=Paper=
{{Paper
|id=Vol-3140/degiorgis
|storemode=property
|title=The Racing Mind and the Path of Love: Automatic Extraction of Image Schematic Triggers in Knowledge Graphs Generated from Natural Language
|pdfUrl=https://ceur-ws.org/Vol-3140/paper1.pdf
|volume=Vol-3140
|authors=Stefano De Giorgis,Aldo Gangemi,Dagmar Gromann
|dblpUrl=https://dblp.org/rec/conf/isd2/GiorgisGG22
}}
==The Racing Mind and the Path of Love: Automatic Extraction of Image Schematic Triggers in Knowledge Graphs Generated from Natural Language==
https://ceur-ws.org/Vol-3140/paper1.pdf
The Racing Mind and the Path of Love: Automatic
Extraction of Image Schematic Triggers in
Knowledge Graphs Generated from Natural
Language
Stefano De Giorgis1 , Aldo Gangemi1,2 and Dagmar Gromann3
1
University of Bologna - FICLIT Department, Via Zamboni 32, 40126 Bologna (BO), Italy
2
ISTC - CNR, Via S. Martino della Battaglia 44, 00185 Roma (RM), Italy
3
Centre for Translation Studies, University of Vienna, Universitätsring 1, 1010 Vienna, Austria
Abstract
Embodied Cognition and Cognitive Metaphors Theory take their origin from our use of language: senso-
rimotor triggers are disseminated in our daily communication, expression and commonsense knowledge.
We propose, in this work, a first attempt of image-schematic triggers automatic extraction, starting from
knowledge graphs automatically generated from natural language. The methodology proposed here is
conceived as a modular addition integrated in the FRED tool, able to generate knowledge graphs from
natural language, while it has its foundation in querying ImageSchemaNet, the Image Schematic layer
developed on top of FrameNet and integrated in the Framester resource. This methodology allows the
extraction of sensorimotor triggers from WordNet, VerbNet, MetaNet, BabelNet and many more.
Keywords
image schemas, knowledge reprentation, cognitive semantics, embodied cognition, frame semantics
1. Introduction
Image schemas (IS) have been proposed within the tradition of embodied cognition as concep-
tual structures that capture sensorimotor experiences and shape abstract cognition, including
commonsense reasoning and semantic structures of natural language (see e.g. [1, 2]). Image
schemas are defined as internally structured gestalts [3], that is, composed of spatial primitives
(SP) that make up more complex image schemas as unified wholes of meaning [4, 1, 5]. For
example, considering the expression ”my mind is racing”, the ”mind” is conceptualized as an
entity moving along a path and it activates the well-established SOURCE_PATH_GOAL image
schema, which is activated in conceptualizing any kind of movement along a path. This IS, in
order to realize its semantics, is (gestaltically) composed by the SOURCE, PATH, and GOAL
spatial primitives.
The Sixth Image Schema Day (ISD6), March 24–25, 2022, Jönköping University, Sweden
Envelope-Open stefano.degiorgis2@unibo.it (S. De Giorgis); aldo.gangemi@unibo.it (A. Gangemi); dagmar.gromann@gmail.com
(D. Gromann)
GLOBE https://github.com/StenDoipanni/ (S. De Giorgis); https://www.unibo.it/sitoweb/aldo.gangemi (A. Gangemi);
https://dagmargromann.com/ (D. Gromann)
Orcid 0000-0003-4133-3445 (S. De Giorgis); 0000-0001-5568-2684 (A. Gangemi); 0000-0003-0929-6103 (D. Gromann)
© 2022 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
While their existence in natural language has been studied by means of corpus-based (e.g.
[6, 7]) and machine learning methods (e.g. [8, 9, 10]), few approaches to formalize image schemas
(e.g. Image Schema Logic [11]) and connect them to existing resources to capture semantics
exist.
The notion of image schema profile is used here as in [12] and [13] to describe the collection
of IS which are activated by some entity, sentence, situation, or event.
This work proposes a first attempt to build an automatic image schema profile extractor
from natural language, and we explain the working pipeline, providing a possible evaluation
methodology using as a golden standard the ISCAT repository.
2. Preliminaries
To perform automatic image schema profile extraction we rely on the FRED tool and the
ImageSchemaNet ontology, part of the Framester resource.
