Content uploaded by Andrew Manches
Author content
All content in this area was uploaded by Andrew Manches on Feb 20, 2019
Content may be subject to copyright.
Identifying Embodied Metaphors for Computing
Education
Abstract
Computing education is increasing in global importance, with calls for greater understanding of conceptual
development that can inform pedagogy. Here, we report a study investigating elementary computing concepts through
the lens of Embodied Cognition. Sixteen students (9 female) studying university-level computing were asked to explain
their understanding of computing concepts (without materials) in individually video-recorded sessions. We analysed the
gestures generated for three elementary concepts: algorithms, loops, and conditional statements. In total, 368
representational gestures were identified across 48 (16 × 3) explanations, thereby providing evidence that offline
thinking in this domain is embodied. Our analysis of representational gestures showed that participants drew upon two
overarching embodied metaphors in their explanations: 1) Computing Constructs as Physical Objects, in which
participants simulated manipulating physical objects (e.g., pinching) when referring to range of computing constructs,
and 2) Computing Processes as Motion along a Path, whereby participants moved their hands along one of three body-
based axes when referring to temporal sequences. We contrast our findings to similar research in mathematics and
discuss implications for computing pedagogy – namely the role of gesture in the classroom and technologies that can
exploit embodied metaphors.
Keywords
Embodied Cognition; Gesture; Metaphor; Computing Education; Computational Thinking; Representation
Introduction
“An algorithm? Asked Ann. She had never heard of an algorithm for quests. Hope flowed through her. She could handle
algorithms. “It’s simple” started Sir Galwin. “If you have one or more leads, you follow the best one. Otherwise, if you don’t
have any leads, you travel to where you can find more information. Break any ties by flipping a coin.”
(Kubica, 2012, p.17)
Kubica (2012) argues that computing concepts – from Algorithms to Depth-First Searches – are illuminated through
metaphorical storytelling. While the use of metaphors or analogies to teach students computing concepts is
commonplace, it is not usually clear how these mappings are derived, i.e., whether they are informed by any theoretical
understanding of conceptual development in the domain. Increasingly, there have been calls to improve computing
education (Brown, Sentance, Crick, & Humphreys, 2014; Wing, 2008) that have led to curricula change in several
countries. However, concerns have been raised about the theoretical evidence base for the understanding of learning in
this domain. Grover and Pea (2013) make the case for Learning Sciences research to offer such an evi dence base,
especially from the field of Embodied Cognition. As well as addressing fundamental questions concerning the nature of
thinking, Embodied Cognition has also been applied in the design and implementation of learning environments,
notably in the domain of mathematics (Eisenberg, 2009). Given the role of metaphor in Embodied Cognition
perspectives (Lakoff & Johnson, 1980; Núñez & Lakoff, 2000), this field has strong implications for computing, which,
as exemplified above, has historically drawn upon metaphor and analogy in its teaching (e.g., Hui & Umar, 2010;
Pollard & Duvall, 2006). Despite this, there has been little empirical research adopting the theoretical lens of Embodied
Cognition to promote our understanding of the development of computing concepts, or approaches that draw upon
this field to inform computing education.
In this paper, we first present theoretical literature on Embodied Cognition, conceptual metaphor and gesture research
and subsequently relate this work to the domain of computing. We then report our method, adopted from gesture
research, and results, where we present original empirical findings for the embodied nature of computing concepts
through the analysis of gestures generated by individuals explaining three core computing concepts. Furthermore, we
identify representational patterns from these gestures thereby demonstrating the potential of this methodological
approach to provide a deeper understanding of the nature of thinking in this educational domain. Finally, we consider
the implications of this paper for future work and interventions, as well as draw attention to methodological limitations.
Consequently, the core contribution of our work is to help reveal the conceptual foundations of basic computing
concepts through a novel and rigorous research approach. Our findings compliment work aiming to identify the
cognitive foundations in computer science, such as Roman-Gonzalez, Perez-Gonzalez and Jimenez-Fernandez (2017)
who provide evidence for computational thinking, but via psychometric measurements. We draw upon the findings to
describe how this methodological approach may inform pedagogy and the design and im plementation of learning
environments, where instruction needs to be grounded in deep conceptual understanding for science concepts and
education. Here, we share the view with Niebert, Marsch, and Treagust (2012) that ‘understanding needs embodiment’.
1. Literature Review
1.1 Embodied Cognition, Conceptual Metaphors and Gesture Research
1.1.1 Embodied Cognition
Until the 1980s, dominant theories of cognition viewed knowledge as residing in a semantic memory system separate
from the brain’s modal systems for perception (e.g., vision, audition), action (e.g., movement, proprioception), and
introspection (e.g., mental states, affect) (see Barsalou, 2008). So, cognitive science typically assumed a view of the mind
as an abstract information processing system, in which our sensory and motor (sensorimotoric) systems served a
peripheral role conveying information to and from a central cognitive processor (the brain) where high level abstract
thinking took place. Various theoretical perspectives then emerged challenging this position. The work of Ed
Hutchins,(1995), for example, argued that cognition should be considered as distributed across objects, individuals,
artefacts, and tools in the environment. Related to Hutchins’ work is Scaife and Rogers’ (1996) proposal of External
Cognition which emerged from their work examining the role of graphical representations. The authors define external
cognition as a complex iterative relationship between internal and external representations involved in problem solving.
Whilst Distributed and External Cognition theories drew attention to our capacity to externalise, or offload, thinking;
theoretical work also started to recognise how cognitive processes are not confined to the brain but are deeply rooted in
the body’s interaction with the world (Lakoff & Johnson, 1999; Varela, Thompson, & Rosch, 2017): cognition is
embodied.
According to Wilson (2002), Embodied Cognition is an umbrella term capturing various claims, of which she identifies
at least six. Five of these refer to the way we use the environment (and others) as a dynamic resource to reduce ‘online’
cognitive demands – as articulated in other distributed, social and external cognition theories. Wilson contrasts these
‘online’ cognition claims with a more controversial sixth: that our thinking in the absence of ‘relevant stimuli’ – our
‘offline thinking’ - is encoded modally rather than amodally (Barsalou, 1999), and we draw upon these modal, or
‘embodied’, resources to think, reason and communicate different ideas through conceptual mappings. One process in
which we achieve such conceptual mapping is through conceptual metaphors, which Núñez (2000, p. 5) describes as the
“cognitive mechanism by which the abstract is comprehended in terms of the concrete”.
1.1.2 Conceptual Metaphors
Conceptual Metaphor theory emerged from the field of cognitve linguistics, with the proposal that from repeated
embodied experiences with the world, generalised ‘image schemas’ emerge that we map metaphorically to abstract
concepts (Lakoff & Johnson, 1980). For example, our conceptualisation of time is grounded upon experiences of linear
motion in space; for many cultures (not all), the future is seen as something in front of us, and the past behind (Núñez
& Sweetser, 2006).
Similar to analogies (see Gentner, 1983), conceptual metaphors involve mappings between a source domain (e.g.,
movement) and target domain (e.g., time). The key difference is directionality i.e., the target domain is understood, often
unconsciously, in terms of relations that hold in the source domain; whereas the source domain is rooted in everyday
sensorimotoric experience. Conceptual Metaphor theory, therefore, offers an explanation for our ability to think and
reason about abstract concepts, and despite critiques of the theory (see Kövecses, 2008), has provided a valuable
theoretical framework for examining the conceptual underpinnings of abstract concepts, notably in science (e.g., energy
transfer (Close & Scherr, 2015); thermodynamics (Brookes & Etkina, 2015; Jeppsson, Haglund, & Amin, 2015)) and
mathematics (Murphy, 2009; Núñez & Lakoff, 2000; Stacey, Helme, & Steinle, 2001).
Much of the empirical support for conceptual metaphors comes from analysis of language, for example, in
mathematics, language such as “take away” and “break” or “count up” and “next” may indicate how we conceptualise
numbers as collections of objects or as points along a path (see Núñez & Lakoff, 2000). However, when we
communicate such concepts, we do not just use words; we often gesture. And increasing work suggests that these
gestures provide another window into the conceptual metaphors underpinning abstract concepts, thereby supporting
claims that cognition is embodied.
