|
nif:isString
|
-
The study was approved by the local ethics committee of the Faculty of Psychological and Educational Sciences of Ghent University. Participants signed an informed consent. Twenty-five adults with ADHD between 19 and 37 years old (12 males) participated in this study. Data of one male participant were excluded because of the use of physical manipulations during the heartbeat perception task (see heartbeat perception task below). The results of the remaining 24 participants are reported (mean age: M = 23.46 (SD = 4.48), 12 males, one left-handed). The control group consisted of 23 typically developed adults, matched on age and sex (mean age: M = 23.57 (SD = 3.17), 13 males, four left-handed). Groups did not differ for age (F(1, 45) = 0.01, p = .926) or sex ratio (χ2(1) = 0.20, p = .654). The difference in IQ (controls: M = 111 (SD = 12); ADHD: M = 104 (SD = 13) between groups was marginally significant (F(1, 45) = 3.99, p = .052). Individuals with ADHD were recruited through staff members, advertisements, self-support groups for ADHD, and a local database (adults with ADHD who participated in previous research). All adults with ADHD had a formal diagnosis established by a psychiatrist and completed a semi-structured clinical interview (DIVA, Diagnostisch Interview Voor ADHD bij Volwassenen 2.0) [46]. Adults with ADHD using stimulants (methylphenidate: MPH) were asked to stop this medication 48 hours prior to participation in the experiment. Seven participants with ADHD reported using MPH regularly, 12 only occasionally (e.g. for studying exams), 4 used it in the past but not anymore, and 1 person reported to never have used stimulants. Part of the controls were recruited via an online platform, called Experimetrix, at Ghent University, which is an online experiment scheduling system for students. Further for recruitment of controls and adults with ADHD, advertisements were placed on the website of ZitSTil, the Flemish knowledge and expertise centre for ADHD, and on online forums of popular magazines. Exclusion criteria for all participants were an estimated IQ below 80, history of brain-related illness or neurological disorder and a clinical diagnosis of depression or ASD. To asses IQ, both groups completed an abbreviated version [47] of the WAIS-IV (Wechsler Adult Intelligence Scale-IV) [48], except for the individuals with ADHD who were recruited through the local database since they had already completed the same abbreviated version of the WAIS-III in a previous study. Further, control participants had to score below clinical cut-offs in the attentive and hyperactive/impulsive domain, as evaluated by the Zelfrapportage Vragenlijst voor Aandachtsproblemen en Hyperactiviteit (ZVAH) [49], gauging presence of childhood or adulthood ADHD, in order to be included in the study. In the ADHD group, 15 individuals exceeded the cutoff (> 46) of the Wechsler Utah Rating Scale (WURS) [50], a measure of presence of childhood ADHD, while no control participants exceeded the cutoff (controls: M = 16.7 (SD = 7.6); ADHD: M = 50.4 (SD = 13.5). In the ADHD group, according to the ZVAH, presence of childhood ADHD was confirmed for 20 participants (cutoff 6; 8 ADHD predominantly inattention, 12 ADHD combined subtype), while ADHD in adulthood was confirmed in 21 participants (cutoff 4; 7 ADHD predominantly inattention, 14 ADHD combined subtype). The scores on the Adult Self-Report (ASR) [51] revealed, not surprisingly, that the adult ADHD group scored significantly higher on the DSM oriented ADHD scale than the control group (F(1, 45) = 42.19, p < .001) (controls: M = 55.7 (SD = 6.7); ADHD: M = 74.3 (SD = 12.0)). Data on substance abuse, depression and anxiety were also collected via the ASR. No difference in substance abuse, as measured with the DSM oriented scale of the ASR, was observed between groups (F(1, 45) = 0.25, p = .620) (controls: M = 56.0 (SD = 6.9); ADHD: M = 56.9 (SD = 5.7)). The ADHD group scored significantly higher on the DSM oriented depression scale of the ASR compared to the control group (F(1, 45) = 12.55, p = .001) (controls: M = 52.7 (SD = 3.7); ADHD: M = 59.8 (SD = 9.0)), while the group difference on the anxiety scale did not reach significance (F(1, 45) = 2.89, p = 0.10) (controls: M = 53.1 (SD = 5.0); ADHD: M = 55.8 (SD = 6.0)). Several factors that have previously been shown to affect IA and could confound the results were assessed. First, both anxiety and depression have been shown to be (differently) related to IA [52–54]. As mentioned, adults with ADHD scored higher on depression symptoms, but no difference in anxiety symptoms was noticed. Second, alexithymia, which reflects difficulty in identifying and describing feelings and is characterized by externally oriented thinking, has been shown to be negatively correlated with IA [55]. Alexithymia was measured with the Toronto Alexithymia Scale (TAS-20) [56]. The ADHD group scored significantly higher on the TAS-20 in comparison to the control group (F(1, 45) = 7.93, p = .007) (controls: M = 40.1 (SD = 7.0); ADHD: M = 48.6 (SD = 12.7)).
