By Mr. Boukhana Noureddine, EFL teacher, Mohammed VI Qualifying High School
For years I taught vocabulary the way I had been taught it. Write the word on the board, give the meaning, translate if the class looks lost, ask everyone to copy it down. It takes about three minutes per word and it feels productive. Then two weeks later I would ask about "hilarious" or "reluctant" and get blank faces from students who had copied them perfectly.
After more than thirty years in Moroccan classrooms, I no longer blame the students for that. The method gave them nothing to think about, so they thought about nothing. This article describes the alternative I use now: teaching vocabulary through OHEE (Observation, Hypothesis, Experiment, Extension), a guided discovery sequence. I explain why it works, show a full 55-minute lesson, and say honestly where it struggles.
Why the "word on the board" lesson fails
The research on this is older than most of our students. Craik and Lockhart (1972) argued that how deeply you process information decides how well you remember it. Copying a translation is shallow processing. Deciding what a word means from the sentence around it is deep.
Laufer and Hulstijn (2001) turned this into something a teacher can use: the Involvement Load Hypothesis. A vocabulary task is more effective when the learner feels a need for the word, has to search for its meaning, and has to evaluate whether the word fits a context. A glossary gives the learner none of the three.
There is a second problem. Nation (2001) describes knowing a word as knowing its form, its meaning and its use: which words it sits next to, which grammar it takes, how formal it is. A translation covers meaning only. A student can "know" that sarcastic means mockeur or its Arabic equivalent and still write "he is sarcastic with the joke" in an exam paragraph.
What "discovery" should and should not mean
Bruner (1961) championed discovery learning, and the idea has been misread ever since as "leave the students alone and they will work it out." They will not. Mayer (2004) and Kirschner, Sweller and Clark (2006) reviewed the evidence and found that unguided discovery performs poorly, especially with beginners.
OHEE avoids that trap. The teacher chooses the examples, controls how much students see, and confirms or corrects every conclusion. Students do the noticing and the guessing. I do the designing and the checking. That is guided discovery, and for vocabulary it fits well because words usually arrive with context that can be read for clues.
The five stages for vocabulary
My standard 55-minute lesson runs like this:
| Stage | Time | What happens |
|---|---|---|
| Warm-up | 5 min | Students talk about the topic and produce words they already know |
| Observation | 10 min | Students study target words in context, with no definitions given |
| Hypothesizing | 15 min | Three moments: receptive, constructive, productive |
| Experiment | 15 min | Students test and revise their hypotheses in new contexts |
| Extension | 10 min | Students use the words on something from their own lives |
The warm-up stands alone. It is not part of Observation. Its only job is to switch on what students already know about the topic, and to give me a first semantic map to build on.
A worked example: the unit "Humour"
Target words: hilarious, witty, sarcastic, a pun, a punchline. Five words is plenty for one lesson. When I try ten, I get ten half-learned words.
Warm-up (5 minutes)
I ask: "What was the last thing that made you laugh?" Students tell a partner, then a few share with the class. I write whatever vocabulary they produce (funny, joke, laugh, comedian) on one side of the board. This becomes the seed of a semantic map. The new words will be added to it later.
Observation (10 minutes)
I project five short contexts with the target words in bold and no definitions:
- The comedian was hilarious. The audience could not stop laughing.
- My uncle is very witty. He always has a clever answer, and everybody listens when he talks.
- "Great, another Monday," she said sarcastically, rolling her eyes.
- "I used to be a baker, but I couldn't make enough dough." That is a pun.
- Everyone waited for the punchline, but the story just stopped.
Students work in pairs and answer three guiding questions. Is the word positive, negative or neutral? Does it describe a person, a thing or an action? Which words nearby gave you a clue? I walk around and say almost nothing. The temptation to explain is strong. I resist it.
Hypothesizing (15 minutes)
This is the stage most teachers collapse into one step. I split it into three distinct moments, because each one asks something different of the student.
