Dance as Embodied Analogy – Designing an Empirical Research Study

There is a growing body of research into concepts of embodiment and embodied cognition but little attention has been paid to the rich well of knowledge permeating dance practices. One particular aspect that can shine light on how our bodies and minds are integrated relates to the widespread use of ‘mental imagery’ in dance. The meaning making and comprehension involved in the interplay between the concepts underpinning mental imagery and their embodiment in dance can be understood in terms of cross-modal analogising. This is the hypothesis at the heart of the research study outlined in this article.

This is an independent research project, in affiliation with the Multimodal Language and Cognition group, Radboud University, located at the Max Planck Institute for Psycholinguistics (MPI) in Nijmegen, The Netherlands. It is framed within a linguistics and (embodied) cognitive psychology research context. My background is in contemporary/Post Modern dance, including performance, education, community dance and research. I have been intrigued about the relationship between bodily movement and ‘meaning’ for thirty-five years and am now bringing my inter-related interests together.

Themes and context
Hofstadter and Sander (2013) argue that analogy is the core of conceptualisation, that we understand new experiences and build a personal understanding of the world through comparison with past experiences. This is a constant, ongoing process and can occur across sensory and communicative modalities, and contextual domains. For example, when we observe a field of wild flowers swaying in the breeze we can paint a picture of what we see, we can imagine what it feels like to ‘be the flowers’ and we can dance as if we are the field. None of these interpretations are the thing itself, they are all forms of analogy, each echoing pertinent features of the source image.
This research is framed within the fast growing fields of grounded and embodied cognition, which argue that bodily experiences play a significant role in concept formation and acts of comprehension (see Barsalou, 2008). As humans we have an innate ability to explore and understand our environment through bodily experiences, which enables us to construct concepts, categories and interwoven bodies of knowledge.

Imagery use in dance is an example of this process, in which our conceptualisation of the world around us informs our bodies, which in turn alters and expands our understanding of the world (see figure 1). Investigating the physical, embodied analogising that occurs in dance can provide valuable insight into the body’s role in how we think.

The proposed research has two strands: 1: Movement to Interpretation; 2: Concepts to Embodiment (Movement). In the former, experiments will use videoed dance phrases as stimuli to elicit responses in linguistic and gestural modalities. The latter will investigate the ways that concepts, presented in various modalities are embodied in movement. This paper focuses on the design of Strand 1.

Useful concepts: analogies, categorisation, gestalts and framing
Analogy involves comparing two or more ‘entities’, identifying similarities and shared characteristics at multiple levels, which is a process known as structure-mapping (Gentner, 1983). Shared surface characteristics such as colour, shape and size are mapped quickly but are by nature superficial and tell us little about the deeper nature of either entity. An example of this would be embodying a tree by standing virtually still with arms spread, wearing a brown and green costume.

Effective and informative analogies work at a more complex level in which elements ‘hold like roles’ (Gentner & Linsey, 2013, p.4). For example: a rocket booster provides the energy for launch; a plié provides the energy for a vertical jump. Viewing these entities (rocket booster and plié) through the lens of analogy, we can say that their surface characteristics are largely dissimilar – one is cylindrical and made of metal whereas the other is bi-pedal, made of flesh, muscle and bone. Their similarities lie at the relational level in that both provide the energy for launch. The similarity between them is in the role that they play in ensuring “take-off”; the relationship that they have with propelling an object off the ground.

Analogising goes beyond simple comparisons to reveal inferences. That is, ‘because this ‘happens’ with A, perhaps the same is true for B’. For example, from knowing that vertical alignment and more fuel will facilitate a powerful and accurate launch due to the increased downward thrust, we can infer that vertical alignment and a deeper plié will improve jump height. The complex reasoning involved in producing inferences appears to be able to take place subconsciously.

Analogies including their sub-set, metaphors, can often be considered as having a source and target domain. The source is the entity to which the target is compared or which inspires the production of an analogy. The roles are not innate but depend on context. Taking the example above, ‘the rocket’ (engineering domain) can be a source image to inform the jump (human movement domain) or ‘the jump’ can inspire the rocket interpretation and our understanding of it.

