Abstract concepts, language and sociality:

from acquisition to inner speech

rstb.royalsocietypublishing.org
Anna M. Borghi1,2, Laura Barca2, Ferdinand Binkofski3 and Luca Tummolini2
1
Department of Dynamical and Clinical Psychology, Sapienza University of Rome, Via dei Marsi 78, Rome 00185,
Italy
2
Institute of Cognitive Sciences and Technologies, Italian National Research Council, Via San Martino della
Review Battaglia 44, Rome 00185, Italy
3
Division for Clinical Cognitive Sciences, University Hospital Aachen, Pauwelsstrasse 17, 52074 Aachen, Germany
Cite this article: Borghi AM, Barca L,
AMB, 0000-0001-9455-8408; LB, 0000-0002-8553-7717; FB, 0000-0002-6750-943X;
Binkofski F, Tummolini L. 2018 Abstract LT, 0000-0001-7791-0457
concepts, language and sociality: from
acquisition to inner speech. Phil. Trans. R. Soc. The problem of representation of abstract concepts, such as ‘freedom’ and
B 373: 20170134. ‘justice’, has become particularly crucial in recent years, owing to the
increased success of embodied and grounded views of cognition. We will
http://dx.doi.org/10.1098/rstb.2017.0134
present a novel view on abstract concepts and abstract words. Since abstract
concepts do not have single objects as referents, children and adults might
Accepted: 9 January 2018 rely more on input from others to learn them; we, therefore, suggest that lin-
guistic and social experience play an important role for abstract concepts. We
will discuss evidence obtained in our and other laboratories showing that
One contribution of 23 to a theme issue
processing of abstract concepts evokes linguistic interaction and social
‘Varieties of abstract concepts: development, experiences, leading to the activation of the mouth motor system. We will
use, and representation in the brain’. discuss the possible mechanisms that underlie this activation. Mouth
motor system activation can be due to re-enactment of the experience of con-
Subject Areas: ceptual acquisition, which occurred through the mediation of language.
Alternatively, it could be due to the re-explanation of the word meaning,
cognition, behaviour
possibly through inner speech. Finally, it can be due to a metacognitive pro-
cess revealing low confidence in the meaning of our concepts. This process
Keywords: induces in us the need to rely on others to ask/negotiate conceptual mean-
abstract concepts, social interaction, linguistic ing. We conclude that with abstract concepts language works as a social tool:
experience, grounded cognition, inner speech, it extends our thinking abilities and pushes us to rely on others to integrate
our knowledge.
metacognition
This article is part of the theme issue ‘Varieties of abstract concepts:
development, use, and representation in the brain’.
Author for correspondence:
Anna M. Borghi
e-mail: anna.borghi@uniroma1.it
1. Introduction
Abstract concepts (ACs) are quite heterogeneous: they differ in degree of
abstractness, varying from more to less abstract and they belong to subtypes
(e.g. emotional, social, mathematical concepts) that can be distinguished in
terms of content and neural representation [1–6].
Despite these differences, when compared with concrete concepts like
‘table’ and ‘dog’, they have much in common: they are generally difficult to
associate with a single image, they do not have a single object/entity as refer-
ent, and more often refer to complex situations with multiple objects/entities
[7]. Furthermore, even if they are grounded in sensorial modalities, they activate
less the five senses than concrete concepts [8] and their content is more variable
both within and across participants, as testified for example by the feature
variability in property generation tasks [9].
In the first part of the paper we will discuss and substantiate with evidence
two claims derived from the words as social tools (WAT) proposal on ACs
[9,10]. We state that the more abstract concepts are, (1) the more linguistic
and social/interactive input from others is needed to acquire them; (2) the
more they are represented activating linguistic brain networks, and engage
the mouth motor system and the acoustic system. In the second part we will dis-
cuss inner grounding of ACs, and focus on metacognition. We will propose a

& 2018 The Author(s) Published by the Royal Society. All rights reserved.
link between activation of the mouth motor representation (c) Evidence on mouth motor system activation during 2
and metacognitive feeling that we need others’ competence
abstract concept learning and processing

rstb.royalsocietypublishing.org
to enrich our concepts [11–14].
In this section, we will overview studies performed to test the
hypothesis that AC processing involves the mouth more than
does concrete concept processing. First, we found facilitation
2. Abstractness, linguistic and social input and of mouth responses in studies on learning of novel/artificial
mouth motor system involvement categories in adults. We operationalized concrete categories
as having a single, concrete object as referent, rich in percep-
(a) Why language is important for abstract concepts tual features; abstract categories were instead operationalized
The view that language is important for AC representation is by multiple interacting elements (e.g. small cylinders moving
not new. According to the dual-coding theory [15], concrete in spirals and then knocking against each other) [28]. Once

