Large-scale brain networks in affective and social neuroscience: towards an integrative functional architecture of the brain

Abstract

Understanding how a human brain creates a human mind ultimately depends on mapping psychological categories and concepts to physical measurements of neural response. Although it has long been assumed that emotional, social, and cognitive phenomena are realized in the operations of separate brain regions or brain networks, we demonstrate that it is possible to understand the body of neuroimaging evidence using a framework that relies on domain general, distributed structure-function mappings. We review current research in affective and social neuroscience and argue that the emerging science of large-scale intrinsic brain networks provides a coherent framework for a domain-general functional architecture of the human brain.

Figure 1. Relative frequency of mental functions attributed to activation in four networks

Each pie charts depict the relative frequency with which various mental functions are discussed in the context of activation in the “executive control” network (top left), “salience” network (top right), the “mirroring” network (bottom left), and the “mentalizing network” (bottom right) as obtained using the Neurosynth database including over 6,000 publications from over 50 journals [107].

Figure 2. The progression from a faculty psychology/modular approach towards a constructionist/distributed structure approach to brain-function inferences

In (a), individual brain regions specifically compute a domain-specific psychological faculty that could be isolated with a domain-specific behavioral task. In (b), domain-specific brain networks interact to produce responses within domain-specific tasks. (c) Functional motifs within domain-general intrinsic brain networks interact to produce a wide variety of tasks; dotted lines indicate that every network is not necessary engaged to support every task response. Constellations of subprocesses likely underlie each higher order functional description (e.g. “motor movements” can be broken down into layers of motor selection, involving lateral inhibition, etc.). We utilize the higher order process descriptions in this figure to highlight the point that even at a higher order of description, the processes comprising “somatovisceral regulation” are domain-general processes that are not specific to emotion, the processes comprising “mentalizing” are not specific to social cognition, and so on.