2.1. Framester, ImageSchemaNet and the Frame Semantics Approach
Frames in a most general notion are (cognitive) representations of typical features of a situa-
tion. Fillmore’s frame semantics [14] has been most influential as a combination of linguistic
descriptions and characterisation of related knowledge structures to describe cognitive phenom-
ena. Words or phrases are associated with frames based on the common scene they evoke. In
FrameNet, frames are also explained as situation types. In Framester semantics [15] observed/re-
called/anticipated/imagined situations are consequently occurrences of frames.
Fillmore explicitly compares frames to other notions, such as experiential gestalt [16], stating
that frames can refer to a unified framework of knowledge or a coherent schematization of
experience. Thus, widely acknowledged frames provide a theoretically well-founded and
practically validated basis for commonsense knowledge patterns.
Framester [17, 15] is a linked data hub that provides a formal semantics for frames [15],
based on Fillmore’s frame semantics [14]. It creates/reengineers linked data versions of lin-
guistic resources, such as WordNet [18], OntoWordNet [19], VerbNet [20], BabelNet [21], etc,
jointly with factual knowledge bases (e.g. DBpedia [22], YAGO [23]). Framester also includes
ImageSchemaNet [24].
ImageSchemaNet [24] is a formal and re-usable representation of image schemas as Semantic
Web technology in form of an ontological layer. It presents a formal representation of image
schemas as a new layer of the Framester hub. Since a major flaw in current image schema
theory was the lack of agreement about the lexical coverage of image schemas, ImageSchemaNet
introduces an image-schematic layer linking IS and SP to FrameNet frames and frame elements,
WordNet synsets (sets of contextual synonyms) and word supersenses, VerbNet verbs, etc.,
thereby creating a formal, lexicalized integration of cognitive semantics, enactive theories, and
frame semantics. Currently, ImageSchemaNet provides lexical coverage and formalization for the
following six image schemas: SOURCE_PATH_GOAL, CONTAINMENT, CENTER_PERIPHERY,
PART_WHOLE, SUPPORT, and BLOCKAGE.
Figure 1: Knowledge graph produced by FRED for My mind is racing
Framester can be used to jointly query (via a SPARQL endpoint1 ) all the resources aligned to
its formal frame ontology2 . Framester has been used [25] to formalize the MetaNet resource
of conceptual metaphors3 , based on FrameNet frames as metaphor sources and targets (frame-
based), as well as to uncover semantic puzzles emerging from a logical treatment of frame-based
metaphors.
2.2. FRED Tool
FRED [26] is a hybrid knowledge extraction system to generate knowledge graphs directly
from natural language text input. It is built with a pipeline that includes both statistical and
rule-based components, aimed at producing RDF and OWL knowledge graphs, with embedded
entity linking, word-sense disambiguation, and frame/semantic role detection, aligning entities
in the produced graph directly to entities from Framester resource.
3. Automatic Image Schema Profile Extraction
The workflow to extract IS and SP from natural language is composed as follows: the first step
is to take a sentence and pass it to FRED tool to generate a knowledge graph from text. Let’s
take as example “My mind is racing”, which relates to the cognitive metaphor ”THE BODY IS A
MACHINE” and shows the activation of SOURCE_PATH_GOAL. Figure 1 shows the knowledge
graph automatically generated as output by FRED4 .
The second step consists in taking all the subjects, predicates, and objects of all triples, namely
all nodes and arches in the graph, to systematically interrogate the Framester repository, in
particular the ImageSchemaNet graph, via SPARQL queries, to check if there is an activation of
some Image Schema or Spatial Primitive5 . The ImageSchemaNet graph contains more than 40k
triples which take as subject entities from resources in the Framester hub (FrameNet, VerbNet,
1
http://etna.istc.cnr.it/framester2/sparql
2
The Framester Schema is available at: https://w3id.org/framester/schema/
3
The MetaNet schema in Framester’s OWL is at https://w3id.org/framester/metanet/schema/.
4
This graph can be reproduced using the My mind is racing sentence as input for the FRED online demo, available
here: http://wit.istc.cnr.it/stlab-tools/fred/demo/ and ticking the “Align concepts to Framester” option.