Phil. Trans. R. Soc. B 373: 20170134

concepts evoke imagistic representations, and ACs verbal they learned these novel categories, participants were
ones. In line with Paivio [15], we contend that concepts acti- taught their names. Results of a subsequent property verifica-
vate multiple representations. Differently from Paivio, we do tion task revealed a facilitation of microphone (mouth)
not equate imageability with concreteness, because the two responses to abstract words. This facilitation was more pro-
dimensions have proven to be correlated but not equivalent nounced when the word meaning had been explained to
[16]. In developmental psychology, the syntactic bootstrap- the participants. Responses to concrete words were instead
ping hypothesis (e.g. [17]) proposes that children learn facilitated when participants responded by pressing a key
‘easy words’ by mapping the words to the objects/entities with the hand (figure 1a).
in the world. Once a substantial number of words have Second, facilitation of the mouth responses occurred also
been acquired, learning proceeds by adding structure to the in the processing of real abstract words. Borghi & Zarcone
original machinery: this allows acquisition of ‘hard words’ [29] asked participants to decide whether abstract and con-
such as abstract ones. In this process syntax and semantics crete prime-definitions matched with Italian concrete and
are strictly linked: syntax cannot substitute semantics, but it abstract target-words: to respond they pressed a key with
can help learners to narrow their hypotheses on the possible the hand or a device with the teeth. The advantage of hand
word meaning. Recently, proponents of multiple represen- over mouth responses disappeared with abstract words, con-
tation views have emphasized the importance of language firming that AC processing activates the mouth motor system
for AC learning [9,18,19] and representation [10,16,20 –23]. (figure 1b). The finding is consistent with results from explicit
Why should language be important for ACs? First of all, it ratings with words and sentences. Participants think that
is critical for their acquisition, and this might impact their rep- abstract words involve the mouth more and concrete words
resentation. Different justice situations are far more the hand more [31]. Similarly, the content of mental state sen-
heterogeneous than different cups, and using a common tences was more associated with mouth actions than with
label helps us assemble them in a category. Furthermore, hand or leg actions, while the content of maths-related sen-
explanations of the conceptual meaning can be more crucial tences was more associated with the hand [3], likely due to
to form the concept of ‘justice’, for instance, than that of the activation of finger-counting habits [32].
‘cup’. In order to learn ACs we might also need to actively The studies illustrated so far provide evidence of facili-
ask for definitions/contributions from competent community tation with mouth responses when processing novel and
members [12,13] or to resort to recognized information sources real abstract words. If the activation of the mouth motor
(e.g. Wikipedia). Finally, because ACs are heterogeneous and system is not simply a by-product of AC processing, then
complex, language can help us process them. Language aug- occupying the mouth should interfere with word acquisition
ments our cognitive abilities, computational resources, and and processing. We performed two developmental studies
problem-solving capacities [24 –26]; this extension might be investigating whether pacifier use influences language acqui-
pivotal to use of concepts that are increasingly abstract. sition and processing later in life, and one study investigating
whether actively moving the mouth (gum chewing) affects
(b) Language and vocalization adults’ online processing of ACs. Barca et al. [33] asked
When we are exposed to words, their sub-vocal pronuncia- first-grade children to define abstract (agreement), concrete
tion is activated [27]. This mechanism underlies the (flag) and emotional ( pain) words. Children were divided
exposure to all words, but we propose that it is more pro- in groups depending on how long they had been using the
nounced in the case of abstract words. Owing to the pacifier (never, until age 2, until age 3, beyond 3 years). Defi-
heterogeneity of their members, we contend that in order to nitions were then evaluated in terms of accuracy and the
acquire ACs the scaffolding of the physical environment has conceptual relations provided to define the words were
to be complemented by linguistic information, for example, coded (i.e. perceptual, spatial, action –function, emotion, situ-
labels and explanations. This information can be conveyed ation, experiential, superordinate, subordinate, norms, free
either by others (e.g. parents, experts) or by written sources association). Pacifier overuse did not influence accuracy of
(e.g. books, Internet), depending on expertise and age. In definitions. However, the pattern of conceptual relations eli-
our view the activation of linguistic experience has an embo- cited by the three kinds of words clearly differed
died counterpart: it should lead to a sub-vocal pronunciation depending on pacifier use. The distinction between emotional
of the words and/or of ( part of ) their explanations. and concrete words was less marked for children who used a
Responses with the mouth should thus be facilitated during pacifier for more than 3 years, and the distinction between
AC processing. Conversely, occupying the mouth during concrete and abstract words became progressively less
AC acquisition and processing should induce an interference, marked for children who used it until 3 years and beyond.
impairing acquisition and slowing down responses. Independently from the type of word they had to define,
 (a) 3
(b)
1400

RT difference M – H (ms)

rstb.royalsocietypublishing.org
1200
RTs (ms) 1000 150
800
keyboard
600 microphone
400 50
200
0
abstract words concrete words abstract words concrete words