1.1.3 Gestures
During problem-solving or explaining, we move our hands, even when there is no listener present (e.g., while on the
telephone) or even when both the listener and the speaker are blind (Iverson & Goldin-Meadow, 1997). Such work
suggests that the primary function of gestures is to support the speaker; although gestures do also support the listener’s
comprehension. According to McNeill (1992, p. 37), gestures refer to “idiosyncratic spontaneous movement[s] of the
hands and arms accompanying speech”. Whilst speech is more analytic, gestures are more global and image-based,
providing a unique lens to examine meaning (Parrill & Sweetser, 2004).
Gestures support cognitive activity by enabling the speaker to externalise dynamic visual imagery, and in doing so, they
provide additional information about the speaker’s conceptualisation. Gestures are frequently used by both children and
teachers (Alibali & Nathan, 2007; Flevares & Perry, 2001) and so provide a rich source of data with which to examine
cognition, as well as influence learning. Indeed, the last twenty years has created a substantial body of literature
documenting the potential of gesture to support learning (Goldin-Meadow, 2009). Teachers are able to use children’s
gestures to assess learning (Kelly, Singer, Hicks, & Goldin-Meadow, 2002) and identify readiness to learn new concepts
(Pine, Lufkin, & Messer, 2004). It is also possible to support learning directly by encouraging children to gesture,
possibly because this can reduce cognitive load in tasks (Goldin-Meadow, Nusbaum, Kelly, & Wagner, 2001), make
implicit knowledge explicit (Broaders, Cook, Mitchell, & Goldin-Meadow, 2007) or provide new action representations
that children can later draw upon (Goldin-Meadow, Cook, & Mitchell, 2009). Furthermore, teachers’ gestures can
support learning (Valenzeno, Alibali, & Klatzky, 2003). Whilst teachers tend to gesture spontaneously, work has shown
the value of training teachers to use gestures (Hostetter, Bieda, Alibali, Nathan, & Knuth, 2006).
According to Hostetter and Alibali (2008), gestures emerge from perceptual and motor simulations that underlie
embodied language and mental imagery, i.e., gestures are simulated actions. The authors focus on representational
gestures, i.e., those gestures that represent the content of speech by pointing to a referent (deictic gestures), those
depicting a referent or its trajectory (iconic gestures), or those depicting a concrete referent or its trajectory for an
abstract idea (metaphoric gestures). The authors contend that metaphoric gestures arise from perceptual and motor
simulations of spatial image schemas on which conceptual metaphors are based. Therefore, analysing these gestures is a
way to examine conceptual metaphors empirically (Amin, Jeppsson, & Haglund, 2015). Edwards (2009) for example,
drew upon Conceptual Metaphor theory to analyse the gestures generated by student teachers when asked to explain
their understanding of fractions. In the study, 251 gestures were generated by the 12 participants (student teachers), of
which 81 referred to fractions. From her data, Edwards offers a simple topology distinguishing gestures that draw upon
experiences with tangibles objects (e.g., cutting objects into equal partitions) and those that simulate writing of
mathematical procedures or spatial locations of symbols.
Given the potential of examining metaphorical gestures to develop our understanding of thinking in abstract domains
such as mathematics and science, it is necessary to ask whether such gestures are evident when individuals
communicate their thinking in the domain of computing. Yet, whilst there is anecdotal evidence to suggest so (e.g.,
online video explanations of computing concepts
1
) our review of the literature did not reveal relevant empirical work;
although we are not the first to identify the significance of embodiment for computing education (e.g., Landy, Trninic,
Soylu, Kehoe, & Fishwick, 2014).
1.2 Embodied Cognition and Computing concepts
Computing literature emphasizes the abstract nature of thinking in this domain. For example, Kramer (2007, p. 38)
states that ‘abstraction’ is key to com puting, both the “ability to perform abstract thinking and to exhibit abstraction
skills”. According to Wing (2008, p.3717), “the essence of computational thinking is abstraction. In computing, we
abstract notions beyond the physical dimensions of time and space”. From this perspective, the role of metaphors and
analogies is simply pedagogical – to provide ways (e.g., visual imagery) to help students to access abstract notions. For
example, many programming environments (e.g., Figure 1) have adopted block building as a visual metaphor to help
children understand the syntactic nature of code. There are even physical activities that have been developed that aim to
engage children with body-based analogies of computing concepts, such as the work of Computer Science Unplugged
(Bell, Lambert, & Marghitu, 2012), which provides a range of non-computer based activities and games such as running
around lines in a playground to explore the concept of sorting networks. According to Cortina (2015), such activities are
effective because “by being physically part of the solution to a problem as it is being solved, kids learn from
observations and experiences”. However, no clear theoretical framework is provided for why certain body-based
experiences may develop thinking, beyond motivation.
1
E.g., Computing at School explanation videos https://www.youtube.com/watch?v=VowsYScyWNg&t=33s
Fig. 1. Blocks representing construction of code (Arduino)
Embodiment offers a different perspective on metaphors in computing cognition. Rather than just making abstract
concepts more tractable, metaphors may be important in the way we conceptualise certain computing concepts. That
is, conceptual metaphors might be the very mechanisms that we use to make sense of, reason about, and communicate
computing concepts (similar to claims in mathematical cognition). If embodied metaphors do underpin the way
individuals conceptualise certain computing notions, then these metaphors may be evident when people are asked to
communicate their thinking. Such evidence would have important implications for computing education.
1.3 Embodied Metaphors and Instruction
According to Amin (2015), investigating conceptual metaphors has the following implications for education:
• characterizing scientist and learner conceptions and identifying obstacles to learning;
• understanding the process of conceptual change; and
• suggesting productive pedagogical strategies.
With respect to pedagogical strategies, an embodied approach suggests the use of particular gestures to support learning
(Goldin-Meadow, 2015), the use of instructional materials (Pouw, van Gog, & Paas, 2014), or the development of
learning designs leveraging emerging technologies (e.g., gesture recognition devices, tangible technologies) that detect
and dynamically link body-based actions to digital representations (Abrahamson, 2009; Lindgren et al., 2016). While
research has informed other STEM subjects (notably mathematics and science), we propose that this field can
contribute to computing education. This research is especially timely in light of recent calls for increased research into
computational practice and perspectives in formal education (Lye & Koh, 2014).
Here, we investigate the role of Embodied Cognition in computing concepts by examining if, and how, individuals
generate representational gestures when explaining computing concepts – and if so, how these gestures provide a
window into underlying conceptual metaphors. As the first study we are aware of examining gestures in computing
science, this study is exploratory, with our overarching research question being: what types of embodied metaphors, if any, do
individuals generate when explaining elementary computing concepts?
2. Methodology
2.1 Participants
This study adhered to British Psychological Society ethical guidelines and ethical clearance was provided through the
University’s ethical committee. With the intention of examining spontaneous, rather than more purposeful and
rehearsed, gestures, we recruited university students who were familiar with elementary computing concepts (so were
able to communicate effectively); yet were unlikely to have rehearsed verbal explanations of concepts. Participants were
16 computer science students aged 18 to 37 years (M=23.1, SD=5.1) from three universities in the Edinburgh area of
the UK. Data from all subjects were analysed; although there was a video recording error for one explanation. This
sample size is comparable in similar exploratory in-depth gesture analysis work (Cooperrider, Gentner, & Goldin-
Meadow, 2016; Edwards, 2009; Flevares & Perry, 2001; Núñez & Marghetis, 2014), and was therefore considered
adequate. All students were competent English speakers, although English was not the first language for half of the
participants. Students were mainly undergraduates, in the second semester of their first year (n = 8), with several in their
second year (n = 3). Three were master’s students and two were doctoral students. Nine participants were female (56%)
and all participants were right-handed.
