Language as a disruptive technology:

abstract concepts, embodiment and

rstb.royalsocietypublishing.org the flexible mind
Guy Dove
Department of Philosophy, University of Louisville, Louisville, KY, USA
Review
GD, 0000-0003-0470-7006
Cite this article: Dove G. 2018 Language as a
disruptive technology: abstract concepts, A growing body of evidence suggests that cognition is embodied and
grounded. Abstract concepts, though, remain a significant theoretical chal-
embodiment and the flexible mind. Phil.
lenge. A number of researchers have proposed that language makes an
Trans. R. Soc. B 373: 20170135.
important contribution to our capacity to acquire and employ concepts,
http://dx.doi.org/10.1098/rstb.2017.0135 particularly abstract ones. In this essay, I critically examine this suggestion
and ultimately defend a version of it. I argue that a successful account of
Accepted: 12 December 2017 how language augments cognition should emphasize its symbolic properties
and incorporate a view of embodiment that recognizes the flexible, multi-
modal and task-related nature of action, emotion and perception systems.
One contribution of 23 to a theme issue On this view, language is an ontogenetically disruptive cognitive technology
‘Varieties of abstract concepts: development, that expands our conceptual reach.
use and representation in the brain’. This article is part of the theme issue ‘Varieties of abstract concepts:
development, use and representation in the brain’.

Subject Areas:
neuroscience, cognition, behaviour
1. Introduction
Keywords: What is the role of the language system in embodied cognition? This paper offers a
concepts, embodied cognition, grounded theoretical framework for answering this pressing question. Building on Andy
cognition, language, semantic memory Clark’s suggestion that we are natural-born cyborgs [1], it proposes that language
can be thought of as a disruptive cognitive technology that transforms the embo-
died mind. Just as the adoption of new technologies often upends our social,
Author for correspondence:
cultural and economic lives, the acquisition of a natural language alters a child’s
Guy Dove cognitive purview. It disrupts embodied cognition by offering a new medium
e-mail: [email protected] through which to capture experience [2]. Experience with language leads to the
development of a distributed neural system able to manipulate linguistic symbols
in a compositional and productive fashion. The neurologically realized language
system amounts to a distributed action/perception control system that likely
relies on hierarchically organized network hubs. Linguistic forms themselves are
grounded because they involve actions, sights and sounds, but they are free to
capture content in a manner that is not tied to their grounding [3].
On this view, language is an external symbol system—one that has the
computational features associated with amodal symbol systems—that we learn
to manipulate in an embodied and grounded way. It is just one of the externally
sourced symbol technologies that we may acquire [4]. For example, learning how
to perform long division on paper requires a similar grounded manipulation of,
and interaction with, physical symbols [5,6]. The specialness of language has to do
with the pervasive role that it plays in our cognitive lives and the way in which it
complements embodied cognition by enhancing our capacity to encode infor-
mation about the world that goes beyond our immediate experience. This
proposal creates a number of predictions. First and foremost, it predicts that
much of our conceptual system is not grounded in language but is instead directly
grounded in action, emotion and perception systems. Importantly, such thinking
without words has its own compositionality and productivity [7,8]. Second, while
language is likely to contribute to all types of concepts, it is more likely to be help-
ful with abstract ones. Third, as a cognitive tool, the role of language should be
flexible, context-sensitive and experience-dependent. Finally, because a natural

& 2018 The Author(s) Published by the Royal Society. All rights reserved.
language is an acquired neuroenhancement, its influence concepts such as DEMOCRACY, FREEDOM, LEPTON, 2
should change over the course of development. NUMBER and TRUTH may simply represent one end of a spec-

rstb.royalsocietypublishing.org
The purpose of this essay is to outline and defend the dis- trum. Researchers often demarcate abstract concepts by one of
ruptive technology view. The argument proceeds at two several measures, including body–object interaction [20], concre-
levels: the general and the specific. While much of the essay teness [21], context-availability [22], emotional valence [23],
is aimed at the big picture and endeavours to show that the imageability [24], semantic richness [25] and strength of percep-
neuroenhancement view integrates and unifies seemingly tual experience [26]. Importantly, while these measures correlate
disparate threads of current research, the last section exam- to some extent, they are not equivalent [23]. Such divergence
ines the way in which it offers a promising explanation of a suggests that abstract concepts may form a heterogeneous
particular linguistic/conceptual phenomenon—metaphor. class. Indeed, a cogent argument can be made that researchers
Together these elements provide a compelling case for think- have been too cavalier in assuming that abstract concepts are
ing that language augments and extends the cognitive reach homogeneous [27]. In this paper, I shall not make this assump-

Phil. Trans. R. Soc. B 373: 20170135

of the embodied mind. tion. My argument will simply be that language has an
important role to play in concepts in general and abstract con-
cepts in particular. This role may ultimately contribute to a
2. Embodiment pluralistic account of abstract concepts.
When compared with concrete concepts, abstract concepts
The idea that our concepts are fundamentally embodied has
tend to refer to entities or events that are harder to perceive
gained a great deal of currency in the psychological and
with our senses or manipulate with our actions [28], to involve
brain sciences. Many hold that the neural mechanisms typi-
more complex relations, introspective features or social inter-
cally used to experience the world are also used to think
actions [7,9] and to exhibit greater variability across contexts
about it. By these lights, cognition involves the selective reuse
[29]. Evidence suggests that they may be processed in a differ-
of action, emotion and perception systems to carry out situated
ent manner from other concepts. For instance, abstract words in
simulations of our experience [9,10]. Because of their distal con-
a semantic categorization task are associated with a particu-
nection to experience, abstract concepts represent a particular
larly widespread pattern of cortical activation that includes
challenge for this approach [11]. In this essay, I explore and
temporal, parietal and frontal regions [30]. This distributed pat-
defend the notion that language provides an especially impor-
tern could be explained by the reliance of abstract concepts on a
tant scaffold for embodied concepts in general and abstract
network of association areas [31].
ones in particular.
All of this raises a difficult question: How can one capture
A diverse body of evidence supports the thesis that our con-
abstract content using grounded mechanisms? Certainly, one
cepts are embodied and grounded [9,10,12]. For example,
of the purported benefits of embodiment [7,32,33] is its ability
Pecher et al. [13] find a modality-switching cost associated
to overcome the symbol grounding problem [34]. This problem
with a property verification task. Participants verified verbally
arises because a system containing only abstract symbols and
expressed facts involving one modality (such as the fact that
their interrelations struggles to explain how individual rep-
leaves rustle) more rapidly after verifying a fact involving the
resentations come to be associated with objects and events in
same modality (such as the fact that blenders make noise) than
the world. Supporters of embodiment propose that this pro-
after verifying a fact involving a different modality (such as
blem is overcome by the experiential connections between the
the fact that cranberries are tart). Hearing motion-related verbs
representations of modality-specific sensorimotor systems and
interferes with visual motion processing [14] and visual
our external environment (both physical and social). This
motion processing interferes with the processing of motion-
benefit may come with a cost, however, because represen-
related verbs [15]. Neuroimaging data provide further evi-
tational systems containing only modality-specific symbols
dence of conceptual embodiment. Reading odour-related
face a corresponding symbol ungrounding problem [11]: that is,
words (e.g. cinnamon, garlic and jasmine) elicits increased acti-
any theory that posits a central role for experiential mechanisms
vation in the primary olfactory cortex relative to neutral control
in our concepts must explain how we are able to acquire and
words [16], and reading action words (e.g. lick, pick and kick)
understand concepts that go beyond our experience [35–37].
elicits increased activation in the cortical regions associated
with performing the relevant movements [17]. The specificity
of the modulated activity can be quite fine-grained. Right-
and left-handers exhibit increased activation in the premotor 4. The role of language
areas that are contralateral to their dominant hands on lexical
Although, much of the initial research implicating sensorimo-
decisions involving manual action verbs [18]. In addition, the
tor and affective systems has focused on concrete concepts,
degree to which expert hockey players comprehend hockey-
researchers have begun to investigate tasks involving abstract
action sentences better than controls correlates positively
ones. Evidence has come to light that implicates action
with activity in the left dorsal premotor cortex [19].
[38,39] and emotion [23] systems with the processing of these
concepts. As things stand, though, there is insufficient reason
to think that abstract concepts rely exclusively on affective
3. The trouble with abstract concepts and motor activations. A number of theories propose that the
All concepts involve abstraction. Horizontal generalization from language system, or at least our experience of language, plays
individual exemplars (e.g. specific dogs) to categories (e.g. the a significant role in our conceptual system. Examples include
category of dog) lies at the very heart of conceptualization. Ver- embodied conceptual combination theory (ECCo) [40],
tical generalization linking categories together creates conceptual language and situated simulation theory (LASS) [41], symbol
hierarchies (e.g. pugs are a type of dog and dogs are a type of interdependency theory [42] and word as social tool theory
animal). The ubiquity of abstraction suggests that abstract (WAT) [43]. Because I do not have the space to critically
evaluate each of these theories (for reviews, see [28,44]), my that differences in the modality of acquisition should lead to 3
strategy instead will be to examine the core generalizations differences in how concepts are neurobiologically realized.