Phil. Trans. R. Soc. B 373: 20170134

(c)
categorization time (ms) 1800 >3 years

1700

1600

1500 never

2–3 years
1400
<2 years

1300
abstract concrete emotional
words words words

Figure 1. Interactions between response effector and kind of concept showing a facilitation of the mouth responses with abstract concepts. (a) Property verification
task (from Borghi et al. [28]). Interaction obtained with novel categories. (b) Definition– word matching task (from Borghi & Zarcone [29]). Interaction obtained
with real categories. (c) Semantic categorization task (from Barca et al. [30]). Interaction between pacifier use and kind of concept, revealing longer response times
(RTs) with abstract concepts for late-users of pacifier. (Online version in colour.)

children who overused the pacifier produced fewer free The role played by the two mechanisms can be disen-
associations (the only feature more frequent in abstract than tangled with the help of a further study [30]. We asked
in other concepts), fewer emotional, experiential and interac- 8-year-olds with a different history of pacifier use to perform
tive relations, and slightly more subordinates. The results a semantic categorization task: they had to press a different
suggest that overuse of pacifier during language acquisition, key on the keyboard if the word displayed on the screen
even if it does not affect accuracy in defining words, influ- referred to an animal or to something else. The non-animal
ences later conceptualization: it is namely correlated with a words were our critical items: abstract, concrete and
less marked distinction between emotional and concrete emotional words. Response times analyses revealed that
words, and between concrete and abstract words. children who overused the pacifier processed abstract
Two different mechanisms might be responsible for this words slower than all other children. Given that the effect
phenomenon: mouth motor system interference and facial was specific for ACs and did not involve emotional ones,
mimicry. The mouth motor system interference one is based the result is consistent with the view that pacifier overuse
on the WAT proposal: by keeping the mouth actively occu- interferes with processing of ACs and influences their
pied, the pacifier would hinder the benefits of linguistic representation.
input, which is particularly crucial to learn ACs, and to The interference effect elicited by active engagement
rehearse the word and/or its explanation via inner speech. of the mouth during AC processing is also revealed by
Inner speech can also serve as vehicle to metacognition, pre- another recent study [35] in which adult participants rated
paring children to interact with others when they feel scarcely complexity and pleasantness of abstract and concrete
confident on their knowledge (see later discussion on concepts. When participants chewed gum, perceived plea-
metacognition). santness of ACs decreased and perceived complexity of
Alternatively, the reduced facial mimicry induced by ACs increased compared with concrete concepts. Gum chew-
pacifier use might reduce motor resonance with others, ing has been associated with reduction of nociception and
thereby making the grounding of emotional concepts more increase of alertness, while it is debated whether it increases
difficult. Importantly, the two mechanisms are not in contrast calmness and contentedness (e.g. [36]). If the mood change
and might even overlap. However, we are inclined to think induced by gum chewing had an effect on performance, we
that they play a different role for abstract and emotional should find a general increase of pleasantness ratings and a
words. We hypothesize that the mouth motor system interfer- decrease of difficulty ratings, but we found the opposite,
ence mechanism concerns specifically abstract words, in line and the effect was specific for ACs. We are, therefore, inclined
with the WAT view, while the facial mimicry one concerns to believe that the effect is not due to mood or alertness modi-
emotional words, in keeping with evidence showing a fications but rather, as predicted by WAT, to the interference
relationship between pacifier use and development of induced by actively moving the mouth, which conflicts with
emotional competence [34]. activation of the mouth during AC processing.
(d) Mechanisms underlying mouth motor system neural evidence on the role of both semantic and phono- 4
articulatory aspects of language for ACs. Consistently,
activation

rstb.royalsocietypublishing.org
meta-analyses and functional magnetic resonance imaging
The mouth activation with abstract words, leading to facili- (fMRI) studies have shown a higher left-hemispheric acti-
tation if the mouth is a response effector or to interference if vation, determined by stronger activation of the left inferior
the mouth is actively occupied during word processing, frontal gyrus, typically associated with phonological working
could be due to three different mechanisms: (1) the re- memory, left middle temporal gyrus [41], anterior superior
enactment of the experience of conceptual acquisition, which temporal sulcus [42], and superior temporal gyrus, associated
typically occurred through the linguistic mediation, in a with acoustic experience [43].
social context. (2) The re-explanation to oneself of the word
meaning, possibly through inner speech [26]. (3) The
preparation to ask others for information on word meanings 3. Kinds of concepts, inner grounding and

