Research Bites: Speech Perception, Speech Production, and Corrective Feedback

Among the numerous factors that influence pronunciation, many have argued that listening – in particular, listening discrimination, plays an important role.

Lee and Lyster (2016) explore this connect between how listening – namely, speech perception, influences speech production. This idea, known as the perception-first view, is well-supported by empirical studies, though it is not without some contention. Lee and Lyster in particular focus on speech perception training and its effect on phonological production (pronunciation). Reviewing a number of studies, the authors indicated that a common training element was corrective feedback. Their study presented below looks at the possible role corrective feedback (CF) may play in moving from accurate speech perception to accurate speech production.

Lee, A. H., & Lyster, R. (2016). Can corrective feedback on second language speech perception errors affect production accuracy?. Applied Psycholinguistics, 1-23. Retrieved from

They conducted their research with 100 Korean learners of English. They divided them into five groups, each of which underwent speech perception training for eight sessions during two weeks through specially designed software. The training included listening to various words that represented words with trouble vowels for Koreans: /i/–/ɪ/ and /ɛ/–/æ/. They were able to listen to each word as many times as they wanted and then they had to choose the word orally represented. For example, they heard /ʃɪp/ and had to choose between “ship” and “sheep”. Based on their answer and the group they were in, they received the following corrective feedback:

Target Group “No, s/he said ‘ship’.” Yes
Nontarget Group “No, not ‘sheep’.” Yes
Combination Group No, s/he said ‘ship’ not ‘sheep’.” Yes
Wrong Group Wrong Right
Control Group None None

The participants were audiorecorded three times (pre, post, delayed post). They had to produce sentences that included the trained words, as well as some untrained words. Analysis of these recordings was done using native English speakers and acoustic analysis software.

Lee and Lyster found the following:

  • Target Group:
    • Production accuracy was significantly higher for trained words at both the post- and delayed posttest;
    • Production accuracy for /ɛ/–/æ/ untrained words was higher at both posttests;
    • Production accuracy for /i/–/ɪ/ untrained words was higher for the immediate posttest only.
  • Nontarget Group:
    • No significant changes for /i/–/ɪ/
    • Production accuracy was higher for /ɛ/–/æ/ trained words at both postests
    • Production accuracy was higher for /i/–/ɪ/ and /ɛ/–/æ/ untrained words at the immediate posttest
  • Combination, Wrong, Control:
    • No significant changes

Overall, they reinforced the idea in the relationship between speech perception and speech production, but CF type was a major factor. They found that providing target feedback (which is akin to a ‘recast’) is more effective than providing negative feedback (which is akin to prompts). That is, giving the target form in response to incorrect perception was better than simply telling them which sound was wrong.

How does this influence improved speech production? The researchers noticed that both target and nontarget groups would verbally respond to CF by trying to produce the correct utterances. The target groups did this more often than the nontarget group, and the other three groups, by the nature of the CF type, did not engage in this behavior. Therefore, speech perception alone is important, but “opportunities for noticing, awareness, and practice, in addition to CF” might be necessary (p. 18).

Research Bites: Gesture and Pronunciation

Smotrova, T. (2015), Making pronunciation visible: Gesture in teaching pronunciation. TESOL Quarterly.

Research shows that body movement and speech are intimately linked, with some theorizing that they are from the same cognitive source. Whether this is true or not, what is known is that body movements and speech are unconsciously coordinated and that these gestures are not random but coordinate to meaning. What is also known is that gesture often occurs alongside not only conversation but instruction in the classroom. Smotrova first looks at how gesture has been used in pronunciation teaching, pointing out it is one of the least researched aspects of language instruction. Clapping, rubber bands, mirroring and imitation, and even some gesture-systems such as the “essential, haptic-integrated English pronunciation (EHIEP) framework“, have been employed in pronunciation teaching, many of which have been shown to be effective. However, overall there is a paucity of research in this area. In this article, Smotrova analyzed gestures as they occurred in a classroom during pronunciation instruction. Her analysis is in-depth and concludes with two important implications: 1) teachers should be made aware of the importance of gestures and utilize them, perhaps systematically, in their instruction; and 2) students should use the teacher’s gestures because they are beneficial and effective to their learning.