rstb.royalsocietypublishing.org
behind them and then offer an overarching theory that inte- In particular, WAT theory predicts that abstract concepts
grates these generalizations and exhibits both explanatory should exhibit a greater tendency to engage language areas
and predictive power. [59,60]. The third tenet is that these neuroanatomical differ-
Embodied cognition posits an intimate link between ences should lead to differences in embodiment: namely, the
cognition and experience, and a great deal of our experience sensorimotor systems associated with speech production and
is with language itself. Some have proposed that this raises perception should be more engaged by abstract concepts [61].
the possibility of merging embodied and distributional Finally, because of their greater reliance on linguistic
approaches to word meaning [42,45–47]. Traditionally, these input, abstract concepts should exhibit greater cross-linguistic
approaches have been viewed as competitors; embodied variability than concrete ones.
accounts have focused on situated interactions with the world In sum, there are at least three distinct general conceptions

Phil. Trans. R. Soc. B 373: 20170135

and distributional accounts have focused on formal relation- in the literature of how language may augment our embodied
ships between symbols [39]. Distributional models treat cognitive abilities and help with abstract concepts. The first
concepts in terms of knowledge of statistical patterns derived focuses on our language-based experience as an additional
from spoken and written language. In addition to being particu- source of information about our physical and social worlds.
larly effective at capturing abstract concepts, they have enjoyed On this conception, implicit knowledge of distributional pat-
some success in explaining performance on both lexical access terns may scaffold certain cognitive activities. The second
and lexical similarity tasks [48]. However, because they focuses on the way in which linguistic forms can facilitate
depend on the statistical relationships between abstract and organize the neural implementation of our concepts.
symbols, they struggle to overcome the symbol grounding pro- On this conception, language transforms the very neural
blem [32]. Recognizing that linguistic and non-linguistic mechanisms responsible for cognition. The third focuses on
experiences can be treated as independent, yet complementary, the social dimension of language acquisition. On this concep-
sources of information about the world, several researchers tion, language leverages our intersubjective experience to
have proposed that these approaches can be combined expand our cognitive horizons. Below I offer an account
[42,45,47]. Indeed, there have been several demonstrations of how language augments cognition that combines and
that hybrid embodied/distributional models can outperform integrates these conceptions.
similar models that limit themselves to either embodied or dis-
tributional information alone [49–51]. Furthermore, several
behavioural studies identify independent language-based and
embodied factors in conceptual processing [41,42]. 5. A theoretical dilemma
Other theorists have explored the idea that linguistic forms The task before us is to provide a theoretical framework for
themselves might influence embodied conceptualization. In understanding the contribution of the language system to
one of the most detailed attempts to model the neurological our concepts. I suggest that previous accounts face something
mechanisms responsible for connecting the language and con- of a theoretical dilemma: they tend to be guilty of either sins
ceptual systems, Pulvermüller [52,53] proposes that linguistic of omission or sins of commission with respect to the funda-
forms play a constitutive role in the formation of action percep- mental properties of the language system. The relevant sins of
tion circuits. Learning a language, on this account, leads to the omission generally involve a failure to provide a rich enough
formation of these distributed circuits by means of Hebbian account of what makes language special. Too often embodied
and anti-Hebbian mechanisms. In other words, linguistic theories make little mention of the structural properties of the
forms serve as a means of stabilizing and organizing grounded language system and their connection to its ability to capture
representations. Lupyan & Bergen [54] similarly argue that semantic content. Sins of commission are often associated
language acts as a control system that, in their words, ‘pro- with fuller accounts of the language system, because they
grammes the mind’ by enabling the active manipulation of tend to involve, either explicitly or implicitly, amodal rep-
sensorimotor representations. This conception builds on pre- resentations that capture these all important structural
vious behavioural data demonstrating that verbal cues (such properties. Such amodal representations seem incompatible
as the spoken word dog) activate more general representations with the basic tenets of embodied cognition [62].
than non-verbal cues (such as the sound of a dog barking) I am going to adopt a twofold strategy in response to this
[55,56]. What unites the different instances of this second dilemma. The first part involves outlining an expanded
type of approach is the recognition that an important feedback notion of embodiment that emphasizes the flexible character
relationship may exist between linguistic forms and sensorimo- of the distributed representations employed in conceptual
tor simulations [57]. tasks. This expanded notion requires an embrace of what has
A third approach emphasizes the role that the social experi- become known as weak embodiment [63]. In strongly embo-
ence of language plays in shaping our concepts. The most died theories, sensorimotor systems are directly implicated in
prominent version of this approach is the WAT theory [27]. conceptual processing [64,65]. Completely disembodied theo-
This theory has four main tenets [28]. The first is that ries, on the other hand, locate concepts in amodal systems
language—broadly construed to include pragmatic and and view sensorimotor activations as epiphenomenal conse-
discourse-related elements—is likely to play a greater scaffold- quences of conceptual processing [37,66]. Weakly embodied
ing role in abstract concepts than in concrete ones. This tenet theories retain a commitment to the proposition that concep-
accords with psycholinguistic evidence on modality of acqui- tual representations are constituted at least in part by activity
sition indicating that the acquisition of abstract words tends in sensorimotor systems while granting that dynamically co-
to rely more on linguistic input and less on sensorimotor ordinated activity across multiple distributed regions is
experience than other words [58]. The second tenet is central to cognition. Such theories often include activity in
cross-modal areas such as convergence zones [67] or network warrant characterizing it as form of embodied cognition. 4
hubs [68,69]. This perspective should not be seen as completely Moreover, there is precedence for this ascription. Anderson

rstb.royalsocietypublishing.org
novel or radical. Indeed, several recent reviews [11,63,70] [85], for instance, articulates a radically interactive view of
suggest that weak embodiment has become the favoured neural reuse while explicitly remaining committed to embo-
view among supporters of embodied and grounded cognition. diment. Nevertheless, one may think that the dependence
The second part of my strategy involves developing a theoreti- of this approach on intermediate representations undermines
cal account of the way in which our experience with language the theoretical bite of embodiment [86]. My response to this
augments our concepts. The key idea will be that language not worry is similar to the one offered above: what ultimately
only provides access to new sources of information about the matters is getting the theory right. If the conjunction of the
world, but also transforms us as thinkers. flexible mind hypothesis and the appeal to the language
system amounts to an abandonment of authentic embodied
cognition, then so be it.