Receptive (5 minutes). Students match each word to one of several possible meanings I provide. They are recognising, not producing. Some options are deliberately close: hilarious against "quite funny" and "extremely funny." This is where arguments start, which is good.
Constructive (5 minutes). Now they write their own definition, in their own words, in English or in a mix of English and Arabic or French if that is what it takes. They also complete short frames: "A person who is witty can ______." "A pun plays with ______." They are building the meaning, not receiving it.
Productive (5 minutes). Each student writes one original sentence using one of the words, about a person or situation they actually know. Pairs swap and read. I collect a few sentences, including wrong ones, and write them on the board. A sentence like "The film was sarcastic" is a useful mistake: we discuss why a film can be funny but a person is usually sarcastic.
Only after this do I confirm the meanings and correct the wrong hypotheses. By that point students want the answer, because they have staked a guess on it.
Experiment (15 minutes)
Students now test their hypotheses against new material. I rotate through activities rather than doing all of them. For this set I would pick three:
- Gap-fill in a short text. A 6-line dialogue about a family dinner with five gaps. Students must choose the right word and justify it.
- Odd one out. hilarious, amusing, witty, tragic. Which does not belong, and why? Then a harder set where two answers are defensible.
- Complete the semantic map. Students add the new words to the warm-up map, linking them to funny, joke and so on, and mark which ones describe people and which describe things.
A student whose hypothesis was wrong will usually feel it here, when the word will not fit the gap. That small discomfort teaches more than my correction would.
Extension (10 minutes)
The extension has to leave the classroom, otherwise it is just a recap. I ask students to interview a family member before the next lesson: "Who is the wittiest person in the family? What is the funniest thing they said? Can you remember a pun, in any language?" They write three to four sentences using at least three target words. Next lesson, five minutes of sharing opens the session.
This is also where students decide whether the word is theirs. Describing an uncle who really is witty is different from completing an exercise about a fictional Tom.
Where it gets difficult
Time. Fifty-five minutes is tight. If the warm-up runs long, something has to give, and it should be the number of words, never the Hypothesizing stage.
Class size. With large classes, pair and group work does most of the work, and I cannot hear every hypothesis. I listen to three or four pairs closely and use their ideas to represent the class. It is imperfect.
Level. Common Core students need more scaffolding than Bac students: fewer words, simpler contexts, more options in the matching task. The structure stays the same. The difficulty of the data changes.
Word type. Abstract words and words with no context to lean on do not discover well. Function words and very high-frequency items should simply be taught directly. OHEE is for content words that students are going to need to use, such as the vocabulary of a unit.
Preparation. Writing good observation data takes time. A bad example (one where the clue is missing or the sentence could mean five things) wastes the stage. I now keep a folder of tested contexts and reuse them.
Checking that the words stayed
A gap-fill at the end of the lesson proves very little. It measures what students remember for twenty minutes. I do a one-minute oral or written check at the start of the next lesson, then recycle the same words in a reading passage a week later. If students recognise the word but cannot use it, the Extension stage needs more work.
A last practical suggestion
If you want to try this, do not rebuild your whole scheme of work. Take one unit, pick five words you would normally put on the board, and design only the Observation stage: five contexts, three guiding questions. Run it once and see what your students say. In my experience the first reaction is confusion, the second is argument, and the third, by about the tenth minute, is a student saying "Sir, I know what it means." That is the moment the method pays for itself.
References
Bruner, J. S. (1961). The act of discovery. Harvard Educational Review, 31(1), 21-32.
Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11(6), 671-684.
Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance during instruction does not work. Educational Psychologist, 41(2), 75-86.
Laufer, B., & Hulstijn, J. (2001). Incidental vocabulary acquisition in a second language: The construct of task-induced involvement. Applied Linguistics, 22(1), 1-26.
Mayer, R. E. (2004). Should there be a three-strikes rule against pure discovery learning? American Psychologist, 59(1), 14-19.
Nation, I. S. P. (2001). Learning Vocabulary in Another Language. Cambridge University Press.