Participants were recruited from lectures, targeted email lists, as well as generic school wide emails, where female
participants were particularly encouraged to ensure gender representative data. All participants gave written consent and
received a small monetary incentive. Thirteen participants consented to their images being used to illustrate the data.
2.2 Design
The study drew upon the theoretical framework of Embodied Cognition and the methodological tools of cognitive
linguistics and gesture research to examine the gestures generated by participants during their verbal explanations. We
interpret meaning from individuals’ gestures from context (e.g., the structured interview) and corresponding sp eech
(Parrill & Sweetser, 2004).
The study was conducted as an interview where participants were asked to explain their understanding of computer
science terms. Although less naturalistic than work in this field capturing everyday interaction (Flevares & Perry, 2001;
Nemirovsky, Rasmussen, Sweeney, & Wawro, 2012), this approach echoes comparable work (e.g, Cooperrider et al.,
2016; Edwards, 2009) as it provides a means to generate rich data over a short period as well as consistency between
participants. Despite limitations of sampling discussed later in the paper, the methodological approach was considered
ideal for addressing the research question, namely, to explore the embodied metaphors that individuals’ produce, if any,
in their conceptual explanations of computing concepts.
All 16 participants were given the same concepts in the same order (the sample was too small to counterbalance order).
Importantly, participants were not given relevant external resources (e.g., physical materials, whiteboard, computer
screen) in order to limit the use of context-specific deictic gestures (and actions) directed toward these resources (i.e.,
online cognitive processing; Wilson, 2002). This contrasts with other studies (e.g., Herbert & Pierce, 2007) where
resources provided are likely to have influenced gestures.
2.2.1 Computing concepts
Participants were asked to explain the following three elementary computing concepts: algorithms, loops and conditional
statements. We selected these concepts by a) examination of the Association for Computing Machinery 2013 guidelines
for university computer science curriculum
2
b) discussions with computing education experts that the concepts were
meaningful across ages of computing education (from Primary to Higher Education).
2.3 Procedure and Materials
Participants were interviewed individually by the 2nd author of the paper in a quiet room within each participant’s
university. Interviews were video recorded using a Panasonic HD video camera mounted on a tripod in the corner of
the interview room. At the beginning of the interview, the interviewer briefly explained the purpose of the research in
terms of examining experts’ understanding of different computing concepts. The focus of gesture in the study was not
revealed at this stage (or in any information prior to the study). However, all participants were fully debriefed after
taking part.
Participants were interviewed standing because during piloting we found that sitting could sometimes hinder and/or
obscure gesturing. Current initiatives across universities encourage less sitting at work, so, arguably standing would seem
natural and not influence explanations. No participants said they perceived this as extraordinary when questioned
afterwards.
2
https://www.acm.org/binaries/content/assets/education/cs2013_web_final.pdf
The interviewer then read out various computing terms and for each asked participants to explain “as clearly as you can”.
For the first computing term, the interviewer asked: “Can you explain your understanding of [pause] “Algorithm?”. Thereafter,
the interviewer simply stated the next term.
The interviewer judged an explanation to be finished when participants gave an explicit cue, either verbally or otherwise
(e.g., nod of head). The interviewer limited feedback (beyond general positivity) in order to minimize any influence on
the participants’ explanations. Furthermore, the interviewer did not gesture because this may have primed participants’
embodied representations (i.e., led to mirroring). While this may have created a slightly unnatural socio-communicative
context, the influence was not obvious. If participants asked for feedback during the interview, such as asking whether
they should use an example, the interviewer would reiterate the task by saying “if you just explain your understanding as clearly
as you can”.
After the participant had finished explaining the final computing concept, the interviewer debriefed the participant
about video recording – i.e., the focus on was on their use of gesture as well as speech. The interviewer also asked the
participants whether they remembered using gestures and, if so, why they think they used gestures. Only one participant
stated suspecting the focus was on gesture.
2.4 Measures and Data Analysis
Videos of participants’ explanations were transcribed and analysed using video annotation software (ELAN;
Wittenburg, Brugman, Russel, Klassmann, & Sloetjes, 2006) to provide an overview of the proportion o f time different
individuals gestured for concept explanations. For each explanation, two coders (1st and 2nd author) independently
coded all representational gestures, drawing upon spoken language to guide interpretation. A representational gesture
was defined as a gesture – handshape or motion trajectory – depicting aspects of their meaning, either literally or
metaphorically (Alibali & Nathan, 2011). There was a moderate agreement between coders (Cohen’s Kappa = 0.52),
which is line with other studies (e.g., Cienki, 2005). After coding, coders resolved discrepancies, which were mostly
about the start and end point of individual gestures within the flow of hand movement.
From the representational gestures identified, the first and second author then worked together to identify patterns
across participants of conceptual mappings – mappings between gesture and conceptual entities evident in speech.
Although such interpretation is subjective by nature, interpretation evolved from discussion between coders and has
been made explicit in the findings section (see Parrill & Sweetser, 2004). This interpretative process was facilitated by the
focused interview approach (i.e., the context constrained what types of con cepts were being discussed).
3. Results
3.1 Individuals’ use of gesture
3.1.1 Gesturing time
All 16 participants provided a verbal explanation for each concept, and used gesture for at least one of their three
explanations; indeed, gestures were used on 43 out of 48 explanations (16 x 3). Table 1 summarises the participants’
speech and gestures for each concept. Non-parametric analyses revealed no significant differences in the amount of
gesture or speech time between concepts.
Table 1. Descriptive summary of time (secs.) for speech and gesture for each concept
Speech
Gesture
Gesture/Speech
Range
Median
Total
Range
Median
Total
%
Algorithm
8-128
22
481
0-62
10
229
48
Loops
12-85
23
491
0-47
15
294
60
Conditional
3-41
15
275
0-28
8
148
54
Overall
42-214
67
1246
7-102
33
671
54
Note. Gesture/Speech refers to the proportion of time participants gestured during speech.
3.1.2 Representational gestures
In total, 368 representational gestures were coded across a total of 48 explanations (16 participants with 3 concept
explanations). Table 2 illustrates the number of representational gestures created by each participant for each concept.
The total number of gestures for participants ranged from 3 to 76 (Median = 25).
Table 2. Number of representational gestures generated by each participant for each concept
Participant
Algorithm
Loops
Conditional Statement
Total
1
5
3
0
8
2
12
7
11
30
3
0
0
3
3
4
6
7
7
20
5
10
11
11
32
6
13
6
6
25
7
0
22
2
24
8
2
7
5
14
9
43
33
Missing data
76
10
8
12
7
27
11
4
10
0
14
12
3
12
5
20
13
7
8
7
22
14
5
8
11
24
15
6
7
5
18
16
2
4
5
11
Total
126
157
85
368
3.2 Embodied metaphors
During explanations of the concepts: algorithm, loop and conditional statements, the 16 participants created a total of
368 representational gestures used to refer to a range of computing constructs including but not limited to: data, code,
process, input, execution, output, conditions. For some gestures, there was a direct mapping between the gesture and an external
representation, e.g., lines of code on a vertical graphical interface. For example, Participant 8 used her fingers to denote
the bracketed boundaries of an IF statement (Figure 2a), or participants pointing in steps from top to bottom while
explaining an algorithm (Figure 2b). These gestures might be described as iconic, where the concrete referent is coding
script. Yet, such external representations are designed around various metaphors, blurring the distinction between
iconic and metaphoric gestures.
(a) P8 (b) P1
Fig. 2. Gestures with iconic mapping to code on screen (a) P8 and (b) P1
P8: “If the statement in the IF condition is met”
3
P1: “You give steps” [RH moves down in 4 steps]
From data across all participants, many gestures appeared to be grounded upon two key image schemas (cognitive
structures arising from repeated embodied experiences): the container schema, consisting of three parts: an interior, a
boundary and an exterior, and a path-source-goal schema. These two schemas are described by Núñez & Lakoff (2000)
as underpinning key concepts in mathematics, and it is perhaps unsurprising they are evident here given the strong
domain correspondence between computing and mathematics (Knuth, 1974). These schemas underpin two
conceptual metaphors we identified from the data and are used to structure findings.