Phil. Trans. R. Soc. B 373: 20170135

6. The flexible mind
One of the challenges facing embodied cognition is that embodi-
ment means different things to different people. Broadly 7. Language as a disruptive technology
speaking, embodied approaches can be divided into one of Zwaan & Madden [87] famously use a pair of analogies to
two major categories: those that emphasize the influence of highlight the difference between traditional computational
the body on the mind and those that emphasize the importance views of cognition [88,89] and embodied ones. They liken the
of body–world couplings. What we need is a framework for computational mind to a bricklayer that assembles structures
understanding embodied cognition that integrates ideas about out of well-defined mental units and the embodied mind to a
embodiment that focus on body–mind connections with ideas beachcomber that builds structures out of whatever has
about embodiment that focus on body–world connections. I washed up on shore. While beachcombers may shape and
offer two core conjectures: (i) that adaptive neural reuse is a cen- modify what they find, much of the original character remains.
tral feature of the brain mechanisms responsible for cognition Situated sensorimotor simulations are similarly likely to pre-
and (ii) that manipulation of external resources often serves to serve aspects of their experiential origins. Without making
scaffold our cognitive endeavours. Rather than focusing exclu- too much of the analogy itself (after all even driftwood is a
sively on the contribution of sensory and motor areas, my structure composed of smaller parts), it is worth pointing out
approach focuses also on the context-sensitivity assumed by that one of the things that washes ashore is a collection of
many embodied accounts of concepts [71,72]. It fits well with bricks (i.e. language). A supporter of embodied cognition
the growing evidence that conceptual representations may thus faces a choice: either maintain that the language system
vary with stimulus [73,74], task [75–77] and context [78,79]. is completely separate from our conceptual system or provide
This framework identifies a number of characteristic fea- some explanation of how the two are integrated. Although
tures of the mechanisms responsible for our concepts. First, the first of these seems contrary to the interactive spirit of
they are fundamentally multimodal. Not only is the interplay embodied cognition, it has been the standard approach—
between modalities essential to how we perceive and act on words, phrases and sentences have been treated as mere elici-
the world, it is also important to how we conceptualize its con- tors of simulations. As we have seen, though, there has been
tents. Second, this interplay often depends on mechanisms some recent movement towards adopting the second option.
associated with the ongoing evaluation of incoming sensory What I propose is that language is a disruptive technology
input relative to the predictions generated by the motor that transforms the embodied mind. This idea is intended to fit
system [80]. Hard-won experiential knowledge plays an with, and build upon, earlier proposals. Vygotsky [90] proposes
important dynamical role in embodied simulations. Third, that internalized language can serve as scaffold for learning. On
the selective nature of embodied simulations requires a hier- his view, inner speech can help the child organize, plan and
archical neuroanatomical organization, both internal to, and remember actions [91]. Clark [92] emphasizes the degree to
across, specific modalities [70,81]. Finally, this approach also which language is a physical transformation of our ‘cognitive
holds that the degree and form of embodiment is likely to niche’ that extends the abilities of the embodied mind. The act
change over the course of development [82,83]. of labelling, for instance, may help learners become attuned to
There will be some that argue that these features have been perceptual commonalities and overcome the inherent complex-
part of their conception of embodiment all along. After all, ity and noisiness of perceptual inputs [93,94]. More broadly,
Barsalou [7] cites the ability to explain flexibility as a major language creates a novel set of perceptual objects and targets
benefit of his approach. Moreover, Connell & Lynott [84] for action. This enables us to model the world by means of the
contend that the dynamic influences that the body, the environ- manipulation of an external and shared symbol system.
ment, the relevant goals and the task have on our conceptual The suggestion on the table is that language augments
representations imply that ‘you can’t represent the same embodied cognition. Part of the impetus for this proposal is
concept twice’. Wilson & Golonka [71] propose that task- that the symbolic character of language—the fact that it is an
dependence is a central component of embodied cognition. externally derived symbol system that is both compositional
My intention is not to claim exclusive priority but merely to and productive—offers a number of potential cognitive
codify what I see as the best approach. benefits. One of these is the common absence of a direct con-
There will be others, though, who claim that the view out- nection between linguistic representations and their referents.
lined in this essay amounts to a disavowal of embodied This semantic arbitrariness may help them anchor embodied
cognition [62]. While I disagree with this assessment, not and grounded knowledge. Giving voice to this idea, Zwaan
much hangs on this. The central role played by situated sen- [84] hypothesizes that distributed linguistic representations
sorimotor simulations in this account seems sufficient to serve as symbolic placeholders for multimodal simulations.
An additional cognitive benefit may arise from the fact that phenomenon in which language appears to scaffold our cogni- 5
linguistic symbols are syntactically re-combinable in a way tive efforts. Metaphor is important for the purposes of this

rstb.royalsocietypublishing.org
that is independent of the combinatorial properties of essay not only because it may play a role in the acquisition of
non-linguistic embodied and grounded cognition. This inde- some abstract concepts, but also because it serves as a useful
pendent structural flexibility may make it easier to generate test case for the disruptive technology approach.
new thoughts and encode unexpected connections between Metaphor has traditionally been seen as both a source of
thoughts [95,96]. Developing this idea further, Lynott & evidence for embodiment and a potential means of solving
Connell [97] propose that conceptual combination arises from the problem abstract concepts. Working from observations
the interaction between the linguistic and simulation systems. concerning language use, cognitive linguists have shown that
Some researchers who acknowledge that linguistic represen- a great deal of our discourse is organized around experiential
tations have a role to play in conceptual tasks suggest that this metaphors [101,102]. Several have proposed that we rely on
role is typically more superficial and less effective than that of embodied conceptual metaphors to understand abstract con-

