For speech-language pathologists
What is minimal pairs therapy, and how do you choose pairs?
Published · Updated
The short answer
Minimal pairs are words that differ by one sound that changes the meaning, such as “key” and “tea”. In minimal pairs therapy, a child who says both words the same way plays games where saying the wrong sound gets the wrong picture, so they learn the contrast matters. Variants choose pairs differently: maximal oppositions, multiple oppositions and complexity-based targets. A 2022 tutorial recommends conventional minimal pairs for children with few error patterns, and maximal or multiple oppositions for more severe disorders.
What minimal pairs are
ASHA defines minimal pairs (minimal oppositions) as pairs of words that differ by only one phoneme or single feature that changes the meaning. They’re used to establish contrasts the child’s phonological system doesn’t yet have (ASHA). The focus is on sound rather than spelling, and the contrasting sounds can be at the start, middle or end of the word (Storkel, 2022).
- Velar fronting (k → t): key / tea.
- Velar fronting (g → d): go / dough.
- Final consonant deletion: bow / boat.
- S and SH: seat / sheet.
For printable sets by contrast, including k/t, g/d, s/sh and final consonant deletion, see the free minimal pairs word lists. For which patterns are typical at which ages, see phonological processes by age.
Minimal pairs, maximal and multiple oppositions, and complexity
| Approach | How targets are chosen | Who it suits |
|---|---|---|
| Conventional minimal pairs | The target is paired with the child’s own substitute or deletion, so the child’s error makes the two words sound the same (Storkel, 2022) | Children with a small number of error patterns: older children or those with mild disorders (Storkel, 2022) |
| Maximal oppositions (Gierut) | The target is paired with a sound that differs on many features (voice, place and manner), ideally across major sound classes; treatment of the “empty set” pairs two sounds the child doesn’t know (ASHA; Storkel, 2022) | Children with several errors across different sound classes; studied from age 3 to 8 (Storkel, 2022) |
| Multiple oppositions (Williams) | One to four targets from a phoneme collapse are contrasted with the single substitute at once, chosen to cover the breadth of the collapse and to differ maximally from the substitute (Storkel, 2022) | Children with a global phoneme collapse, where one sound replaces many, usually younger children or those with moderate to severe disorders (Storkel, 2022) |
| Complexity approach | Targets are more complex, linguistically marked sounds not yet in the child’s system, such as clusters, fricatives and affricates, or the least stimulable sounds, to spark wider generalization (ASHA; Gierut, 2007) | Preschoolers with phonological disorders (Storkel, 2018); ASHA lists it as a way of selecting targets rather than a treatment procedure (ASHA) |
In the first maximal opposition study, a child who left out word-initial consonants learned 16 of them after treatment on three maximally contrasting sets (Gierut, 1989). Later work described a hierarchy of minimal pair efficacy based on the number of new sounds and the number and type of feature differences being taught (Barlow & Gierut, 2002).
How to choose pairs
- Sample broadly, then probe deeply. Start with a standardized test or a conversational sample to find error patterns, then give a deep probe of each error sound across many words and positions to see whether it’s consistent and exactly what replaces it (Storkel, 2022). See assessments at a glance.
- Match the approach to the profile. One pattern points to conventional minimal pairs, errors across many sound classes to maximal oppositions, and one sound replacing many to multiple oppositions (Storkel, 2022).
- Final consonant deletion. Storkel describes it as a global collapse, with many final consonants collapsed into nothing, so contrasting an open syllable with several closed ones (for example “bee” with “beak”, “beef” and “beach”) suits a multiple oppositions approach (Storkel, 2022). See final consonant deletion therapy.
- Simple or complex targets? This isn’t settled. A review of complexity research concluded that treating more complex targets leads to greater phonological gains (Gierut, 2001). But in a study of 48 children, those treated on early-developing sounds they partly knew made more progress on the treated sounds than those treated on late-developing ones, with no difference in generalization, and parents were more satisfied (Rvachew & Nowak, 2001).
