Parkinson’s disease (PD) is characterised by enhanced synchronisation of beta (13-30Hz) oscillations in the motor cortex and basal ganglian networks. A slowing EEG rhythm characterised by an enhanced alpha (8-12 Hz) peak in PD at 8 Hz has also been observed as a marker of cognitive decline as indicated by ACE scores. Current literature focusses mostly on observations from subcortical activity during hand movements or cortical measurements at rest. Here, we investigate movement and cortical neural activity using real world neuroscience, recording PD patients in a domestic environment. 
We observed participants’ (13 PD, 24 control) gait dynamics during walking and whilst they made a cup of tea. PD patients repeat the tasks a few hours apart to capture the natural fluctuations of their medication schedule. We observe elevated low beta power (13-20 Hz) in PD participants OFF medication compared to ON and age matched controls, but not between ON and controls. Conversely, we observe elevated alpha power both OFF and ON relative to controls, but not between the ON and OFF conditions indicating distinct motor and cognitive functions for alpha and beta power.
Alpha power for ON and OFF were both significantly higher than controls during walking (t = 3.5, p=0.006; t=2.6, p=0.02) and during the tea task (t=4.2, p = 0.001; t=2.7, p=0.02). Low beta power was greater for OFF vs ON and OFF vs controls during walking (t=2.9, p=0.02; t=2.3, p=0.03) and during the tea task (t=3.2, p=0.007; t=2.5, p=0.025). No significant difference was found between ON and controls, suggesting beta levels are restored to a healthy level. ACE scores correlated negatively with low alpha (7-10Hz) power during the 4m walk (r=-0.49, p=0.03), indicating a negative association with cognitive abilities. These results are in line with previous studies suggesting that beta power is modulated by medication and suggest that alpha power is elevated in PD patients relative to controls but not modulated by medication. We also observed a case study patient across 7 monthly visits. Beta power and UPDRS scores were elevated OFF medication for 6 of 7 visits. We also analysed the gait stages during walking calculating the power ratio between the support and swing phases of the gait cycle which showed a greater beta ratio ON medication. This indicates greater desynchronisation of beta during movement, reflective of healthy physiological gait.
Our results suggest that digital neural markers can be reliably measured during free behaviour to capture both motor and cognitive functioning. Furthermore, they indicate the potential of tracking neural markers relative to individual baseline fluctuations. Here we observe a snapshot of the average power, but future research can look at the temporal dynamics of the tea making task to uncover the interactions between alpha and beta power changes across the task’s many sub-elements.