Phil. Trans. R. Soc. B 373: 20170135

other multimodal simulations [41,84,98]. In particular, linguistic cepts [103–105]. Typically this is thought to depend on
symbols are seen as a quick and dirty heuristic that can be used mappings from grounded conceptual domains to abstract
when conditions do not require complex task performance. Of ones. Perhaps the most well-attested embodied metaphor
course, the notion of effectiveness is itself notoriously context- involves understanding the passage of time as a motion follow-
sensitive. Indeed, there are at least three problems with the ing a linear path along the back-to-front axis [106,107] or along
dismissive assessment offered by these theorists: First, as the left-to-right axis [108–110]. Evidence of other embodied
symbolic placeholders, linguistic representations may be par- metaphors—such as understanding morality in terms of clean-
ticularly effective at resolving the problem of generalization. liness [111], power in terms of verticality [112,113] and
Recall the studies by Lupyan and co-workers [54–57] demon- similarity in terms of closeness [114,115]—has also been found.
strating that verbal cues are more effective at eliciting general Viewing language as a neuroenhancement predicts that the
representations than modality-specific cues. Second, linguistic cognitive scaffolding due to conceptual metaphor will be
representations are associated with external symbols and are somewhat circumscribed, because it treats metaphor as just
thus able to leverage the social character of language. Philoso- one of the ways in which language can augment cognition.
phers of language emphasize the degree to which the This is supported by developmental psycholinguistic research
linguistic function of labels depends in part on their ability to indicating that abstract concepts are part of the vocabularies of
track referents by means of socially determined causal links very young children but metaphors are not [36,116,117].
[99,100]. Abstract concepts would seem to be particularly Indeed, children’s comprehension of metaphor appears to
good candidates for this sort of reliance on external support. remain poor until they reach 8–10 years of age [118]. Further-
Third, given the fact that abstract concepts often involve com- more, there are also groups of people, such as high functioning
plex situations and relational properties, it is far from clear individuals with an autism spectrum disorder, that acquire
that linguistic representations are eliminable. Although abstract concepts despite experiencing pronounced difficulties
researchers have found some surprising evidence that embo- with metaphors [119].
died and grounded representations are activated with abstract According to the proposal under consideration, our use of
concepts during certain tasks, this does not demonstrate that lin- metaphoric simulations should be context-sensitive and task-
guistic representations are uninvolved. Furthermore, reviews of specific. Some available neural evidence supports this predic-
functional brain imaging research implicate language-related tion. A number of studies have found that metaphors and
areas of the cortex in the processing of abstract words [59,60]. abstract concepts elicit different patterns of activation [120–
Finally, there is reason to think that linguistic representations 123]. In an event-related potential (ERP) study [124], partici-
may be needed for concepts that directly involve language use pants made upward or downward movements with marbles
(e.g. ASSERT, CAJOLE and PROMISE). as they read words that had literal (ascend and descend) or
Few would deny that language provides a means to gain metaphorical (inspire and defeat) vertical associations. Con-
information about objects and events in the absence of direct gruency effects were found with both types of words when
experience, and most recognize that language enables us to the associations matched the direction of the movements, but
leverage the knowledge of others. Much of the impetus for their time signatures were different: the effects emerged at
pluralistic embodied approaches that incorporate the language 200–300 ms after word onset with the literal movement
system is the idea that language itself can be a rich source of words but after 500 ms with the metaphoric movement
information about our physical and social environment. words. The delay with the metaphors suggests that the relevant
What distinguishes the current proposal from others is the sensorimotor simulations are not automatically engaged in the
explicit claim that the structural properties of language are same the way that they are with the literal action words. In
central to its ability to augment cognition. keeping with the context-sensitivity and task-specificity
found generally in embodied concepts [72], attention appears
to influence congruency effects between affective evaluation
and vertical space [125]. Boroditsky & Ramscar [107] find
8. A case study that people at an airport who are about to fly out or who
Thus far, I have defended the proposal that language augments have just arrived tend to employ an ego-moving perspective
cognition by outlining its broad theoretical promise and pointing on time (to think of themselves moving through time) when
to its success at integrating other current views. This proposal is answering questions about moving temporal events ‘forward’
intended to provide an overarching framework for understand- while those who are just waiting to pick someone up tend to
ing the role of language in our concepts. Given this, one might employ a time-moving perspective.
wonder whether or not it has any predictive bite. In this section, An additional prediction associated with the disruptive
I hope to demonstrate that it does by examining a particular technology view—one that is not typically associated with
embodied accounts of metaphor—is that some uses of metaphor supported by a recent experiment: after introducing a novel 6
may not engage sensorimotor simulations but rather depend on metaphor connecting time and weight (the past is heavy and

rstb.royalsocietypublishing.org
linguistic associations. This prediction fits with the hypothesis the present is light), congruency effects were found in the
that some metaphors may undergo a gradual process of conven- weight judgements of books that appeared new or old [133].
tionalization as they become more familiar [126]. Desai et al. It is also supported by the fact that providing participants
[127] examine the brain activation elicited by four types of sen- with a brief exposure to mirror-reversed orthography can
tences: literal action sentences (The instructor is grasping the reverse the orientation of the congruency effects on temporal
steering wheel very tightly.), non-idiomatic metaphor sentences judgements associated with a particular language [134]. Some
(The congress is grasping the state of affairs.), idiomatic meta- recent evidence also suggests that language-specific metaphors
phor sentences (The congress is grasping at straws in the may build upon preexisting non-linguistic embodied map-
crisis.) and abstract sentences (The congress is causing a big pings. Whereas speakers of Dutch tend to talk of musical
trade deficit again.). They found that higher-level sensorimotor pitch in terms of height (the way that we do in English), speak-

Phil. Trans. R. Soc. B 373: 20170135

regions associated with the described actions showed increased ers of Farsi tend to talk of it in terms of thickness [135]. These
activation with both the action and the non-idiomatic metaphor different linguistic metaphors appear to influence how Dutch
sentences but not with the idiomatic metaphor or abstract sen- and Farsi speakers reproduce recently heard musical pitches
tences. This suggests that sensorimotor simulations are not in the presence of irrelevant spatial information involving
essential for the semantic processing of these idioms and fits either height or thickness [136]. A follow-up study finds that
with the notion that they are ‘frozen’ metaphors whose content prelinguistic infants are sensitive to both the pitch–height and
is stored by means of linguistic associations. the pitch–thickness mappings [137].
One potential objection to the disruptive technology view is Metaphor is often taken as just another data point in the
that it seems to require a commitment to linguistic relativism larger case for embodied cognition. Treating it as the outcome
(the idea that the natural language one possesses influences of the interaction between an inherently flexible embodied
the thoughts one is likely to have). While I do not have the cognitive system and an internalized language system
space to fully address this concern, some remarks seem war- enables us to go beyond the observation that some metaphors
ranted. First, a growing body of evidence supports at least a engage action, emotion and perception systems. In particular,
weak form linguistic relativism [128–130]. Second, given the it predicts that metaphor should emerge gradually in devel-
centrality of non-linguistic embodied simulations to the current opment; be circumscribed in scope, context-sensitive and
account, its commitment to relativism need not be full-throated. task-specific; and involve both sensorimotor simulations of
Third, even critics of linguistic relativism often grant that bodily experience and linguistic associations.
language influences thinking-for-speaking [131]. On the current
proposal, though, the distinction between thinking and
thinking-for-speaking is blurred [2]. In sum, a compelling argu-
ment can be made that the approach advocated in this essay 9. Conclusion
strikes the right balance with respect to the influence of Abstract concepts represent a significant challenge for em-
language on thought. Moreover, research on embodied meta- bodied cognition. The notion that language might help
phors provides support for this generalization. For instance, grounded agents acquire and use concepts in general, and
some behavioural studies implicate language-specific meta- abstract concepts in particular, has recently gained traction.
phors. Examples can be found in studies examining the time- Extant accounts, though, tend to commit one of two errors:
as-space metaphor. Research indicates that speakers of Spanish they either treat language as just another experiential source
tend to conceptualize time from left to right [110,132]. Hypothe- of information or offer a conception of the language system
sizing that this might be due in part to the orientation of their that is incompatible with embodiment and grounding. This
writing system, Ouellet et al. [109] examine the responses of essay provides an account of how language scaffolds the
speakers of Hebrew (which is read from right to left) and speak- embodied mind in which the symbolic character of language
ers of Spanish to words presented auditorily in a temporal (underwritten by its combinatorial structural properties) is a
judgement task. Speakers of Spanish responded quicker when feature not a bug. The acquisition of a natural language not
responding to words associated with the past with their left only expands our access to information about the world,
hand and words associated with the future with their right but also serves a neuroenhancement by providing a new
hand while speakers of Hebrew exhibited the opposite pattern. medium of embodied thought.
While the approach advocated here predicts that experience
with particular natural languages should result in differences in Data accessibility. This article has no additional data.
embodied metaphors, it also predicts that embodied metaphors Competing interests. I declare I have no competing interests.
should be somewhat flexible and experience-dependent. This is Funding. I received no funding for this study.

