology must--immediately and transparently--help us teach better than we do already. If it doesn't, we simply shouldn't use it. In addition, Norman says, for any new technology to be widely adopted, it must appeal to the emotions as well as to reason. If people don't enjoy using a particular technology, no matter how logically useful it may be, they will tend to shun it.

Perhaps because as language teachers we tend to favor eclecticism, we will often throw any emerging technology into the mix as a "helpful resource." As Doughty and Long (2003) point out, teachers often do not distinguish between new technological tools that are innovative but not actually helpful and those which are innovative and genuinely helpful. In my own instructional design, I have identified three "intervention phases" in the listening process: decoding, comprehension, and interpretation. Before we assume any new technology or intervention is actually going to be supportive, I believe we need to understand the learners' goals during these listening processes. What actually motivates the learners towards achieving these goals is what ultimately will be useful.

PERSPECTIVES ON ARTICLES IN THIS ISSUE

The current issue of Language Learning & Technology offers three articles that provide frameworks for evaluating technology in the teaching of listening, in that they examine some of the variables that affect quality of instruction.

In the first article, "Help options and multimedia listening," Grgurovic and Hegelheimer provide a study of input, task and feedback modifications for a recorded academic lecture. An operational goal of their study is documenting how the frequency and time of use of different help options affect learner comprehension. This study confirms the current position in much CALL research that the additional "interactions" with support options not only tend to aid text comprehension, as demonstrated by increase in post-listening test scores, but also promote language acquisition, as inferred through the input-interaction hypothesis (that is, if interaction, specifically repair-motivated interaction, promotes comprehension, and if comprehension promotes acquisition, then interaction promotes acquisition). While the audio-video input in the study, an Astronomy lecture, itself is not modified or elaborated, the opportunities for processing input are amplified through the optional use of repeated viewings, subtitles, transcripts, lexical pushdowns, and feedback on responses to comprehension questions.

The main value of the Grgurovic and Hegelheimer study, in my view, is not so much the attempt to substantiate the position that increased interaction (human-human or human-machine) tends to promote comprehension and acquisition. For me, a more pragmatic value is in the authors' investigation of the learners' patterns of navigation for the support options. The authors suggest that navigation patterns in use of subtitles, transcripts, dictionary look ups, and explanations are related to proficiency: More proficient learners make more use of the additional options. This is not surprising, of course. Lower proficiency students generally seek less input because if they are processing input inaccurately or incompletely, more input usually leads to more confusion. Because of this "help option dilemma", teachers and media designers need a qualitative instructional paradigm that introduces support options in ways that are intuitively supportive to the learners. For instance, Martinez (2001) suggests that different types of learners ("transforming learners," "performing learners," "conforming learners") prefer different sequencing and alternative representations of "support" during a task. If we want to assist learners as they listen, we cannot assume simply that more intervention leads to better learning outcomes.

In the second article of this volume, "Are They Watching?", Wagner provides a study of listener behavior in video-based test taking situations. Prior to the description of his own study, the author provides a valuable survey of recent studies on the use of video to teach listening. Wagner leads up to the now axiomatic claim that multimedia experiences provide learners with richer, more authentic and more memorable encounters with the target language. The well-established arguments for teaching listening through subtraction of all but the audio channel are outweighed by the value of authentic, engaging input. (There is however a ceiling effect on the value of richness in multimodal input for teaching purposes, particularly concerning simultaneous presentation of graphics, text, and audio. See Clark and Mayer (2002) for a discussion of how "seductive details" in multimedia can depress learning.)

The hub of the Wagner study is an observation about viewer attention that previously had not been clearly documented. Wagner investigates the link between visual input display and attention to input. He finds that learners in his study do consistently attend to the video portion of the input in his constructed testing situations, "orienting" to the video screen on average about 70% of total time on task. This is a useful starting point for probing the notion of learner attention, though, in my view, there's not sufficient accounting for the non-verbal version of the Observer's Paradox. The students in the study had a video camera pointed at them from the top of each monitor during the entire test. This feature could well have stimulated additional orientation to the screen, in effect encouraging the behavior the researcher was trying to measure.

One of Wagner's queries about his research results is especially intriguing. He wonders why learners oriented less to the video screen during dialogue input than they did during monologue input (there was about a 10% time differential). He points out that in dialogue settings a great deal of social information needed for comprehension is transmitted visually, so it would seem that listeners should use their visual channel more when processing this kind of conversational input.

My own understanding of the work on bimodal processing is that attending to multiple modes provides greater redundancy. Listeners need redundancy of all sorts. It is an essential condition for effective language processing. However, redundancy is not purely an additive-subtractive construct (Moreno & Mayer, 2002; Paivio, 1986; Reed, 2006). Once two or more modalities are combined during input processing, they cannot be separated out. The observation of this phenomenon can be traced to experimental studies of perception in the 1970s, first described by McGurk and McDonald (1976), and latter dubbed the "McGurk effect." This phenomenon demonstrates that human information processing tends to be visually dominated, but the information we perceive through hearing and through sight are coordinated, interrelated, and irreversible. In Wagner's study, once the learners start watching and listening, and have ongoing access to both input channels, they will be utilizing both sight and hearing simultaneously. My sense is that the viewer experiences a "fused perception", which does not occur in an alternating or additive fashion, even when the viewer is seemingly ignoring the video input. Once viewers have become engaged in processing meaning from the video (images + sound), even when they temporarily turn away from the monitor, they are still "seeing" images, and when they shut out the audio input for a moment, they are still "hearing" the associated sound.