Related Links:

Research Bites: Reading, Pronunciation, and the Phonological Loop

(Thanks to Gianfranco Conti for making me aware of this article.)

In “Phonology in Second Language Reading: Not an Optional Extra”, Walter (2008) argues that the ability to distinguish and create the sounds of the L2 serves as an important underlying mechanism for L2 reading comprehension. In other words, good pronunciation skills help students become good readers.

The implications of Walter’s study points to “evidence that the development of a reliable phonological repertoire in L2 provides an important basis for skillful reading” (p. 469). She argues that teaching comprehension is not enough. Poor comprehenders (who may be good single-sentence decoders) “need to be better at mentally representing spoken language” (p. 470). More exposure to the target language’s sounds as well as reading at or below level with audio may be one way to address this problem. In addition, minimal pair work, dictations with confusable words, and other activities that explicitly help students differentiate sounds can help build a reliable L2 phonological repertoire. Thus, reading instruction must include the distinguishing of sounds. Continue reading

Be Explicit: The Power of Teaching about How to Learn

I’m still trying to process all the wonderful information picked up from sessions at this year’s Southeast TESOL conference in New Orleans. As I think about my experiences and look over my notes (summary blog posts coming soon), I am struck by a common theme I noticed in a number of sessions I attended. This theme wasn’t the main focus of any of the sessions, nor do I think the presenters were necessarily emphasizing this particular idea, but it is one I have seemed to synthesize from the talks, discussions, research, pedagogy, and experiences that filled the air: be explicit.

Now, I’m not talking about explicit grammar teaching (the jury is still out on that one). I’m talking about explicitly teaching students about learning – how they learn and the best ways the learn.

Teach the Brain

The most interesting and exciting session I went to was by educational neuroscientist Janet Zadina, “Empowering English Language Learners: Insights from Neuroscience”. This was a fascinating talk – one which I plan on dissecting more of when I have the time. Besides dispelling numerous neuromyths and explaining how learning works in the brain, Dr. Zadina mentioned a curious idea: teach students how the brain learns. Now, I cannot find the reference she gave for this information. However, she stated that teaching students how their brains learn and the best strategies to utilize for learning has been shown to actually improve learning!

Dr. Zadina briefly spells it out in this talk: Teaching students how the brain works includes teaching them that the brain can change, they learn through neural networks, and how to strengthen neural networks when they are having difficulties. A quick glance at Dr. Zadina’s book chapter titles give even more information on what can be taught to students about their own brains.  For even more resources, Larry Ferlazzo has a great post on this. And check out this great post on Edutopia.

Teach the Mind

Zadina’s research no doubt has a connection to Carol Dweck. Dweck was brought up several times this week, including during Carol Read’s wonderful presentation. Dweck, for years, has been researching the concept of the fixed and growth mindset. People with a fixed mindset believe that intelligence is fixed. They react negatively to mistakes and errors and don’t use them as opportunities to improve because they don’t believe they can. Growth-minded people believe one’s intelligence can grow. They put the effort into learning (and learning from mistakes) because they understand learning is a process that requires work. Read, echoing Dweck, calls for teachers to explain to students these mindsets and help them switch to a growth mindset where students value the process of learning, not just the end result. These calls are also supported by much of Dweck’s research, which shows that switching to this mindset contributes to actual gains in learning.

Teach the Language

There was a speaker from the Defense Language Institute English Language Institute who discussed learning strategies to promote communication in the classroom. Learning strategies are a heavily researched field within ELT and both quantitative and qualitative research have shown that strategy instruction is valuable in the language classroom. There are many strategies that can be taught, and many ways to teach them, but much of the literature reinforces what the speaker stated: be explicit. Naming the strategies was among the different parts of the framework the speaker was discussing. Students should not only learn that there are strategies, and why/how they should use them, but they should be able to refer to these strategies as well. This can facilitate planning, monitoring, and evaluating strategy usage, which in turn facilitates language learning.

The National Capitol Language Resource Center has a great free online book called “Developing Autonomy in Language Learning” which deals heavily with language learning strategies, including how to teach them.