3.2.1 Computing Constructs As Physical Objects
When explaining concepts, 14 out of 16 participants simulated a grasping or pinching action at least once. Such hand
forms are described in work focusing on mathematical concepts (Edwards, 2005; Núñez & Marghetis, 2014).
However, whereas numbers are often represented physically, it is less likely that participants were drawing upon
experiences manipulating tangible representations of computing concepts.
3
Bold text indicates creation of gesture; underlined text duration that gesture form is held. LH: Left Hand; RH: Right Hand; BH:
Both hands
The form of the hand, and whether one or two hands were used, suggested differences in the perceived size of the
imaginary object. This is particularly interesting because physical size has no literal meaning in relation to computing
concepts. For example, a two-handed grasp was used by P6 and P8 (Figure 3a) when referring to an algorithm. P2 used a
similar gesture when referring to code (Figure 3b), and again when referring to a condition, as did P5. P12 (Figure 3c) used
the gesture to refer to a loop and P15 used the gesture when referring to a loop as a “set of tasks”.
(a) P8 (b) P2 (c) P12
Fig 3. Two-handed grasping gesture (a) P8, (b) P2 and (c) P12)
P8: “An IF Statement is part of the algorithm that contains…”
P6: “Use over a large amount of data, in order to sort it, find something or make it more usable”
P12: “…doing something with anything you put into a loop”
For some gestures, participants created a grasp hand form with two hands but their hands were cupped upwards and
spaced apart suggesting they were holding two separate objects. Two participants used this gesture when referring to
data, both moving each hand up and down alternatively as if weighing the simulated objects. In a related example,
several participants created two cupped hands when discussing IF statements, where one hand (the left) simulated holding
the IF part of a statement and the other (right) holding, or indicating the other part of the statement (Figure 4a; 4b). P6
created a similar cupped hand when saying IF, but then used her right hand to represent the condition, which she
referred to as “Blank” (Figure 4c). One participant (P10) only used one cupped hand denoting the IF statement. A
cupped hand was also used by P8 when referring to a program and P7 referring to a “bug in the program”.
(a) P14 (b) P13 (c) P6
Fig. 4. Hands holding separate objects (a) P14 (b) P13 and (c) P6
P14: “An else if or whatever”
P13: [Gesture created before speech] “if something is true...” ...
P6: “If blank / then…”
In contrast to a grasping or cupping action, many participants created a pinch in their explanations, as if manipulating a
smaller object. This hand form was used when describing verbally a range of constructs including IF statement (P2-
Figure 5a), (algorithmic) steps (P1; P4), condition (P4), Loop (P4; P15); True (P4), Command (P8), Section of code
(P11- Figure 5b), Bytes of code (P9 – Figure 5c), Else (P14); and Attribute (P16).
(a) P2 (b) P11 (c) P9
Fig. 5. Pinching action (a) P2 (b) P11 and (c) P9
P2: “When you say this is an If statement….”
P11: “Section of code”
P9: “Bytes of code”
As well as hand form, there were gesture movements that presented computing concepts as objects. These gestures
would often involve one hand acting as a ‘placeholder’ (Figure 6a) or boundary of an object, and then the other hand
arcing into this hand (Figure 6b), or arcing away from this hand. Often there would be associated language around
adding or taking away (e.g., Figure 6c).
(a) P13 (b) P14 (c) P16
Fig. 6. Placing or taking away an object (a) P13 (b) P14 and (c) P16
P13: “Extract some kind of feature” [Grasping RH arcs away from body]
P14: “Whatever’s inside the code” [LH swoops down]
P16: “Set of instructions” [RH fingers placed inside LH which closes slightly]
3.2.2 Computing Processes As Motion Along A Path
Longitudinal axis: pathway up to down in front of body
In relation to the body, there are three main axes: longi tudinal, transverse, and frontal. Similar to reading text, individuals
read and write code from top to bottom on a vertical screen: a longitudinal axis. Therefore, gestures referring to time-
based computing processes might move downwards in a longitudinal plane. Indeed, this gesture was observed in
several participants (Figures 7a; 7b; 7c) discussing algorithmic steps/instructions, with most marking steps going
downwards with a finger or side of the hand (P1, P2, P5, P7, P8, P9 P13, P15).
(a) P9 (b) P8 (c) P13
Fig. 7. Sequential steps down on a longitudinal plane (a) P9 (b) P8 and (c) P13
P9: “What steps we are following” [RH moves arcs three points downwards]
P8: “or set of instructions” [LH arcs three points downwards]
P13: “instead of having to write, do A, do A, do A, in a sequential way” [LH finger arcs three times downwards]
Transversal axis: pathway left to right across the body
Again, similar to text, code is typically written from left to right, which suggest that gestures communicating
computational processes might trace a similar transversal axis. Ten participants (P3, P4, P5, P6, P7, P10, P11, P13, P15,
P16) created at least one gesture where they moved one or both hands from left to right while describing a process.
However, the participants utterances were referring to general processes (e.g., output; ‘executing code’), rather than
individual lines of codes. Therefore, this general left to right schema for time based processes is comparable to the same
left to right ordering found for concepts such as number (Dehaene, Bossini, & Giraux, 1993).
Participants also showed transversal gestures to communicate processes. For example, gestures involved small left to
right arcs when they referred to algorithmic steps. In contrast, several participants created a single horizontal gesture or
arc to the right; typically, when referring to a more general resultant process, such as completing instructions (Figure 8a)
or achieving “output” (Figure 8b). Interestingly, two participants who created a gesture to the right when discussing a
resultant process, used a subsequent arcing gesture back to the left when describing back to a previous step in the
process – for example, a failed condition (Figure 8c).
(a) P16 (b) P13 (c) P3
Fig. 8. Left to right transversal gestures communicating time-based processes (a) P16 (b) P13 and (c) P3
P16: “What a computer takes in and completes [BH in triangular form arc to the right, with body turning to right]
P10: “to achieve an output” [RH moves horizontally to the right]
P3: “if it fails the condition.” [RH arcs from right to move slightly to the left and point left]
The left to right transverse as a metaphor was seen as time-based processes in participants’ gestures for loops. In total,
14 out of 16 participants moved one or both hands in a circle when discussing loops (as well as other iterative
processes). For 9 participants (P2; P4; P5; P6 P7; P8; P9; P11; P16), this circular gesture was clockwise on a transversal
axis (Figures 9a; 9b, 9c.) These clockwise gestures were interpreted as left to right because a) gestures began by moving
left to right, b) the left to right arc was often more pronounced, c) the hand would often move to the right whilst
rotating clockwise. There was, nevertheless, one exception to the clockwise gesture: P7 created an anticlockwise gesture.
This participant was talking about “loops in a program where the program does kind of screwing up [sic]”, and moreover, they later
created a clockwise gesture when discussing loops.
(a) P11 (b) P6 (c) P16
Fig. 9. Clockwise rotating gesture for loops ( (a) P11 (b) P16 and (c) P16
P11: “Kinda similar to an algorithm” [RH Finger rotates three times clockwise]
P6: “when a program goes through the same process over and over again” [RH with extended finger rotates
clockwise from body to the side]
P16: “when you are calling the instruction over and over at the same time” [BH rotating clockwise in front of
body]
Frontal axis: pathway forwards from the body
Although the clockwise gestures for loops described in the previous section were predominately transversal, many did
angle slightly outwards. Moreover, for 7 participants (P1; P5; P9; P10; P12; P13; P14), the circular gesture for loops was
projected forward away from the body on a frontal axis (Figure 10a), where the forward motion was emphasized.
Similar to the transversal circular gesture, this gesture was used not only to talk about loops but other processes (Figure
10b).
As well as rotating circular gestures, participants often displayed simpler straight or arced gestures in front of themselves
when describing computing processes. In the couple of instances when participants created a gesture moving back
toward their body, their language corresponded to a process of checking back on a process (Figure 10c). Consequently,
this supports the proposition that participants were drawing upon a general body-based spatial metaphor of time
(future is in front of us) to conceptualize computing processes.
