Predicting cognitive decline using a recognition memory task

In 2020, the global prevalence of dementia exceeded 55 million individuals (World Health Organization, 2024). Research has found that having a Mild Cognitive Impairment (MCI) is a precursor of dementia (Janoutová et al., 2015) and that the medial temporal lobe is one of the first to deteriorate with Alzheimer's disease (Pasquini et al., 2019). Dementia impacts wellbeing. Detecting dementia not only enables the patient to live a better life but also lessens the burden on the NHS. Nonetheless, there is an eight-year time frame where the risk of dementia developing from MCI is high (Dang et al., 2019). Therefore, this PhD aims to forecast cognitive decline specifically in individuals with MCI, rather than dementia.

Previous research has shown that a task may be able to predict the likelihood of developing dementia from MCI (Zola et al., 2013). However, this task is ambiguous as there are suggested two memory processes which are self- and retrieval-generated priming (Wagner, 1981) and both could be operating within this task. One study with healthy participants has shown evidence of this theory by manipulating times and finding different effects on relative recency and objects in place (Nitka et al., 2020). This reflects self- and retrieval-generated priming and distinguishes them.

My PhD aims to explore a similar task as there were limitations of Nitka et al's (2020) task for example the images used were all easy to verbally distinguish (e.g. violin and hand), which means it could be testing verbal memory as well as recognition memory. Therefore, one of my studies will explore a similar task to this with healthy individuals however the 6 images within each trial will be more similar (e.g. in one trial all of the 6 images will be of the moon). Hence, this study will assess whether self- and retrieval-generated priming can be more accurately measured through relative recency and object-in-place. This approach offers a purer gauge of recognition memory compared to prior methods, particularly considering the increased difficulty in verbally discriminating the images.

The potential following study could explore whether this computerised task could predict cognitive decline in individuals with MCIs. The use of relative recency and object-in-place may detect a selective deficit. This would mean that this task would presumably be more sensitive to a decline in cognitive performance compared to previous studies because a single affected form of priming could be compensated by the other form. In order to conduct this research, I will administer the ACE-3 with the task and again one year later to measure their cognitive function and decline.

Additionally, this research may explore the eye movement data from the eye trackers to predict cognitive decline in individuals with MCI. As previous research found individuals with Alzheimer's disease have different eye movement data compared to healthy controls (Pereira et al., 2014).

If the results are significant, this task could be utilised within GP settings to help medical staff and the patient know who is at risk of cognitive decline in the next year. Earlier detection would allow timely interventions and enrolment into clinical trials when treatments are most effective at slowing progression.