Brixton test data in bvFTD and SD

Published: 16 September 2020| Version 1 | DOI: 10.17632/pjpxdmw84h.1


The Brixton Spatial Anticipation Test is a widely used test of executive function. There has hitherto been surprisingly little systematic examination of the usefulness of the test in behavioural variant frontotemporal dementia (bvFTD), the prototypical degenerative disorder of the frontal lobes. The dataset relates to a study that explored the sensitivity and specificity of the Brixton test through a direct comparison of 76 patients with bvFTD and 34 with semantic dementia (SD), characterised by degeneration of the temporal lobes. All patients had been referred to a specialist cognitive clinic within a neuroscience centre and the Brixton test had been administered as part of their clinical diagnostic work-up. The study of patients' Brixton performance investigated both quantitative (error scores, sets achieved) and qualitative (error types) aspects of performance. bvFTD patients, as predicted, performed significantly more poorly than SD patients and achieved fewer sets (i.e. identified fewer rule shifts). bvFTD patients were distinguished too by more frequent perseverative and random errors. Perseverations could be subdivided into those representing repetition of an immediately preceding response, an immediately preceding rule, or an earlier rule that is not the immediately preceding one. A distinction could also be drawn between perseverations of the immediately preceding rule that take note of the position of the blue circle (stimulus-related) and those that do not (stimulus unrelated). bvFTD patients could be distinguished from SD by the presence of the latter. The data confirm the value of the Brixton test in the clinical assessment of bvFTD and point to the clinical and theoretical value of analysis of errors.


Steps to reproduce

We evaluated test score sheets blind to diagnosis to avoid bias. In a pilot examination of error types we recognised potential ambiguities in classification because responses could fulfil definitions of more than one type of error. To ensure consistency we developed detailed criteria for classification and a flow chart, published in our article on the Brixton test in bvFTD and SD Journal of Neuropsychology, 2020 (first author Rao, SP)


The University of Manchester