How cognition has been assessed in research with people with aphasia: a systematic scoping review

ABSTRACT

Cognitive deficits beyond language may be present in aphasia, and accurately assessing them is fundamental for treatment planning. However, it is unclear which tools are currently used to assess cognition in aphasia, how much language they require, and which cognitive domains should be the focus of assessment. This study aimed to determine which cognitive domains have been the focus of assessment, and systematically appraise the linguistic requirements of these tests. We followed the PRISMA-ScR guidelines and systematically searched the literature from 2010–2020 for the tests used to assess cognition in research with people with aphasia. 480 studies were included and 389 tests were appraised for their linguistic load. Results showed that cognitive testing focused mostly on executive functions and memory. Over half the tests used to assess cognition in aphasia had some degree of language load (i.e., verbal stimuli, language production or required mapping stimuli to its meaning). This review highlights the difficulty in accurately diagnosing concomitant cognitive deficits in aphasia due to the language requirements in commonly used cognitive assessments. This review underlines the need to be mindful of the linguistic requirements in cognitive assessments. Finally, we offer some recommendations of which tests may be the most appropriate to assess cognition in aphasia.

KEYWORDS:

• Aphasia
• cognitive assessment
• language load

Disclosure statement

No potential conflict of interest was reported by the author(s ).

Notes

1 Measures may also be referred as tests or tasks. In this paper we refer to tests to indicate specific tests such as the tower tests, and tasks to refer to the actions required by a test (such as recall digits in the Digit Span test).

2 We did not evaluate the language load of the instructions as: (a) There are many ways one could evaluate the language load, for example, linguistic complexity (how syntactically and semantically complex are they?), length of instructions, or the amount of inference and pragmatic knowledge required. (b) Many of the tasks did not include standard instructions (mainly the experimental tasks). (c) The complexity of instructions often correlate with the complexity of the task, for example the Wisconsin Card Sort Test has been argued to have complex instructions (Siegal & Varley, 2006), and it is also a complex task; (d) Tessaro et al. (in preparation) found that 88% of respondents to a survey reported changing their approach when assessing cognition in people with aphasia, and respondents indicated that they make modifications to the task instructions (e.g., reduce the linguistic complexity, invalidating any attempt to measure instruction complexity.

3 There is a degree of variability in the administration of the clock drawing task which affects the memory load. Those tasks where the instructions are only provided in spoken form will inevitably have a higher load, than those where the instructions are written and remain in view. For the sake of this analysis we have not discriminated between these versions.

Additional information

Funding

This work was supported by Australian Research Council [DP190101490]; Macquarie University [20191485]; Newcastle University [Postgraduate Scholarship].