References
1. Clark A. 2003 Natural-born cyborgs. New York, NY: 3. Dove G. 2011 On the need for embodied and dis- 5. Landy D, Allen C, Zednik C. 2014 A perceptual
Oxford University Press. embodied cognition. Front. Psychol. 1, 242. (doi:10. account of symbolic reasoning. Front. Psychol. 5,
2. Dove G. 2014 Thinking in words: language 3389/fpsyg.2010.00242) 275. (doi:10.3389/fpsyg.2014.00275)
as an embodied medium of thought. 4. Clark A. 2011 Supersizing the mind: embodiment, 6. Landy D, Goldstone RL. 2007 How abstract is symbolic
Topics Cogn. Sci. 6, 371–389. (doi:10.1111/ action, and cognitive extension. New York, NY: thought? J. Exp. Psychol. Learn. Mem. Cogn. 33,
tops.12102) Oxford University Press. 720–733. (doi:10.1037/0278-7393.33.4.720)
7. Barsalou L. 1999 Perceptual symbol systems. Behav. Verb. Behav. 16, 217– 231. (doi:10.1016/S0022- language modulates motor system activity. Q. 7
Brain Sci. 22, 577 –609. (doi:10.1017/S0140525X 5371(77)80048-0) J. Exp. Psychol. 61, 905–919. (doi:10.1080/

rstb.royalsocietypublishing.org
9002149) 22. Schwanenflugel PJ, Shoben E. 1983 Differential 17470210701625550)
8. Barsalou LW, Prinz JJ. 1997 Mundane creativity in context effects in the comprehension of abstract and 39. Guan CQ, Meng W, Yao R, Glenberg AM. 2013
perceptual symbol systems. In Creative thought: an concrete verbal materials. J. Exp. Psychol. 9, 82 – The motor system contributes to comprehension of
investigation of conceptual structures and processes 102. (doi:10.1037/0278-7393.9.1.82) abstract language. PLoS ONE 8, e75183. (doi:10.
(eds TB Ward, SM Smith, J Vaid), pp. 267– 307. 23. Kousta S-T, Vigliocco G, Vinson DP, Andrews M, Del 1371/journal.pone.0075183)
Washington, DC: American Psychological Campo E. 2011 The representation of abstract 40. Lynott D, Connell L. 2010 Embodied conceptual
Association. words: why emotion matters. J. Exp. Psychol. 140, combination. Front. Psychol. 1, 212. (doi:10.3389/
9. Barsalou LW. 2008 Grounded cognition. Annu. Rev. 14 –34. (doi:10.1037/a0021446) fpsyg.2010.00212)
Psychol. 59, 617 –645. (doi:10.1146/annurev.psych. 24. Paivio A. 1986 Mental representations: a dual coding 41. Barsalou LW, Santos A, Simmons KW, Wilson CD.
59.103006.093639) approach. Oxford, UK: Oxford University Press. 2008 Language and simulations in conceptual