Teach the Mouth

I also attended a workshop called “Speech Conditioning for the Second Language Learner”. It focused on a framework for teaching pronunciation starting with rhythm, then vowel work, and then intonation – all with the goal of increasing intelligibility, not accent reduction. The framework that was introduced follows ideas of muscle training in any sport. Pronunciation involves a re-learning of the many muscles of the mouth. To assist in this muscle work, the speaker explicitly taught students about how pronunciation works. She starts her classes with discussions of this Wikipedia article on isochrony, helping students understand the difference between syllable-, mora-, and stress-timed languages and where they fall in this categorization. This allows students to notice, first, that these differences even exist, and second, what kind of speaking pattern English has. In addition to rhythm work, she teaching mouth shapes and tongue positions to assist in learning vowels. This type of explicit instruction is not done everyday. Rather, it is frontloaded in the course so that students have this metalinguistic foundational knowledge to base their future practice on.

Explicit pronunciation teaching is nothing new. I watched a very interesting IATEFL presentation about a very similar idea called “What to teach before you teach pronunciation.” Piers Messum, the speaker, takes a similar, explicit approach in helping learners understand the muscles involved in speaking English. Knowledge of consciousness of these is said to lead to better and clearer speech. I know that as a learner of French, Korean, and Polish, explicit knowledge of the articulatory processes did help me understand pronunciation more, and extending this knowledge into practice does seem to help better cement the tricky tongue work and mouth movements needed to master a foreign language.

Warning: Explicit Language

So, the take away: be explicit in order to empower students. This explicit knowledge is like a flashlight. Students who have this kind of knowledge of the underlying processes of learning and language have a great advantage to those who are stumbling in the darkness that is the complex phenomenon of learning. Anything to help illuminate the path will be beneficial.

Getting in Bed with TED (Some ideas for using TED Talks)

I haven’t always been a fan of TED Talks. I’ve always watched them with a bit of apprehension, having never liked the compression of ideas into a few minutes, the scripted and overenthusiastic applause, or the elitist atmosphere of the speakers and talks. However, I have come around. I realized, first with surprise and then with admiration, that my students actually enjoy TED a lot more than I do. They apparently don’t have the cynicism that I do. So, while I still find some of the well-timed cheers and applause gag-worthy, I can now see the promise of TED, both in the dissemination of ideas (hence their tagline: “Ideas Worth Spreading”) and in the utility of this medium for my students.

TED Talks now form a solid core of my upper level listening courses because my students find the topics interesting, and because the videos are extremely exploitable, that is, they lend themselves to a lot of adaptation and adoption. So, I wanted to share some activities I have done and some ideas I have had for using TED Talks in the classroom.

First, a quick breakdown of why TED Talks are so exploitable:

  • They include a transcript
  • They often include multiple translations
  • You can download the video for offline viewing/editing
  • You can download only the audio for offline listening/editing
  • They represent enough topics to get anyone interested
  • They are relatively short, from between 5-20 minutes
  • Creative Commons license

Keep in mind that TED Talks may only be useful for intermediate and above learners, and usually for adolescents and up. So, not for really for everyone. But, for those of us who teach older and more proficient learners (especially in more academic environments), TED Talks offers a treasure trove of material which only requires a little creativity and imagination.


TEDxESL offers ESL lessons built around TED Talks. Most of the lessons are grammar and vocabulary heavy with some communicative tasks and activities, but they are a great place to start if you are looking for ways to exploit TED.


TED Talks serve as a great way to anchor any lesson, whether it is grammar, vocabulary, speaking, writing, etc. Anchoring a lesson means putting your lesson into an interesting, meaningful and relevant context. It builds interest, motivation, and helps to activate (or build) background knowledge. In other words, anchoring a lesson lays important groundwork for learning.

If you can find a connection between what you are teaching and a TED Talk, then the talk can serve as a great way to build background knowledge and interest in your topic. It could serve as a springboard for generating discussion, as context for several grammatical points of vocabulary, or as a muse for essay idea generation.

