Bottom-Up Listening Repair

Students listen to a short extract and repair their mishearings by reprocessing individual sounds and words at the decoding level — training the perception skills that top-down strategies cannot compensate for.

Procedure

1. Choose a 30–60 second authentic audio clip with clear phonological challenge (reduced vowels, elision, assimilation between words).
2. Play the clip once. Students write down everything they heard — gaps and all, using dashes for anything missed.
3. Reveal a partial transcript with 6–8 words blanked out. Students fill in what they think they heard.
4. Play again. Students compare their guesses with what they now hear — do they hear it differently?
5. Reveal the full transcript. Discuss specific decoding failures: which sounds were reduced, linked, or dropped, and why.

Tips

• Choose blanks that expose real phonological processes (schwa reduction, /t/ elision, consonant assimilation) rather than arbitrary words. The goal is consciousness-raising about connected speech, not a general listening test.
• The comparison between Step 3 and Step 4 is the learning moment: students experience their own perception shifting as they attend to the signal differently. This is what makes it "repair" rather than testing.
• Resist playing the audio more than twice before the transcript reveal — the cognitive effort of uncertainty is productive. Too many listens trains dependence on repetition.
• Field distinguishes this approach from traditional listening comprehension: the goal is improving the decoding process itself, not extracting meaning from text.