(a) P10 (b) P6 (c) P9
Fig. 10. Gesturing using fontal axis as a metaphor for time ( (a) P10 (b) P6 and (c) P9
P10: “which has to be true if the loop is to be repeated” [RH rotating forward]
P6: “or counting up to something” [BH rotating forward]
P9: “checks afterwards” [RH finger points back to body]
3.2.3 Conceptual integration: computing constructs within a process
Our findings show two overarching conceptual metaphors evident in participants’ explanations: Computing Constructs as
Physical Objects, and Computing Processes as Motion along a Path. The participants also demonstrated gestures that were an
integration of both these metaphors. In these gestures, participants would simulate grasping or pinching an object then,
while maintaining this hand form, trace a trajectory along one of the axes discussed. For example, Participant 11, when
explaining what an algorithm was, created a pinch gesture for “steps”, and then circled this hand forward when talking
about these repeating (Figure 11a). The same participant later simulated grasping a larger object when referring to an ‘IF
statement’, then simulated moving this object to her right side when describing how it functioned. Similarly, Participant
10 created a pinch hand form for loop and rotated his hand (Figure 11b). P15 used both hands to simulate pinching a
loop, then moved this in steps to the right of her body. Other examples, including participants P1 (Figures 11c), P11
and P4 described algorithmic steps by pinching an imaginary object and moving the pinched hand down in progressive
steps.
(a) P12 (b) P11 (c) P1
Fig. 11. Integrating metaphors in gestures ( (a) P12 (b) P11 and (c) P1
P11: “a certain amount of steps that you keep on repeating” [LH forms a grasp then rotates forward from the body
twice]
P10: “Loop are section (sic) of code that execute” [RH creates a pinch then rotates forward from body in a single circle]
P1: “each step will be distinct” [RH creates a sequence of four steps one below the other]
4. Discussion
To address the research question: what types of embodied metaphors, if any, do individuals generate when explaining elementary
computing concepts? we asked 16 computer science students to explain their understanding of three elementary computing
concepts, and analysed the spontaneous gestures generated in these explanations. Although the sample was relatively
small, the study demonstrated the rich source of data generated by participants in the form of representational gesture.
Whilst much work has adopted the methodological approach of this paper in other STEM areas, notably mathematics
and science, there has been a notable absence in computer science, possibly attributable to its more recent development
as a conceptual domain. Hence, a key contribution of this paper is draw attention to the potential of gesture research to
develop our understanding of the conceptual underpinnings of computer science and how this may be supported.
In a total of 48 explanations (16 participants × 3 concepts), the study identified 368 representational gestures generated
in the absence of ‘relevant stimuli’, e.g., a screen/board to point to. Because computing concepts have often been
presented as abstract, our findings supports a key claim of Embodied Cognition, that offline thinking involves mental
simulations of perception and action (Alibali & Nathan, 2011; Wilson, 2002). Importantly, the representational nature
of these gestures provides additional information to speech with which to analyse the nature of thinking in this domain .
The findings from our work supports suggestions that the Learning Sciences, and more specifically Embodied
Cognition, may offer insight into computing education (e.g., Grover & Pea, 2013; Landy et al., 2014). Computing
science education requires greater understanding of conceptual development, and as Niebert et al (2012) claim,
“understanding needs embodiment”.
The participants in this study generated an encouraging number of gestures (although not normally distributed across
participants). This enabled various patterns of gestures to be identified which warrant further investigation with a larger,
more representative sample. Patterns included the use of a circular gesture to communicate re-iteration, spatial groups
to represent balancing of clauses, or different hands grasps to refer to different sized constructs. Many of these gesture
patterns can be related to work in other domains, for example, the pin ched hand gesture is illustrated as an example of a
“factor reference” gesture in Cooperrider et al’s (2016) investigation of individuals’ complex relational reasoning.
We have argued in this paper that the range of representational gestures appear grounded upon two image schemas, the
container and path-source-goal schema. Perhaps unsurprisingly, these image schemas are proposed to underpin
mathematical concepts (Núñez & Lakoff, 2000), and the metaphors identified from this study are closely relatable to
two conceptual metaphors proposed for arithmetic: Arithmetic as Object Collection and Arithmetic as Motion along a Path. Such
metaphorical similarity is, arguably, to be expected. Indeed, two participants explicitly compared computing constructs
(e.g., formulae) to mathematics in their explanations. Yet, differences in the metaphors are also revealing. For example,
participants’ gestures related to holding single objects of different sizes, rather than simulating manipulation of object
collections. When participants did simulate manipulating more than one object, their actions suggest a more complex
relationship between constructs; for example, putting one object inside another object (e.g., into a program), holding
two objects simultaneously (e.g., IF and THEN), or using both hands to collect ‘input’ toward the body. Future work
may further investigate these spatial relationships in computing concepts.
In contrast to gestures manipulating objects, gestures in which participants moved one or both hands along a linear axis
typically referred to a computing process (rather than discrete constructs). Revealingly, gestures marked both the start
and end points of a process, and a sequence of steps along a process. The axis of the delineated path is also interesting,
where a longitudinal axis corresponds with vertical lines of code, a transversal axis corresponds with a cultural left-right
direction of time/magnitude seen in other domains from reading to mathematics, and a frontal axis appeared to
correspond to a cultural metaphor of time in relation to the body. Although gestures along these axes seemed to
simulate processes, it is interesting to note the points made along the trajectory because they often corresponded to
algorithmic steps. Further, two participants moved their hands in counter direction (i.e., right-left) when referring to
programs having bugs or looping back to previous instructions.
Because the metaphors we identified refer to constructs and processes, we would expect some gestures to integrate
these metaphors to communicate a construct within a process. For example, four participants gestures illustrated segments of
code progressing in steps one after another, or input (as a bounded object) preceding a similarly bounded output. There
may well be other examples of integrated metaphors that may communicate computational constructs within a process
for example, how data are processed, or when sensors are triggered.
4.1 Implications for Instruction
We investigated the representational gestures used by individuals when explaining concepts but did not extend to
evaluating if and how such gestures supported learning. While some have proposed that such work can develop
understanding in science domains (Niebert et al., 2012), we agree that caution is needed in leaping to pedagogical
guidelines. However, it is worth looking other fields to consider how our findings may inform pedagogical research in
computing education.
4.1.1 Learners’ gestures (and actions)
While attending to learners’ gestures in a busy classroom is challenging, research has shown that teachers can assess
understanding by attending to gesture (Kelly et al., 2002). Therefore, the research in this paper provides an indication of
how it may be possible to train teachers to glean information about learning and understanding from their students’
gestures in computing. Teachers may use this additional source of information to guide their practice. For example,
mismatches between learners’ speech and gesture can often indicate readiness to learn from instruction (Church, 1999).
As well as observing individuals’ spontaneous use of gesture, there may be benefits from explicitly encouraging
gesturing by making implicit knowledge explicit (Broaders et al., 2007). It would be interesting therefore to examine
whether encouraging novice learners to use particular gestures, such as those identified in this paper, supports their
understanding and learning. Such support is significant when considering how some curricula, such England’s
Department of Education’s (2013) Computing Curriculum, expect children as young as 5 to learn concepts such as
algorithms. It would be worth investigating therefore whether encouraging children to use particular gestures, such as
those identified in this paper, could support learning
It is also important to consider other body-based actions beyond hand gestures. Rather than represent ideas just by
using the hands, it may be possible for learners, particularly younger learners, to act out various spatial representations,
for example, carrying out a sequence of steps using physical steps, and then arcing around to the start again to re-iterate
the steps. Indeed, such physical activity in computing education was previously discussed in relation to Computer
Science Unplugged (Bell et al., 2012). There is consequently much potential to link our work to ‘unplugged’ methods
for computing education by considering the congruency between certain embodied activities and the target computing
concepts (see Skulmowski & Rey, 2018). The contribution of this research therefore may be to encourage greater
reflection of the mappings between certain physical actions and the computing concepts being learnt, and the potential
to internalise these actions through gestures. As argued by Roth and Welzel (2001), gesture may be the bridge between
action experiences and scientific language.
