The idea that certain postures and movements are associated with specific emotions is not new: The concept of expressing emotions through body language dates as far back as Aristotle (Lee, 2008). While it is widely accepted that emotions are expressed in movement and posture, evidence suggests that the connection between movement and emotion is bidirectional: that we can affect emotional state by changing posture and movement. This concept originates in Darwin's ideas (Darwin, 1882) and the James-Lang theory (James, 1884), which postulate that bodily responses to stimuli are necessary for emotional experience, and therefore feelings are not the causes of autonomic system activation and emotional behavior, but rather are the consequence of them.

Subsequent theorists in the field of emotion followed this idea, proposing that sensory feedback from facial and postural movements contributes significantly to emotional experience (Tomkins, 1962; Laird, 1974; Izard, 1993). In recent years, this theory has been re-formulated in neurophysiological terms by Antonio Damasio, who asserted that emotions are evoked by interoceptive and proprioceptive feedback from the body and our conscious feelings result from our perception of this somatic input (Damasio et al., 2000; Damasio and Carvalho, 2013). The uncovering of neuronal underpinnings of interoception (Craig, 2002; Critchley, 2005) and identification of the anterior insular cortex as the brain region in which representation of internal bodily states becomes available to conscious awareness (Craig, 2009; Harrison et al., 2010), provide plausible neurocircuits in support of this hypothesis (Critchley and Harrison, 2013). Evidence suggesting that the effects of facial expressions on affective state are attained through proprioception comes from mimicry and botulin toxin studies: Mimicry, which involves facial muscle contraction, more strongly activated brain emotional processing regions compared to mere observation of facial expression (Carr et al., 2003), and mimicry of emotional but not ingestive (chewing, licking) facial expressions activated brain emotional processing regions, where the magnitude of facial movement predicted responses within the right insula (Lee et al., 2006). Reduced muscle activation during emotional facial expressions following botulin toxin treatment and corresponding reduction in proprioceptive feedback, attenuated neural activation in the amygdala (Hennenlotter et al., 2009) and weakened emotional experience (Davis et al., 2010). Although fMRI data cannot be collected during whole-body emotionally expressive movements, observation of such movements has been shown to activate emotional processing regions (de Gelder et al., 2004; Pichon et al., 2008) and to modify affective state (Shafir et al., 2013). Based on the embodied simulation theory (Niedenthal et al., 2010; Gallese and Sinigaglia, 2011), movement-related brain activation during motor execution is similar to that during motor observation of the same movements. Thus, brain activation during motor observation may imply what happens during motor execution. Motor observation creates neuronal activity in the observer's motor system which is likely related to simulation of the observed movement (Borgomaneri et al., 2012), and sensory cortex activation during movement observation (Gazzola and Keysers, 2009) indicates that this simulation probably also includes simulation of the expected proprioceptive feedback from the simulated movement (Valchev et al., 2016), which probably elicits the feelings associated with those emotional movements (Bastiaansen et al., 2009). People tend to perceive certain movements as expressing specific emotions (Dael et al., 2012; Kleinsmith and Bianchi-Berthouze, 2013), which suggests the existence of associations in our brain between certain movements and specific emotions. Thus, similar to what happens during observation of bodily emotional expressions, the proprioception from emotional movements during execution of such expressions probably activates emotional processing regions to enhance the associated feelings.

Damasio's theory implies that by deliberately choosing our motor behavior, we can affect our feelings. This was stated by Riskind as early as 1984 (Riskind, 1984) and demonstrated by subsequent studies (for review, see Shafir, 2015). While these studies used specific, scripted movements to demonstrate the effects of movement on emotions, in life, people use a variety of movements to express their emotions. Accordingly, Shafir et al. (2016) investigated complex, non-scripted improvised movements, using Laban Movement Analysis (LMA) (a comprehensive method for describing and documenting human movement) and identified unique sets of Laban movement components which can elicit the feelings of happiness, sadness, fear, or anger. These findings concur with the motor components of scripted movements and postures studied in Duclos et al. (1989), Flack et al. (1999), Duclos and Laird (2001), Koch et al. (2007), Shafir et al. (2013) and other studies. While the associations between emotions and specific motor components have been mostly used in the past for diagnosis or for emotion recognition, we are the first, to the best of our knowledge, to suggest specific techniques for using these associations for emotion regulation.

