*To whom correspondence may be addressed:
Email: R.Binney@Bangor.ac.uk
Establishing a Role of the Semantic Control Network in Social Cognitive
Processing: A Meta-analysis of Functional Neuroimaging Studies
Veronica Diveica
1
, Kami Koldewyn
1
& Richard J. Binney
1*
1
School of Psychology, Bangor University, Bangor, Gwynedd, Wales, LL57 2AS
Abstract
Most leading models of socio-cognitive processing devote little discussion to the nature and
neuroanatomical correlates of cognitive control mechanisms. Recently, it has been proposed
that the regulation of social behaviours could rely on brain regions specialised in the
controlled retrieval of semantic information, namely the anterior inferior frontal gyrus (IFG)
and posterior middle temporal gyrus. Accordingly, we set out to investigate whether the
neural activation commonly found in social functional neuroimaging studies extends to these
‘semantic control’ regions. We conducted five coordinate-based meta-analyses to combine
results of over 500 fMRI/PET experiments and identified the brain regions consistently
involved in semantic control, as well as four social abilities: theory of mind, trait inference,
empathy and moral reasoning. This allowed an unprecedented parallel review of the neural
networks associated with each of these cognitive domains. The results confirmed that the
anterior left IFG region involved in semantic control is reliably engaged in all four social
domains. This suggests that social cognition could be partly regulated by the neurocognitive
system underpinning semantic control.
Keywords: social cognition; semantic cognition; cognitive control; empathy; theory of mind;
moral reasoning; trait inference; meta-analysis.
.CC-BY 4.0 International licenseavailable under a
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted April 2, 2021. ; https://doi.org/10.1101/2021.04.01.437961doi: bioRxiv preprint
2
1. Introduction
1
The ability to comprehend and respond appropriately to the behaviour of others is
2
essential for humans to survive and thrive. A major challenge for the cognitive sciences,
3
therefore, is to characterise how we understand others and coordinate our behaviour to
4
achieve mutually beneficial outcomes, and what can cause this ability to break down (Frith,
5
2007). There is an indubitable requirement for systems that control, or regulate, the cognitive
6
processes underpinning social interactions. This is because social interactions are intricate
7
and fraught with the potential for misunderstandings and faux pas; first, the everyday social
8
signals to which we are exposed are typically complex, often ambiguous and sometimes
9
conflicting. This is compounded by the fact that the meaning of a given gesture, expression or
10
utterance can vary across contexts (Barrett et al., 2011; Rodd, 2020). Moreover, once we
11
have settled upon an interpretation of these signals, we are then faced with the additional
12
challenge of selecting an appropriate response, and inhibiting others which might, for
13
example, be utilitarian but socially insensitive or even damaging. In order to undergo social
14
interactions that are coherent, effective and context-appropriate, we must carefully regulate
15
both our comprehension of, and response to, the intentions and actions of others (Binney and
16
Ramsey, 2020; Fujita et al., 2014; Gilbert and Burgess, 2008; Ramsey and Ward, 2020).
17
Despite there being a wealth of literature describing executive functions involved in
18
general cognition (Assem et al., 2020; Diamond, 2013; Duncan, 2013, 2010; Fedorenko et
19
al., 2013; Petersen and Posner, 2012), prominent models of socio-cognitive processing are
20
under-specified in terms of the contribution and neural basis of cognitive control mechanisms
21
(e.g., Adolphs, 2009, 2010; Frith & Frith, 2012; Lieberman, 2007). For example, Adolphs
22
(2009; 2010) only very briefly refers to cognitive processes involved in ‘social regulation’
23
and largely within the limited context of emotional regulation. Likewise, Frith and Frith
24
(2012) refer to a “supervisory system” which has the characteristic features of executive
25
.CC-BY 4.0 International licenseavailable under a
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted April 2, 2021. ; https://doi.org/10.1101/2021.04.01.437961doi: bioRxiv preprint
3
control, but its functional and anatomical descriptions lack detail important for generating
26
testable hypotheses. However, research into specific social phenomena, such as prejudice
27
(Amodio, 2014; Amodio and Cikara, 2021) and automatic imitation (Cross et al., 2013;
28
Darda and Ramsey, 2019) has recently begun to give the matter of cognitive control greater
29
attention. Of particular interest has been the contribution of the domain-general multiple-
30
demand network (MDN), a set of brain areas engaged by cognitively-challenging tasks
31
irrespective of the cognitive domain (Assem et al., 2020; Duncan, 2010; Fedorenko et al.,
32
2013; Hugdahl et al., 2015). MDN activity increases with many kinds of general task
33
demand, including working memory load and task switching, and it has been suggested that
34
this reflects the implementation of top-down attentional control and the optimal allocation of
35
cognitive resources to meet immediate goals (Duncan, 2013, 2010). The MDN is comprised
36
of parts of the precentral gyrus, the middle frontal gyrus (MFG), the intraparietal sulcus
37
(IPS), insular cortex, the pre-supplementary motor area (pre-SMA) and the adjacent cingulate
38
cortex (Assem et al., 2020; Fedorenko et al., 2013), some of which have been implicated in
39
controlled social processing such as, for example, working memory for social content (Meyer
40
et al., 2012), social conflict resolution (Zaki et al., 2010), inhibition of automatic imitation
41
(Darda and Ramsey, 2019) and mental state inference or theory of mind (ToM) (e.g.