5
Some useful explorative queries are available at https://github.com/StenDoipanni/ISAAC/tree/main/
ImageSchemaNet
Table 1
PATH-family subset accuracy
ISCAT Processed Extracted SOURCE_PATH_GOAL SOURCE PATH GOAL
390 383 262 187 0 65 8
WordNet etc.) and declare the activation of some IS or SP. ImageSchemaNet knowledge graph
generation is shown in [24].
Finally, for each occurrence of activation retrieved, a triple is added to the original graph declar-
ing the activation. In the above-mentioned example, shown in Figure 3, IS activators retrieved
in the ImageSchemaNet graph are: the frames f s d a t a : S e l f M o t i o n and f s d a t a : M o t i o n , and the
VerbNet verb v n . d a t a : R a c e _ 5 1 0 3 2 0 0 0 , resulting in three occurrences of SOURCE_PATH_GOAL
activation.
4. Evaluation
We propose here an evaluation setting and some preliminary results, even though the evaluation
process poses some problems, as exposed in the following sections.
4.1. Evaluation Setting
The evaluation setting uses the ISCAT dataset6 [27], and both above mentioned FRED tool
as state-of-the-art for frame detection from natural language and ImageSchemaNet ontology.
ISCAT is a repository of sentences manually annotated with one image schema per sentence.
All examples are taken from a large variety of original sources, mainly from literature (e.g.
[28, 3]), but also from some online sources (e.g. MetaNet, newspaper articles), which are listed
in the cleaned version of the repository7 . Please be reminded that multiple image schemas
might be correct for a single sentence, and no inconsistency or incompatibility could come from
co-location of different image schemas in the same sentence, even if each sentence in the ISCAT
dataset is manually annotated with only one.
4.2. Evaluation Results
Table 1 shows some first results, in particular the column “ISCAT” is the amount of sentences
in the ISCAT repository which were labeled as activating SOURCE_PATH_GOAL. Column
“Processed” shows the actual number of sentences for which FRED successfully generated a
knowledge graph. The 7 missing entries produced blank files, probably due to the brevity of
the sentences, and were for this reason left apart. Column “Extracted” shows the number of
actual sentences for which our method individuated the activation of at least one of the six IS
covered by ImageSchemaNet ontology. Finally, columns “SOURCE_PATH_GOAL”, “SOURCE”,
6
Image Schema Database procured by Jörn Hurtienne
7
Available here https://github.com/dgromann/ImageSchemaRepository
Table 2
CONTAINMENT coactivation in PATH-family subset
CONTAINMENT CONTAINER INSIDE
61 44 0
Table 3
PART_WHOLE coactivation in PATH-family subset
PART_WHOLE PART WHOLE
56 0 0
Table 4
CENTER_PERIPHERY coactivation in PATH-family subset
CENTER_PERIPHERY CENTER PERIPHERY
48 0 5
Table 5
BLOCKAGE coactivation in PATH-family subset
BLOCKAGE BLOCKER BLOCKED
16 0 0
“PATH”, and “GOAL” show the number of occurrences in which it was retrieved the activation
of SOURCE_PATH_GOAL or one of its spatial primitives.
Tables 2, 3, 4, 5, and 6 show the number of occurrences of coactivation of one of the other IS
in the subset of sentences that in the ISCAT repository were originally labeled as activating
only SOURCE_PATH_GOAL.
4.3. Evaluation Discussion
Although the results are promising, a precise evaluation criterion is still lacking, mainly due to
the fact that our method, as previously shown by tables, is strongly inclined to detect many
image schemas per sentence, while the golden standard considers only one activation per
sentence. This tendency is better explained in Figure 2 that shows an excerpt of the knowledge
graph8 generated for the sentence “Our relationship hit rock bottom and we were stuck there
without moving forwards or backwards” and enriched with IS and SP activation.
In this highly metaphoric example, our methodology retrieves the activation of BLOCKAGE
by some Framester frames, namely f s : C a u s e I m p a c t , f s : H i t T a r g e t and f s : I m p a c t , and by the
WordNet synset w n : s y n s e t - h i t - v e r b - 2 ; locative references like the f s : P l a c i n g frame and the
w n : s u p e r s e n s e - n o u n - l o c a t i o n activate CONTAINMENT and the CONTAINER spatial primitive,
8
The full graph is available in Appendix A in Figure A.
