Abstract Attention network theory proposes three distinct types of attention - alerting, orienting, and control - that are supported by separate brain networks and modulated by different neurotransmitters, i.e., noradrenaline, acetylcholine, and dopamine. Here, we explore the extent of cortical, genetic, and molecular dissociation of these three attention systems using multimodal neuroimaging. We evaluated the spatial overlap between fMRI activation maps from the attention network test (ANT) and cortex-wide gene expression data from the Allen Human Brain Atlas. The goal was to identify genes associated with each of the attention networks in order to determine whether specific groups of genes were co-expressed with the corresponding attention networks. Furthermore, we analysed publicly available PET-maps of neurotransmitter receptors and transporters to investigate their spatial overlap with the attention networks. Our analyses revealed a substantial number of genes (3871 for alerting, 6905 for orienting, 2556 for control) whose cortex-wide expression co-varied with the activation maps, prioritizing several molecular functions such as the regulation of protein biosynthesis, phosphorylation, and receptor binding. Contrary to the hypothesized associations, the ANT activation maps neither aligned with the distribution of noradrenaline, acetylcholine, and dopamine receptor and transporter molecules, nor with transcriptomic profiles that would suggest clearly separable networks. Independence of the attention networks appeared additionally constrained by a high level of spatial dependency between the network maps. Future work may need to re-conceptualize the attention networks in terms of their segregation and re-evaluate the presumed independence at the neural and neurochemical level.
Attention network theory proposes three distinct types of attention-alerting, orienting, and control-that are supported by separate brain networks and modulated by different neurotransmitters, that is, norepinephrine, acetylcholine, and dopamine. Here, we explore the extent of cortical, genetic, and molecular dissociation of these three attention systems using multimodal neuroimaging. We evaluated the spatial overlap between fMRI activation maps from the attention network test (ANT) and cortex-wide gene expression data from the Allen Human Brain Atlas. The goal was to identify genes associated with each of the attention networks in order to determine whether specific groups of genes were co-expressed with the corresponding attention networks. Furthermore, we analyzed publicly available PET-maps of neurotransmitter receptors and transporters to investigate their spatial overlap with the attention networks. Our analyses revealed a substantial number of genes (3871 for alerting, 6905 for orienting, 2556 for control) whose cortex-wide expression co-varied with the activation maps, prioritizing several molecular functions such as the regulation of protein biosynthesis, phosphorylation, and receptor binding. Contrary to the hypothesized associations, the ANT activation maps neither aligned with the distribution of norepinephrine, acetylcholine, and dopamine receptor and transporter molecules, nor with transcriptomic profiles that would suggest clearly separable networks. Independence of the attention networks appeared additionally constrained by a high level of spatial dependency between the network maps. Future work may need to reconceptualize the attention networks in terms of their segregation and reevaluate the presumed independence at the neural and neurochemical level.
Multitasking is a common feature of many modern work and home environments, and this study investigated the relationship between multitasking performance in two different paradigms: a more controlled task-switching paradigm (TSWP) and a more complex, semi-ecologically valid multitasking paradigm (SynWin). The study also explored whether parallel processors may have performance advantages in a complex dual-task environment. Results showed no significant correlation between individual multitasking efficiency in the TSWP and SynWin paradigms. Additionally, the results indicated that the combination of subtasks was the primary factor affecting performance in dual-task variants of the SynWin, rather than the use of a parallel processing mode. We conclude that there may be constraints with respect to the experimental conditions necessary to generalize findings from controlled multitasking paradigms to semi-ecologically valid tasks scenarios. Future research should prioritize efforts to understand how people multitask in more realistic settings and the underlying cognitive processes involved.