4.1.2 Teachers’ gestures
Gesture research in other domains has demonstrated how teachers themselves frequently gesture, naturally adjust their
gestures to scaffold student understanding (Alibali & Nathan, 2007) and that their gestures can improve students’
understanding (Valenzeno et al., 2003). Our scan of online videos of expert explanations and observations in
computing science classes indicates that gestures are indeed common in this domain, although further research may
carry out more naturalistic work to evaluate such anecdotal evidence. The implications of the work in this paper are,
firstly, to encourage teachers to reflect upon their own gestures, and secondly, to suggest they may be more purposeful
in their gestures, possibly being trained to use particular gestures to support understanding, in a similar way to how they
may learn to use certain language or visual representations.
4.1.3 Educational materials
Educational software for computing draws upon many visual metaphors to support learners, most notably the use of
block-based coding environments as illustrated in Figure 1. An interesting question is whether such environments help
generate conceptual metaphors such as Computing Constructs as Physical Objects, or are designed to exploit these metaphors.
Arguably, there will be an interplay between computing environments and computing metaphors, in which metaphors
are used to design environments and then shape the metaphorical thinking of those using them.
The findings from this paper suggests that the design of educational materials in this domain may help learners by
exploiting conceptual metaphors. If so, this conceptual lens may be applied to inform further design guidelines, for
example, the spatial layout of conditional statements. The lens may also help identify possible conceptual challenges,
such as how to visually represent a dynamic representation of a loop.
4.1.4 Embodied Technologies
The capacity for emerging technologies to detect and respond to users’ direct, body-based, physical interaction, coupled
with increasing research indicating the role of the body in learning, has inspired a new design-based research field
exploring the potential of ‘embodied learning technologies’. According to Bakker, Antle and Van Den Hoven (2012), a
first step in designing a novel embodied learning technology is to identify the underlying embodied metaphors in the
target domain. While their work has focused on music, this approach to developing technologies based on prior
empirical work identifying embodied metaphors from gestures has been adopted by others in the fi eld; for example,
pioneering work by Abrahamson and colleagues (Abrahamson, 2007; Abrahamson & Trninic, 2011) has built on
gesture research to inform the development of gesture recognition technology to support learners’ concepts of ratio.
While several authors refer to embodiment in their design of innovative computing education applications (e.g., Daily,
Leonard, Jörg, Babu, & Gundersen, 2014; Leonard et al., 2015), embodiment seems to be interpreted in terms creating
programming sequences through an engaging body-based interface, rather than drawing upon knowledge of how
particular computing constructs may be physically represented.
The findings from this paper suggest two overarching ‘embodied metaphors’: Computing Constructs as Physical Objects and
Computing Processes as Motion along a Path. There are many ways such metaphors could be leveraged through different
technologies. Indeed, tangible programming blocks (Bers & Horn, 2009; Sipitakiat & Nusen, 2012) arguably achieve
this by using physical objects to represent different constructs (although they may be limited in representing different
constructs through different sizes and relationships, e.g., containership). It would also be possible to design gesture-
based technologies that enabled children to manipulate computing constructs through grasping/pinching gestures.
Such technologies can also be programmed to respond to whole body movements, gesturing a hand forwards to
execute the stages of a simple code, or gesturing a loop to create this programming instruction.
4.2 Limitations and future steps
This study was exploratory where it was not certain initially if students would reveal meaning through gesture or
whether patterns would emerge. Whilst the prevalence of representational gestures across participants validated the
aims of this research study, there are important limitations to note with regard to interpretation of meanings. Some of
these limitations can be addressed in future work.
The first limitation to highlight is the lack of other measures of conceptual understanding or procedural ability with
which to help interpret the gestures generated by participants. Consequently, there is no means with which to know if,
and how, certain gestures relate to ability. Such work is fundamental in evaluating the potential to leverage gestures to
assess and support learning. This is particularly the case where concepts comprise of multiple metaphors of increasing
complexity, as evident in other scientific concepts such as ‘energy’ (Close & Scherr, 2015). Further work should
therefore adopt a more developmental lens and examine how learners’ gestures evolve over time.
As well as examine gestures over time, further work should seek to expand the range and diversity of participants. The
sample for this study was relatively small (16) and heterogenous in gender and age and first language (although relatively
homogeneous in occupation (computing science students). Therefore, it is unclear from this this study whether
explanations and gestures generalise to a wider population, or what factors shaped representational thinking. Certainly, it
is important to challenge any assumption that younger learners draw upon, or even benefit from, the same metaphors
despite the focus on elementary concepts. As in other domains, it is possible that different metaphors serve to support
understanding at different levels.
It is also likely that patterns across participants reflect similar curricula, or the forms of tools and external representations
used, which will likely vary substantially across contexts . Similarly, gestures co-occurred with speech, hence it is likely
that there is a strong linguistic influence (English). Indeed, much care is needed in generalising findings using the
methodological approach of this paper as gesture meanings can differ significantly between cultures (Kendon, 1997).
However, whilst understanding these relationships was beyond the remit of this paper, the findings demonstrate the
value of such investigation in future work. Indeed, the patterns of gestures across participants in this study does suggest
the potential to investigate overlap and differences in shared meanings.
A further limitation of this study that should be considered from a theoretical perspective is the removal of relevant
stimuli in interviews to focus on offline thinking. In more naturalistic settings, cognition is likely to be distributed across
the environment (e.g., computer visualisations, drawings) or people (e.g., others’ gestures). It is therefore important for
further work to investigate how embodied meanings shape, and are shaped by, context. How, for example, do
visualisations (which are common in computing education) shape gesture production? Does the introduction of various
visual metaphors (e.g., block-based coding) shape conceptual thinking? The analysis of learners’ gestures during, and
after, interaction with different external representations can help address these questions.
It is also important to note that the work reported only focuses on three computing concepts, and whilst these concepts
were chosen to be representative of elementary computing concepts, they warrant critical reflection. One key point to
note is how two terms: ‘loops’ and ‘conditional statements’ have quite familiar meanings independently of computing
science, and it is likely that these terms have evolved clearer spatial representations than o ther computing terms. Yet,
data from the concept algorithm suggest that individuals will draw upon embodied experiences to conceptualise terms
even where spatial relationships are less clear. Further work should examine and compare a more comprehensive range
of concepts, and in doing so identify what factors influence the embodied nature of different computing concepts, and
the possible influence of linguistic meaning beyond the domain.
In summary, therefore, future work has much potential to address limitations of the reported study and adopt the
methodological lens to examine how gestures vary over time between learners with different cultural and educational
experiences. Such work can then provide a foundation to evaluate and inform the design of existing and novel
educational interventions. Although exploratory and limited to three concepts, the reported findings can also directly
inform future work which may examine how children and their teachers communicate their understanding of core
terms such as algorithm, how pedagogical approaches (bo th materials and teacher communicati on) devel op
understanding of these concepts, and the effectiveness of designed interventions – from informing teachers’ gesture to
novel learning technologies.
5. Conclusion
Existing research has examined the role and implications of Embodied Cognition for STEM focusing on mathematics
and science education. The contribution of this paper is to provide empirical evidence for the embodied nature of
computing concepts, and in doing so, draw attention to the potential of this line of investigation for this field. The paper
also demonstrates the potential to identify patterns and variation in how individuals express their understanding
through gesture. However, caution must be made in generalising the findings in light of participant sampling. Further
work may seek to extend this research with larger participant groups, as well as empirically examine the potential to
leverage embodied mechanisms in pedagogy and design.
As argued by Grover and Pea (2013), the socio-economic drive to accelerate computing education has created a much-
needed gap in theoretical understanding in this domain, a gap they believe can be addressed through Learning Sciences
research including in the field of Embodied Cognition. We hope this paper contributes to efforts addressing this gap.
