Intraverbal Assessment for Persons with Aphasia or Other Acquired Brain Injury

Abstract

An intraverbal assessment was administered to older adults with aphasia, using a hierarchy of questions that required increasingly complex verbal discriminative stimulus control. Five categories of errors were defined and analyzed for putative stimulus control, with the aim to identify requisite assessment components leading to more efficient and effective treatments. Evocative control over intraverbal error responses was evident throughout the database, as shown by commonalities within four distinct categories of errors; a fifth category, representing a narrow majority of errors, was less clear in terms of functional control over responses. Generally, questions requiring increasingly complex intraverbal stimulus control resulted in weaker verbal performance for those with aphasia. A new 9-point intraverbal assessment model is proposed, based on Skinner's functional analysis of verbal behavior. The study underscores that loss or disruption of a formerly sophisticated language repertoire presents differently than the fledgling language skills and errors of new learners, such as typically developing children and those with autism or developmental disabilities. Thus, we would do well to consider that rehabilitation may require a different approach to intervention than habilitation. We offer several thematic topics for future research in this area.

Keywords: Aphasia; Brain injury; Language skills; Speech; Stimulus control.

Fig. 1

Total score for each participant. Note. Participants are ordered from lowest to highest score and not in order of test completion. The numbers above each bar represent the total score

Fig. 2

Response error distribution. Note. This figure shows distribution of error responses in each test set, according to error category. Error categories refer to possible stimulus control for responses. Aggregate errors per test set appear in bold above each test set

Fig. 3

Total errors by classification. Note. This figure shows error frequency across types of error responses (n = 497). Bars indicate total for all participants, according to defined error categories