Dance movement (psycho)therapy (DMT) has long relied upon movement to both facilitate emotional expression and change emotional state. Movement responses evoked by DMT have potential to help people cope more effectively with their external world (Smallwood, 1978) and they positively affect mood and wellbeing (Koch et al., 2014). One established DMT method, developed by Blanche Evan, stimulates emotional transformation and psychological growth by teaching new movement qualities to expand movement vocabulary, and by exploring, through movement improvisation, personally meaningful images, ideas, statements, or new responses to difficult situations, to affect emotional state and increase resiliency (Bernstein, 1995). Evan's interventions are based on her psychophysical concept that (1) all experiences occur concurrently on psychological and physical (bodily) levels; and (2) the body impacts the psyche as the psyche impacts the body. This psychophysical concept is consistent with Damasio's theories. The goal underlying DMT movement suggestions is to facilitate moving in new ways not regularly used, to support emotional shift.

If feedback from movement affects emotional state, then reducing the frequency of Laban motor components shown to enhance unwanted emotional state may also empower emotional resiliency. Established somatic movement techniques for moment-to-moment movement awareness, such as the Feldenkrais and Alexander Techniques, empower self-modification of daily movement, and thus could be used to affect emotional state, for example, by learning to avoid a slouched posture through Alexander Technique. Increased body awareness to one's daily posture and movements through increased proprioceptive awareness is attained also through Yoga (Schmalzl et al., 2015). The increased proprioceptive awareness can help also to increase daily usage of desired components: Alexander Technique has also been demonstrated to facilitate lightness (Cohen et al., 2015), which can enhance feelings of happiness (Shafir et al., 2016).

LMA classifies movement components in four main categories: Body, Effort, Space, and Shape (Table 2). Components from these categories change during movement in real time, making LMA particularly useful for observing and noting movement nuance and phrasing in everyday situations and in therapy (for a full description of LMA, see Fernandes, 2015). Its descriptive language identifies qualitative and quantitative aspects of movement with words such as light, strong, sudden, forward, sink, or retreat. It is relatively easy for people to understand, and familiar to most dance movement therapists (Dayanim et al., 2006; White, 2009). LMA has been used to describe movements associated with diverse topics, from personality and emotional state (Levy and Duke, 2003; Koch, 2007) to communicative gestures (Zhao, 2001) and rat behavior (Foroud and Pellis, 2003), and some of its components were successfully measured with a single accelerometer (Kikhia et al., 2014) and identified by a Kinect camera (Bernstein et al., 2015).

In Shafir et al. (2016), participants improvised moving different combinations of LMA components and rated which emotion they felt while moving. A logistic regression analysis of these ratings yielded sets of Laban movement components which elicited feeling the emotions of anger, fear, sadness, or happiness: Anger was elicited by strong, sudden, advancing, or direct movements. Fear was elicited by retreating, condensing, enclosing, binding, or moving back. Sadness was elicited by passive weight, bringing the arms to the upper body (chest, shoulder or face), sinking, or dropping the head. Happiness was elicited by jumping, rising, spreading, and free, light, upward, or rhythmic movements. (video examples in Supplementary Material, Supplementary Videos 1–8). In this paper, we theorize how these findings might be used to influence affective state through movement.