Analogies are deeply entwined with categorisation. Something will only be considered to be part of a category if it shares significant characteristics with, that is, it is analogous to, other members of that category. Almost every object or act that we perceive can be categorised in multiple ways. However, this does not mean that anything can be categorised in any way. The attributes and their relationships to each other underpin their potential categorisations even if these are not explicitly recognised.

To clarify the idea of multi-layered categorisation, consider Nijinsky’s exit in Le Spectre de la Rose (Fokine, 1911). This can be perceived as a spirit escaping into infinity, a leap through a window, a dance move in a performance, a combination of ballet steps, contraction and release of muscle fibres triggered by electrical signals in nerve cells and so forth. None of these are more or less correct than the others and their value is dependent on the context.

A dancer attempting to perform this complex action could use many different conceptualisations to enhance their performance. Once the mechanics of the action have been mastered, it is often poetic or metaphorical gestalt1 images and concepts that are the most valuable. Examples of the use and impact of imagery abound within dance education, performance and interpretation (Fisher, in press). It is this fairly broad, gestalt level of categorisation that is the focus of this study. Whilst each person’s conceptual world is complex and unique, there is significant overlap between people, based on shared knowledge and contexts, which allows communication and mutual understanding to occur.

Framing refers to the ways that we interact with the world. According to Tannen, ‘on the basis of one’s experiences of the world in a given culture…, one organises knowledge about the world and uses this knowledge to predict interpretations and relationships regarding… information, events and experiences.’ (1993 p.16). Each of us is able to frame our experiences in a variety of ways. By investigating how gestalt concepts are understood in and through bodily movement, and vice versa, this study frames dance movement as embodied analogy.

Sign Language, (co-speech) gesture and the notion of iconicity
Iconicity is the similarity between a sign and the act or object to which it refers. Verbs in sign languages often look like the action to which they refer such as WALKING, using fingers to represent legs. The hands sometimes re-enact their roles such as holding a TOOTHBRUSH. Alternatively, body parts often become (part of) the object that they represent such as SCISSORS or WINGS (BIRD). Iconicity in sign language and gesture sometimes involves drawing or moulding the shape or other physical characteristics of an object (such as the shape of a HOUSE) (Ortega, Sümer & Özyürek et al., 2016; Stefanini, Bello, Caselli, Iverson & Volterra, 2009).
It has been proposed that the relationship between iconic signs (in sign languages) or gestures and the concept to which they refer, is best understood in terms of analogical structure-mapping (Emmorey, 2014; Taub, 2001). Sign languages have a communicative purpose and therefore there is a bias towards efficiency. This is manifest in simple mapping, for example BIRD is communicated either through a finger and thumb representation of the beak (in American Sign Language) or hands flapping at the side of the body, like wings (in Turkish Sign Language) (Emmorey, 2014, p.2). In both cases the embodiment uses a fairly superficial, schematised model. The relationship between concept and embodiment in dance is more diverse and multi-faceted but methodologies used in multi-modal linguistic research are highly pertinent.

Sensorimotor neuronal involvement in “thinking” about concepts
Research in recent decades has shown that neurones in muscles, sensorimotor and pre-motor regions in the brain are activated when we imagine concepts and past experiences related to movement and bodily sensation (Iachini, 2011). This gives empirical support to the embodied cognition theory that our bodily experiences are intrinsically related to both establishing conceptual knowledge and ongoing thought processes (Barsalou, 1999, 2008; Schmidt & Lee, 2011; Brunel, Vallet, Riou & Rey, 2016).

