60 / 218
58
Chapter 3
between the bromocriptine and placebo session. [Reward x Switching x
DAT1
x Drug: F(1,46)
= 1.5, p > .1], [Reward x Switching x Drug: F(1,46) < 1]. The degree to which reward affected
performance irrespective of task switching [main effect Reward: response time: F(1,46) =
19.4, p < .001; error rate: F(1,46) = 20.6, p < .001] was also not modulated by bromocriptine
(error rate and response time: [Reward x Drug: F(1,46) < 1], [Reward x Drug x
DAT1
: F(1,46)
= 1.6 , p > .2]).
Neuropsychological assessments
There were no differences between the two
DAT1
genotype groups in term of age, gender, IQ,
trait impulsivity (BIS-11), depression (BDI), trait anxiety (STAI) or working memory capacity
(listening span; all p > .2) (
supplementary results: table S3.3a
), and there were no significant
correlations between any of these trait measures and drug-induced changes in performance
(all p > .2).
Listening span
Working memory capacity (measured with listening span; Daneman and Carpenter 1980,
Salthouse and Babcock 1991) has been associated with striatal dopamine synthesis capacity
(Cools et al., 2008) Moreover, previous studies have shown that dopaminergic drug effects
can be predicted from working memory capacity (Luciana et al., 1992; Kimberg et al., 1997;
Luciana and Collins, 1997; Mehta et al., 2001; Mehta et al., 2003; Mehta et al., 2004; Gibbs
and D’Esposito, 2005b, a; Cools et al., 2007a; Mehta et al., 2008). Accordingly, we assessed
drug effects as a function of listening span. To this end, we divided the group into low-span
participants (n = 23) and high-span participants (n = 25), using a median-split analysis
(
supplemental table S3.3b
). Consistent with prior work, and like the 10R
DAT1
genotype
group, the low-span group was sensitive to the beneficial effects of bromocriptine on task-
switching (Drug x Switching: F(1,22) = 4.457, p = .046). Conversely, the high-span group was
not sensitive to the effect of bromocriptine (Drug x Switching: F(1,24) < 1), similar to the
DAT1
9R group (Drug x Switching x Span: F(1, 46) = 1.2, p > .2). Moreover, in the subgroup
of low-span participants that took part in all four drug sessions, bromocriptine also reduced
the switch cost (Drug x Switching F(1,9) = 5.466, p = .044), an effect that was not present after
pre-treatment with sulpiride (Drug x Switching: F(1,9) < 1) nor after sulpiride alone (F(1,9)
= 1.06, p = .33).
Effects of drugs on mood ratings and prolactin levels
Participants reported no drug-induced changes in mood in the large sample (all p > .3) or in
the subgroup (all p > .05). There were also no significant correlations between drug-induced
changes in task performance and drug-induced changes in mood (all p > .3)
Furthermore, bromocriptine decreased plasma prolactin levels relative to placebo, whereas