Aphasia with Anomia in Mandarin Chinese: An Exploratory Pilot Study Based on Phonological Awareness Therapeutic Strategies

Anomia, a pivotal symptom of aphasia, impedes functional communication. In tonal languages such as Mandarin, phonological awareness plays a critical role in lexical retrieval. This study developed and evaluated a phonological awareness-based intervention for anomia in Mandarin-speaking people with aphasia (PWA). Ten post-stroke Mandarin-speaking PWA with left-hemisphere damage and anomia (5M/5F; 4 non-fluent, 6 fluent aphasia) received a three-stage phonological awareness intervention (phoneme–nonword–real word). Pre- and post-intervention assessments covered naming ability, phonological awareness, and initial consonant articulation. After treatment, phonological awareness improved significantly across all measures (adjusted p < 0.05; partial η2 = 0.733–0.932). Both groups showed significant pre-to-post improvements in rhyme, alliteration, tone, phoneme awareness, syllable decoding, and syllable plausibility judgments (all adjusted p < 0.05); syllable awareness improved significantly only in the FA group (MD = 41.67, 95% CI = 24.85–58.48, adjusted p = 0.002). For naming accuracy, both groups improved significantly on trained words (FA: MD = 25.76, 95% CI = 13.08–38.44, adjusted p = 0.006; NFA: MD = 38.64, 95% CI = 29.30–47.98, adjusted p = 0.002). On the standardised test and untrained generalization words, only the FA group improved significantly (MD = 8.33, 95% CI = 4.05–12.62, adjusted p = 0.008; and MD = 11.37, 95% CI = 6.36–16.37, adjusted p = 0.004, respectively). Initial consonant articulation accuracy improved significantly in both groups (FA: MD = 17.46, 95% CI = 7.14–27.78, adjusted p = 0.014; NFA: MD = 25.00, 95% CI = 8.20–41.80, adjusted p = 0.036). Regarding naming errors, NFA showed significantly higher pre-treatment omission errors (median = 20.00% vs. 6.67%, adjusted p = 0.010) and phonological errors (median = 11.67% vs. 3.33%, adjusted p = 0.020) than FA on the standardized test; on trained words, NFA had higher pre-treatment omission errors (median = 15.91% vs. 6.82%, adjusted p = 0.048), while FA had higher post-treatment word formation errors (median = 9.09% vs. 0.00%, adjusted p = 0.038). No significant treatment-related changes in the occurrence rate of naming errors were detected on untrained generalization words. In conclusion, this exploratory pilot study provides preliminary evidence that a phonological awareness-based intervention may improve phonological awareness and trained-word naming ability in Mandarin-speaking PWA. Improvements on untrained words were limited and observed only in the FA group. Therefore, the intervention’s generalization remains uncertain. These findings cautiously support a language-specific rehabilitation approach attuned to Mandarin phonology.