Phil. Trans. R. Soc. B 373: 20170135

10. Fischer MH, Zwaan RA. 2008 Embodied language: a 25. Recchia G, Jones MN. 2012 The semantic richness of processing. In Symbols, embodiment and meaning
review of the motor system in language abstract concepts. Front. Human Neurosci. 6, 315. (eds M De Vega, AM Glenberg, AC Graesser),
comprehension. Q. J. Exp. Psychol. 6, 825– 850. (doi:10.3389/fnhum.2012.00315) pp. 245 – 283. Oxford, UK: Oxford University
(doi:10.1080/17470210701623605) 26. Connell L, Lynott D. 2012 Strength of perceptual Press.
11. Dove G. 2016 Three symbol ungrounding problems: experience predicts word processing performance 42. Louwerse MM, Jeuniaux P. 2010 The linguistic and
abstract concepts and the future of embodied better than concreteness or imageability. Cognition embodied nature of conceptual processing.
cognition. Psychon. Bull. Rev. 23, 1109 –1121. 125, 452–465. (doi:10.1016/j.cognition.2012.07.010) Cognition 114, 96 –104. (doi:10.1016/j.cognition.
(doi:10.3758/s13423-015-0825-4) 27. Borghi A, Binkofski F. 2014 Words as social tools: an 2009.09.002)
12. Kemmerer D. 2010 How words capture visual embodied view on abstract concepts. New York, NY: 43. Borghi AM, Cimatti F. 2009 Words as tools and the
experience: the perspective from cognitive Springer. problem of abstract words meanings. In Proc. 31st
neuroscience. In Words and the world: how words 28. Borghi AM, Binkofski F, Castelfranchi C, Cimatti F, Ann. Conf. Cogn. Sci. Soc. (eds N Taatgen, H van
capture human experience (eds B Malt, P Wolff ), Scorolli C, Tummolini L. 2017 The challenge of Rijn), pp. 2304–2309. Amsterdam, The
pp. 289 –329. New York, NY: Oxford University abstract concepts. Psychol. Bull. 143, 263 –292. Netherlands: Cognitive Science Society.
Press. (doi:10.1037/bul0000089) 44. Pecher D, Boot I, van Dantzig S. 2011 Abstract
13. Pecher D, Zeelenberg R, Barsalou L. 2003 Verifying 29. Barsalou L. 1987 The instability of graded structure: concepts: sensory motor grounding, metaphors, and
properties from different modalities for concepts implications for the nature of concepts. In Concepts beyond. In The psychology of learning and
produces switching costs. Psychol. Sci. 14, 119 – and conceptual development: ecological and motivation, vol. 54 (ed. B Ross), pp. 217–248.
124. (doi:10.1111/1467-9280.t01-1-01429) intellectual factors in categorization (ed. U Neisser), Burlington, NJ: Academic Press.
14. Meteyard L, Bahrami B, Vigliocco G. 2007 Motion pp. 101 –140. Cambridge, UK: Cambridge University 45. Andrews M, Frank S, Vigliocco G. 2014 Reconciling
detection and motion verbs. Language affects low- Press. embodied and distributional accounts of meaning in
level visual perception. Psychol. Sci. 18, 1007 – 30. Pexman PM, Hargreaves IS, Edwards JD, Henry LC, language. Topics Cogn. Sci. 6, 359–370. (doi:10.
1013. (doi:10.1111/j.1467-9280.2007.02016.x) Goodyear BG. 2007 Neural correlates of concreteness 1111/tops.12096)
15. Meteyard L, Zokaei N, Bahrami B, Vigliocco G. 2008 in semantic categorization. J. Cogn. Neurosci. 19, 46. Johns BT, Jones MN. 2012 Perceptual inference
Visual motion interferes with lexical decision on 1407 –1419. (doi:10.1162/jocn.2007.19.8.1407) from global lexical similarity. Topics Cogn. Sci.
motion verbs. Curr. Biol. 18, R732 –R733. (doi:10. 31. Fernandino L, Binder JR, Desai RH, Pendl SL, 4, 103 –120. (doi:10.1111/j.1756-8765.2011.
1016/j.cub.2008.07.016) Humphries CJ, Gross WL, Conant LL, Seidenberg MS. 01176.x)
16. Gonzalez J, Barros-Loscertales A, Pulvermuller F, 2015 Concept representation reflects multimodal 47. Riordan B, Jones MN. 2010 Redundancy in linguistic
Meseguer V, Sanjuan A, Belloch V, Avila C. 2006 abstraction: a framework for embodied semantics. and perceptual experience: comparing distributional
Reading cinnamon activates olfactory brain regions. Cereb. Cortex 26, 2018– 2034. (doi:10.1093/cercor/ and feature-based models of semantic
Neuroimage 32, 906 –912. (doi:10.1016/j. bhv020) representation. Topics Cogn. Sci. 3, 303–345.
neuroimage.2006.03.037) 32. Glenberg AM, Robertson DA. 2000 Symbol (doi:10.1111/j.1756-8765.2010.01111.x)
17. Hauk O, Johnsrtude I, Pulvermüller F. 2004 grounding and meaning: a comparison of high- 48. Jones MN, Dye M, Johns BT. 2017 Context as an
Somatopic representation of action words in human dimensional and embodied theories of meaning. organizing principle of the lexicon. In The
motor and premotor cortex. Neuron 41, 301–307. J. Mem. Lang. 43, 379 –401. (doi:10.1006/jmla. psychology of learning and motivation, vol. 67 (ed. B
(doi:10.1016/S0896-6273(03)00838-9) 2000.2714) Ross), pp. 239– 283. Amsterdam, The Netherlands:
18. Willems RM, Hagoort P, Casasanto D. 2010 Body- 33. Prinz JJ. 2002 Furnishing the mind: concepts and Elsevier.
specific representation of action words: neural their perceptual basis. Cambridge, MA: MIT Press. 49. Andrews M, Vigliocco G, Vinson DP. 2009
evidence from right- and left-handers. Psychol. Sci. 34. Harnad S. 1990 The symbol grounding problem. Integrating experiential and distributional data to
21, 67 –74. (doi:10.1177/0956797609354072) Physica D 42, 335 –346. (doi:10.1016/0167- learn semantic representations. Psychol. Rev. 116,
19. Beilock SL, Lyons IM, Mattarella-Micke A, Nusbaum 2789(90)90087-6) 463–498. (doi:10.1037/a0016261)
HC, Small SL. 2008 Sports experience changes the 35. Chatterjee A. 2010 Disembodying cognition. Lang. 50. Bruni E, Tran GB, Baroni M. 2011 Distributional
neural processing of action language. Proc. Natl Cogn. 2, 79 –116. (doi:10.1515/langcog.2010.004) semantics from texts and images. In Proc. EMNLP
Acad. Sci. USA 105, 13 269 –13 273. (doi:10.1073/ 36. Dove G. 2009 Beyond perceptual symbols: a call for 2011 Geometrical Models for Natural Langauge
pnas.0803424105) representational pluralism. Cognition 110, 412– Semantics (GEMS 2011) Workshop (eds S Pado, Y
20. Siakaluk P, Pexman P, Sears C, Wilson K, Locheed K, 431. (doi:10.1016/j.cognition.2008.11.016) Peirsman), pp. 22 –32. Stroudsburg, PA: Association
Owen W. 2008 The benefits of sensorimotor 37. Mahon BZ, Caramazza A. 2008 A critical look at the for Computational Linguistics.
knowledge: body– object interaction facilitates embodied cognition hypothesis and a new proposal 51. Steyvers M. 2010 Combining feature norms and text
semantic processing. Cogn. Sci. 32, 591– 605. for grounding conceptual content. J. Physiol. Paris data with topic models. Acta Psychol. 133,
(doi:10.1080/03640210802035399) 102, 59 –70. (doi:10.1016/j.jphysparis.2008.03.004) 234–243. (doi:10.1016/j.actpsy.2009.10.010)
21. Marschark M, Paivio A. 1977 Integrative processing 38. Glenberg AM, Sato M, Cattaneo L, Riggio L, 52. Pulvermüller F. 2013 How neurons make meaning:
of concrete and abstract sentences. J. Verb. Learn. Palumbo D, Buccino G. 2008 Processing abstract brain mechanisms for embodied and abstract-
 symbolic semantics. Trends Cogn. Sci. 17, 458–470. 69. van den Heuvel MP, Sporns O. 2013 Network hubs 87. Zwaan RA, Madden CJ. 2005 Embodied sentence 8
(doi:10.1016/j.tics.2013.06.004) in the human brain. Trends Cogn. Sci. 17, 683– 696. comprehension. In Grounding cognition: the role of

rstb.royalsocietypublishing.org
53. Pulvermüller F. 2017 Neural reuse of action (doi:10.1016/j.tics.2013.09.012) perception and action in memory, language, and
perception circuits for language, concepts, and 70. Barsalou LW. 2016 On staying grounded and thought (eds D Pecher, R Zwaan), pp. 224–245.
communication. Prog. Neurobiol. 160, 1–44. avoiding quixotic dead ends. Psychon. Bull. Rev. 23, New York, NY: Cambridge University Press.
(doi:10.1016/j.pneurobio.2017.07.001) 1122 –1142. (doi:10.3758/s13423-016-1028-3) 88. Fodor JA. 1975 The language of thought.
54. Lupyan G, Bergen B. 2016 How language programs 71. Wilson AD, Golonka S. 2013 Embodied cognition is Cambridge, MA: Harvard University Press.
the mind. Topics Cogn. Sci. 8, 408 –424. not what you think it is. Front. Psychol. 4, 58. 89. Pylyshyn ZW. 1989 Computation and cognition:
55. Edmiston P, Lupyan G. 2015 What makes words (doi:10.3389/fpsyg.2013.00058) toward a foundation for cognitive science.
special? Words as unmotivated cues. Cognition 143, 72. Yee E, Thompson-Schill SL. 2016 Putting concepts Cambridge, MA: MIT Press.
93 –100. (doi:10.1016/j.cognition.2015.06.008) into context. Psychon. Bull. Rev. 23, 1015–1027. 90. Vygotsky L. 1986 Thought and language.
56. Lupyan G, Thompson-Schill SL. 2012 The evocative (doi:10.3758/s13423-015-0948-7) Cambridge, MA: MIT Press.

