Up to 70% of stroke survivors experience cognitive impairments. Assessing cognitive skills after stroke is critical to diagnose, to educate, and to tailor rehabilitation to maximise functional outcomes. The neuropsychological tests used to assess cognition are typically pen-and-paper based, which are often dependent on language skills for completion.Language impairments observed in stroke survivors with aphasia may confound non-linguistic cognitive performance in these pen-and-paper tests. The primary objective of this research was to develop a clinimetrically sound cognitive assessment for stroke survivors with or without aphasia.A systematic review was conducted to determine whether studies that evaluated cognitive assessments in stroke included participants that represented the stroke population.Approximately two-thirds of the studies excluded individuals with communication problems, with a similar percentage of studies excluding individuals with cognitive impairments. This review highlighted the need for new and more feasible methods to assess cognition in the wider stroke population; particularly those with aphasia or cognitive impairment.We developed a non-immersive virtual reality cognitive assessment for individuals with stroke (with or without aphasia) -the Cognitive Assessment for Aphasia App (C3A). This "aphasia-friendly" assessment incorporated computerised audio and visual feedback, practice opportunities, and intuitive functional tasks to minimise dependency on language skills for completion. Developing the C3A required the input from a multidisciplinary clinical and research team and stroke survivors. We targeted cognitive domains that are commonly affected in stroke and that influence functional outcomes. The C3A included a simple reaction time task to assess psychomotor skills and attention, a visual search task to assess visuospatial skills, a sequence copy task to assess visual memory, and a kitchen task to assess executive functioning.A total of 97 participants were recruited (36 with aphasia, 29 stroke non-aphasia, 32 controls) from acute and inpatient rehabilitation settings and the community. All participants were assessed on the C3A and a battery of standard pen-and-paper cognitive tests typically used in stroke. Participants with aphasia and controls undertook the auditory comprehension sub-tests of the Comprehensive Aphasia Test. C3A performance iii measures included latency and errors. Following successful trials of the C3A we sought to determine its feasibility and validity in stroke and controls.Before analysing the C3A results, the association between language performance and the pen-and-paper cognitive tests in participants with aphasia and controls was explored. The association between language and a real-life measure of cognition (the Kettle Test) was also tested. All cognitive tests were significantly associated with auditory comprehension and naming performance (with up to 78% of variance explained), except for Star Cancellation and the Kettle Test. To measure non-lingu...