Elsevier

Neuropsychologia

Volume 50, Issue 12, October 2012, Pages 2745-2751
Neuropsychologia

Sensitivity of alpha and beta oscillations to sensorimotor characteristics of action: An EEG study of action production and gesture observation

https://doi.org/10.1016/j.neuropsychologia.2012.08.005Get rights and content

Abstract

The sensorimotor experiences we gain when performing an action have been found to influence how our own motor systems are activated when we observe others performing that same action. Here we asked whether this phenomenon applies to the observation of gesture. Would the sensorimotor experiences we gain when performing an action on an object influence activation in our own motor systems when we observe others performing a gesture for that object? Participants were given sensorimotor experience with objects that varied in weight, and then observed video clips of an actor producing gestures for those objects. Electroencephalography (EEG) was recorded while participants first observed either an iconic gesture (pantomiming lifting an object) or a deictic gesture (pointing to an object) for an object, and then grasped and lifted the object indicated by the gesture. We analyzed EEG during gesture observation to determine whether oscillatory activity was affected by the observer's sensorimotor experiences with the object represented in the gesture. Seeing a gesture for an object previously experienced as light was associated with a suppression of power in alpha and beta frequency bands, particularly at posterior electrodes. A similar pattern was found when participants lifted the light object, but over more diffuse electrodes. Moreover, alpha and beta bands at right parieto-occipital electrodes were sensitive to the type of gesture observed (iconic vs. deictic). These results demonstrate that sensorimotor experience with an object affects how a gesture for that object is processed, as measured by the gesture-observer's EEG, and suggest that different types of gestures recruit the observer's own motor system in different ways.

Highlights

► Participants gain sensorimotor experiences with objects (heavy or light). ► We analyze EEG during observation of gestures directed to similar-looking objects. ► We analyze EEG during action execution: lifting heavy and light objects. ► Gesture type (iconic/deictic) changes alpha/beta power during gesture observation. ► Prior sensorimotor experiences change alpha/beta power during gesture observation.

Introduction

Our own actions and our perceptions of others' actions are closely linked. There is ample evidence suggesting that our own experiences with action can influence perception in such a way that we become especially sensitive to observing similar actions performed by others (Hecht, Vogt & Prinz, 2001; Schütz-Bosbach & Prinz, 2007) For example, dancers trained in ballet show greater premotor cortex activation when observing ballet dancing than when observing other dances with which they have not had extensive motor experience (Calvo-Merino, Glaser, Grezes, Passingham, & Haggard, 2005). This type of evidence has been used to support the idea that action and perception rely on similar neural substrates.

One prominent hypothesis about the relation between perception of action and production of action concerns a putative human mirroring system (Decety & Grezes, 1999), composed of ventral and dorsal premotor cortex, the anterior inferior parietal lobule, somatosensory areas such as BA2, and the middle temporal gyrus (Gazzola & Keysers, 2009). The mirroring hypothesis postulates that, when an action is observed, the brain regions involved in performing that action are activated—as though the observer were performing the action herself (Gazzola & Keysers, 2009); this vicarious simulation of observed action then has the potential to facilitate the interpretation and understanding of others' actions (Rizzolatti & Sinigaglia, 2010). Mirroring processes may also allow for fluid social interactions by enabling social partners to prepare appropriate responses to observed actions (Gallagher, 2008).

Evidence suggests that the putative human mirroring system may also be involved in observing gestures (Emmorey et al., 2010, Goldenberg and Hagmann, 1997, Hostetter and Alibali, 2008). Gestures are actions, but they do not have a direct effect on the world the way most actions do—instead, gestures are representational. Gestures have been hypothesized to ground thought in action (Beilock and Goldin-Meadow, 2010, Cartmill et al., 2012, Goldin-Meadow and Beilock, 2010) and experiments to support this hypothesis have shown that perception of gesture, like perception of action, recruits the observer's sensorimotor system (Enticott et al., 2010, Villarreal et al., 2008). If observed gestures are, at least to some degree, represented as actions, then the neural systems underlying gesture perception should be sensitive to characteristics of the actions represented in those gestures; in other words, observing a gesture representing an action should evoke similar neural responses as executing the action itself.

In the current study, we compared two types of gestures to determine whether they resulted in different patterns of neural activity when observed: (1) deictic gestures, which draw attention to objects (e.g. pointing at an object), and (2) iconic gestures, which display characteristics of action as if the gesturer were performing the action himself (e.g., moving the hand as though grasping and lifting an object, i.e., character viewpoint gestures (Cartmill et al., 2012; McNeill, 1992)). These two gestures were selected because they vary in how closely they mimic action. We hypothesized that observing iconic gestures, which closely mimic an action performed on an object, would result in greater activation of sensorimotor cortex than observing deictic gestures, which are static and serve primarily to indicate the location of an object.