6. Acknowledgements
We would like to thank the participants involved in this study and the Carnegie Trust for the Universities of Scotland
(Grant Ref: 70187) for funding the research. We would also like to thank the reviewers of the journal whose reviews
greatly contributed to the thinking and writing of this paper.
7. References
Abrahamson, D. (2007). From Gesture to Design: Building Cognitively Ergonomic Learning Tools. Paper presented at the Annual
meeting of the International Society for Gesture Studies, Evanston, IL: Northwestern University.
Abrahamson, D. (2009). Embodied design: Constructing means for constructing meaning. Educational studies in
mathematics, 70(1), 27-47. doi:10.1007/s 10649-008-9137-1
Abrahamson, D., & Trninic, D. (2011). Toward an Embodied-Interaction Design Framework for Mathematical Concepts. Paper
presented at the Proceedings of the 10th International Conference on Interaction Design and Children, Ann
Arbor, USA.
Alibali, M. W., & Nathan, M. J. (2007). Teachers' gestures as a means of scaffolding students' understanding: Evidence
from an early algebra lesson. In R. Goldman, R. Pea, B. J. Barron, & S. J. Derry (Eds.), Video research in the
learning sciences. (pp. 349–365). Mahwah, NJ: Erlbaum.
Alibali, M. W., & Nathan, M. J. (2011). Embodiment in mathematics teaching and learning: evidence from learners' and
teachers' gestures. Journal of the learning sciences, 21(2), 247-286. doi:10.1080/10508406.2011.611446
Amin, T. G. (2015). Conceptual Metaphor and the study of conceptual change: Research synthesis and future
directions. International Journal of Science Education, 37(5-6), 966-991. doi:10.1080/09500693.2015.1025313
Amin, T. G., Jeppsson, F., & Haglund, J. (2015). Conceptual Metaphor and Embodied Cognition in science learning:
introduction to special issue. International Journal of Science Education, 37(5-6), 745-758.
doi:10.1080/09500693.2015.1025245
Bakker, S., Antle, A. N., & Van Den Hoven, E. (2012). Embodied metaphors in tangible interaction design. Personal and
Ubiquitous Computing, 16(4), 433-449. doi:10.1007/s00779-011-0410-4
Barsalou, L. (1999). Perceptual symbol systems. Behavioral and Brain Sciences, 22(4), 577-660.
doi:10.1017/S0140525X99002149
Barsalou, L. (2008). Grounded cognition. Annual Review of Psychology, 59, 617-645.
doi:10.1146/annurev.psych.59.103006.093639
Bell, T. C., Lambert, L., & Marghitu, D. (2012). CS unplugged, outreach and CS kinesthetic activities. Paper presented at the
Proceedings of the 43rd ACM technical symposium on Computer Science Education (SIGCSE '12), New
York, NY, USA.
Bers, M., & Horn, M. S. (2009). Tangible programming in early childhood: Revisiting developmental assumptions
through new technologies. In I. R. Berson & M. J. Berson (Eds.), High-tech tots: Childhood in a digital world (pp.
49-70). Greenwich, CT: Information Age Publishing.
Broaders, S. C., Cook, S. W., Mitchell, Z., & Goldin-Meadow, S. (2007). Making children gesture brings out implicit
knowledge and leads to learning. Journal of Experimental Psychology-General, 136(4), 539-550. doi:10.1037/0096-
3445.136.4.539
Brookes, D. T., & Etkina, E. (2015). The importance of language in students' reasoning about heat in thermodynamic
processes. International Journal of Science Education, 37(5-6), 759-779. doi:10.1080/09500693.2015.1025246
Brown, N. C., Sentance, S., Crick, T., & Humphreys, S. (2014). Restart: The resurgence of computer science in UK
schools. ACM Transactions on Computing Education (TOCE), 14(2), 9. doi:10.1145/2602484
Church, R. (1999). Using the relationship between gesture and speech to capture transitions in learning. Child
Development, 14(2), 313-342. doi:10.1016/S0885-2014(99)00007-6
Cienki, A. (2005). Image schemas and gesture. From perception to meaning: Image schemas in cognitive linguistics, 29, 421-442.
doi:10.1515/9783110197532.5.421
Close, H. G., & Scherr, R. E. (2015). Enacting Conceptual Metaphor through Blending: Learning activities embodying
the substance metaphor for energy. International Journal of Science Education, 37(5-6), 839-866.
doi:10.1080/09500693.2015.1025307
Cooperrider, K., Gentner, D., & Goldin-Meadow, S. (2016). Gesture reveals spatial analogies during complex relational reasoning.
Paper presented at the 38th Annual Meeting of the Cognitive Science Society, Austin, TX.
Cortina, T. J. (2015). Reaching a broader population of students through unplugged activities. Communications of the
ACM, 58(3), 25-27. doi:10.1145/2723671
Daily, S. B., Leonard, A. E., Jörg, S., Babu, S., & Gundersen, K. (2014). Dancing alice: Exploring embodied pedagogical strategies
for learning computational thinking. Paper presented at the Proceedings of the 45th ACM technical symposium on
Computer science education, Atlanta, GA, USA.
Dehaene, S., Bossini, S., & Giraux, P. (1993). The mental representation of parity and number magnitude. Journal of
Experimental Psychology: General, 122(3), 371-396. doi:10.1037//0096-3445.122.3.371
Department for Education. (2013). Computing programmes of study: key stages 1 and 2. Retrieved from
https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/23903
3/PRIMARY_national_curriculum_-_Computing.pdf
Edwards, L. D. (2005). Metaphors and gestures in fraction talk. Paper presented at the 4th Congress of the European Society
for Research in Mathematics Education, Sant Feliu de Guixols, Spain.
Edwards, L. D. (2009). Gestures and conceptual integration in mathematical talk. Educational studies in mathematics, 70(2),
127-141. doi:10.1007/s10649-008-9124-6
Eisenberg, M. (2009). Embodiment as a Strategy for Mathematics Education. Paper presented at the 8th Interaction Design and
Children conference, Como, Italy.
Flevares, L. M., & Perry, M. (2001). How many do you see? The use of nonspoken representations in first-grade
mathematics lessons. Journal of Educational Psychology, 93(2), 330-345. doi:10.1037/0022-0663.93.2.330
Gentner, D. (1983). Structure-Mapping - a Theoretical Framework for Analogy. Cognitive Science, 7(2), 155-170.
doi:10.1016/S0364-0213(83)80009-3
Goldin-Meadow, S. (2009). How gesture promotes learning throughout childhood. Child Development Perspectives, 3(2),
106-111. doi:10.1111/j.1750-8606.2009.00088.x Ci
Goldin-Meadow, S. (2015). From action to abstraction: Gesture as a mechanism of change. Developmental Review, 38,
167-184. doi:10.1016/j.dr.2015.07.007
Goldin-Meadow, S., Cook, S. W., & Mitchell, Z. A. (2009). Gesturing gives children new ideas about math. Psychological
Science, 20(3), 267. doi:10.1111/j.1467-9280.2009.02297.x
Goldin-Meadow, S., Nusbaum, H., Kelly, S. D., & Wagner, S. (2001). Explaining math: Gesturing lightens the load.
Psychological Science, 12(6), 516-522. doi:10.1111/1467-9280.00395
Grover, S., & Pea, R. (2013). Computational Thinking in K–12 A Review of the State of the Field. Educational Researcher,
42(1), 38-43. doi:10.3102/0013189X12463051
Herbert, S., & Pierce, R. (2007). Video Evidence: What Gestures Tell us About Students’ Understanding of Rate of Change. Paper
presented at the Annual Conference of the Mathematics Education Research Group of Australasia,
MERGA.
Hostetter, A. B., Bieda, K., Alibali, M. W., Nathan, M. J., & Knuth, E. J. (2006). Don’t just tell them, show them! Teachers can
intentionally alter their instructional gestures. Paper presented at the Proceedings of the 28th Annual Conference of
the Cognitive Science Society.
