Word discrimination therapy: A new technique for the treatment of a phonologically based word‐finding impairment

Abstract

Background: The majority of cognitive neuropsychological treatments for impairments in phonological encoding utilise some form of phonological cueing or support. Most commonly, the participant is provided with supporting information about the phonological form of a particular word (either partial or complete, auditory or written), and must then generate the word orally. The provision of target phonology facilitates initial production of the word, which can then be repeatedly trained in order to induce relearning of the appropriate lexical–phonological associations.

Aims: The aim of the current case study was to examine the effectiveness of an indirect cueing technique for the treatment of mild impairments in phonological encoding. In this treatment the participant must name pictures without assistance, but the picture names themselves share a phonological relationship (e.g., magnet, mattress, macaroni). These phonologically similar triplets incorporate a form of phonological cueing; however, the cues do not involve the direct provision of the word form.

Methods & Procedures: The case reported here, patient TV, suffers from a mild phonological encoding impairment: he produces phonemic paraphasias and other form‐related errors on longer and/or less frequent words in spontaneous speech and other spoken word production tasks. In the treatment, TV was asked to name triplets of pictures whose names were either phonologically related (sharing either initial or final phonemes) or unrelated. This procedure was repeated across 12 therapy sessions, all of which trained the same set of words. TV's pre‐treatment naming accuracy on the training words was assessed using a repeated baseline design, and was continuously reassessed throughout treatment, and also at 3 months post‐therapy. The key comparison was between words trained within phonologically related triplets and control words trained in unrelated triplets. The design also included multiple assessments on an untrained set of words to assess generalisation.

Outcomes & Results: For this mild case, naming accuracy for all words trained improved significantly. However, gains were significantly faster for the words that were trained within phonologically related triplets than those trained in unrelated triplets. Significant generalisation to untreated items was also observed, and gains on both treatment items and generalisation effects were maintained at 3 months post therapy.

Conclusion: Results from this case study suggest that this type of indirect cueing technique may be a particularly efficient method for treating mild impairments in phonological coding.

Notes

We are very grateful for the time and effort provided by TV in his participation in this project. Thank you also to Carolina Lukkien for her assistance. This research was submitted as part of the first author’s doctoral thesis and was funded by a Monash Research Graduate Scholarship.

1. The response was considered phonologically related if it shared the target word’s first or last phoneme, or at least 30% of its phonemes in any position.