Do picture‐naming tests provide a valid assessment of lexical retrieval in conversation in aphasia?

Abstract

Background: Word‐finding problems commonly occur in aphasia and can significantly affect communication. Assessment of this deficit typically involves naming pictures. However, this method has been criticised as lacking ecological validity. Alternative methods include the measurement of lexical retrieval in narration or conversation, although few published studies have quantified word finding in the latter.

Aims: We aimed to identify a reliable and valid assessment of lexical retrieval in conversation, and to elucidate the nature of the relationship between lexical retrieval in picture naming and in conversation.

Methods and Procedures: We developed a quantitative measure of word finding in conversation in aphasia and established the reliability and stability of the method. We compared the scores of a group of people with aphasia on this measure with their scores on a picture‐naming test.

Outcomes and Results: We found significant relationships between picture‐naming scores and a number of key variables analysed in the conversation measure. We propose that scores on picture naming relate to the ability to retrieve nouns in everyday conversation for the people with aphasia who took part in this study.

Conclusions: The use of picture‐naming tasks is justified, providing a valid and rich means of assessment of lexical retrieval. Further research is required to replicate these results with more people with aphasia. We offer the quantitative assessment of conversation developed here for use in research and clinical spheres.

We wish to thank the 10 people with aphasia and their conversation partners who gave their time to the research. We also are grateful to two anonymous reviewers for their comments. The work described here was carried out with the support of a grant from the Tavistock Trust for Aphasia.

Notes

1. On the other hand there are conversational contexts where a single‐word response is acceptable, for example the use of ellipsis is a common conversation device in responding to closed questions. Moreover conversation analysis of normal speakers' everyday talk has dispelled the myth that speech is constructed in grammatically correct sentences (Helasvuo, 2001).

2. Mayer and Murray (2003) found strong correlations between spoken picture naming and percentage word retrieval in conversation and picture description. The authors reject these findings however, as when the group (n  =  14) were split into two on the basis of severity, the correlations were no longer so strong. This is a straightforward and predictable result of range effects

3. However, it is the case that certain aspects of the conversation were lost to the analysis. In particular, gestures and pointing demonstrating the meaning of the word being sought, and other non‐verbal indications crucial to the interaction, were not available to the analysis.

4. Within the turns of the people with aphasia, pauses of more than 1.5 seconds may occur due to word retrieval problems or difficulties with sentence construction. The subsequent speech was not coded as a new turn, if the preceding speech indicated that the pause was due to a search for a word or a grammatical construction, and the other speaker did not start a turn at this point.