These same or related sensorimotor brain regions are triggered when observing the movements of others. The cells involved are known as Mirror Neurons and they form part of the Action Observation Network2 (Zentgraf, 2011). This system has been shown to be active in Kinesthetic Empathy, the internal simulation of movement sensations during movement observation (Hagendoorn, 2004; Watching Dance website). Greater familiarity with particular movement patterns results in stronger activation (Calvo-Merino, Glaser, Grèzes, Passingham & Haggard, 2005, 2006).
Further research using linguistic priming techniques, found that sensorimotor re-activation also occurs in response to lexical items (words), which are semantically associated with the concepts (Pulvermüller, 2005; Lam, 2015, 2016; Willems & Hagoort, 2007; Willems, Ozyurek & Hagoort, 2006). Also, words referring to concepts with shared action features prime each other (Myung, Blumstein, & Sedivy, 2006). These findings point towards strong links between bodily movement and conceptualisation.

Overview of the research project
This research study bridges scientific and artistic domains and is therefore open to criticism from multiple directions – artists may contend that experimental materials are so manipulated that they lose the artistic essence, whilst scientists are apt to criticise materials for having too many variables. The aim in designing this study is to use ‘real-world’ dance material in a way that employs controls to minimise undesired variables.

Hypothesis: The embodiment of gestalt concepts in dance movement can be understood in terms of analogising.

The research project has two strands: movement to interpretation and concepts to embodiment. The second strand presents significant challenges for empirical research, most notably related to methods of data collection and analysis. This strand is not discussed explicitly in this paper.

Movement to interpretation
The process of analogising from movement (observation) to conceptual interpretations, communicated through spoken language can be characterised as follows: Dance movement is perceived through visual senses, which triggers activation in the motor regions of the brain (within the Action Observation Network). This is comprehended via comparison with prior knowledge of both movement patterns and associations with other ‘things’ with a range of shared characteristics. The observer is thus able to categorise the observed movement in numerous, simultaneously co-existing ways which are then expressed through symbolic signs in speech (figure 2).

Research questions and related behavioural experimental paradigms3 include:
RQ1a: Are observed dance movements mapped to primed concepts across domains? RQ1b: Are dancers more sensitive to embodiment of concepts than non-dancers?
Expt. 1: Categorisation by forced choice (expanded in Part 3);

RQ2a: Do specific dance stimuli elicit an overlapping range of interpretations that demonstrate significant characteristics of analogy?
RQ2b: What are the dominant features that inform interpretations?
RQ2c: Do dancers produce more complex, extensive and/or sophisticated interpretations than non-dancers?
Expt. 2.1: Free description of controlled stimuli (dance videos), in verbal and co-speech gesture;
Expt. 2.2: Structured descriptions (“This movement is like…”); Data from these experiments will be analysed using structural mapping techniques.
Expt. 2.3: rating the validity of lexical items (words) produced in experiments 2.1 & 2.2, as descriptors for the stimuli materials (Perry, Perlman, & Lupyan, 2015);4

RQ3: Do observed dance movements prime related concepts?
Expt. 3: cross-modal primed word/non-word semantic decision task (Yap et al., 2011);

RQ4: Can the qualitative features of dance performance (as shaped by gestalt imagery) subconsciously influence categorisation choices?
Expt. 4: Sorting/categorisation task (Malt et al., 2014).5

Design of the stimulus materials The core stimulus materials to be used across these experiments, consist of video recordings of two movement phrases (AA and BB), choreographed by the author. Each of these is comprised of six sub-phrases (1-6). Each phrase is approximately 13.5 to 15 seconds long and shares significant characteristics of action, space and temporal features. The sub-phrases are also recombined to produce two further phrases (AB and BA). In this way, the vocabulary is limited whilst maximising the overall number of distinct phrases6:

phrase AA: A, A, A, A, A, A;
phrase BB: B, B, B, B, B, B;
phrase AB: A, B, A, B, A, B;
phrase BA: B, A, B, A, B, A.

The phrases were choreographed with the following goals in mind: to use modern dance vocabulary; to limit use of conventionalised, easily “nameable” vocabulary; to allow for variations in performance quality, arising from the application of imagery.
Spatial features: The phrases follow the same basic floorplan (see figure 3). In general, the movements take place within a within a relaxed kinesphere and a mid-height range.