- Real words or nonwords. Much maximal opposition research used nonwords given a meaning, and studies suggest nonwords work at least as well as real words. Real-word sets can include words a child doesn’t know, so explain to families that words are chosen to highlight a contrast, not to build vocabulary (Storkel, 2022).
- Check the child’s dialect. A pattern that’s part of the child’s dialect isn’t an error (ASHA).
A session structure
Each research team runs these approaches a little differently (Storkel, 2022). A typical plan, drawn from Storkel’s tutorial:
- Choose the starting point. If the child can’t produce the target yet, establish it in isolation or syllables first; stimulability is often defined as at least 3 correct of 10 imitated attempts. If they can produce it with cues, start with imitation of the paired words; if readily stimulable, start with spontaneous production (Storkel, 2022).
- Contrast activities. In conventional minimal pairs, the child asks the SLP for pictures; if they say the substitute, the SLP picks up the other picture and then explains which sound to use (Storkel, 2022, describing Weiner’s original procedure). In maximal oppositions, the child says the two words of each pair in turn, with matching and, later, sorting by sound. Multiple oppositions moves from drill-play towards communicative games (Storkel, 2022).
- Criteria. Imitation to about 70–75% accuracy, then spontaneous production to about 70–90%, across several sets or sessions (Storkel, 2022).
- Probe untreated words. Williams’ decision rules, as summarized by Storkel: below 90% on untreated words, keep working at word level with new words; at 90% or more, take a conversation sample; below 50% in conversation, train beyond word level; at 50% or more, reassess (Storkel, 2022). See progress monitoring.
- Dose. Conventional minimal pair studies used 1 to 4 sessions a week of 30 to 90 minutes with 25 to 150 trials; multiple oppositions studies used 65 to 100 trials, and Williams recommends at least 50 trials a session, or 70 for more severe disorders (Storkel, 2022). In an RCT of 54 preschoolers, multiple oppositions three times a week beat once a week for the same number of sessions (Allen, 2013). See treatment intensity and dosage.
What the evidence says
- A large but mostly small-scale literature. Conventional minimal pairs have been used in over 40 to 50 studies (Storkel, 2022). A review of 134 speech sound intervention studies from 1979 to 2009 found most were quasi-experimental (41.5%) or case studies (32.6%), and called for rigorous comparisons of approaches (Baker & McLeod, 2011).
- Multiple oppositions. Williams found conventional minimal pairs worked poorly for a global collapse, while contrasting several targets with the substitute at once produced stronger learning (Williams, 2000, as described by Storkel, 2022). Allen’s RCT showed it works best at higher frequency (Allen, 2013), and in a five-child study where parents delivered it at home alongside weekly SLP sessions, parents did so with good fidelity and all children improved on treated words (Sugden et al., 2020).
- Comparisons. In a clinic-based trial, 19 children were randomized to conventional minimal pairs or contrasts differing across a range of features, 6 hours over a school term. Both groups progressed, with no difference between them, and gains were maintained at follow-up (Dodd et al., 2008). In an early study of four children, minimal pairs and a modified cycles approach were both effective (Tyler, Edwards & Saxman, 1987). Storkel notes that Brazilian Portuguese studies found no difference between maximal oppositions and approaches such as cycles (Storkel, 2022).
- Later reviews. A systematic review of preschool speech sound interventions found most evidence was lower-graded (Wren et al., 2018). A 2024 meta-analysis of 23 group-design studies found medium-high average effects on speech accuracy across a range of interventions, from only seven controlled studies and with heterogeneity, and called for better reporting (Kunnari et al., 2024).
We didn’t find a systematic review that pools minimal pair studies on their own or compares the contrast approaches head to head in large trials. Choosing between them still rests mainly on the child’s error profile.
Questions SLPs ask
What are minimal pairs in speech therapy?
Pairs of words that differ by one sound that changes the meaning, such as “key” and “tea”. Therapy uses them so a child learns that the sound they’re substituting or leaving out changes the word.
Can you use minimal pairs for final consonant deletion?