Grammar Study

The sky’s the limit when it comes to grammar. You can find most grammar points, simple and complex, used throughout the hundreds of hours TED Talks. The transcripts will really help with this. The question becomes which to focus on? My recommendation would be to find a highly interesting talk and pull out the grammar points that you think the students a) need to review, b) need to learn, or c) will struggle with. Another idea is to have students follow along with the transcript and underline any grammatical structures that cause confusion. Have them write the sentences on some slips of paper and give them to you. You now have a week’s worth of lessons and ideas. You could also use the translated subtitles to offer translation practice (for budding translators, or as another method of grammar learning) or compare the grammar and nuanced meaning of the two languages. Need to find grammar points fast? Try the TED Corpus Search Engine or analyze a script with AntConc!

Vocab / The Academic Word List

Just like grammar, there is a plethora of vocabulary that can be gleamed from TED Talks. You can focus on common words or words from the academic word list. I would copy and past the script into the Vocab Grabber and focus on less common words. And like grammar, you can also have students generate the lessons. Have them choose some words from the script. They can then look them up and teach them to their group members. Afterwards, they can write a dialogue or script using those words and perform their script for the class. They can also recycle this vocabulary into presentations or essays.

Audacity / Audio Editing

Because TED offers the audio-only version of any talk, these talks lend themselves to audio editing. Using Audacity, you can quickly pull out excerpts, speed up or slow down the audio, or create loops of specific words or sounds you want students to focus on.

Listening Journals

I am a big fan of listening journals – getting students to do both extensive and intensive listening practice while at the same time reflecting on their listening experiences. TED Talks make excellent sources for listening journals. I couple the extensive listening of TED Talks with gap-fills or transcription activities from the students’ favorite parts.


There are a lot of different speakers from all over the world on TED. You can help students explore and be exposed to different accents by listening to various TED speakers. Listen to a different one each week or get them to explore TED accents on their own. Here is a list of 25 videos featuring different accents

TED Corpus Search Engine

Yoichiro Hasebe, a professor of linguistics from Japan, has create the multi-lingual, multi-modal TED Corpus Search Engine, which allows you to search for words or phrases and see these words in their textual context as well as their aural/visual context (i.e. you can see the part of the TED talk they actually come from). In addition, you can select translations (taken from TED subtitles) which can be displayed alongside the text. There is probably a lot you can do with this, from analyzing lexical and grammatical items, noticing usage patterns, or hearing a word’s pronunciation in context.


Gap fills may seem like a traditional activity, but they have stuck around so long because they are effective at helping students work on their decoding skills. I get students to do gap-fills along with their listening journals, but there is no reason why they couldn’t do it in class, especially if you get students to reflect afterwards on why they misheard certain words. I typically make it so my gaps can be one or more words, which forces students to focus not just on single words but on multiple word utterances that may have gone through so elision or another connective process.

Likewise, TED Talks would be very useful for transcriptions and dictoglosses and I would definitely use these in any class that has a listening focus.


Any TED Talk could be used to help students with their pronunciation. These talks may serve as an excellent model from which students can practice individual words, thought groups, word stress, etc. I would choose a short excerpt (10-20 seconds, perhaps one they have completed for a gap-fill) and have students do a mimicry exercise where they try to record themselves saying the exact same thing in the exact same way as the speaker. I have had students actually record themselves first play the clip, hit pause, say their part, and repeat – but getting them to do the whole thing at once is better. This mimicry exercise is great because it gets them (hopefully) to not only pronounce their segmentals clearly, but has them practicing suprasegmentals such as word stress, elision, intonation, pauses, etc. In addition, it gets them to analyze their own pronunciation by comparing it to the model and judging whether or not it is similar/intelligible enough.


No EAP class would be complete without note-taking practice. TED, while straying from the format of the traditional academic lecture students may encounter, does give students a chance to listen and take-notes, hopefully following the Cornell method.

Academic Speaking Circles

This is an idea I have modified from Tyson Seburn’sAcademic Reading Circles“. In the original concept, students read an academic text and are given specific roles with which they to use to have a discussion following the reading. The roles include a leader (they gauge groups understanding, ask comprehension questions), a contextualizer (they research topics and concepts from the text), a visualizer (they find things from the text that could be visually represented), a connector (they ask questions to draw connections between the text other lessons, courses, or everyday life and experiences), and a highlighter (they focus on interesting linguistic structures like grammar and vocab).