An important unanswered question with respect to gesture perception is whether an observer's prior experience with the action represented in a gesture changes how that gesture is processed. This question has important implications for observational learning and, more specifically, for how gesture is used in teaching situations. Recent evidence suggests that mirroring in the observer is sensitive to the somatosensory and motor characteristics of the observed action, and also to the amount of prior experience the observer has had with the observed action (Calvo-Merino et al., 2005, Kilner et al., 2004, Orgs et al., 2008, Quandt et al., 2011). In the current study, we extended this line of reasoning to ask whether one's prior somatosensory or motor experiences with specific objects affect the subsequent processing of others' gestures towards those objects.

Alpha- and beta-range rhythms in the electroencephalogram (EEG) have been examined in studies of action processing (Muthukumaraswamy, Johnson, & McNair, 2004; Pfurtscheller, Neuper, & Krausz, 2000). Rhythms in these frequency ranges typically show a regional decrease in power in response to both executing and observing action, suggesting that they may be related to the common neural coding of action and perception (Perry and Bentin, 2009, Press et al., 2011). While it is not clear precisely how EEG rhythms relate to specific cognitive processes, it is thought that alpha and beta bands are closely tied to the allocation of visuospatial attention (Mathewson et al., 2011) and the activation of sensory (van Ede, de Lange, Jensen, & Maris, 2011) and/or motor cortex (Perry, Stein, & Bentin, 2011). The primary focus of the current study is using alpha and beta rhythms to explore activation of sensorimotor cortex during action observation and production. We performed separate analyses on the lower alpha (8–10 Hz), upper alpha (11–13 Hz), and beta (14–30 Hz) bands. We were particularly interested in the upper alpha band response during gesture observation, given evidence that this frequency band is sensitive to previous experience with actions (Marshall et al., 2009, Quandt et al., 2011).

To our knowledge, no prior studies have examined the relation between patterns of cortical activity elicited when a communicative gesture is observed and patterns of cortical activity elicited when the corresponding action is executed. We designed an experiment in which participants were first given sensorimotor experience with different objects, after which they observed gestures referring to those objects. We were interested in whether specific experience with a set of objects would change EEG activity elicited when observing gestures referring to those objects. We were also were interested in whether EEG activity would vary as a function of the type of gesture observed (iconic vs. deictic).

At the outset of the experiment, and throughout the experimental session, each participant received sensorimotor experience with one set of objects: either heavy/yellow and light/blue objects OR heavy/blue and light/yellow objects. We collected EEG while participants observed video clips of an actor performing either an iconic or deictic gesture toward the yellow or blue object. Each participant saw four video clips in total: iconic/yellow (i.e., iconic gesture directed toward a yellow object), iconic/blue, deictic/yellow, and deictic/blue. After each video clip, the participant reached for, grasped, and lifted an object of the same color as the object indicated by the gesture. Thus, if they saw an iconic (or deictic) gesture directed toward a blue object, they lifted the blue object that was in front of them. Importantly, the actor in the video clips never touched either object, and the objects remained stationary during the entire video clip, so there was never any information regarding the weight of the objects gestured to in the video clips. This aspect of the design allowed us to relate differences in the EEG during gesture observation to the participants' expectations about the relative weights of the objects (which would be based on their own experience interacting with the objects). We also examined differences in alpha and beta power during participants' execution of the grasping and lifting actions on the same objects.

Our analyses tested three hypotheses: (1) EEG responses elicited when executing an action on objects and EEG responses elicited when observing a gesture referring to those objects will show similar modulation by object weight (light vs. heavy during action execution, and expected light vs. expected heavy during gesture observation). (2) EEG alpha and beta range rhythms will show greater reactivity when observing iconic gestures than when observing deictic gestures. (3) When observing a gesture referring to an object, participants' expectations about the sensorimotor consequences of lifting that object (which are based on their own previous experiences lifting the object) will modulate alpha and beta rhythm activity. If this last hypothesis is supported, it would provide evidence that prior experience producing an action modulates the way a gesture related to that action is processed. This result, in turn, would support the idea that gesture perception is embodied, in the sense that prior experiences with objects modulate how we process gestures referring to those objects.

Section snippets

Participants

Thirty-seven right-handed (Oldfield, 1971) undergraduates (19 females; mean age=21.7, SD=3.8) took part in the study in exchange for course credit. All participants gave their informed consent prior to the experimental session, and the university Institutional Review Board had approved the study protocol.

Stimuli

Two pairs of objects were created out of opaque, identically-sized cylindrical metal containers (15.5 cm tall×7.0 cm diameter) that varied in weight (heavy, 1150 g, or light, 125 g) and color

Execution of action

For the action execution condition, we performed repeated-measures ANOVAs for lower-alpha, upper-alpha, and beta power using the factors of Weight (heavy, light), Region (mid-frontal F3/F4, lateral frontal F7/F8, central C3/C4, mid-parietal P3/P4, lateral parietal P7/P8, temporal T7/T8, and occipital O1/O2), and Hemisphere (left, right). Probability values reported for all main effects and interactions have been adjusted using the Greenhouse–Geisser correction factor (epsilon).

