Examining practice effects within the GENFI pre‐symptomatic FTD cohort
Kerala L Adams‐Carr, Rhian S Convery, Arabella Bouzigues, Sophie E Goldsmith, Jonathan D. Rohrer, Lucy L. Russell- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Practice effects can be observed on repeated cognitive testing and are often viewed as potential sources of bias. However, there is mounting evidence that their absence or reduction can indicate early changes in cognition, with decreased ‘learning over repeated exposure’ seen in those with mild cognitive impairment. A recent GENFI study demonstrated practice effects are diminished in C9orf72 and GRN mutation carriers proximal to disease onset compared to controls at testing intervals of one year. The aim of this study is to explore how practice effects differ in the GENFI cohort at shorter time intervals using Ignite, an iPad‐based set of cognitive tests, with the aim of exploiting this understanding to detect subtle cognitive impairment within clinical trials.
Method
36 individuals from the UCL GENFI cohort took part, 26 of whom were presymptomatic mutation carriers and 10 were non‐carrier controls. They completed Ignite at baseline and at 3 months. Change scores were calculated for each task, and linear regressions were performed comparing mutation carriers with non‐carriers, controlling for significant co‐variates of age, education, and sex.
Result
There was a trend towards improvement on repeat testing in both the carrier and non‐carrier groups, but a difference emerged on the Think Back Level 1 task, a test of working memory analogous to the ‘N‐back’ task, where the non‐carriers improved on a speed‐accuracy trade‐off score to a greater degree than mutation carriers (mean improvement 16.4 vs 5.2 respectively, p = 0.026). There was also a trend towards a difference on the Path Finder Level 1 and 2 tasks, adaptations of the Trail Making Tests Parts A and B, where the non‐carriers improved more than the mutation carriers.
Conclusion
Practice effects were observed across the majority of tasks in both the carrier and non‐carrier groups, except for a working memory task, where a difference was observed between the groups. This was a small pilot study, and a larger scale study of ‘burst testing’ within the GENFI cohort is underway to examine practice effects in greater detail, at shorter time intervals. Subanalyses to stratify by genetic subtype and proximity to disease onset are required.