Phil. Trans. R. Soc. B 373: 20170134

[12–14]. Results on acquisition of novel categories showing a
facilitation of mouth responses and extended-in-time interfer- mouth motor system activation
ence effects of pacifier overuse on processing of ACs are As anticipated in the Introduction, although ACs have typi-
compatible with the re-enactment explanation: likely children cally been considered a unitary whole, they might greatly
were not able to fully benefit from the input of others owing differ in their content. Consistently, fMRI studies have
to the pacifier, and this might influence the re-enactment of shown that, for example, the concept ‘arithmetic’ activates
the acquisition experience. Results showing facilitation of numerical cognition areas, while the mental state concept
mouth responses with ACs (e.g. [28,29]) and interference ‘convince’ engages mentalizing and social cognition areas
when the mouth is actively involved in gum chewing are [4,44]. Other concepts like ‘confidence’ and ‘self-esteem’
more compatible with the second and third mechanisms, i.e. might instead be grounded on metacognition. Along similar
re-explanation to oneself and/or the preparation to ask others lines, Barsalou [45] proposed that the grounding of ACs
for information. Importantly, however, these mechanisms such as ‘truth’ depends on the monitoring of internal, inferen-
could coexist; further research is needed in order to better disen- tial processes, such as the successful/unsuccessful matching
tangle them. Notice that all mechanisms help grounding, of expectations and sensations. We will here contend that
because they assist us in tracking conceptual referents. metacognition not only is relevant to explain the differences
These three mechanisms might be implemented through in content between ACs but also can offer a new mechanism.
inner articulation of the label and inner speech, viewed as a We have extensively discussed one of the crucial mechan-
more complex and possibly discourse-like internal articulation isms involved in AC processing, namely linguistic activation.
of the word sound. We will now illustrate a second mechanism that in our view
characterizes AC representation, and propose a possible
(i) Inner articulation of the label relationship between metacognition and language covert
While silently reading words, we covertly articulate their activation.
sound, producing a mouth movement [27,37] or activating
the speech-motor cortex [38]. Pronouncing the name of an (a) Inner grounding
object and internally articulating its label is useful to memor- Inner grounding is the process of grounding concepts on
ize it but it also improves object perception. It can indeed perceptual systems tracking states and processes in the
facilitate object finding in a visual search task [39]: using environment inside the body. Internal perceptual systems
the word ‘tree’ is better than simply thinking about a tree can be contrasted with perceptual systems that track infor-
to perceptually detect trees. We propose an extension of mation in the environment outside the body or at the
this hypothesis: in our view the effect of labels is more crucial interface between body and external environment (e.g.
with ACs. Pronouncing/hearing their labels can help us to vision, audition, touch). Some concepts are grounded
put together dissimilar category exemplars, thus reducing mostly on perceptual systems that track information about
the working memory overload linked to the generation of what happens in the outside world, while other concepts
many hypotheses. In addition, labels can help us to select are more related to the inside world. Different perceptual
the relevant cues: while we observe a scene with people systems are tuned to track information on internal states
running in a field we interpret it differently if we hear/ and processes. Interoception tracks information about the
pronounce the word ‘freedom’ or the word ‘grass’. state and processes of the body organs, while proprioception
tracks information about the state and processes of the body
(ii) Inner speech itself and its parts. We consider metacognition as another
We propose that with ACs the need to speak to ourselves or kind of internal perceptual process, tuned to track
to prepare to interact with others is stronger than with information about cognitive states and brain processes.
concrete concepts. Inner speech can help us retrieve infor- We hypothesize that concepts that are grounded on meta-
mation on the category exemplars, reflect on the word cognitive processes are perceived as more abstract than those
meaning, reconstruct the linguistic explanations we received, grounded on interoception and proprioception. A possible
predict what we will need to search [40] and prepare our- reason is that the information tracked by proprioception
selves to interact with others when we feel scarcely and interoception could also be partly picked up by extero-
confident in our own knowledge. We propose that the prob- ception. Only metacognition tracks precisely the kind of
ability of using inner speech becomes higher with the information that cannot be picked up by any external percep-
increase of abstractness and complexity of concepts. tual system, much like colour information cannot be tracked
The proposal that AC processing likely involves inner by the haptic system and temperature cannot be tracked by
articulation of the label and inner speech is compatible with the visual system.
 MoA
(a) (b) 5

AoA
AoA

rstb.royalsocietypublishing.org
EMONEG
ABS
ABS

EMOPOS
EMONEG

SOCIAL
EMOPOS

SOCIAL

META

TASTE
META ABS

OLFACTION
TASTE MoA CNR

HEAR
HEAR
BOI AoA

Phil. Trans. R. Soc. B 373: 20170134

OLFACTION HEA EMO

EMO
IMG

TOUCH
EMO

TOUCH MET INT

CNR
CNR SOC
abstractness

INT
INT

SIGHT
SIG EMON
inner grounding
SIGHT
valence

IMG
IMG TAS TOU
senses
BOI
OLF EMOP
social
.0
.9
.8
.7
.6
.5
.4
.3
.2
.1
0
1
2
3
4
5
6
7
8
9
0
0.
0.
0.
0.
0.
0.
0.
0.
0.
1.
–1
–0
–0
–0
–0
–0
–0
–0
–0
–0

Figure 2. From Villani et al. [6]. (a) Correlation matrix describing correlations between each of the individual dimensions. (b) Network representing latent factors
and correlations. AoA, age of acquisition; ABS, abstractness; EMONEG/EMON, negative emotion; EMOPOS/EMOP, positive emotion; SOCIAL/SOC, social valence; META/
MET, metacognition; HEAR/HEA, hearing; EMO, emotional valence; CNR, concreteness; INT, interoception; SIGHT, vision; IMG, imageability; BOI, body – object
interaction; SIG, sight; TOU, touch; TAS, taste; OLF, olfaction. (Online version in colour.)