Temporal and dynamic features: all the basic sequences use a steady breath based rhythm and have one relatively fast and sharp sub-phrase.

The rationale behind restricting the vocabulary, spatial and temporal features of the basic sequences, was to ensure that the effect of imagery on the movement material was the most significant and distinguishing feature. For example, if one basic sequence included large jumps or was dynamically more varied than the others, that might bias observers towards a particular, non-neutral interpretation.

The variations: each movement phrase was videoed in three variations – NEUTRAL7 (as choreographed, without reference to specific imagery); WATER; and INSECT. The initial instructions to the dancer were broad, to inspire an instinctive response. During the preparation process, the following more specific images emerged:

WATER: ‘Moving as if both part of and surrounded by a warm, gently moving body of water’. Breath flow also influenced the movement quality. Additional descriptors used during filming included ‘buoyant’, ‘floating’ and ‘supported’.

INSECT: A large, black and white beetle with distinctive antennae (see figure 4: Asian long-horned beetle). The dancer referred to fast, sudden movements, lack of emotion and a clicking sound.

Refining instructions were given to tweak the dancer’s movements so that the overall spatial and temporal structure remained consistent. In the stimulus videos, the dancer’s face has been blurred to avoid the influence of and attention to facial expressions.

Rationale for image selection
This study is investigating whether or not a structural mapping approach can be applied to the relationship between representations of concepts in different modalities (verbal language and human movement). It is important that the target concepts (images) are significantly distinct from one another, so that their embodiment would also be distinct. Just as an INSECT would not be categorised as a type of WATER, dance movement which is informed by the concept of INSECT should not be interpreted as embodying the idea of WATER. This lack of overlap is aided by the two concepts coming from different contextual domains, that is, Natural Phenomena and Living Creatures. Both images have rich movement potential and powerful metaphorical implications.

Figure 5 (Comparison of characteristics) provides a general comparison between the concepts of WATER and INSECT, as they were used in making the stimuli materials. This is provided as a non-comprehensive guide to illustrate the distinctions between them.

Due to the scope of the initial project, the metaphorical associations of the two images, such as flowing water representing a journey through life, will not be discussed in detail here but may be revisited at a later date as they promise a rich layer of insight into complex aspects of human meaning making.

Experiment 1: forced choice study design
This section focuses on the design of the first experiment – Movement interpretation: responses to dance phrases using a forced choice paradigm.
The research question is: Are observed dance movements mapped to primed concepts across modalities? This requires that the concepts are embodied in perceivable ways that correspond (map) to recognisably similar categories in the minds of the observers.

The following design was used for a pilot study with eight volunteers at the MPI. The experiment is programmed using Presentation®8 and presented via video projection. Introduction and instruction screens are followed by three blocks of all twelve stimuli videos in a randomised order. Each video is preceded by a fixation cross, to engage attention and is followed by three descriptors presented as follows:
This dance phrase is most like:

An insect
Water9
Neither of these
Please circle your selection on the paper provided.

The order of the response options is randomised across videos. These descriptors act to prime the participants into expecting that the stimuli may be reminiscent of one of the concepts listed. As with other linguistic priming experiments (see McNamara, 2004 for a review) it is anticipated that this will trigger mental representations against which the dance phrases will be compared.

Each block of videos is separated by a distractor slide showing a complex geometrical pattern, to reset the participants’ visual field. At the end of the experiment, participants were asked:

What factors influenced your choices?
What other images/association (if any) did you think of?
Other comments?

Background information about the participants was collected after the experiment, including experience of watching and participating in Modern and other dance forms. The experiment was followed by a debriefing session explaining the experimental hypothesis and design, leading into a general discussion.

It is predicted that there will be a high degree of matching between the video phrases and the congruent target word. That is WATER videos: ‘Water’; INSECT videos: ‘An Insect’; NEUTRAL videos: ‘Neither of these’. The NEUTRAL stimuli are likely to elicit the broadest range of responses as observers are asked to select the image/concept that the dance is most like, biasing them towards a positive choice over a generic, negative one.

