Pragmatics and figurative language in individuals with traumatic brain injury: fine-grained assessment and relevance-theoretic considerations

ABSTRACT

Background

Difficulties in the pragmatic aspects of language are considered a hallmark of Traumatic Brain Injury (TBI). However, this claim is mostly based on assessments of discourse production, neglecting another relevant pragmatic aspect, namely the comprehension of figurative expressions such as idioms, metaphors, and proverbs. Accounts developed in the framework of Relevance Theory might help framing an investigation in this domain, by describing a continuum of loose uses for idioms and metaphors and the echoing of general wisdom in proverbs.

Aims

First, we aimed at sketching a detailed pragmatic profile of individuals with TBI, by using the Assessment of Pragmatic Abilities and Cognitive Substrates (APACS) test, a standardized tool evaluating both production and comprehension. Second, we aimed at doing a fine-grained investigation of figurative language comprehension skills, by taking into account different figurative types and task-related issues.

Methods and Procedures

Thirty-nine individuals with TBI were administered the APACS test and compared to a sample of matched controls. Six subtasks considering three figurative types (idioms, metaphors, and proverbs) in two response-formats (multiple-choice and verbal explanation) were derived from the APACS test and analyzed with a mixed model approach.

Outcomes and Results

Results showed a diffuse pragmatic language disorder both in the expressive and receptive modality, with 56% of individuals with TBI performing below normative cut-off in the APACS Total score. The analysis of the figurative language subtasks confirmed the difficulties of the TBI group with figurative expressions compared to controls. A further model restricted to the TBI group showed that verbal explanation was more difficult than the multiple-choice format. However, we did not observe robust differences between figurative types. The only noticeable effect was related to proverbs in the verbal explanation format, which were more difficult than all other subtasks.

Conclusions

This study shows two main aspects. First, the pragmatic language disorder of people with TBI encompasses not only the well-known difficulties in discourse production but also a disruption of figurative language understanding. Second, such a disruption affects different figurative types, indicating a general problem in inferring the speaker’s meaning from non-literal language. Relevance-theoretic considerations helped us highlighting the commonalities across figurative types, rather than the differences, and the enhanced meta-representational efforts linked to proverbs in open response formats. Intervention could capitalize on these findings and implement training programs that target pragmatic inferences across different types of expressions to promote the most successful return to communication in its variety of situations.

KEYWORDS:

• Pragmatic language disorder
• cognitive-communication disorder
• figurative language
• APACS
• Relevance Theory

Supplementary material

Supplemental data for this article can be accessed here.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. The interview was held in Italian. The examples give the literal English translation, with English equivalent forms of the proverbs in parentheses.

2. In the original version of APACS, Figurative Language 1 task has two possible scores (0 = incorrect response, 1 = correct response), whereas Figurative Language 2 task has three possible scores (0 = incorrect response, 1 = partially correct response, 2 = correct response). To allow for a direct comparison between the subtasks derived from Figurative Language 1 and the subtasks derived from Figurative Language 2, we binarized the response to the subtasks derived from Figurative Language 2, as follows: 0 = incorrect or partially correct response, 1 = correct response. Importantly, in a preliminary step, we compared the results using this method with those obtained using ordinal mixed model (allowing also for maintaining the intermediate score for subtasks derived from Figurative Language 2), but the two strategies yielded virtually the same results.