(b) Metacognition especially in abstract domains, such as God, death and life,
Inner grounding in general, and metacognition in particular, and to monitor their competence [47]. We hypothesize that this
can be important mechanisms useful to distinguish between form of metacognition increases with the increase of the abstract-
abstract and concrete concepts, and also to identify different ness level, and that it might explain the activation of the mouth
kinds of ACs. We identify two forms of metacognition that motor representation. The idea that language can provide access
might characterize ACs. General metacognition concerns more to grounding is compatible also with the recent fMRI finding [48]
generally the tracking of states and processes occurring in that ACs activate both linguistic contextual information and
the brain. For example, to classify an experience as a recall/ semantic features, but in different brain regions: linguistic con-
memory, rather than as mere imagery, we need to interpret textual information is reflected mostly in high-level linguistic
the image as deriving from a memory trace. Instead, if the areas, whereas semantics engages distributed brain areas.
image is a ‘fantasy’, we keep trace of having built it within
our mind. These inner brain operations share many aspects (c) Inner grounding and language: evidence
with introspective properties that, according to Barsalou & Preliminary results of a recent study can give some hints on the
Wiemer-Hastings [46], characterize ACs; they need not have role played by language and inner grounding for ACs [6]. Par-
a specific feeling associated with them, and are likely uncon- ticipants rated 425 Italian abstract concepts on a variety of
scious. We hypothesize that general metacognition involves the dimensions, from the most classical ones, typically used to
majority of ACs, but especially mental state ones. identify ACs (abstractness, concreteness, imageability), to
There is another dimension to metacognition that we more novel ones, such as emotional valence [16], interoception
propose to be crucial for grounding of ACs. This dimension is [49], body–object interaction (BOI: ease with which a human
inspired by the distinction introduced by Shea [14] between body can physically interact with a word’s referent; [50]),
implicit deference—the disposition to rely on others while social valence [1,9], perceptual strength in sight, hearing,
using a concept—and explicit deference—a judgment on our touch, taste, olfaction [8], age of acquisition (AoA), modality
own mental states, for example leading us to decide that our con- of acquisition (MoA: based more on experience or more on
cept is not adequate. This distinction is at the basis of the notion language; [51]), and general-metacognition (reliance on internal
of social metacognition we propose. One of the reasons why mental/cognitive processes). Interestingly, abstractness was
language and sociality are important for ACs, is that we might positively correlated with modality of acquisition (MoA) and
need to rely on others to fix reference. This process is based on metacognition, and negatively correlated with imageability,
a metacognitive assessment: we reflect on our own concepts, rea- BOI, concreteness, touch and sight (figure 2a). Hence, ACs are
lizing that to fully capture their meaning we need the associated with linguistic acquisition, which occurs late. Five
contribution of others. We use the term social metacognition to latent factors underlie the original dimensions (figure 2b). Fac-
suggest a link between an internal process, the awareness of tors aggregated as follows: (1) abstractness, including the
our knowledge inadequacy, and the need to activate actions opposition between concreteness, BOI, vision/sight and image-
directed toward others. Studies on testimony reveal the increas- ability to abstractness, AoA and MoA; (2) inner grounding,
ing ability of children to rely on others as information sources characterized by the positive correlations between
interoception, emotion and general-metacognition; (3) senses, this proposal, showing that language activation engages the 6
i.e. the four external perceptual modalities, except hearing mouth motor system. Mouth involvement seems to play a

rstb.royalsocietypublishing.org
likely because highly correlated with social valence; (4) emotion substantial role, as interference effects ( pacifier overuse in
polarity (negative and positive emotions); (5) social dimension, infancy, gum chewing during online processing) reveal.
including social valence and audition. Thus inner grounding The reviewed evidence suggests that this activation is due
factor is independent from abstractness, while linguistic MoA to the inner articulation of the label and/or to inner speech,
is not. This distinction is reflected in analyses on concept contributing either to the re-enactment of the acquisition
kinds. We classified the terms into 11 categories, following the experience, to the re-explanation to ourselves of the concep-
major classifications of the literature (e.g. emotions, mental tual meaning and/or to the information request on the
states, institutional concepts). Further cluster analyses will conceptual meaning to others. Across the three mechanisms,
allow us to test our hypotheses on how items/words group phonological and semantic information are strictly interwo-
together. We performed ANOVAs to ascertain whether and ven. Importantly, all mechanisms contribute to grounding,