Yes: pair a word without the final sound with one that has it, such as “bow” and “boat”. Because many final consonants are deleted at once, a 2022 tutorial suggests multiple oppositions may suit it well.
What are minimal pairs for k and g?
For velar fronting, pair the target with the child’s substitute, such as “key” and “tea” or “go” and “dough”. The free minimal pairs word lists include full sets for k/t and g/d.
What is the difference between minimal pairs and multiple oppositions?
Minimal pairs contrast one target with the child’s substitute. Multiple oppositions contrast several targets with the same substitute at once, for children who use one sound in place of many.
Do minimal pairs work for lisps or distorted R?
Contrast approaches target substitutions and deletions rather than distortions. Distortions usually call for motor-based work: see traditional articulation therapy.
Sources
- ASHA Practice Portal: Speech sound disorders (articulation and phonology).
- Storkel, H. L. (2022). Minimal, maximal, or multiple: Which contrastive intervention approach to use with children with speech sound disorders? Language, Speech, and Hearing Services in Schools, 53(3), 632–645.
- Gierut, J. A. (2007). Phonological complexity and language learnability. American Journal of Speech-Language Pathology, 16(1), 6–17.
- Storkel, H. L. (2018). The complexity approach to phonological treatment: How to select treatment targets. Language, Speech, and Hearing Services in Schools, 49(3), 463–481.
- Gierut, J. A. (1989). Maximal opposition approach to phonological treatment. Journal of Speech and Hearing Disorders, 54(1), 9–19.
- Barlow, J. A., & Gierut, J. A. (2002). Minimal pair approaches to phonological remediation. Seminars in Speech and Language, 23(1), 57–68.
- Gierut, J. A. (2001). Complexity in phonological treatment: Clinical factors. Language, Speech, and Hearing Services in Schools, 32(4), 229–241.
- Rvachew, S., & Nowak, M. (2001). The effect of target-selection strategy on phonological learning. Journal of Speech, Language, and Hearing Research, 44(3), 610–623.
- Allen, M. M. (2013). Intervention efficacy and intensity for children with speech sound disorder. Journal of Speech, Language, and Hearing Research, 56(3), 865–877.
- Baker, E., & McLeod, S. (2011). Evidence-based practice for children with speech sound disorders: Part 1 narrative review. Language, Speech, and Hearing Services in Schools, 42(2), 102–139.
- Williams, A. L. (2000). Multiple oppositions: Case studies of variables in phonological intervention. American Journal of Speech-Language Pathology, 9(4), 289–299.
- Sugden, E., Baker, E., Williams, A. L., Munro, N., & Trivette, C. M. (2020). Evaluation of parent- and speech-language pathologist-delivered multiple oppositions intervention for children with phonological impairment: A multiple-baseline design study. American Journal of Speech-Language Pathology, 29(1), 111–126.
- Dodd, B., Crosbie, S., McIntosh, B., Holm, A., Harvey, C., Liddy, M., Fontyne, K., Pinchin, B., & Rigby, H. (2008). The impact of selecting different contrasts in phonological therapy. International Journal of Speech-Language Pathology, 10(5), 334–345.
- Tyler, A. A., Edwards, M. L., & Saxman, J. H. (1987). Clinical application of two phonologically based treatment procedures. Journal of Speech and Hearing Disorders, 52(4), 393–409.
- Wren, Y., Harding, S., Goldbart, J., & Roulstone, S. (2018). A systematic review and classification of interventions for speech-sound disorder in preschool children. International Journal of Language & Communication Disorders, 53(3), 446–467.
- Kunnari, S., Sanduvete-Chaves, S., Chacón-Moscoso, S., Alves, D. C., Ozbič, M., Petinou, K., Tolonen, A. K., Zajdó, K., Frizelle, P., Murphy, C. A., Saldaña, D., & Laasonen, M. (2024). Intervention studies with group design targeting expressive phonology for children with developmental speech and language disorder: A systematic review and meta-analysis. International Journal of Language & Communication Disorders, 59(6), 2686–2705.