Table 6
SUPPORT coactivation in PATH-family subset
SUPPORT SUPPORTER SUPPORTED
17 0 0
while the aspect of movement is captured by f s : S e l f M o t i o n , f s : T r a v e l , f s : B o d y M o v e m e n t and
f s : M o t i o n frames, as well as by the VerbNet entity v n : M o v e _ 1 1 0 2 0 0 0 0 . Interestingly the WordNet
synset w n : s y n s e t - b o t t o m - n o u n - 1 activates both PART_WHOLE and CENTER_PERIPHERY. This
double activation could be explained considering the gestaltic nature of image schemas: to
conceive the “bottom” of some entity is necessary to conceive the whole entity and then to
focus on some part, and if the considered entity is some kind of container, then its “bottom”,
as well as its top border, could be seen as the periphery of the area occupied by the container.
A final positive aspect of this frame-based approach is that, due to the graph structure of the
output, the image schema profile extraction is fully explainable and it is possible to navigate
through the graph from the activation occurrence to the trigger entity, back to the lexical unit
in the original sentence.
5. Conclusion and Future Works
We proposed here a first version of an automated frame-based image schema profile extractor
which operates in an explainable way, enriching knowledge graphs with an image schematic
lexical layer. Actual ongoing improvements of this approach include extending image schematic
coverage, including also other IS left aside in the current version of the ImageSchemaNet on-
tology. Furthermore, we are working on frame role-labeling mechanism to enhance IS and SP
matching to natural language statements. Future developments could include, on the side of the
evaluation golden standard, a further manual revision of the proposed image schematic labeling
approach. On the technical side, we are considering the integration of more recent neural frame
labeling approaches, in order to allow the reuse of image schemas detector for other knowledge
layers e.g. emotionality, moral values, etc. expressed in natural language grounded in embodied
cognition.
Appendix
A. The Path of Love
Here you find the graph produced by FRED for a sample sentence automatically enriched with
image schemas and spatial primitives activation, expressed in Turtle syntax. The locus of
activation is in this way clear and explainable via linking the IS or SP activation directly to the
lexical unit from natural language text.
Figure 2: An excerpt of the knowledge graph produced by FRED and enriched with Image Schema and
Spatial Primitive activation retrieved by querying ImageSchemaNet for Our relationship hit rock bottom
and we were stuck there without moving forwards or backwards. The full graph is available in Appendix A
in Figure A.
Our relationship hit rock bottom and we were stuck there without moving forwards or backwards.
@prefix framestercore: .
@prefix ns0: .
@prefix ns1: .
@prefix ns2: .
@prefix ns3: .
@prefix ns4: .
@prefix ns5: .
@prefix ns6: .
@prefix owl: .
@prefix rdfs: .
@prefix vndata: .
@prefix wn30instances: .
ns5:hit_1 a ns5:Hit ;
ns4:Agent ns5:relationship_1 ;
ns4:Theme ns5:bottom_1 .
ns5:move_2 a ns5:Move ;
ns3:hasTruthValue ns3:False ;
ns5:union ns5:move_1 ;
ns4:Agent ns5:person_1 ;
ns4:Theme ns5:forward_1 .
ns5:rock_1 a ns5:Rock ;
ns1:associatedWith ns5:bottom_1 .
ns5:stick_1 a ns5:Stick ;
ns1:hasQuality ns5:There ;
ns4:Theme ns5:person_1 .
ns5:Bottom rdfs:subClassOf ,
framestercore:PartOrientational,
wn30instances:supersense-noun_location ;
owl:equivalentClass wn30instances:synset-bottom-noun-1 .
ns5:Hit rdfs:subClassOf ns1:Event,
framestercore:CauseHarm,
framestercore:CauseImpact,
framestercore:ExperienceBodilyHarm,
framestercore:HitTarget,
framestercore:Impact ;
owl:equivalentClass vndata:Hit_17010100,
wn30instances:synset-hit-verb-2 .
ns5:Relationship rdfs:subClassOf ns1:Description,
framestercore:CognitiveConnection,
wn30instances:supersense-noun_relation ;
owl:equivalentClass wn30instances:synset-relationship-noun-1 .
ns5:RockBottom ns1:associatedWith ns5:Rock ;
rdfs:subClassOf ns5:Bottom .
ns5:Stick rdfs:subClassOf ns1:Event,
framestercore:Attaching,
framestercore:BeingAttached,
framestercore:CauseMotion,
framestercore:InchoativeAttaching,
framestercore:Placing ;
owl:equivalentClass vndata:Stick_9070100 .