Phil. Trans. R. Soc. B 373: 20170134

how the categories identified by the experimenters differed because through different strategies they help us to search
from each other on each dimension. We found a major distinc- for conceptual referents.
tion between ACs such as ‘emotions’ and ‘bodily states’, The activation of language and of the mouth motor system
characterized by higher scores in interoception and BOI, and does not occur to the same extent with all concepts. We argued
‘institutional’ and ‘knowledge domains’ (e.g. philosophy) that ACs are characterized by inner grounding. While for more
concepts, judged to be linguistically acquired (higher MoA). ‘embodied’ ACs interoception and proprioception play a
This study suggests that perceived differences in concept major role, for less ‘embodied’ ACs we might need language
kinds depend not only on content but also on mechanisms, and metacognition. We propose that even abstract meta-
such as interoception and language activation. Interestingly, cognitive processes are accomplished with the help of a
the more concepts are grounded in external and internal per- bodily medium, the mouth. Sub-vocalizations and inner
ception (interoception), the less linguistic experience weighs. speech can namely facilitate tracking of information about
Consistently, recent fMRI evidence revealed that mental state cognitive states and brain processes. We predict that the
concepts activate the face motor system in the brain more mouth motor system activation is particularly pronounced
than emotional concepts [52]. when we feel uncertain of our concepts and need to rely on
As to metacognition, as expected, general-metacognition others (social metacognition).
was more strongly correlated to abstractness than emotion In conclusion, words are social tools in many senses: in
(interoception was not correlated to abstractness), but ratings the most obvious sense that they help us to communicate
did not differ between concepts kinds. However, ratings might with others and change the state of the world, and in two
be limited in accessing a likely largely unconscious dimension. most subtle senses discussed in this paper: words extend
The case is different for social-metacognition, not tested in this our thinking abilities, and they open us to sociality when
preliminary study. Social-metacognition might be more explicit we doubt our knowledge and we need to share and align
because it leads to an action with others. In future work we with others [53].
plan to investigate how it characterizes ACs and whether it
Data accessibility. This article has no additional data.
interacts with mouth motor system activation. We predict that
Authors’ contributions. A.M.B., L.B. and L.T. conceived the paper. A.M.B.
the more concepts are abstract and complex, the more social- drafted the text and all authors provided critical revisions to it. All
metacognition weighs, and the more the mouth is activated, authors approved the final version.
either to re-explain to ourselves conceptual meaning or to Competing interests. We have no competing interests.
prepare ourselves to ask/negotiate it to/with others. Funding. Project ‘Abstract concepts, language and sociality’, protocol
no. RG11715C7F1549F7, Sapienza University of Rome, 2017.
Acknowledgements. We would like to thank Cristiano Castelfranchi,
4. Conclusion Felice Cimatti, Elena Daprati, Marco Tullio Liuzza, Luisa Lugli, Clau-
dia Mazzuca, Daniele Nico, Domenico Parisi, Giovanni Pezzulo,
The WAT view proposes that ACs activate linguistic and Claudia Scorolli, Caterina Villani and Giandaniele Zannino for
social experience. We overviewed recent studies supporting their collaboration and continual discussions on this topic.

References
1. Crutch SJ, Troche J, Reilly J, Ridgway GR. 2013 and concrete domains. PLoS ONE 8, e0067090. multiple dimensions: norms on more than 400
Abstract conceptual feature ratings: the role of (doi:10.1371/journal.pone.0067090) abstract words.
emotion, magnitude, and other cognitive domains 4. Mellem MS, Jasmin KM, Peng C, Martin A. 2016 7. Barsalou LW. 2003 Abstraction in perceptual symbol
in the organization of abstract conceptual Sentence processing in anterior superior temporal systems. Phil. Trans. R. Soc. Lond. B 358, 1177–
knowledge. Front. Human Neurosci. 7, 186. (doi:10. cortex shows a social-emotional bias. 1187. (doi:10.1098/rstb.2003.1319)
3389/fnhum.2013.00186) Neuropsychologia 89, 217– 224. (doi:10.1016/j. 8. Connell L, Lynott D. 2012 Strength of perceptual
2. Dreyer FR, Frey D, Arana S, von Saldern S, Picht T, neuropsychologia.2016.06.019) experience predicts word processing performance
Vajkoczy P, Pulvermüller F. 2015 Is the motor system 5. Setti A, Caramelli N. 2005 Different domains in better than concreteness or imageability.
necessary for processing action and abstract emotion abstract concepts. In Proc. XXVII Ann. Conf. Cognitive Cognition 125, 452–465. (doi:10.1016/j.cognition.
words? Evidence from focal brain lesions. Front. Science (eds B Bara, B Barsalou, M Bucciarelli), pp. 2012.07.010)
Psychol. 6, 1661. (doi:10.3389/fpsyg.2015.01661) 1997 –2002. Mahwah, NJ: Erlbaum. 9. Borghi AM, Binkofski F. 2014 Words as social tools:
3. Ghio M, Vaghi MMS, Tettamanti M. 2013 Fine- 6. Villani C, Lugli L, Liuzza MT, Borghi AM. In an embodied view on abstract concepts. Berlin,
grained semantic categorization across the abstract preparation. Kinds of abstract concepts and their Germany: Springer.
10. Borghi AM, Binkofski F, Cimatti F, Scorolli C, Tummolini simulations. J. Exp. Psychol. 35, 423–433. (doi:10. 41. Wang J, Conder JA, Blitzer DN, Shinkareva SV. 2010 7
L. 2017 The challenge of abstract concepts. Psychol. 1037/a0014504) Neural representation of abstract and concrete

