Smartphone-Based Assessment of Daily Cognitive Fluctuations in Patients with Parkinson’s Disease

Due to increasing life expectancy and the rising average age of the population, the prevalence of neurodegenerative diseases such as Parkinson’s disease is also increasing. While early Parkinson’s research primarily focused on dopamine deficiency and associated motor symptoms (e.g., bradykinesia, rigidity, tremor, postural instability), scientific interest in recent years has increasingly shifted toward the investigation of non-motor symptoms (i.e. depression, anxiety, sleep disorders, fatigue, constipation, Chaudhuri et al., 2006). A widely reported non-motor symptom is cognitive impairment. Studies have shown that up to 50% of patients with early Parkinson’s disease develop cognitive limitations within the first five years after diagnosis, and up to 80% of patients with Parkinson’s disease experience dementia in later stages of the disease (Pigott et al., 2015). In Parkinson’s disease, particularly the cognitive domains of memory, executive function, attention visuo-spatial abilities 2021) are affected (Aarsland et al., 2021) In the research on Parkinson’s research , the assessment of intraindividual variability (IIV) in cognitive performance is gaining increasing importance. IIV refers to systematic fluctuations of an individual around their own performance level. According to Kuhlmann et al. (2023), IIV can be operationalized through within-person variance across repeated measurements, where individuals may differ both in their average performance and in the extent of their fluctuations. A recent longitudinal study with newly diagnosed patients with Parkinson’s disease (n = 423) demonstrated that higher IIV over a five-year period was associated with a greater risk of mild cognitive impairment or dementia—regardless of traditional cognitive test scores (Jones et al., 2020). Another study (n=1,1228; Davis et al., 2025) further showed that patients with Parkinson’s disease exhibit significantly higher intraindividual variability in cognitive performance compared to healthy controls. This increased IIV was also an independent predictor of functional impairment. These findings suggest that IIV may serve as an early marker for cognitive decline in Parkinson’s patients and could thus be relevant for both diagnostic and prognostic purposes. However, intraindividual variability in cognition has so far primarily been assessed in previous studies using a classical longitudinal design over several years. Znadowski et al. (2022) have shown, however, that cognitive performance in older adults is subject to substantial short-term fluctuations, which are particularly pronounced in patient groups with neurological disorders. Intraindividual variability may therefore represent a crucial early marker for cognitive impairment in Parkinson’s patients and could help identify individuals at particularly high risk for cognitive decline. Yet, single-point assessments are prone to misinterpretation, as they inadequately or entirely fail to capture these short-term fluctuations, potentially leading to erroneous assumptions about the course of cognitive impairment. An intensive longitudinal research design over several days is therefore better suited for capturing short-term cognitive fluctuations. As introduced above, Parkinson’s disease is a multisymptomatic condition with potentially interacting relationships among symptoms. Therefore, the possible associations between cognitive fluctuations and other temporally variable parameters—such as motor symptoms and psychological well-being—are also being investigated. Studies suggest a link between cognitive impairment and specific motor symptoms in Parkinson’s disease: bradykinesia was linked to working memory and mental flexibility, while axial signs were linked to episodic memory and visuospatial functioning (Domellöf et al., 2011). This is thought to stem from the interconnection of cognitive and motor networks within the basal ganglia, which integrate cognitive control and motor execution (Schneider et al., 2015). It has been shown that axial motor symptoms—particularly posture, gait, and balance—are most strongly associated with cognitive impairments, followed by speech, facial expression, bradykinesia, and rigidity. In more advanced stages of the disease, clear associations have also been observed between axial symptoms and visuo-spatial impairments, as well as between bradykinesia and executive functions (Schneider et al., 2015). However, whether there is also a relationship between short-term fluctuations in motor symptoms and cognitive performance in Parkinson’s patients remains unexplored. In addition to cognitive impairments and motor symptoms, changes in psychological well-being and mood are among the most common symptoms in Parkinson’s disease and can significantly affect perceived quality of life (Tan, 2012). A longitudinal study over a four-year period demonstrated associations between a decline in dopaminergic binding in the striatum and worsening in both cognition and mood in Parkinson’s patients. However, associations between fluctuations in psychological well-being and cognitive abilities have not yet been investigated. The exploration of these relationships is possible using the method of ambulatory assessment (AA), which involves collecting data in individuals’ natural living environments under real-life conditions and in real time (Trull & Ebner-Priemer, 2013). Data can include self-reports, observations, and biological, physiological, and behavioral measurements, often collected via digital sensors and smartphones. Therefore, we aim to investigate the extent of IIV in cognitive performance – specifically on the SDMT – and its association with both average performance as well as further symptom interactions (affect, motor symptoms) in patients with Parkinson’s disease using an intensive longitudinal design. By capturing short-term fluctuations in real-world via ambulatory assessment, our study seeks to clarify whether IIV in cognition can be a potential early marker for cognitive decline and help to identify patients especially at risk. IIV in cognition will thereby primarily be assessed via a short, digital version of the Symbol Digit Modalities Test (SDMT: Smith, 1973). The SDMT is a measure of attention and information processing, in which participants need to match as many symbols to digits as possible according to a predefined key in 90 seconds. It has been administered for assessing attention and information processing speed in patients with Parkinson’s disease in previous studies (i.e. (Cerasa et al., 2014) and was chosen, as it is relatively short and easy to implement on a smartphone. Additionally, three other digital tests are administered and treated as secondary outcomes: Word recall, Stroop test (Stroop, 1935), adapted version of the Judgement of Line Orientation (Benton et al., 1978).