Phil. Trans. R. Soc. B 373: 20170135

power of words: activation of concepts by verbal 73. Sidhu DM, Kwan R, Pexman PM, Siakaluk PD. 2014 91. Alderson-Day B, Fernyhough C. 2015 Inner speech:
and nonverbal means. J. Exp. Psychol. 141, Effects of relative embodiment in lexical and development, cognitive functions, phenomenology,
170–186. (doi:10.1037/a0024904) semantic processing of verbs. Acta Psychol. 149, and neurobiology. Psychol. Bull. 141, 931 –965.
57. Lupyan G. 2012 Linguistically modulated perception 32 –39. (doi:10.1016/j.actpsy.2014.02.009) (doi:10.1037/bul0000021)
and cognition: the label feedback hypothesis. Front. 74. Taikh A, Hargreaves IS, Yap M, Pexman PM. 2015 92. Clark A. 2006 Language, embodiment, and the
Cogn. 3, 54. (doi:10.3389/fpsyg.2012.00054) Semantic classification of pictures and words. cognitive niche. Trends Cogn. Sci. 10, 370–374.
58. Wauters L, Tellings A, Van Bon W, Van Haaften A. Q. J. Exp. Psychol. 68, 1502–1518. (doi:10.1080/ (doi:10.1016/j.tics.2006.06.012)
2003 Mode of acquisition of word meanings: 17470218.2014.975728) 93. Clark A. 1998 Magic words: how language
the viability of a theoretical construct. Appl. 75. Connell L, Lynott D. 2014 I see/hear what you augments human computation. In Language and
Psycholing. 24, 385– 406. (doi:10.1017/S01427 mean: semantic activation in visual word thought: interdisciplinary themes (eds P Carruthers,
16403000201) recognition depends on perceptual attention. J Boucher), pp. 162 –183. Cambridge, UK:
59. Binder JR., Desai RH, Graves WW, Conant LL. 2009 J. Exp. Psychol. 143, 527–533. (doi:10.1037/ Cambridge University Press.
Where is the semantic system? A critical review and a0034626) 94. Lupyan G, Clark A. 2015 Words and the world:
meta-analysis of 120 functional neuroimaging 76. Pexman PM, Hargreaves IS, Siakaluk PD, Bodner GE, predictive coding and the language-perception-
studies. Cereb. Cortex 19, 2767– 2796. (doi:10.1093/ Pope J. 2008 There are many ways to be rich: cognition interface. Curr. Dir. Psychol. Sci. 24,
cercor/bhp055) effects of three measures of semantic richness on 279–284. (doi:10.1177/0963721415570732)
60. Wang J, Conder JA, Blitzer DN, Shinkareva SV. 2010 visual word recognition. Psychon. Bull. Rev. 15, 95. Camp E. 2009 Putting thoughts to work: concepts,
Neural representation of abstract and concrete 161 –167. (doi:10.3758/PBR.15.1.161) systematicity, and stimulus-independence. Phil.
concepts: a meta-analysis of imaging studies. Hum. 77. Sidhu DM, Heard A, Pexman PM. 2016 Is more Phenomenol. Res. 78, 275 –311. (doi:10.1111/j.
Brain Mapp. 31, 1459 –1468. (doi:10.1002/hbm. always better for verbs? Semantic richness effects 1933-1592.2009.00245.x)
20950) and verb meaning. Front. Psychol. 7, 798. (doi:10. 96. Weiskopf D. 2010 Embodied cognition and linguistic
61. Borghi AM, Zarcone E. 2016 Grounding abstractness: 3389/fpsyg.2016.00798) comprehension. Stud. Hist. Philos. Sci. 41, 294–
abstract concepts and the activation of the mouth. 78. Lebois LAM, Wilson-Mendenhall CD, Barsalou LW. 304. (doi:10.1016/j.shpsa.2010.07.005)
Front. Psychol. 7, 1498. (doi:10.3389/fpsyg.2016. 2015 Putting everything in context. Cogn. Sci. 39, 97. Lynott D, Connell L. 2010 Embodied conceptual
01498) 1987 –1995. (doi:10.1111/cogs.12295) combination. Front. Psychol. 1, 1– 14. (doi:10.3389/
62. Arbib MA, Gasser B, Barres V. 2014 Language is 79. Yee E, Ahmed S, Thompson-Schill SL. 2012 Colorless fpsyg.2010.00212)
handy but is it embodied? Neuropsychologia 55, green ideas (can) prime furiously. Psychol. Sci. 23, 98. Connell L, Lynott D. 2013 Flexible and fast:
57 –70. (doi:10.1016/j.neuropsychologia.2013.11. 364 –369. (doi:10.1177/0956797 611430691) linguistic shortcut affects both shallow and deep
004) 80. Pickering MJ, Andy C. 2014 Getting ahead: forward conceptual processing. Psychon. Bull. Rev. 20, 542–
63. Meteyard L, Cuadrado SR, Bahrami B, Vigliocco G. models and their place in cognitive architecture. 550. (doi:10.3758/s13423-012-0368-x)
2012 Coming of age: a review of embodiment and Trends Cogn. Sci. 18, 451 –456. (doi:10.1016/j.tics. 99. Saul K. 1980 Naming and necessity. Cambridge, MA:
the neuroscience of semantics. Cortex 48, 788–804. 2014.05.006) Harvard University Press.
(doi:10.1016/j.cortex.2010.11.002) 81. Binder JR. 2016 In defense of abstract conceptual 100. Putnam H. 1975 The meaning of meaning. In Mind,
64. Gallese V, Lakoff G. 2005 The brain’s concepts: the representations. Psychon. Bull. Rev. 23, 1096–1108. language and reality; philosophical papers, vol. 2,
role of the sensory-motor system in conceptual (doi:10.3758/s13423-015-0909-1) pp. 215–271. Cambridge, UK: Cambridge University
knowledge. Cogn. Neuropsychol. 22, 455–479. 82. Kontra C, Goldin-Meadow S, Beilock SL. 2012 Press.
(doi:10.1080/02643290442000310) Embodied learning across the life span. Topics Cogn. 101. Lakoff G, Johnson M. 1980 Metaphors we live by.
65. Glenberg AM. 2015 Few believe the world is flat: Sci. 4, 731 –739. (doi:10.1111/j.1756-8765.2012. Chicago, IL: University of Chicago Press.
how embodiment is changing the scientific 01221.x) 102. Gibbs RWJ. 2006 Embodiment and cognitive science.
understanding of cognition. Can. J. Exp. Psychol. 69, 83. Pexman PM. In press. The role of embodiment in Cambridge, UK: Cambridge University Press.
165–171. (doi:10.1037/cep0000056) conceptual development. Lang. Cogn. Neurosci. 103. Lakoff G, Johnson M. 1999 Philosophy in the flesh:
66. Mahon BZ. 2015 What is embodied about (doi:10.1080/23273798.2017.1303522) the embodied mind and its challenge to western
cognition? Lang. Cogn. Neurosci. 30, 420 –429. 84. Connell L, Lynott D. 2014 Principles of thought. New York, NY: Basic Books.
(doi:10.1080/23273798.2014.987791) representation: why you can’t represent the same 104. Lakoff G. 2014 Mapping the brain’s metaphor
67. Damasio A.R. 1989 The brain binds entities and concept twice. Topics Cogn. Sci. 6, 390–406. circuitry: metaphorical thought in everyday reason.
events by multiregional activation from convergence (doi:10.1111/tops.12097) Front. Human Neurosci. 8, 958. (doi:10.3389/fnhum.
zones. Neural Comput. 1, 123 –132. (doi:10.1162/ 85. Anderson ML. 2014 After phrenology: neural reuse 2014.00958)
neco.1989.1.1.123) and the interactive brain. Cambridge, MA: MIT Press. 105. Williams LE, Huang JY, Bargh JA. 2009 The
68. Sporns O, Honey CJ, Kotter R. 2007 Identification 86. Dove G. 2013 Intermediate representations exclude scaffolded mind: higher mental processes grounded
and classification of hubs in brain networks. PLoS embodiment. Behav. Brain Sci. 36, 353 –354. in early experience of the physical world. Eur. J. Soc.
ONE 2, e1049. (doi:10.1371/journal.pone.0001049) (doi:10.1017/S0140525X12002555) Psychol. 39, 1257 –1267. (doi:10.1002/ejsp.665)
106. Boroditsky L. 2000 Metaphoric structuring: 118. Winner E, Rosenstiel AK, Gardner H. 1976 The Neuroimage 83, 862– 869. (doi:10.1016/j. 9
understanding time through spatial metaphors. development of metaphoric understanding. Dev. neuroimage.2013.07.044)

