Background Visual disturbances are increasingly recognized as common non‐motor symptoms in Parkinson's disease (PD). In PD patients, intermittent diplopia has been found to be associated with the presence of visual hallucinations and the Parkinson's psychosis spectrum. Here, we investigated whether diplopia in PD is associated with other non‐motor traits and cognitive impairment. Methods We investigated 50 non‐demented PD patients with and without intermittent diplopia and 24 healthy controls for visual disturbances, as well as motor and non‐motor symptoms. All participants underwent a neuropsychological test battery; visuospatial abilities were further evaluated with subtests of the Visual Object and Space Perception Battery (VOSP). The two PD patient groups did not differ significantly in age, symptom duration, motor symptom severity, frequency of visual hallucinations, or visual sensory efficiency. Results PD patients with diplopia reported more frequent non‐motor symptoms including more subjective cognitive problems and apathy without changes in global cognition measures compared to those without diplopia. PD patients with diplopia had greater impairment in several tests of visuospatial function (pentagon copying p = .002; number location p = .001; cube analysis p < .02) and object perception (p < .001) compared to PD patients without diplopia and healthy controls. By contrast, no consistent group differences were observed in executive function, memory, or language. Conclusions PD patients with diplopia have a greater non‐motor symptom burden and deficits in visuospatial function compared to PD patients without diplopia. PD patients with diplopia might be prone to a cortical phenotype with cognitive decline and apathy associated with worse prognosis.

Parkinson's disease (PD) is characterised by motor deficits as well as cognitive alterations, particularly concerning frontal lobe control. Here, we were interested in whether executive function is abnormal already early in PD, as well as whether this dysfunction worsens as a part of the dementia in PD. The following groups engaged in tasks addressing action control: PD patients with mild and advanced motor symptoms (aPD) without dementia, PD patients with dementia (PDD), patients with Alzheimer's disease (AD) and healthy subjects (CON). Subjects either had to perform or inhibit button presses upon go and no-go cues, respectively. These cues were preceded by pre-cues, either randomly instructive of right or left hand preparation (switch condition), or repetitively instructive for one side only (non-switch condition). PDD and aPD omitted more go responses than CON. Furthermore, PDD disproportionally committed failures upon no-go cues compared to CON. In the non-switch condition, PDD performed worse than AD, whose deficits increased to the level of PDD in the switch condition. Over all PD patients, task performance correlated with disease severity. Under the switch condition, task performance was low in both PDD and AD. In the non-switch condition, this also held true for advanced PD patients (with and without dementia), but not for AD. Thus, the deficits evident in PDD appear to develop from imbalanced inhibitory-to-excitatory action control generally inherent to PD. These results specify the concept of dysexecution in PD and differentiate the cognitive profile of PDD from that of AD patients.