There may also be bias towards more-or-less even response selections and attempts try to match answers with previously ones based on in-task experience or learning. Controls are employed in the instructions and general design to minimise these effects.

For this fairly closed experimental design, distinctions between dancers and non-dancers (the two subject groups) are expected to be minimal although those more attuned to dance may be more sensitive to variations.

The supporting questions will provide insight into the dimensions that observers are most aware of influencing their interpretations. A dance cohort may offer more detailed and/or varied feedback resulting in part from their prior exposure to dance specific vocabulary and analytical approaches.10

Results and discussion
The participants in the pilot study were self-selected and many had extensive knowledge of relevant research methodologies. Some participants had some experience with Modern and/or Latin dance as both participants and observers. None had training in dance analysis.
Initial analysis indicates highly significant effects at this exploratory level.

The ‘Water’ descriptor was selected 79% of the time, for the WATER stimuli, which is highly significant. Similarly, the INSECT stimuli garnered 74% ‘Insect’ responses. The NEUTRAL videos drew a much more varied response as predicted – ‘Neither’ and ‘Water’ were chosen just over one third of the time and ‘Insect’ just under.

Although ‘Water’ was the most highly selected response overall at 41%, it was very rarely chosen in relation to INSECT stimuli. The ‘Insect’ descriptor had the fewest false positives, especially for the WATER stimuli. Individual participants’ accuracy rates for WATER and INSECT stimuli ranged between 59-100% which is substantially above chance (33%)11. All these factors point to clear differentiation in embodied imagery between the stimuli, verifying their validity as experimental resources.
The factors which influenced choices were dynamic qualities of speed (slow = water; fast = insect), smoothness, flow, staccato/abruptness, force, power; spatial features included angularity, hand shapes and the extent of high-low shifts. Most people did not think of other imagery because their focus was on the primed concepts. Other associations which were noted included wild animals such as ‘languid movement of a big cat’, a horse, a caterpillar, a small stream and wind (in the trees).

Participants found the dance phrases to be a little long and many commented that they based their decisions on the opening movements. The question arose as to whether the focus is on perceived imagery, which is dependent on the viewers’ conceptualisations or on interpreting the dancer’s intent. Refinements will be made to the instructions to clarify these points and guide participant expectations more effectively before running the full experiment with two distinct participant groups consisting of approximately twenty dance and twenty non-dance students.

Conclusion
This article has provided an overview of the development of an ongoing research study into the notion of dance as embodied analogy, using methodologies and associated theories that are well developed in psycholinguistics, gesture and other fields of embodiment research. The results from the pilot study suggest that dance movements can be perceived in ways that map to primed concepts in the minds of observers. It is hoped that findings from the initial experiments will provide a solid foundation for rich and fruitful research which will shed light on the relationship between embodiment and conceptualisation.

***

This article is based on: Dance as Embodied Analogy – designing an empirical research study This is an independent research project being run in affiliation with the Multimodal Language and Cognition group, which is part of the Centre for Language Sciences at Radboud University, located at the Max Planck Institute for Psycholinguistics in Nijmegen.