ns5:backward_1 a ns5:Backward ;
ns6:hasQuantifier ns6:multiple .
ns5:forward_1 a ns5:Forward ;
ns6:hasQuantifier ns6:multiple .
ns5:move_1 a ns5:Move ;
ns3:hasTruthValue ns3:False ;
ns4:Agent ns5:person_1 ;
ns4:Theme ns5:backward_1 .
ns5:relationship_1 a ns5:Relationship ;
ns5:relationshipOf ns5:person_1 .
ns5:Move rdfs:subClassOf ns1:Event,
framestercore:BodyMovement,
framestercore:CauseChangeOfPositionOnAScale,
framestercore:CauseMotion,
framestercore:ChangePositionOnAScale,
framestercore:Motion,
framestercore:SelfMotion,
framestercore:Travel ;
owl:equivalentClass vndata:Move_11020000 .
ns5:Rock rdfs:subClassOf ns1:PhysicalObject,
framestercore:CauseHarm,
wn30instances:supersense-noun_object ;
owl:equivalentClass wn30instances:synset-rock-noun-1 .
ns5:bottom_1 a ns5:RockBottom .
ns5:person_1 a ns5:Person .
framestercore:BodyMovement ns0:activates ns2:SOURCE_PATH_GOAL .
framestercore:CauseImpact ns0:activates ns2:BLOCKAGE .
framestercore:HitTarget ns0:activates ns2:BLOCKAGE .
framestercore:Impact ns0:activates ns2:BLOCKAGE .
framestercore:Motion ns0:activates ns2:SOURCE_PATH_GOAL .
framestercore:Placing ns0:activates ns2:CONTAINMENT .
framestercore:SelfMotion ns0:activates ns2:SOURCE_PATH_GOAL .
framestercore:Travel ns0:activates ns2:SOURCE_PATH_GOAL .
vndata:Move_11020000 ns0:closeMatchActivation ns2:SOURCE_PATH_GOAL .
wn30instances:supersense-noun_location ns0:lexicalSenseActivation ns2:CONTAINER .
wn30instances:synset-bottom-noun-1 ns0:lexicalSenseActivation ns2:CENTER_PERIPHERY,
ns2:PART_WHOLE .
wn30instances:synset-hit-verb-2 ns0:lexicalSenseActivation ns2:BLOCKAGE .
Figure 3: Knowledge graph produced by FRED and enriched with Image Schema and Spatial Primitives
activation from ImageSchemaNet for the sentence:
Our relationship hit rock bottom and we were stuck there without moving forwards or backwards
(Visualization realized with Linked Data Finland RDF Grapher service available here: https://www.ldf.fi/
service/rdf-grapher).
References
[1] J. M. Mandler, C. Pagán Cánovas, On defining image schemas, Language and Cognition
(2014) 1–23. doi:1 0 . 1 0 1 7 / l a n g c o g . 2 0 1 4 . 1 4 .
[2] L. Talmy, The fundamental system of spatial schemas in language, in: B. Hampe, J. E. Grady
(Eds.), From perception to meaning: Image schemas in cognitive linguistics, volume 29 of
Cognitive Linguistics Research, Walter de Gruyter, 2005, pp. 199–234.
[3] M. Johnson, The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason,
The University of Chicago Press, Chicago and London, 1987.
[4] B. Hampe, Image schemas in cognitive linguistics: Introduction, From perception to
meaning: Image schemas in cognitive linguistics 29 (2005) 1–14.
[5] M. M. Hedblom, O. Kutz, F. Neuhaus, Choosing the right path: image schema theory as a
foundation for concept invention, Journal of Artificial General Intelligence 6 (2015) 21–54.
[6] A. Papafragou, C. Massey, L. Gleitman, When English proposes what Greek presupposes:
The cross-linguistic encoding of motion events, Cognition 98 (2006) B75–B87.
[7] J. A. Prieto Velasco, M. Tercedor Sánchez, The embodied nature of medical concepts:
image schemas and language for pain., Cognitive processing (2014). doi:1 0 . 1 0 0 7 / s 1 0 3 3 9 -
013-0594-9.
[8] D. Gromann, M. M. Hedblom, Body-mind-language: Multilingual knowledge extraction
based on embodied cognition, in: AIC, 2017, pp. 20–33.
