Page 56 - The prevalence of the Val66Met polymorphism in musicians: Possible evidence for compensatory neuroplasticity from a pilot study

S. Si et al. NeuroImage 213 (2020) 116681
male/female ¼ 101/326) took part in the present study. All participants perceived parenting styles of our sample, an adapted version of PAQ was
were of Han Chinese origin. This study was approved by the institutional employed. It consists of 21 items and yields permissive, authoritarian and
review board of Shandong Normal University; and a written consent form authoritative scores for both the fathers and the mothers. Each item was
was obtained from each participant after a full explanation of the study scored on a Likert-type scale from 1 (strongly disagree) to 5 (strongly
procedure. Participants first completed psychometric tests (creativity agree). The construct validity of the adapted version of PAQ was inves-
test, parental authority questionnaire and general intelligence test) tigated by examining the confirmation of the dimensionality of the PAQ
which were administered in a randomized order to control for potential using a confirmatory factor analysis (CFA) with maximum likelihood
test order effects, then peripheral venous blood samples were collected estimation in LISREL version 8.70, for the fathers and mothers. According
for genotyping. to the fit indices, the 3-factor structure model of the adapted PAQ did
display acceptable fit to the sample of the present study (father: RMSEA
2.2. Measures ¼ 0.16, NNFI ¼ 0.76, CFI ¼ 0.78, GFI ¼ 0.58; mothers: RMSEA¼.067,
NNFI¼.94, CFI¼.95, GFI¼.89). All subscales of the adapted PAQ
2.2.1. Creativity demonstrate good reliability: mother authoritative parenting α ¼ 0.81,
Creativity levels were measured with three figural divergent thinking father authoritative parenting α ¼ 0.86, mother authoritarian parenting
(DT) tasks which were selected from the Runco Creativity Assessment α ¼ 0.83, father authoritarian parenting α ¼ 0.86, mother permissive
Battery (rCAB; Creativity Testing Service, creativitytestingservices.com). parenting α ¼ 0.70, father permissive parenting α ¼ 0.72.
They were widely used in previous creativity research (Murphy et al.,
2013; Zhang et al., 2014a) and were commonly identified as culture-free 2.2.3. Intelligence
tasks. DT tasks provide a measure of (mainly) capacity for idea genera- General fluid intelligence was measured by Raven’s Advanced Pro-
tion. Since idea generation is critical to creativity across domains, DT task gressive Matrices (APM; Raven et al., 1998). This test is a
scores can be understood as indicators of creative potential, and they well-established culture-free measure of general intelligence. In this test,
have been found to be predictive of creative achievement (Guilford, participants were asked to choose the correct entry to complete the 3 3
1966; Kim, 2008; Plucker, 1999; Runco et al., 2010). Moreover, as DT matrix for each item. Participants were given 20 min to complete the 36
tasks suffer from a poor alternate-form reliability (Barbot et al., 2016; items, and the score was calculated as the total number of items correctly
Barbot, 2019), the importance of relying on a range of DT tasks (rather answered. It has been demonstrated that the 20 min-timed score was a
than a single one) has been highlighted. Accordingly, we used three tasks reasonable and adequate predictor of the untimed APM score (Hamel and
simultaneously. Schmittmann, 2006). The IQ scores range 9–31 (M ¼ 20.56,SD ¼ 3.968)
Participants were required to complete the tasks according to the and no participant’ IQ was 3 SD below or above the mean. So no
following constructions: This is a visual game. Below, you will find a line participant was excluded.
drawing. We would like you to make a list of all of the things that each
drawing could be. The more things you write down, the better! There are 2.2.4. SNP selection
no incorrect answers, no grades, no points, and spelling does not matter. The selections of DRD2 SNPs and COMT SNPs have been described in
This is a game, NOT a test. You have 12 min to complete these questions detail previously (Zhang et al., 2014a, 2014b). They were selected based
(Four min. for each task). Now look at the first figure below and write on both allele frequency and function.
