2.2. Materials

The Eriksen flanker task [27] was used as a measure of cognitive control during the neuroscientific experiment. Contrary to other cognitive control tasks like the Stroop Task it does not use language stimuli but non-linguistic and non-emotional visual conflict stimuli. In Figure 3, a typical example of the task is shown. The arrow version of the flanker task was also administered in this study. The central target arrowhead is flanked by non-target arrowheads, which follow the same direction (congruent flankers) or opposite direction of the target (incongruent flankers). In this study design, flankers were presented in white in the middle of a dark gray screen, serving better contrast and less eye fatigue, respectively. The size of all five arrowheads on the screen was 2 × 18 cm (0.79 × 7.1 inch). The Erikson flanker task lasted about 5 min, including 240 trials (60 congruent left direction, 60 congruent right direction, 60 incongruent left direction and 60 incongruent right direction). Since at least 30 trials per condition are needed to provide reliable measurements, a total number of 240 trials guaranteed conservative, reliable response time, and N200 amplitude measurements [85]. The participants’ response times and accuracy rates in dependence of both congruent and incongruent trials were assessed. In this study, the flanker task design was developed to be suitable for a simultaneous measurement of both EEG and functional near-infrared spectroscopy (fNIRS). The sluggish vascular response of the fNIRS, tending to overlap in time when stimuli are presented shortly one after another, thus had to be taken into consideration. Therefore, a mini-block design was introduced. It comprised 10 mini-blocks, each consisting of 24 trials with an equal distribution of right and left directed target arrowheads. Furthermore, 5 out of the 10 mini-blocks comprised a sequence of 75% congruent trials and 25% incongruent trials (i.e., mainly congruent blocks) and the other 5 mini-blocks 75% incongruent trials and 25% congruent trials (i.e., mainly incongruent blocks). Each trial consisted of the stimulus presentation (i.e., white arrowheads) on a gray screen background (duration of 250 ms), followed by a constant inter-stimulus-interval (ISI) (duration of 750 ms; see Figure 4) in which only a gray blank screen was visible. Over the entire course of the trial (1.000 ms), participants were instructed to press one of the arrow keys on the keyboard to indicate the direction of the central target arrowhead. The trial duration was fixed, hence, participants did not automatically jump to the next trial after giving an answer. In case of multiple answers per trial, only the first one was counted. If no answer was given, a missing value was logged, not an error. After each mini-block, there was a variable longer inter-block-interval (IBI) lasting 10 s on average (range 6–14 s). This extended IBI was introduced to minimize systematic overlap of the sluggish hemodynamic response assessed by the fNIRS [86]. The mini-blocks was pseudorandomized each within itself and among each other (in their order of appearance) resulting in 4 differently pseudorandomized versions for the measurements. Pseudorandomization rules within each block included maximally three consecutive congruent or incongruent trials and maximally three left or right directed target stimuli in succession, whereas pseudorandomization rules across mini-blocks included maximally three consecutive congruent or incongruent mini-blocks. In the beginning of the experiment, there were two additional training blocks of 10 trials each, which were not analyzed.

Flanker task conditions.

Trial sequence of the flanker task.

Before the neuroscientific assessment of the flanker task, all participants underwent the Mini DIPS interview [80] to verify their conformity to the control or patient group. The Mini DIPS is a German short structured clinical interview, which serves to diagnose the most common mental disorders requiring therapeutic intervention. As a short version of the DIPS interview it can be conducted in 30 min and is a widely used diagnostic tool in clinical practice, as well as scientific studies [87,88,89]. In the present project it was conducted by specially trained collaborators [90].

Additional psychological tests and questionnaires were used as control variables for neuroscientific measurements: Beck Depression Inventory-II (BDI-II; [81,82]) served as an assessment of participant’s depression levels and potential comorbidities. The Brief Inventory of Thriving (BIT; [83]) was used as a more general comparison of psychological well-being between both groups. The BIT is a short form of the CIT (Comprehensive Inventory of Thriving; [83]) comprising 10 of the CIT items which makes it a short screening tool for mental health. To ensure the same state of alertness in both groups before the cognitive control task, all participants completed the Karolinska Sleepiness Scale (KSS; [84]), ranging from 1 to 10 with 1 = “extremely alert” to 10 = “extremely sleepy, can’t keep awake”). Two self-developed additional items were implemented to assess the participants’ caffeine intake and possible tetrahydrocannabinol (THC) consumption less than 24 h before the experiment. They were supposed to serve as additional control variables as general drug consumption was already previously assessed in the Mini DIPS interview. The two items were designed as follows: “Did you have any coffee, energy drinks or other drinks containing caffeine before this experiment?” The possible answers only broadly categorized the subjects into three options: no consumption, less than 4 h before, less than 8 h before. The item regarding THC (“Did you recently consume tetrahydrocannabinol (THC)?”) included the options: no consumption, less than 12 h before, less than 24 h before. The given time windows were based on common assumptions about the acute psychoactive effects of these substances. While caffeine consumption was equally distributed among groups it did not serve as an exclusion criterion, contrary to THC consumption, where subjects with any consumption within the last 24 h were excluded (n = 1). Participants also reported their current subjective anxiety levels directly before the task on a simple numeric analogue scale (NAS) ranging from 0 to 10 (0 = “no anxiety at all”, 10 = “highest anxiety level imaginable”) to assess possible baseline differences between patients and controls affecting task performance. A further questionnaire included demographic data and the patients’ clinical history.

During the flanker task, response times and accuracy rates were assessed in addition to the neuroscientific data.