As an augmented approach to psychotherapy, we suggest that understanding the movement components associated with each emotion, moving them and experiencing their effects during movement therapy, then integrating those components consciously into everyday life can expand motor vocabulary, open new possibilities of action and consequently emotional spectrum. Likewise, when desired and appropriate, the capacity to diminish the prevalence of certain components might provide a tool for self-regulation through weakening pre-set responses and their associated feelings. For this process of change to unfold smoothly we suggest three LMA theories to support a gradual process of transformation:

The developmental progression of movement components suggests the order of introducing new elements for expanding movement range.

The trained movement therapist can use these theories to attend to the client's movement and verbal cues for readiness to change movement components, observe subtle expression of emotion, share these observations with the client, and suggest how to use movement for resiliency in everyday life (for examples, see Tsachor, 2013).

The developmental progression of acquiring and mastering LMA movement components is laid out in published works by Eddy (Eddy, 1991, 2012, 2016), Amighi et al. (1999), Hackney (Hackney and Weeks, 2002), Tortora (2006), and Tsachor (2013) and shown in Table 1. This perspective suggests that components mastered in each stage of development underlie the motor skills of the next stage. Accordingly, introducing movements from earlier stages provides the support necessary to develop mastery of movements that develop in later stages. First published by Tsachor (2013), the table is revised here to outline a developmental order of the appearance of those components associated with basic emotions. Used in conjunction with findings from Shafir et al. (2016), the chart can point to the order in which to develop movement skills supporting resilient emotional responses.

Using the developmental approach, a person may move primarily with Shape components from one stage (such as directional movement, first mastered at 6–12 months), but rarely move components from a later stage (e.g., spreading, advancing, or retreating, first mastered at 18–36 months). Understanding that when we don't progress beyond one stage, it may be because movement from the previous stage was not mastered, we might choose to first reinforce skills from the preceding stage, such as Shape Flow (first mastered at 0–3 months), by, for example, easily expanding and condensing with the rhythms of breathing. This may then facilitate our ability to change shape in the core, eventually permitting spreading and rising (components found to be associated with happiness) or retreating and enclosing (which were associated with fear) all mastered first at 18–36 months. Non-judgmental naming of the developmental movement progression can encourage learning new movement options, facilitating choice in responding to one's own particular challenges.

LMA identifies four main categories of Movement: Body, Effort, Space, and Shape. The teaching of Effort, Space and Shape has been shown to be effective for changing stuck movement patterns (Bales, 2006; de Souza, 2016). Laban's theory of “Space Harmony” posits that moving in specific directions in Space naturally affines with specific Efforts (movement dynamics components, such as light, strong, sudden, sustained) and Shape components (changes in the body's configuration, such as sinking, rising, spreading, retreating). For example, reaching upward (in space) often lengthens the torso, which takes one into rising (a Shape component) and supports lightness (an Effort component), while downward flexion (in Space) may shorten the torso with sinking (a Shape component), and facilitates strength (an Effort component). Each row in Table 2 shows a set of affinities. These affinities, established from observation of “how the body and its limbs are able to execute certain dynamic nuances in movement toward certain areas in space better than toward others” (Laban and Ullmann, 1966 p.30) relate specific intention in Space to specific Effort and Shape components. Moving in the Space direction affined to a desired Effort or Shape component can make it easier to learn (Dell et al., 1977; Fernandes, 2015). Using the previous example: moving up may facilitate learning to rise and to move lightly (first row in Table 2). Since most components related to emotions found by Shafir et al. (2016) are Effort and Shape components, we can facilitate changes in Effort and Shape by using the affined directions in Space, which can be easier to move, thereby potentially affecting emotions.

People with any movement capacity can expand movement range via affined components: A sequence such as up (affined with rise and light), down (affined with sink and strong), etc. can be done with any body part, and at any size of reach space, so people can begin practicing the new components with skills already within their repertoire.