The Mental Tracking Method [41], a well-validated task with good psychometric values [57] and repeatedly used to measure IA [10,16,18, 22], was chosen as an objective measure of IA. The heartbeat perception task was programmed using E-Prime 2.0 software (http://www.pstnet.com/products/e-prime/) and presented on a 19-inch CRT monitor with 640x480 screen resolution and a 60 Hz refresh rate. All sessions took place during daytime for all participants. In accord with earlier studies using the Mental Tracking method [16,22,41], participants were instructed to focus on their own cardiac activity and silently count the number of heartbeats during three separate intervals. There were three different intervals (25s, 35s, 45s) and the order of presentation of these three intervals was random. Participants were explained beforehand that there would be three intervals during which they had to silently count their own heartbeats. The length of the intervals was not communicated to the participants. They were explained that an interval was indicated by a start and stop tone (coming from the PC speaker) corresponding with the beginning and end of an interval, and that they had to verbally report the number of counted heartbeats. Every trial started with a question on the screen, asking whether participants were ready. After this prompt, the start sound was presented together with a blank screen. After the stop signal, participants verbally reported the number of counted heartbeats during a resting period of 30 s, after which the prompt of the following interval was presented anew. The reported number was written down by the experimenter out of view of the participant. Participants did not receive feedback on their performance. Importantly, participants were instructed not to use physical manipulations (e.g. taking their pulse) to ease the counting and the experimenter monitored the participants through a camera. A heartbeat perception score was derived in keeping with previous studies [16,22,41]: per interval a difference score of the number of recorded and counted heartbeats was calculated. These difference scores were then divided by the number of recorded heartbeats, subtracted from 1, summed and averaged by the number of intervals. Due to this formula: 1/3 Σ (1 –(|recorded heartbeats–counted heartbeats|) / recorded heartbeats), the heartbeat perception score could vary between 0 and 1, with higher scores indicating higher IA and thus a small difference between counted and recorded heartbeats. The heartbeat perception task was administered twice, at the beginning and end of the total testing session (see procedure for details). A mean heartbeat perception score was calculated by summing both heartbeat perception scores (acquired at the beginning and end of the testing session), and dividing that sum by two (hence based on in total 6 heartbeat counting intervals). The electrocardiogram was recorded via two external electrodes from the Biosemi ActiveTwo system (Biosemi, Amsterdam, The Netherlands); one electrode was placed on the left lower rib cage and one electrode was placed on the left upper rib cage. R-waves were counted offline by means of a custom-made R-top algorithm in Brain Vision Analyzer 2 software.
Participants completed the Dutch translation of the BPQ [42], a subjective self-report measure of IA. We used an official back-translated version, approved by the authors of the original BPQ (translated version by Godefroid, Dhar & Wiersema, available at stevenporges.com). This questionnaire consists of four different subscales with a total of 96 items. The subscale of interest for our study was the awareness subscale, which consists of 45 items (Cronbach’s α: .97 for both groups), questioning how aware participants are of their autonomic signals (e.g., swallowing frequently, how hard my heart is beating). Research has shown good reliability and validity measures for the awareness scale [45]. Although this subscale was of main interest for our study, the other subscales (stress response; reactivity of the autonomic nervous system; stress style) of the BPQ were included to check for specificity. Items are rated on a five-point Likert scale, ranging from 1 (never) to 5 (always). The mean score of each subscale was obtained by summing all responses and dividing the sum by the number of items in the subscale.
Participants were seated in a sound-attenuated and dimly lit room, sitting approximately 60 cm in front of the computer screen. Each participant signed an informed consent prior to participation in the experiment and received monetary compensation for their participation. Then participants were explained about the tasks that would follow. This study was part of a larger experimental set-up. Two other behavioral tasks with a total duration of 50 min were administered in between both administrations of the heartbeat perception task; the results of these tasks will be reported elsewhere. Verbal as well as written instructions were given prior to the start of each task. This study was approved by the local ethics committee.
ANOVAs with group (ADHD vs. control) as between-subjects factor were performed to compare performance between groups on the heartbeat perception task, and the subjective measure of IA as indexed by the mean score of the awareness subscale of the BPQ. To check for specificity, we also performed ANOVAs on the mean scores of the other subscales of the BPQ. Finally, links between the IA indices and symptoms of anxiety, depression, and alexithymia were explored by additional correlational analyses.
|
![[This page as RDF]](http://data.gesis.org/somesci/static/rdf-icon.gif){kind=icon}