Seburn’s Circles talk about a “text” and in ELT jargon, a “text” is not something read but it is a form of input. Clearly, TED Talks are a form of input and thus a form of text. Therefore, it would be very easy to adapt Seburn’s idea to Academic Discussion Circles based on a TED Talk (I am trying this out this term). Some modifications might be to alter or combine the role of visualizer (since TED often includes visuals) and add pronunciation features to what the highlighter should be looking for.

Presentation Skills

I have stopped getting my students to do presentations in my classes because 1) I don’t really feel it prepares them for the few classes they may have to give presentations in while at university and 2) I don’t feel like I have enough time to teach good presentation techniques. However, for those who do wish to teach these skills, TED offers a number of great speakers who students can watch, analyze, and model. Getting students to do a TED-style presentation would be very fun, very rewarding and most likely payoff down the line of students’ academic careers.

Mindsets and Meta

Fall, 2014: I had just learned about the fixed vs growth mindset and found a wonderful talk by this concept’s main researcher, Carol Dweck. I used this TED Talk not only to practice some listening at the beginning of the term, but also to establish the concept of fixed vs growth mindset. For some of my students, this was their last term before they would possibly enter university and they needed to make progress. Figuring out those who had a fixed mindset and instilling into them the idea of a growth was an important first week challenge. In the first week of class, we learned (or relearned) about the power of errors, mistakes, and failing – and that they could in fact achieve their goals and make their English grow. Throughout the term I constantly referred to the fixed/growth mindset and for some of my students, I could see that it really hit home.

TED could be an excellent source for this kind of metacognitive priming, and it could be using as a reference (and inspiration) point throughout a course. A colleague of mine told me about a gaol-setting video she had her students watch. When she saw students struggling, she would refer back to the video and get them to focus on a specific goal in order to overcome their struggles. This seemed to work well for her.

TED has many, many inspirational talks that can be tied directly into helping change a class’ mindset, perspective, or way of learning.

More Ideas

IPA (Phonetics) Resources

If I were a beer drinker, I would make some joke about the IPA. Unfortunately, I don’t drink beer and take myself too seriously to besmudge the fine acronym IPA, which stands for the International Phonetic Alphabet – a quite marvelous invention meant to represent the sounds of human speech. IPA is also a useful tool for teachers and students – it allows you to see dictionary and spoken forms of words, compare accents and dialogues, and its just cool. So, here are several IPA resources that I frequently use:

Sounds: The Pronunciation App

Available for Android and iOS, this app is useful for learning the English sounds of either British or American English. It comes with a simple IPA chart, some IPA exercises (read and write, listen and write) and a minimal dictionary for examples. I recommend the free version.


Type IPA

I use this site all the time to type IPA. It gives you the option of only using an English IPA keyboard or you can choose the full IPA keyboard.



PhoTransEdit’s Text to Phonetics page allows you to type regular English words, choose between British (RP) or General American pronunciation, and then see the IPA version of the word. Very useful!



Wikipedia offers an indepth IPA chart for a range of English accents. You can click on most of the symbols and hear a sample. They also offer the complete IPA chart for all languages.


Learner’s Dictionary

I recommend this dictionary to all students. It has the simplest definitions of words, which are of very high quality. It also offers IPA and audio pronunciation.


Phonetics from the University of Iowa

See American English sound animations. Really useful and cool website. They have also just released a mobile version for Android and iOS. Check it out!

Listening, Pronunciation, and Connected Speech

I stumbled upon a very creative video today called “How English Sounds to Non-English Speakers“. The video centers around a couple having a conversation in what sounds like English, but is not very comprehensible. While watching, you catch glimpses of English words, but with little context. If you have ever studied a foreign language, thus experience will be very familiar to you: you know the individual words, but when they are strung together in natural, connected speech, the only thing you can pick out are keywords – and these may or may not lead to comprehension.

I couldn’t have found this video at a better time. I am currently teaching about word and sentence stress in my graduate pronunciation class.  In this video, the only words that are clear are ones that receive the main stress in a sentence. The words that carry the most important meaning are often the words that are stressed the most in a sentence, and thus are pronounced clearly while other sounds get linked, changed, or even omitted.