rstb.royalsocietypublishing.org
Bull. 3, 263–292. (doi:10.1037/bul0000089) 28. Borghi AM, Flumini A, Cimatti F, Marocco D, Scorolli concepts: a meta-analysis of neuroimaging studies.
11. Burge T. 1979 Individualism and the mental. C. 2011 Manipulating objects and telling words: a Hum. Brain Mapp. 31, 1459–1468. (doi:10.1002/
Midwest Stud. Phil. 4, 73 –121. (doi:10.1111/j.1475- study on concrete and abstract words acquisition. hbm.20950)
4975.1979.tb00374.x) Front. Psychol. 2, 15. (doi:10.3389/fpsyg.2011. 42. Binder JR, Desai RH, Graves WW, Conant LL. 2009
12. Prinz JJ. 2002 Furnishing the mind: concepts and 00015) Where is the semantic system? A critical review and
their perceptual basis. Cambridge, MA: MIT Press. 29. Borghi AM, Zarcone E. 2016 Grounding abstractness: meta-analysis of 120 functional neuroimaging
13. Prinz JJ. 2012 Beyond human nature. How culture abstract concepts and the activation of the mouth. studies. Cereb. Cortex 19, 2767–2796. (doi:10.1093/
and experience shape our lives. New York, NY: Front. Psychol. 7, 1498. (doi:10.3389/fpsyg.2016. cercor/bhp055)
Penguin. 01498) 43. Hoffman P, Binney RJ, Lambon RMA. 2015 Differing
14. Shea N. 2018 Metacognition and abstract concepts. 30. Barca L, Mazzuca C, Borghi AM. 2017 Pacifier contributions of inferior prefrontal and anterior