rstb.royalsocietypublishing.org
Cognition 36, 1047 –1056. Psychol. 12, 289–297. (doi:10.1037/0012-1649.12. 128. Casasanto D. 2008 Who’s afraid of the big bad
107. Boroditsky L, Ramscar M. 2002 The roles of body 4.289) whorf? Crosslinguistic differences in temporal
and mind in abstract thought. Psychol. Sci. 13, 119. MacKay G, Shaw A. 2004 A comparative study of language and thought. Lang. Learn. 58(Suppl. 1),
185–188. (doi:10.1111/1467-9280.00434) figurative language in children with autistic 63– 79. (doi:10.1111/j.1467-9922.2008.00462.x)
108. Casasanto D, Boroditsky L. 2008 Time in the mind: spectrum disorders. Child Lang. Teach. Therapy 20, 129. Gentner D, Goldin-Meadow S (eds). 2003 Language
using space to think about time. Cognition 106, 13 –32. (doi:10.1191/0265659004ct261oa) in mind: advances in the study of language and
579–593. (doi:10.1016/j.cognition.2007.03.004) 120. Boulenger V, Shtyrov Y, Pulvermüller F. 2012 When thought, pp. 157 –192. Cambridge, MA: MIT Press.
109. Ouellet M, Santiago J, Israeli Z, Gabay S. 2010 Is the do you grasp the idea? MEG evidence for 130. Lupyan G. 2012 What do words do? Toward a
future the right time? Exp. Psychol. 57, 308 –314. instantaneous idiom understanding. Neuroimage theory of language-augmented thought. In The
(doi:10.1027/1618-3169/a000036) 59, 3502–3513. (doi:10.1016/j.neuroimage.2011. psychology of learning and motivation, vol. 57,

Phil. Trans. R. Soc. B 373: 20170135

110. Santiago J, Lupáñez J, Pérez E, Funes MJ. 2007 11.011) (ed. BH Ross), pp. 255 –297.
Time (also) flies from left to right. Psychon. Bull. 121. Desai RH, Binder JR, Conant LL, Mano QR, 131. Slobin DI. 1996 From ‘thought and language’ to
Rev. 14, 512–516. (doi:10.3758/BF03194099) Seidenberg MS. 2011 The neural career of sensory- ‘thinking for speaking’. In Rethinking linguistic
111. Schnall S, Benton J, Harvey S. 2008 With a clean motor metaphors. J. Cogn. Neurosci. 23, 2376– relativity (eds J Gumperz, S Levinson), pp. 70 –96.
conscience: cleanliness reduces the severity of moral 2386. (doi:10.1162/jocn.2010.21596) Cambridge, UK: Cambridge University Press.
judgments. Psychol. Sci. 19, 1219 –1222. (doi:10. 122. Rapp AM, Leube DT, Erb M, Grodd W, Kircher TT. 132. Flumini A, Santiago J. 2013 Time (also) flies from
1111/j.1467-9280.2008.02227.x) 2004 Neural correlates of metaphor processing. left to right if it is needed! In Proc. 36th Ann. Conf.
112. Zanolie K, van Dantzig S, Boot I, Wijnen J, Schubert Cogn. Brain Res. 20, 395– 402. (doi:10.1016/j. Cogn. Sci. Soc. (eds M Knauff, M Pauen, N Sebanz,
TW, Giessner SR, Pecher D. 2012 Mighty metaphors: cogbrainres.2004.03.017) I Wachmuz), pp. 2315–2320. Austin, TX: Cognitive
behavioral and ERP evidence that power shifts 123. Stringaris A, Medford N, Giampietro V, Brammer M, Science Society.
attention on a vertical dimension. Brain Cogn. 78, David A. 2007 Deriving meaning: distinct neural 133. Slepian ML, Ambady N. 2014 Simulating
50 –58. (doi:10.1016/j.bandc.2011.10.006) mechanisms for metaphoric, literal, and non- sensorimotor metaphors: novel metaphors influence
113. Lakens D, Semin GR, Foroni F. 2011 Why your meaningful sentences. Brain Lang. 100, 150 –162. sensory judgments. Cognition 130, 309 –314.
highness needs the people: comparing the absolute (doi:10.1016/j.bandl.2005.08.001) (doi:10.1016/j.cognition.2013.11.006)
and relative representation of power in vertical 124. Bardolph M, Coulson S. 2014 How vertical 134. Casasanto D, Bottini R. 2014 Mirror reading
space. Soc. Psychol. 42, 205–213. (doi:10.1027/ hand movements impact brain activity elicited can reverse the flow of time. J. Exp. Psychol. 143,
1864-9335/a000064) by literally and metaphorically related words: 473–479. (doi:10.1037/a0033297)
114. Boot I, Pecher D. 2010 Similarity is closeness: an ERP study of embodied metaphor. Front. 135. Shayan S, Ozturk O, Sicoli M. 2011 The thickness of
metaphorical mapping in a perceptual task. Human Neurosci. 8, 1031. (doi:10.3389/fnhum. pitch: cross-modal metaphors in Farsi, Turkish and
Q. J. Exp. Psychol. 63, 942–954. (doi:10.1080/ 2014.01031) Zapotec. Senses Soc. 6, 19– 29. (doi:10.2752/
17470210903134351) 125. Santiago J, Ouellet M, Román A, Valenzuela J. 2012 174589311X12893982233911)
115. Casasanto D. 2008 Similarity and proximity: when Attentional factors in conceptual congruency. Cogn. 136. Dolscheid S, Shayan S, Majid A, Casasanto D. 2013
does close in space mean close in mind? Mem. Sci. 36, 1051–1077. (doi:10.1111/j.1551-6709. The thickness of musical pitch: psychophysical
Cogn. 36, 1047 –1056. (doi:10.3758/MC.36.6.1047) 2012.01240.x) evidence for linguistic relativity. Psychol. Sci. 24,
116. Murphy GL. 1996 On metaphoric representation. 126. Bowdle BF, Gentner D. 2005 The career of 613–621. (doi:10.1177/0956797612457374)
Cognition 60, 173– 204. (doi:10.1016/0010- metaphor. Psychol. Rev. 112, 193–216. (doi:10. 137. Dolscheid S, Hunnius S, Casasanto D, Majid A.
0277(96)00711-1) 1037/0033-295X.112.1.193) 2014 Prelinguistic infants are sensitive to space-
117. Murphy GL. 1997 Reasons to doubt the present 127. Desai R, Conant L, Binder J, Park H, Seidenberg M. pitch associations found across cultures. Psychol.
evidence for metaphoric representation. Cognition 2013 A piece of the action: modulation of sensory- Sci. 25, 1256 – 1261. (doi:10.1177/
62, 99 –108. (doi:10.1016/S0010-0277(96)00725-1) motor regions by idioms and metaphors. 0956797614528521)