Fifty-eight patients with PD were recruited between October 2011 and April 2013 from the movement disorders clinic of Hospital of the University of Basel or through advertisements. The patients were participants of a Cognitive training-study (Zimmermann et al., 2014) and underwent neuropsychological, psychiatric and neurological assessment. Inclusion criteria for the study were idiopathic PD according to UK Parkinson‘s Disease Brain Bank Criteria (Gibb and Lees, 1988) and written informed consent. Patients were excluded if they had moderate or severe dementia (MMSE ≤ 24; Folstein et al., 1975), insufficient knowledge of the german language, other severe brain disorders, alcohol or drug dependency. For the calculations of this study, the data of n = 40 (14 PD with mild cognitive deficits and 26 PD with normal cognition) has been analyzed. All patients were on dopaminergic medication and were tested while they were in the ON state.

Apathy was measured with the German version of the Apathy Evaluation Scale (AES) (Lueken et al., 2006) filled out by a relative or a person close to the patients. Symptoms of depression were measured with a self-rating scale, the German version of Beck Depression Inventory II (BDI) (Beck et al., 1961). The severity of motor signs was assessed using Unified Parkinson‘s Disease Rating Scale (UPDRS) (Fahn and Elton, 1987) subscale III, applied by trained neurologists.

Patients were examined with a comprehensive battery of neuropsychological tests (Zimmermann et al., 2014). Raw scores of tests were transformed z-values. According to Table 1 a sum score for initiation was calculated.

EEG was recorded with a 256-channel EEG System (Netstation 300, EGI Inc. Eugene, OR 97403, USA; DC-amplifier; sampling rate: 1,000 Hz; high pass filter: 0.01 Hz; vertex-reference, impedance ≤ 40 kΩ). Subjects were instructed to relax, but to stay awake and to minimize eye and body movements. A continuous EEG with closed eyes was recorded for 12 min. During data acquisition, a subset of electrodes was monitored online by a technician to check for vigilance and artifacts.

All EEG data were automatically preprocessed using TAPEEG, as described in a previous publication (Hatz et al., 2015). In brief, PLI measures were calculated using a Butterworth filter to four predefined frequency bands (theta: 4–8 Hz, alpha1: 8–10 Hz, alpha2: 10–13 Hz and beta: 13–30 Hz) and calculated as described by Stam et al. (2007). PLI calculation was combined with microstates segmentation (msPLI) (Hatz et al., 2016). In a previous publication msPLI showed an increased test-retest-reliability of time in healthy subjects compared to classical PLI (Hatz et al., 2016). For calculation of msPLI the raw EEG were segmented into four microstates (Figure 1) and the Hilbert-transformation was applied to the full-length EEG using a sliding window of 4 s with a 50% Hanning-window. For every microstate class four stitched periods of each 4,000 phase differences (= 4 s) were then extracted, using the time frames indicated by the microstate label vector (Supplementary Figure 1). The number of 4 epochs per microstate, subject and frequency band was selected, as this minimal amount of epochs per microstate was available in almost all EEG given the recording time of EEG data. The 16 resulting matrices (four epochs × four microstates) were averaged per subject. For statistical analysis, electrodes were grouped into 22 regions of interest (ROI) (Supplementary Figure 2), 11 regions per hemisphere excluding electrodes in the midline, neck and face. Graph analysis results were calculated according to Table 2.

The apathy total score was not normally distributed (KSLT: Kolmogorov Smirnow Lilliefors Test; p < 0.01) therefore, this variable was square root transformed, leading to a non-significant p-value (p = 0.34) of the KSLT. After z-score transformation, the executive function variables were averaged; variables indicating good performance in smaller values were reversed (Table 1). To check the validity of the subprocesses of initiation, the intercorrelations between all the collected z-transformed cognitive variables and internal consistency (Cronbach‘s alpha) were calculated. As defined by George et al. (2002), acceptable internal consistency was set at Cronbach‘s α of greater or equal to 0.70.

As the results of connectivity analysis were not normally distributed, resulting msPLI values were log transformed.

Spearman-rank was used to estimate the correlation of connectivity values to neuropsychiatric and neuropsychological outcome measures. Age, gender, education, MMS and LED were used as confounding variables. To correct for those variables for every single connectivity value, a linear regression was calculated using the variables to predict the connectivity value. The residuums of the connectivity values were used for the correlation analyses. To correct for multiple comparisons permutation tests (10'000 permutations) were used (Nichols and Holmes, 2002). Results with p-values < 0.05 were considered significant. Analyses were done using TAPEEG (Hatz et al., 2015).

The summary of demographics and neuropsychology is represented in Table 3.

For analysis of signal-space frequency results, a borderline significant correlation of AES and relative delta bandpower in the right frontal region was found, no correlation to the other frequency bands (Table 4). For connectivity analysis, significant results were found for alpha2-band results only. After correction for multiple comparisons, significant correlations with the AES were found for alpha2-band connectivities involving the left fronto-polar region (p-values ≤ 0.05). Lower connectivities were related to higher apathy scores (Figure 2). Graph measures of alpha2-band networks correlated to AES scores (Figure 3). Analysis of initiation showed only a barely significant correlation to alpha2-band connectivity fronto-parietal left (Figure 4). Using a median-split for the AES results (median = 24.5) and the results for alpha2-msPLI-connectivity at FPL-PLR, a classification of low vs. mildly apathetic PD-patients is possible with a sensitivity of 70%, a specificity of 90% and a AUC of 82.5 (Figure 5).

UG: Support of research from Mach-Gaensslen-Foundation, Gossweiler Foundation, Parkinson Schweiz, Synapsis Foundation, Botnar Foundation, PF: Support of research from Swiss National Science Foundation, Mach-Gaensslen-Foundation, Gossweiler Foundation, Parkinson Schweiz, Synapsis Foundation, Botnar Foundation, Freiwillige Akademische Gesellschaft Basel, Novartis Research Foundation, Novartis, Roche, AbbVie, Hedwig-Widmer-Stiftung. The other authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.