***

1 A gestalt is an organised whole that is perceived as more than the sum of its parts.
2 An alternative term is the Action Perception Networks which has been proposed by Pulvermuller (2005) in placing action at the centre of meaning formation.
3 References are to studies employing similar methodologies
4 Participants will be asked to rate the suitability / relevance of frequent descriptors offered by participants in the free and structured description tasks. For example: “To what extent is this dance phrase like LAVA?”
5 Malt, Armeel, Imai, Gennari & Majid (2014) used categorisation and naming tasks to draw out underlying analogistic features related to human vertical locomotion, within and across four languages. Using complex multifaceted analysis, the researchers found that there was a high degree of consistency within and across cultural-linguistic groups. Within the experimental framework analogous biomechanical features (primarily ‘pendulum’ versus ‘impact-rebound’ actions) were found to be the primary factor in shaping responses. The stimuli design was focused on regular, forward locomotion and in this experiment, effort, function and goals related to the activity were not found to be significant categorising factors. Similar non-linguistic sorting tasks have been used extensively to investigate distinctions between universal and culturally established (relativist) categorisation patterns in relation to phenomena such as colour boundaries (Roberson, Davies, Corbett and Vandervyver, 2005)
6 Further combinations such as AAABBB are possible but it was felt that the overlapping chunks would be too large to maintain the illusion of variety that is desired
7 The stimuli materials will be indicated by the use of UPPER-CASE letters. Descriptor terms will appear ‘in single inverted commas’; concepts (also known as interpretants or signified) will be denoted by an initial Upper-case letter
8 Presentation® experiment control software is commonly used to run experiments. It enables the integration of stimulus files and can collect data from sources such as button-press, eye-tracking and EEG (electroencephalogram)
9 Although the broad concept of Water may evoke a wide range of concepts (such as crashing waves or calm lake), after much discussion it was decided to stick with the broad, generic descriptor which was used to create the stimuli materials. The structure of the experiment is such that the stimuli materials are likely to influence the participants’ attention towards relevant associations.
10 It is anticipated that during the ‘free description’ and ‘structured description’ experiments, dancers responses will be significantly more detailed than non-dancers but may prioritise description of characteristics and attributes over broader ‘interpretations’ as per the ‘Description, Interpretation, Evaluation’ model of Janet Adshead (1988).
11 Excluding one participant who rarely selected ‘Water’ or ‘Insect’ but was 100% accurate when they did.