Hui, T. H., & Umar, I. N. (2010). Metaphors and Pair Programming as a Constructivist Strategy in Computing Education: A
Literature Review. Paper presented at the Emerging Trends in Higher Education Learning and Teaching, Kuala
Lumpur, Malaysia.
Hutchins, E. (1995). Cognition in the Wild. Cambridge, Massachusetts: MIT press.
Iverson, J. M., & Goldin-Meadow, S. (1997). What's communication got to do with it? Gesture in children blind from
birth. Developmental Psychology, 33(3), 453-467. doi:10.1037//0012-1649.33.3.453
Jeppsson, F., Haglund, J., & Amin, T. G. (2015). Varying use of conceptual metaphors across levels of expertise in
thermodynamics. International Journal of Science Education, 37(5-6), 780-805. doi:10.1080/09500693.2015.1025247
Kelly, S. D., Singer, M., Hicks, J., & Goldin-Meadow, S. (2002). A helping hand in assessing children's knowledge:
instructing adults to attend to gesture. Cognition and Instruction, 20(1), 1-26. doi:10.1207/s1532690xci2001_1
Kendon, A. (1997). Gesture. Annual review of anthropology, 26(1), 109-128. doi:10.1146/annurev.anthro.26.1.109
Knuth, D. E. (1974). Computer science and its relation to mathematics. The American Mathematical Monthly, 81(4), 323-
343. doi:10.1300/J025v04n03_17
Kövecses, Z. (2008). Conceptual metaphor theory Some criticisms and alternative proposals. Annual review of cognitive
linguistics, 6(1), 168-184. doi:10.1075/arcl.6.08kov
Kubica, J. (2012). Computational Fairy Tales: CreateSpace Independent Publishing Platform.
Lakoff, G., & Johnson, M. (1980). Metaphors we live by (Vol. 111). University of Chicago Press: Chicago, IL.
Lakoff, G., & Johnson, M. (1999). Philosophy in the Flesh (Vol. 4): New york: Basic books.
Landy, D., Trninic, D., Soylu, F., Kehoe, J., & Fishwick, P. (2014). The Implications of Embodiment for Mathematics and
Computing Education. Paper presented at the The 36th annual meeting of the Cognitive Science Society, Quebec
City, Canada.
Leonard, A. E., Dsouza, N., Babu, S. V., Daily, S. B., Jörg, S., Waddell, C., . . . Boggs, K. (2015). Embodying and
programming a constellation of multimodal literacy practices: Computational thinking, creative movement,
biology, & virtual environment interactions. Journal of Language and Literacy Education, 11(2), 64-93.
Lindgren, R., Manches, A., Abrahamson, D., Price, S., Lee, V., & Tissenbaum, M. (2016). Embodiment and Designing
Learning Environments. Paper presented at the 12th International Conference of the Learning Sciences,
Singapore.
Lye, S. Y., & Koh, J. H. L. (2014). Review on teaching and learning of computational thinking through programming:
What is next for K-12? Computers in Human Behavior, 41, 51-61. doi:10.1016/j.chb.2014.09.012
McNeill, D. (1992). Hand and mind: what gestures reveal about thought. Chicago: University of Chicago Press.
Murphy, C. (2009). The Constructive role of the conceptual metaphor in children's arithmetic: a comparison and
contrast of Piagetian and embodied learning perspectives. Philosophy of Mathematics Education Journal(22).
Nemirovsky, R., Rasmussen, C., Sweeney, G., & Wawro, M. (2012). When the classroom floor becomes the complex
plane: Addition and multiplication as ways of bodily navigation. Journal of the learning sciences, 21(2), 287-323.
doi:10.1080/10508406.2011.611445
Niebert, K., Marsch, S., & Treagust, D. F. (2012). Understanding needs embodiment: A theory-guided reanalysis of the
role of metaphors and analogies in understanding science. Science Education, 96(5), 849-877.
doi:10.1002/sce.21026 Ci
Núñez, R. (2000). Mathematical Idea Analysis: What Embodied Cognitive Science Can Say about the Human Nature of Mathematics.
Paper presented at the Procedings of the Conference of the International Group for the Psychology of
Mathematics Education (PME), Hiroshima, Japan.
Núñez, R., & Lakoff, G. (2000). Where mathematics comes from. New York: Basic books.
Núñez, R., & Marghetis, T. (2014). Cognitive Linguistics and the Concept (s) of Number. In R. C. Kadosh &
A.Dowker (Eds.), Oxford Handbook of Numerical Cognition (pp. 377-401). Oxford: Oxford University Press.
Núñez, R., & Sweetser, E. (2006). With the future behind them: Convergent evidence from Aymara language and
gesture in the crosslinguistic comparison of spatial construals of time. Cognitive Science, 30(3), 401-450.
doi:10.1207/s15516709cog0000_62.
Parrill, F., & Sweetser, E. (2004). What we mean by meaning: Conceptual integration in gesture analysis and
transcription. Gesture, 4(2), 197-219. doi:10.1075/gest.4.2.05par
Pine, K., Lufkin, N., & Messer, D. (2004). More gestures than answers: Children learning about balance. Developmental
Psychology, 40(6), 1059-1067. doi:10.1037/0012-1649.40.6.1059
Pollard, S., & Duvall, R. C. (2006). Everything I needed to know about teaching I learned in kindergarten: bringing elementary education
techniques to undergraduate computer science classes. Paper presented at the ACM SIGCSE Bulletin.
Pouw, W. T., van Gog, T., & Paas, F. (2014). An embedded and embodied cognition review of instructional
manipulatives. Educational Psychology Review, 26(1), 51-72. doi:10.1007/s10648-014-9255-5
Román-González, M., Pérez-González, J.-C., & Jiménez-Fernández, C. (2017). Which cognitive abilities underlie
computational thinking? Criterion validity of the Computational Thinking Test. Computers in Human Behavior,
72, 678-691. doi:10.1016/j.chb.2016.08.047
Roth, W.-M., & Welzel, M. (2001). From activity to gestures and scientific language. Journal of Research in Science Teaching,
38(1), 103-136. doi:10.1002/1098-2736(200101)38:1
Scaife, M., & Rogers, Y. (1996). External cognition: How do graphical representations work? International Journal of
Human-Computer Studies, 45(2), 185-213. doi:10.1006/ijhc.1996.0048
Sipitakiat, A., & Nusen, N. (2012). Robo-Blocks: designing debugging abilities in a tangible programming system for early primary school
children. Paper presented at the 11th International Conference on Interaction Design and Children, Bremen,
Germany.
Skulmowski, A., & Rey, G. D. (2018). Embodied learning: introducing a taxonomy based on bodily engagement and
task integration. Cognitive Research: Principles and Implications, 3(1), 6. doi:10.1186/s41235-018-0092-9
Stacey, K., Helme, S., & Steinle, V. (2001). Confusions between decimals, fractions and negative numbers: A consequence of the mirror as
a conceptual metaphor in three different ways. Paper presented at the Proceedings of the 25th Conference of the
International Group for the Psychology of Mathematics Education, Utrecht, Netherlands.
Valenzeno, L., Alibali, M. W., & Klatzky, R. (2003). Teachers' gestures facilitate students' learning: A lesson in
symmetry. Contemporary Educational Psychology, 28(2), 187-204.
Varela, F. J., Thompson, E., & Rosch, E. (2017). The embodied mind: Cognitive science and human experience: MIT press.
Wilson, M. (2002). Six views of embodied cognition. Psychonomic Bulletin & Review, 9(4), 625-636.
doi:10.1080/10508406.2011.553259
Wing, J. M. (2008). Computational thinking and thinking about computing. Philosophical Transactions of the Royal Society A:
Mathematical, Physical and Engineering Sciences, 366(1881), 3717-3725. doi:10.1098/rsta.2008.0118
Wittenburg, P., Brugman, H., Russel, A., Klassmann, A., & Sloetjes, H. (2006). Elan: a professional framework for
multimodality research. Paper presented at the 5th International Conference on Language Resources and
Evaluation., Paris, France.