This is important information, because it means for anyone trying to understand another language, they need to have a good idea about not only the meaning behind stressed words (the keywords we often ask students to listen for) but also the unstressed connected speech which does in fact carry important meaning.

One way to conceive of this is to think of the prosody pyramid. Gilbert (2008) introduces the idea of the prosody pyramid as a sort of framework for aiding listening and pronunciation. As the first tier of the pyramid is the thought group. Pauses and rising and falling intonation usually signal the beginning and end of a thought group, or what can be called a small chunk of meaning. Think about the commas we use in writing. They are used to not only give the reader a slight break but to signal a shift or change in thought. This could be contrasting two independent clauses (i.e. two thoughts), setting aside information (such as an in a non-restrictive relative clause), or simply listing. Teaching students how to both signal thought groups by using proper intonation, as well as listening for thought groups, is an important skill.

In each thought group, there will be a stream of connected speech with at least one stressed word – the focus word (second tier). This is the word that carries some important meaning. Though there are different levels of stress in a sentence, and all words are important in a sense, the focus word has special importance for a few different reasons. First, for the speaker, it is the word that needs to be pronounced the most clearly. The other words can be somewhat “muddy”. (Proper stress and clear phonemic pronunciation are indicated by the third and fourth tiers respectively.) Second, for the listener, they not only have to pay attention to the focus word, but also be able to make sense of all the unfocused, unstressed speech that occurs around it.

This is where the challenges lie. The student focused on pronunciation must learn unstress and therefore connected speech so that their language flows – not necessarily because native speakers do it. Yes, native speakers use connected speech, but so do non-native speakers. In addition, connected speech occurs because it assists the mouth and tongue, which must rapidly change positions during speech. All languages have connected speech because of this. Though it may fly in the face of the idea of intelligibility, its essential for flow.

Admittedly,  the challenge is not so much for the production but rather the reception of connected speech. Listeners have the real challenge. A great article by Sheila Thorn (2009) discusses this idea at length. She states that most listening tracks are not only unscripted and unauthentic, but are used for modeling and introduction of vocabulary rather than actual listening training. According to Thorn:

The major flaw with this approach is that it has become too successful! Students tend to concentrate too much on constructing meaning from key words, often with a spectacular lack of success. This is because they are not paying attention to other non-content words, many of which contain essential information. (p. 6)

Even when the texts are authentic, they are not used very effectively, as it is automatically assumed learners will just “pick it up” after enough exposure. While relying on stress to pick out the keywords is important, it is not the only important goal of listening comprehension. That gibberish we heard in the video above? That stream of words that explode out of foreign language speakers’ mouths? That is the other goal. And it is note an easy one to achieve:

We need to view this skill as the ultimate objective for our students to attain, whilst accepting that they will only reach this objective at the end of a long learning road. Our role as teachers is to support our students as they take their first steps along this road and help them increase their pace. (p. 7)

Thorn’s article has a number of ideas on how to reach this goal. These include supplementing with authentic text and creating gap-fill activities that do not solely focus on lexis or content words, but rather the words that all students know but probably can’t understand in natural speech. In a class on pronunciation, with a pronunciation focus, or with a teacher who understands the importance of integrating pronunciation into conversation classes, this also means explicit practice in all aspects of connected speech: linking, assimilation, elision, etc. This practice, of course, should be both productive and receptive. Basically, these students should be doing a lot of authentic listening, analyzing and paying attention to connected speech rather than stress.

The peak of the prosody pyramid is contingent on correctly produced stressed focus words. However, without a firm base of connected speech on which to be supported, the peak is fragile. Only a limited sense of meaning can be gleaned from it. Therefore, a balance must be struck between the vast and wide base and the narrow and specific peak. That is to say, a balance between unstress and stress. Though it is quite a challenge, it is a worthy one – one which will likely improve all students’ listening and speaking abilities.


Gilbert, J. B. (2008). Teaching pronunciation using the prosody pyramid. Cambridge University Press. Retrieved from

Thorn, S. (2009). Mining listening texts. Modern English Teacher (18)2, pp. 5-13. Retrieved from