Phil. Trans. R. Soc. B 373: 20170134

Phil. Trans. R. Soc. B 373, 20170133. (doi:10.1098/ overuse interferes with children’s abstract word temporal cortex to concrete and abstract conceptual
rstb.2017.0133) processing. (doi:10.17605/OSF.IO/P7A9U) knowledge. Cortex 63, 250–266. (doi:10.1016/j.
15. Paivio A. 1986 Mental representations: a dual coding 31. Granito C, Scorolli C, Borghi AM. 2015 Naming a cortex.2014.09.001)
approach. Oxford, UK: Oxford University Press. Lego world. The role of language in the acquisition 44. Wilson-Mendenhall CD, Simmons WK, Martin A,
16. Kousta ST, Vigliocco G, Vinson DP, Andrews M, Del of abstract concepts. PLoS ONE 10, e0114615. Barsalou LW. 2013 Contextual processing of abstract
Campo E. 2011 The representation of abstract (doi:10.1371/journal.pone.0114615) concepts reveals neural representations of
words: why emotion matters. J. Exp. Psychol. 140, 32. Fischer MH, Brugger P. 2011 When digits help nonlinguistic semantic content. J. Cogn. Neurosci.
14 –34. (doi:10.1037/a0021446) digits: spatial –numerical associations point to 25, 920 –935. (doi:10.1162/jocn_a_00361)
17. Gleitman LR, Cassidy K, Nappa R, Papafragou A, finger counting as prime example of embodied 45. Barsalou LW. 1999 Perceptions of perceptual
Trueswell JC. 2005 Hard words. Lang. Learn. Dev. 1, cognition. Front. Psychol. 2, 260. (doi:10.3389/fpsyg. symbols. Behav. Brain Sci. 22, 637–660. (doi:10.
23 –64. (doi:10.1207/s15473341lld0101_4) 2011.00260) 1017/S0140525X99532147)
18. Barsalou LW, Santos A, Simmons WK, Wilson CD. 33. Barca L, Mazzuca C, Borghi AM. 2017 Pacifier use 46. Barsalou LW, Wiemer-Hastings K. 2005 Situating
2008 Language and simulation in conceptual and conceptual relations of abstract and emotional abstract concepts. In Grounding cognition: the role of
processing. In Symbols, embodiment, and meaning concepts. Front. Psychol. 8, 2014. (doi:10.3389/ perception and action in memory, language, and
(eds M De Vega, AM Glenberg, AC Graesser), fpsyg.2017.02014) thought (eds D Pecher, R Zwaan), pp. 129–163.
pp. 245 –283. Oxford, UK: Oxford University Press. 34. Niedenthal PM, Augustinova M, Rychlowska M, New York, NY: Cambridge University Press.
19. Ponari M, Norbury CF, Vigliocco G. 2018 Acquisition Droit-Volet S, Zinner L, Knafo A, Brauer M. 2012 47. Harris PL, Koenig MA. 2006 Trust in testimony: how
of abstract concepts is influenced by emotional Negative relations between pacifier use and children learn about science and religion. Child Dev.
valence. Dev. Sci. 21, e12549. (doi:10.1111/desc. emotional competence. Basic Appl. Soc. Psychol. 34, 77, 505 –524. (doi:10.1111/j.1467-8624.2006.
12549) 387 –394. (doi:10.1080/01973533.2012.712019) 00886.x)
20. Dove G. 2009 Beyond perceptual symbols: a call for 35. Borghi AM, Lugli L. In preparation. Abstract 48. Wang X et al. 2017 Organizational principles of
representational pluralism. Cognition 110, 412 – concepts and mouth motor system: evidence with abstract words in the human brain. Cereb. Cortex
431. (doi:10.1016/j.cognition.2008.11.016) an interference paradigm. 23, 1 –14. (doi:10.1093/cercor/bhx283)
21. Dove G. 2011 On the need for embodied and 36. Johnson AJ, Jenks R, Miles C, Albert M, Cox M. 2011 49. Connell L, Lynott D, Banks B. 2018 Interoception:
disembodied cognition. Front. Psychol. 1, 242. Chewing gum moderates multi-task induced shifts the forgotten modality in perceptual grounding of
(doi:10.3389/fpsyg.2010.00242) in stress, mood, and alertness. A re-examination. abstract and concrete concepts. Phil. Trans. R. Soc. B
22. Newcombe PI, Campbell C, Siakaluk PD, Pexman Appetite 56, 408–411. (doi:10.1016/j.appet.2010. 373, 20170143. (doi:10.1098/rstb.2017.0143)
PM. 2012 Effects of emotional and sensorimotor 12.025) 50. Tillotson SM, Siakaluk PD, Pexman PM. 2008
knowledge in semantic processing of concrete and 37. Topolinski S, Maschmann IT, Pecher D, Winkielman Body–object interaction ratings for 1,618
abstract nouns. Front. Human Neurosci. 6, 275. P. 2014 Oral approach –avoidance: affective monosyllabic nouns. Behav. Res. Methods 40,
(doi:10.3389/fnhum.2012.00275) consequences of muscular articulation dynamics. 1075– 1078. (doi:10.3758/BRM.40.4.1075)
23. Recchia G, Jones MN. 2012 The semantic richness of J. Pers. Soc. Psychol. 106, 885. (doi:10.1037/ 51. Wauters LN, Tellings AE, Van Bon WH, Van Haaften
abstract concepts. Front. Human Neurosci. 6, 1– 12. a0036477) AW. 2003 Mode of acquisition of word meanings:
(doi:10.3389/fnhum.2012.00315) 38. Barca L, Cornelissen P, Simpson M, Urooj U, Woods the viability of a theoretical construct. Appl.
24. Clark A. 1998 Magic words: how language W, Ellis AW. 2011 The neural basis of the right Psycholing. 24, 385–406. (doi:10.1017/
augments human computation. In Language and visual field advantage in reading: a MEG analysis S0142716403000201)
thought: interdisciplinary themes (eds P Carruthers, using virtual electrodes. Brain Lang. 118, 53 –71. 52. Dreyer FR, Pulvermüller F. 2018 Abstract semantics
J Boucher), pp. 162–183. Cambridge, UK: (doi:10.1016/j.bandl.2010.09.003) in the motor system?–An event-related fMRI study
Cambridge University Press. 39. Lupyan G, Swingley D. 2012 Self-directed speech on passive reading of semantic word categories
25. Dove G. 2014 Thinking in words: language as an affects visual search performance. Q. J. Exp. Psychol. carrying abstract emotional and mental meaning.
embodied medium of thought. Top. Cogn. Sci. 6, 65, 1068–1085. (doi:10.1080/17470218.2011. Cortex 100, 52 –70. (doi:10.1016/j.cortex.2017.10.
371–389. (doi:10.1111/tops.12102) 647039) 021)
26. Vygotsky LS. 1986 Thought and language, revised 40. Lupyan G, Clark A. 2015 Words and the world 53. Sloman SA, Rabb N. 2016 Your understanding is my
edn. Cambridge, MA: MIT press. predictive coding and the language-perception- understanding: evidence for a community of
27. Topolinski S, Strack F. 2009 Motormouth: mere cognition interface. Curr. Dir. Psychol. Sci. 24, knowledge. Psychol. Sci. 27, 1451–1460. (doi:10.
exposure depends on stimulus-specific motor 279 –284. (doi:10.1177/0963721415570732) 1177/0956797616662271)