Neurochemical and neurophysiological bases of executive control

Abstract

The aim of this work is to further understanding into the biological bases of human executive control through an integration of neuropsychology,psychopharmacology and functional imaging. Executive control includes the prototypical executive functions of response inhibition and error awareness. Response inhibition can be further divided into action cancellation and action restraint, which were assayed with the stop signal (SST) and Error Awareness(EAT) tasks, respectively. The EAT, which is a modified go-no go paradigm, was also used to measure error awareness behaviour. Monoamine neurotransmitters, specifically dopamine, noradrenaline and serotonin, were modulated during task performance with acute doses of methylphenidate (a dopamine and noradrenaline reuptake inhibitor), atomoxetine (a noradrenaline reuptake inhibitor), citalopram (a serotonin reuptake inhibitor) and cabergoline (a D2 agonist). The pharmacoimaging correlates of task performance were also examined with functional magnetic resonance imaging (fMRI). All experiments adhered to a randomised, double blind, placebo-controlled, crossover design and were limited to 18-35 year old healthy right-handed male participants. In Chapter 1, twenty-four (24) participants received acute doses of methylphenidate (30mg), atomoxetine (60mg), citalopram (30mg) and placebo and performed the SST. Methylphenidate, but not atomoxetine or citalopram, led to a reduction in response time variability and stop-signal reaction time indicating enhanced action cancellation compared with the other drug conditions. This enhancement occurred without change to overall response speed suggesting that methylphenidate was modulating core response inhibition processes rather than simply augmenting overall processing speed. This result argued for the prominence of dopaminergic, rather than adrenergic or serotonergic, mechanisms in action cancellation. In Chapter 2, the pharmacoimaging correlates of action restraint aspect were explored. Twenty-seven (27) males received acute doses of methylphenidate(30mg), atomoxetine (60mg), citalopram (30mg) and placebo whilst undertaking the EAT during fMRI. No-go related BOLD activations were then correlated for each of the drug conditions. Although Inhibitory performance under methylphenidate was numerically superior to the other drug conditions,this difference did not achieve statistical significance. On fMRI, methylphenidate activated, versus placebo, the anterior cingulate cortex, right inferior frontal, left middle frontal, left angular and right superior temporal gyri and right caudate. Atomoxetine activated a broad network of cortical regions. Both methylphenidate and atomoxetine, but not citalopram, activated superior temporal, right inferior frontal and left middle frontal clusters. Citalopram only activated the left inferior occipital lobe. Taking each condition’s activations as functionally defined regions of interest, the specificity of no-go related activity was compared across the four conditions. Only methylphenidate demonstrated drug specific effects through increased activation of the anterior cingulate and decreased activation of the caudate. A direct comparison of methylphenidate and atomoxetine showed a similar recruitment in prefrontal regions but specific effects of methylphenidate in the anterior cingulate and caudate. This further supported the notion that dopaminergic processes were making a specific and dissociable contribution during response inhibition. In Chapter 3 the pharmacoimaging correlates of error awareness were examined using the EAT. This study examined the fMRI data obtained whilst testing the cohort described in Chapter 2. Methylphenidate, but not atomoxetine or citalopram, significantly improved error awareness behaviour during the EAT. Methylphenidate, versus the other drug conditions, was also associated with specific BOLD activation differences in the dorsal anterior cingulate cortex and inferior parietal lobe for errors made with versus without awareness. These findings suggested that dopaminergic, rather than noradrenergic or serotonergic, mechanisms might also be important during error awareness. Accordingly, the final experiment in this thesis sought to examine specific dopamine receptor mechanisms of executive control. In Chapter 4 the role of the D2 receptor during response inhibition and error awareness was explored. Twenty-five (25) males received acute doses of cabergoline (1.25mg) and placebo before undertaking the SST and EAT. During the SST cabergoline was able to selectively reduce stop signal reaction time without a concomitant effect on reaction time variability or overall response speed. Cabergoline was, however, unable to improve action restraint in the go no-go component of the EAT. This suggested a specific and dissociable role for the D2 receptor in the action cancellation but not action restraint, which was consistent with the findings of Chapters 1 and 2. Cabergoline was also able to enhance the conscious perception of errors during the EAT, suggesting D2 mechanisms during error awareness, which was supportive of the findings in Chapter 3. In summary, this series of experiments showed that dopaminergic, and potentially D2, mechanisms are specifically implicated in the neurobiology of human executive control. Further experiments with more specific D2 agents, combined imaging paradigms and refined task design would extend these findings.