In order to feel that moving a new component connects authentically to our emotions, the transition to new movement repertoire can be supported by continuing some ongoing, familiar components. During movement, Laban's “Space Harmony” theories organize the way in which change among components occurs, and therefore can serve as a template to structure movement changes in a progressive fashion (Bartenieff, 1972), so changes of expression and emotion will not feel too abrupt. When changing movement components, one common Space Harmony pattern is to (1) Maintain one component from the previous movement, to stabilize the experience, (2) Let go of another component then (3) Add a new (missing) component. This allows new movement to have continuity with the familiar previous movement as it branches into a new expression (Laban and Ullmann, 1966; Dell et al., 1977; Longstaff, 2000).

Shafir et al. (2016) found that 2–4 components were sufficient to bring about the experience of each emotion. Each component predicted only one emotion. Consequently, when a component is not present, the stimuli eliciting that emotion are reduced, making emotional shift easier. Thus, the principles of Space Harmony can support the process of change. For example, our movement behavior may be dominated by the qualities that Shafir et al. (2016) associated with fear: retreating, condensing, enclosing, binding, back. Recognizing that movement often expresses behavior which was adaptive, and when we believe these movements are pervasive, no longer adaptive, or more frequent than useful, we may wish to provide a structured method to gradually diminish the prevalence of some components which reinforce fear.

Based on Space Harmony structure, gradual movement change will be done through several small shifts such as:

Shift 1: Maintain retreating and backwards movement while gradually letting go of binding, and adding rhythmicity (a happy characteristic) such as rocking.

Shift 2: Maintain rhythmicity and backwards, let go of condensing and enclosing, and add some spreading.

Shift 3: Maintain spreading and rhythmicity, eventually releasing the back direction, moving sideward, rhythmically.

In this example, the mover is supported through gradual change, and can attend to emotions and movement impulses, so changes occur when the mover is authentically ready. This change may take place in one session, or gradually over time. In therapy, the individual's experience and clinician's judgment about the timing and readiness for each change are crucial to support self-efficacy, so each person learns to use movement to self-regulate emotions in tandem with psychological insight in therapy.

The methods outlined here for expanding beyond the limits of one's habitual movement components to a fuller range can empower people to exit the grip of intransigent emotions into less emotionally-laden states, or enhance capacity and access to other emotions. They can be applied during individual therapy, group DMT when choosing movements to suggest or reflect back to the group, or become a self-effective by-product of yoga, Alexander Technique, Tai Chi or other mindful movement practices. Moreover, the methods described above have potential to support resiliency through attunement to one's own emotions. By learning the components associated with emotions, and tracking the experience of them in one's body, we can become more cognizant of signals sent “by our body” about emotions. For example, learning to notice slight condensing or binding in response to a situation may precede the conscious perception of “fear” and facilitate recognition of subtle emotional shifts. Such embodied emotional awareness can lead to conscious emotional resiliency.

Changes in movement components can be done through conscious movement responses ranging from full-bodied movement (e.g., during dance or yoga) to unobtrusive qualitative changes during daily activity e.g., lightening up, as in Alexander Technique (Cohen et al., 2015). Behavioral changes and motor learning both require time to integrate and assimilate, and the time required for a somatic movement approach to achieve emotional resiliency through changes in daily motor behavior warrants investigation. Further studies are also needed to determine effectiveness of this approach in specific disorders. Lastly, further study is also warranted to understand and define the role of therapists and teachers in these processes, as professional knowledge is needed to know when it is appropriate to introduce new motor components, in particular for people who may be in a vulnerable/difficult emotional state. We caution against simplification or use of the implied results by those without significant movement training.

Movement is the way in which we humans adapt ourselves to our changing circumstances. We constantly re-configure our Shape and Effort, in order to respond to inner and outer stimuli (Bartenieff and Lewis, 1980). The fact that so much of our movement is voluntary provides an innate mechanism for self-regulation that can be nurtured and strengthened. As we expand choices in our own body's movement, we become empowered to respond to our world using movement as coping strategy adaptive to our needs. Over time, developing our capacity to make subtle movement choices can empower us with embodied skills for emotional resiliency.

All authors have made an equal, substantial, direct and intellectual contribution to the work, and approved it for publication.