Action execution

The lower alpha frequency band showed little reactivity to the weight of the objects during action production. However, power in the upper alpha and beta frequency bands was clearly modulated by object weight. Upper alpha power decreased when the participants lifted the light object, compared to when they lifted the heavy object. Beta power across the scalp was also reduced when lifting the light object, with this effect being particularly strong over the left hemisphere. This lateralization is

Acknowledgments

The authors are grateful to Sarah Sanders, Christina Comalli, Ian Ross, Nhi Tran, Sarah Chaudry, Brieana Viscomi, Yanlian Wang, Hayley Haaf, and Todd Woldoff for help with data collection and video coding. This work was supported by NIH Grant HD-68734 and NSF Grant BCS-0642404 to PJM and TFS, and NSF Grant SBE-0541957 (Spatial Intelligence and Learning Center).

References (54)

  • P.J. Marshall et al.

    Effects of brief imitative experience on EEG desynchronization during action observation

    Neuropsychologia

    (2009)
  • T. Mima et al.

    Force level modulates human cortical oscillatory activities

    Neuroscience Letters

    (1999)
  • S. Muthukumaraswamy et al.

    Mu rhythm modulation during observation of an object-directed grasp

    Cognitive Brain Research

    (2004)
  • R.C. Oldfield

    The assessment and analysis of handedness: The Edinburgh inventory

    Neuropsychologia

    (1971)
  • S. Palva et al.

    New vistas for α-frequency band oscillations

    Trends in Neurosciences

    (2007)
  • A. Perry et al.

    Mirror activity in the human brain while observing hand movements: A comparison between EEG desynchronization in the mu-range and previous fMRI results

    Brain Research

    (2009)
  • A. Perry et al.

    Motor and attentional mechanisms involved in social interaction: Evidence from mu and alpha EEG suppression

    Neuroimage

    (2011)
  • G. Pfurtscheller

    Central beta rhythm during sensorimotor activities in man

    Electroencephalography and Clinical Neurophysiology

    (1981)
  • G. Pfurtscheller et al.

    Functional dissociation of lower and upper frequency mu rhythms in relation to voluntary limb movement

    Clinical Neurophysiology

    (2000)
  • L.C. Quandt et al.

    Experience with novel actions modulates frontal alpha EEG desynchronization

    Neuroscience Letters

    (2011)
  • M. Schieber

    How might the motor cortex individuate movements?

    Trends in Neurosciences

    (1990)
  • M.B. Schippers et al.

    Mapping the flow of information within the putative mirror neuron system during gesture observation

    Neuroimage

    (2011)
  • S. Schütz-Bosbach et al.

    Perceptual resonance: Action induced modulation of perception

    Trends in Cognitive Sciences

    (2007)
  • J. Svoboda et al.

    Effects of muscle contraction on somatosensory event-related EEG power and coherence changes

    Neurophysiologie Clinique/Clinical Neurophysiology

    (2004)
  • M. Villarreal et al.

    The neural substrate of gesture recognition

    Neuropsychologia

    (2008)
  • P. Walla et al.

    Self-awareness and the subconscious effect of personal pronouns on word encoding: A magnetoencephalography (MEG) study

    Neuropsychologia

    (2007)
  • D. Arnstein et al.

    Mu suppression during action observation and execution correlates with BOLD in dorsal premotor, inferior parietal, and SI cortices

    Journal of Neuroscience

    (2011)
  • Cited by (55)

    • Is motor learning of stroke patients in non-immersive virtual environment influenced by laterality of injury? A preliminary study

      2021, Journal of Bodywork and Movement Therapies
      Citation Excerpt :

      The odd numbers of the channels represent left brain areas and the even numbers represent right brain areas. AF3 and AF4 are channels positioned on antero-frontal areas; FC5 and FC6 are channels positioned on fronto-central areas; T7 and T8 are channels positioned on temporal areas; and P7 and P8 are channels positioned on parietal areas (Haufler et al., 2000; Quandt et al., 2012). Besides, it was evaluated the asymmetry score for alpha and beta on each channel.

    • Sensorimotor system engagement during ASL sign perception: An EEG study in deaf signers and hearing non-signers

      2019, Cortex
      Citation Excerpt :

      Studies are exploring the effects of experience on simple actions (i.e., grasping; Braadbaart, Williams, & Waiter, 2013; Drew, Quandt, & Marshall, 2015; Quandt & Marshall, 2014) and complex actions (e.g., dancing; Calvo-Merino et al., 2005; Cross, Hamilton, & Grafton, 2006; Gardner, Goulden, & Cross, 2015; Orgs, Dombrowski, Heil, & Jansen-Osmann, 2008). The AON has been the focus of a wide variety of research that exemplifies its far-reaching involvement in human action processing, such as: biological motion represented as point light displays (Ulloa & Pineda, 2007), gestures (Montgomery, Isenberg, & Haxby, 2007; Quandt, Marshall, Shipley, Beilock, & Goldin-Meadow, 2012), facial expressions (van der Gaag, Minderaa, & Keysers, 2007), and brain computer interfaces (Pineda, Allison, & Vankov, 2000). While research within this field has grown within the past 20 years, there remains much to be understood about the complicated relationship between action experience and action perception.

    View all citing articles on Scopus
    View full text