down as many things as you can for what that figure might be. What does To capture most of the common polymorphisms in DRD2 and COMT,
it look like? What could it be? seven and nine tag SNPs were selected respectively from HapMap
When the tasks were finished, experienced raters who blinded to (http://hapmap.ncbi.nlm.nih.gov) genotype data for the CHB (Han
other information about the participants, received training prior to the Chinese in Beijing, China) population (Data Rel 27 Phase II þ III, Feb09,
final ratings, as well as familiarized with the creativity literature were to on NCBI B36 assembly, dbSNP b126) by applying the Tagger program as
evaluate all responses. Creativity was assessed by three dimensions of the implemented in Haploview (Version 4.2) software (Barrett et al., 2005)
2
task—fluency, flexibility, originality, according to the guideline of with the following criteria: pairwise tagging only, r > 0.80 and minor
Creativity Testing Service. Briefly, fluency scores were calculated by the allele frequency (MAF) > 5%. Specifically, for the genomic region of
number of ideas each participant gave. Irrelevant ideas were ignored and DRD2 (chr11: 112785528.112851091, based on NCBI Genome Build
excluded from scoring. Flexibility referred to the divided categories in 36.3), we selected seven tag SNPs (rs4938019, rs4245148, rs4648319,
one participant’s response. Scoring of flexibility is based on a classifi- rs4436578, rs7122246, rs1076560, rs6279), which captured 59 out of 66
2
cation standard formulated by raters for each task. Originality was the (89%) common alleles (MAF > 5%) with a mean maximal r ¼ .95. With
number of unique ideas given by each participant. Scoring of originality regard to COMT, nine tag SNPs (rs737865, rs174675, rs5993882,
is usually based on the statistical frequencies of each response in the rs5993883, rs4646312, rs6267, rs4680, rs769224, and rs174697) were
study sample (Forthmann et al., 2017; Mouchiroud and Lubart, 2001). In selected which captured 80% of common alleles (MAF > 5%) within the
the present study, unique ideas were those ideas given by less than 5% of genomic region of COMT (chr22:18309309.18336528, based on NCBI
2
the sample. All three indexes demonstrated reliability with internal Genome Build 36.3) with a mean maximal r of 0.966.
consistency αs of 0.88, 0.83 and 0.69, respectively. For each task, two Besides, for DRD2, seven putative functional SNPs (rs1799978,
trained raters (both were psychology graduate students from Shandong rs1799732, rs4648317, rs2283265, rs6277, rs6276, and rs6278) as well
Normal University) were engaged to score all the responses. The as rs1800497 were also genotyped. As for COMT, three putative func-
inter-rater reliabilities for all the three DT indexes were higher than 0.95. tional SNPs (rs6269, rs4633, and rs4818) were selected as well.
2.2.2. Perceived parenting styles 2.2.5. Genotyping
Parental Authority Questionnaire (PAQ) was developed for the pur- Methods for DNA extraction and genotyping have been described
pose of measuring Baumrind (1971) permissive, authoritarian and previously (Zhang et al., 2014a). Briefly, DNA samples were prepared
authoritative parental authority prototypes. PAQ is appropriate for both from peripheral venous blood cells by using the QIAamp DNA Mini Kit
women and men who are older adolescents or young adults (Bud, 1991) (Qiagen, Valencia, CA, USA). Genotyping were performed at the Beijing
and it is useful to measure their current perception of their parents’ Genomics Institute-Shenzhen (BGI-Shenzhen, Shenzhen, China) by using
parenting styles. The PAQ has shown good validity in Asian societies the Sequenom® MassARRAY® iPLEX system (Sequenom, San Diego, CA,
(Ang and Goh, 2006) and has been a reliable indicator of parenting styles USA) according to the manufacturer’s instructions. For quality control,
in Hong Kong adolescents and college students in previous research 5% random DNA samples were genotyped twice for each gene to calcu-
(Chan and Chan, 2009; Lai and McBride-Chang, 2001). To measure late genotyping error. The genotyping accuracy was 100%.

3