42
Rothmayr et al., 2011; Samson et al., 2005; Van der Meer et al., 2011). However, there are at
43
least three key unresolved questions regarding the role of cognitive control in social
44
cognition. First, it remains to be seen whether there could be multiple, distinguishable
45
mechanisms of, and neural systems for, control. Second, it is unclear whether there exists a
46
subset of control systems that are specialised towards processing social information and,
47
third, we have little understanding as to whether certain types of control are necessary for all
48
or only select social behavioural phenomena. Shedding light on these issues has the potential
49
.CC-BY 4.0 International licenseavailable under a
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted April 2, 2021. ; https://doi.org/10.1101/2021.04.01.437961doi: bioRxiv preprint
4
to generate important new hypotheses regarding social behaviour both in the context of health
50
and injury/disease.
51
It has recently been proposed that a relatively specialised form of cognitive control,
52
termed semantic control, could be particularly important for social cognitive processing
53
(Binney and Ramsey, 2020). This follows a broader claim that social cognition and its neural
54
correlates can be understood as a nuanced form of semantic cognition which itself is defined
55
as a set of processes involved in extracting meaning from the environment and using it to
56
guide purposeful and context-appropriate behaviour (Binney and Ramsey, 2020; Lambon
57
Ralph et al., 2017). This framework contrasts with approaches that look upon social
58
processing as a distinct or even special case of cognition (i.e., domain-specific models;
59
Barrett, 2012; Saxe, 2006; but also see Amodio, 2019; Amodio and Cikara, 2021; Schaafsma
60
et al., 2015; Spunt and Adolphs, 2017) and, instead, posits that it is underpinned by two,
61
more domain-general neurocognitive systems. The first system is representational in nature
62
and supports the acquisition and long-term storage of conceptual-level knowledge about
63
objects, people, abstract concepts, and events. The anterior temporal cortices act as a central,
64
supramodal semantic store through interaction with modality-specific and lower-order
65
heteromodal association cortices (Binney et al., 2010; Kuhnke et al., 2021; Lambon Ralph et
66
al., 2017; Patterson et al., 2007; Pobric et al., 2010). The second system, the semantic control
67
system, modulates activation of semantic knowledge to bring to the fore aspects of
68
conceptual information that are relevant to the situational context or the task at hand while
69
inhibiting irrelevant aspects (Chiou et al., 2018; Jefferies, 2013; Lambon Ralph et al., 2017).
70
The reasons why semantic control should be critical for social cognition and
71
interaction are uncomplicated; we retain a vast amount of socially-relevant knowledge
72
including knowledge about familiar people (Greven et al., 2016; Hassabis et al., 2014), about
73
the structure of and relationship between social categories and their associated stereotypes
74
.CC-BY 4.0 International licenseavailable under a
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted April 2, 2021. ; https://doi.org/10.1101/2021.04.01.437961doi: bioRxiv preprint
5
(Freeman and Johnson, 2016; Quinn and Rosenthal, 2012), and an understanding of abstract
75
social concepts, norms and scripts (Frith and Frith, 2003; Van Overwalle, 2009). But only a
76
limited portion of this information is relevant in a given social instance and it would be
77
computationally inefficient to automatically retrieve it all. For example, there is no need to
78
retrieve information about someone’s personality traits, or personal interests and hobbies, if
79
the only task is to pick them out from within a crowd. Moreover, the types and the scope of
80
information we need to retrieve to understand and respond appropriately to certain social
81
signals change according to the context, and irrelevant information could potentially
82
interfere. Therefore, semantic control is essential for limiting semantic retrieval according to
83
the circumstances and avoiding potential social errors.
84
There is a growing body of convergent computational modelling, patient,
85
neuroimaging and neuromodulation evidence that the semantic control system is supported
86
by a neural network that is distinct from that underpinning semantic representation (e.g.,
87
Corbett et al., 2009; Davey et al., 2016, 2015; Jackson, 2021; Jefferies et al., 2008; Jefferies
88
and Lambon Ralph, 2006; Teige et al., 2018). Specifically, semantic control engages regions
89
of the MDN, as well as the semantic control network (SCN) which comprises the anterior
90
IFG and the posterior middle temporal gyrus (pMTG) (Badre et al., 2005; Davey et al., 2016;
91
Jackson, 2021; Noonan et al., 2013). Moreover, while the domain-general MDN is engaged
92
by semantic tasks, and particularly those with high control demands (Jackson, 2021;
93
Thompson et al., 2018), there is evidence that both the anatomy of the SCN and MDN and
94
their functional contributions to controlled semantic processing are at least partially distinct
95
(Gao et al., 2020). In particular, fMRI studies reveal that the mid- to posterior IFG (pars
96
triangularis and pars opercularis), nodes of the MDN, have been shown to increase activity in
97
response to increased ‘semantic selection’ demands, a process that is engaged when
98
automatic retrieval of semantic knowledge results in competition between multiple
99
.CC-BY 4.0 International licenseavailable under a
was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint (whichthis version posted April 2, 2021. ; https://doi.org/10.1101/2021.04.01.437961doi: bioRxiv preprint