Sources

• Adshead, J.(1988). Dance Analysis: Theory and Practice. Surrey, UK: Dance Books Limited
• Barsalou, L.(1999). Perceptual Motor Systems. Behavioural Brain Science, 22, pp.577-660
• Barsalou, L.(2008). Grounded cognition. Annual Review of Psychology, 59, pp.617-645
• Brunel, L., G.T. Vallet, B. Riou, A. Rey and R. Versace.(2016). Conceptual and interactive embodiment: Emergence from sensorimotor interactions. In: Fischer, M.H. and Y. Coello (eds). Foundations of Embodied Cognition, volume 2. Abingdon, UK: Routledge, pp.108-124
• Calvo-Merino, B., J. Grèzes, D.E. Glaser, R.E. Passingham and P. Haggard.(2006). Seeing or doing? Influence of visual and motor familiarity in action observation. Curr Biol, 16(19), pp.1905-1910. doi: 10.1016/j.cub.2006.07.065
• Calvo-Merino, B., D.E. Glaser, J. Grèzes, R.E. Passingham and P. Haggard.(2005). Action observation and acquired motor skills: an FMRI study with expert dancers. Cereb Cortex, 15(8), pp.1243-1249. doi: 10.1093/cercor/bhi007
• Emmorey, K.(2014). Iconicity as structure mapping. Phil. Trans. R. Soc. B 369: 20130301
• Fisher, V.J. (In press). Unfurling the wings of flight: Clarifying ‘the what’ and ‘the why’ of mental imagery use in dance. Research in Dance Education.
• Gentner, D.(1983). Structure-Mapping: A Theoretical Framework for Analogy. Cognitive Science 7, pp.155-170
• Gentner, D. and L.A. Smith.(2013). Analogical learning and reasoning. In: D. Reisberg (ed). The Oxford handbook of Cognitive Psychology. New York, NY: Oxford University Press, pp. 668-681
• Hagendoorn, I.G.(2004). Some Speculative Hypotheses About the Nature and Perception of Dance and Choreography. Journal of Consciousness Studies 11(3-4), pp.79-110
• Hofstadter, D. and E. Sander.(2013). Surfaces and Essences: Analogy as the fuel and fire of thinking. New York: Basic Books
• Iachini, T.(2011). Mental Imagery and Embodied Cognition: a Multimodal Approach. Journal of Mental Imagery 35(3-4)
• Lam, K.J.Y., T. Dijkstra and S-A. Rueschemeyer.(2015). Feature activation during word recognition: action, visual, and associative-semantic priming effects. Frontiers in Psychology11, 6. https://doi.org/10.3389/fpsyg.2015.00659
• Lam, K.J.Y.(2016). Understanding action-related language: sensorimotor contributions to meaning. Ph.D. thesis. Nijmegen: Radboud University
• Malt, B.C., E. Armeel, M. Imai, S.P. Gennari and A. Majid.(2014). Human locomotion in languages: Constraints on moving and meaning. Journal of Memory and Language, 74, pp.107-123
• McNamara, T.P.(2005). Semantic Priming: Perspectives from Memory and Word Recognition. New York: Psychology Press
• Myung, J-Y., S.E. Blumstein and J.C. Sedivy.(2006). Playing on the typewriter, typing on the piano: manipulation knowledge of objects. Cognition, 98(3), pp.223-243
• Ortega, G., B. Sümer and A. Özyürek.(2016). Type of Iconicity Matters in the Vocabulary Development of Signing Children. Developmental Psychology (Advance online publication). doi.org/10.1037/dev0000161
• Oxford Living Dictionaries (n.d.) Gestalt. English: Oxford Living Dictionaries. [online]. [Accessed 16 February 2017] Available from World Wide Web:
• Perry, L.K., M. Perlman and G. Lupyan.(2015). Iconicity in English and Spanish and Its Relation to Lexical Category and Age of Acquisition. PLOS One, Sept 4, 10(9). doi.org/10.1371/journal.pone.0137147
• Pulvermüller, F.(2005). Brain mechanisms linking language and action. Nature Reviews Neuroscience, 6(7), pp.576-582
• Roberson, D., I.R.L. Davies, G.G. Corbett and M. Vandervyver.(2005). Free-Sorting of Colors Across Cultures: Are there Universal Grounds for Grouping? Journal of Cognition and Culture, 5(3), pp.349–386
• Schmidt, R. and T. Lee.(2011). Motor Control and Learning: A Behavioral Emphasis, 5th ed. Champaign, IL: Human Kinetics
• Stefanini, S., A. Bello, M.C. Caselli, J.M. Iverson and V. Volterra.(2009). Co-speech gestures in a naming task: Developmental data. Language and Cognitive Processes, 24, pp.168–189. doi.org/10.1080/01690960802187755
• Taub, S.(2001). Language from the Body: Iconicity and Metaphor in American Sign Language. Cambridge: Cambridge University Press
• Watching Dance (© 2016) What is Kinesthetic Empathy? The Watching Dance Website. [online]. [Accessed 10th October 2016] Available from World Wide Web:
• Willems, R. M. and P. Hagoort.(2007). Neural evidence for the interplay between language, gesture and action: A review. Brain and Language, 101(3), pp.278-298
• Willems, RM., A. Ozyurek and P. Hagoort.(2006). When Language Meets Action: the Neural Integration of Gesture and Speech. Cerebral Cortex, 17(10), pp. 2322-2333
• Yap, DF, W.C. So, M. Yap, Y.Q. Tan and R.S. Teoh.(2011). Iconic gestures prime words. Cognitive Science 35, pp.171–183. doi:10.1111/j.1551-6709.2010.01141.x
• Zentgraf, K., J. Munzert, M. Bischoff and R.D. Newman-Norlund.(2011). Simulation during observation of human actions – Theories, empirical studies, applications. Vision Research, 51, pp.827-835
• Asian long-horned beetle: Google image, source unknown Brain cortex: (undated) Cortex sensorimoteur1.jpg: Pancratderivative work: Iamozy - Own work, This file was derived from: Cortex sensorimoteur1.jpg: CC BY-SA 3.0. [Accessed 20th November 2016]. Available from: Dancer: (2016) photo by Jan-Jaap Zeydner (used with permission) Eye image: (2016) by Ida Visser (aged 13, used with permission) Speech image: (2017) Vicky Fisher