[9] D. Gromann, M. M. Hedblom, Kinesthetic mind reader: A method to identify image
schemas in natural language, in: Proceedings of Advancements in Cogntivie Systems,
2017.
[10] L. Wachowiak, D. Gromann, Systematic analysis of image schemas in naturallanguage
through explainable multilingualneural language processing, in: Proceedings of the 56th
Linguistic Colloquium, Peter Lang Group, Forthcoming.
[11] M. M. Hedblom, O. Kutz, T. Mossakowski, F. Neuhaus, Between contact and support:
Introducing a logic for image schemas and directed movement, in: Conference of the
Italian Association for Artificial Intelligence, Springer, 2017, pp. 256–268.
[12] M. M. Hedblom, O. Kutz, R. Peñaloza, G. Guizzardi, Image schema combinations and
complex events, KI-Künstliche Intelligenz 33 (2019) 279–291.
[13] T. Oakley, Image schemas, The Oxford handbook of cognitive linguistics (2007) 214–235.
[14] C. J. Fillmore, Frame semantics, in: Linguistics in the Morning Calm, Seoul: Hanshin,
1982, pp. 111–138.
[15] A. Gangemi, Closing the loop between knowledge patterns in cognition and the semantic
web, Semantic Web 11 (2020) 139–151.
[16] G. Lakoff, M. Johnson, Metaphors we live by, University of Chicago press, 1980.
[17] A. Gangemi, M. Alam, L. Asprino, V. Presutti, D. R. Recupero, Framester: a wide coverage
linguistic linked data hub, in: European Knowledge Acquisition Workshop, Springer, 2016,
pp. 239–254.
[18] G. A. Miller, Wordnet: A lexical database for english, Communications of the ACM 38
(1995) 39–41.
[19] A. Gangemi, R. Navigli, P. Velardi, The ontowordnet project: extension and axiomatization
of conceptual relations in wordnet, in: OTM Confederated International Conferences” On
the Move to Meaningful Internet Systems”, Springer, 2003, pp. 820–838.
[20] K. K. Schuler, VerbNet: A broad-coverage, comprehensive verb lexicon, University of
Pennsylvania, 2005.
[21] R. Navigli, S. P. Ponzetto, Babelnet: Building a very large multilingual semantic network,
in: Proceedings of the 48th annual meeting of the association for computational linguistics,
2010, pp. 216–225.
[22] S. Auer, C. Bizer, G. Kobilarov, J. Lehmann, R. Cyganiak, Z. Ives, Dbpedia: A nucleus for a
web of open data, in: The semantic web, Springer, 2007, pp. 722–735.
[23] F. M. Suchanek, G. Kasneci, G. Weikum, Yago: a core of semantic knowledge, in: Proceed-
ings of the 16th international conference on World Wide Web, 2007, pp. 697–706.
[24] S. De Giorgis, A. Gangemi, D. Gromann, Imageschemanet: Formalizing embodied com-
monsense knowledge providing an image-schematic layer to framester, Semantic Web
Journal forthcoming (2022).
[25] A. Gangemi, M. Alam, V. Presutti, Amnestic forgery: An ontology of conceptual metaphors,
in: S. Borgo, P. Hitzler, O. Kutz (Eds.), Proceedings of the 10th International Conference,
FOIS 2018, volume 306 of Frontiers in Artificial Intelligence and Applications, IOS Press,
2018, pp. 159–172. URL: https://doi.org/10.3233/978-1-61499-910-2-159. doi:1 0 . 3 2 3 3 / 9 7 8 - 1 -
61499-910-2-159.
[26] A. Gangemi, V. Presutti, D. R. Recupero, A. G. Nuzzolese, F. Draicchio, M. Mongiovì,
Semantic web machine reading with FRED, Semantic Web 8 (2017) 873–893. URL: https:
//doi.org/10.3233/SW-160240. doi:1 0 . 3 2 3 3 / S W - 1 6 0 2 4 0 , 10.3233/SW-160240.
[27] J. Hurtienne, Image schemas and design for intuitive use (2011).
[28] R. W. Gibbs Jr, D. A. Beitel, M. Harrington, P. E. Sanders, Taking a stand on the meanings
of stand: Bodily experience as motivation for polysemy, Journal of Semantics 11 (1994)
231–251.