Motor familiarity: Brain activation when watching kinematic displays of one's own movements☆
Highlights
► We manipulated observers’ familiarity with the presented point-light displays. ► Observers watched four markers representing an opposing table tennis player. ► In some trials the kinematic information originated from strokes of the observers. ► Results showed that observers performed better when observing their “own” strokes. ► BOLD-response was found higher in the angular gyrus and the medial frontal cortex.
Introduction
In daily life, we see other people acting around us and adapt our actions to theirs. However, we hardly ever see ourselves moving from a third-person perspective except when looking in a mirror or watching ourselves on film. Nonetheless, when we do have occasion to view our own actions in this way, we can predict their outcomes better than the outcomes of other people's actions. One possible reason for this enhanced prediction performance is the motor familiarity of own movements. Motor familiarity develops when certain movements are performed frequently, thereby strengthening their proprioceptive and kinematic representations within the respective brain areas. During the last decade, several studies have focused on the effect of motor familiarity on action observation (e.g., Flach et al., 2004, Hohmann et al., 2011, Knoblich and Flach, 2001, Knoblich and Prinz, 2001, Knoblich et al., 2002, Repp and Knoblich, 2004). Knoblich and Flach (2001), for example, asked participants to predict the landing position of darts from videos depicting only the throwing movements, that is, without showing the dart flight itself. Predictions were better when participants observed recorded movements of their own throws compared to those of other people. A second influence on prediction accuracy was how completely the body of the thrower was visible. Accuracy decreased when only the body without the head or when only the arm without the rest of the body could be seen. Further empirical evidence for an advantage of perceiving own movements comes from studies asking participants specifically to recognize either their own or others’ movements. Recognition was better for own movements when writing characters on a graphic tableau (Knoblich & Prinz, 2001), identifying the source of hand clapping (Flach et al., 2004), or playing piano (Repp & Knoblich, 2004).
It has been argued that the clear benefit of motor familiarity is due to predictive simulation via internal models in the brain. These are exactly the same internal models as those used to describe both action control and motor simulation processes such as action observation (Miall, 2003, Ramnani and Miall, 2004, Wolpert and Flanagan, 2001). Predictive forward simulation estimates the anticipated sensory outcomes of a movement. Within this framework, Knoblich and Flach (2001) reasoned that prediction accuracy increases because observing one's own actions provides a better match between the perceiving system and the executing/planning system. They based this interpretation on the classic common coding models of action perception that assume an overlap between representations of visual input and motor representations of the observers’ own action (James, 1890, Prinz, 1997). The integration of motor and sensory aspects of movement into a motor simulation seems to be essential for action observation (see, for reviews, Bubic et al., 2010, Shmuelof and Zohary, 2005, Zentgraf et al., 2011). If sensory consequences of self-generated movements are like they were predicted, they get processed in an attenuated fashion (Bäß et al., 2009, Hesse et al., 2010, Martikainen et al., 2005). Unexpected sensory consequences on the other hand are amplified as error signals (Iwanaga and Nittono, 2010, Waszak and Herwig, 2007). These effects of congruency between prediction expectancy and sensory effects are opposite to the effects of familiarity. Ricciardi et al. (2009) found, that familiar motor sounds activated fronto-parietal areas more than unfamiliar sounds. This fronto-parietal network was sensitive to observed actions in healthy participants and was activated during auditory perception and the execution of actions in both, congenitally blind and healthy blindfolded participants. Familiarity influenced action identification and may thereby be a factor in a further function of the comparison between the predicted sensory consequences and real feedback, which is the attribution of agency (Blakemore, Wolpert, & Frith, 2002).
One relevant structure to which internal models and body representations are attributed within the human motor system is the inferior parietal lobe (IPL). For example, drawing on a study of three apraxic patients with lesions in the left parietal cortex, Sirigu, Daprati, Pradat-Diehl, Franck, & Jeannerod (1999) have proposed that the left parietal lobe holds a kinesthetic representation of own body movements that is used to process sensory motor transformations. A comprehensive study conducted by Daprati, Sirigu and Nico (2010) have demonstrated that both parietal cortices seem to be involved in the integration of the unique, elementary experience of one's own body in motion. However, when observing another person's action, it is often inferior parietal structures that are associated with the recognition of that person (Blakemore and Frith, 2003, Chaminade and Decety, 2002, Decety and Sommerville, 2003, Farrer and Frith, 2002, Iacoboni et al., 1999). Thus, inferior parietal activation sites might also be associated with self- versus other-related processes. Ruby and Decety (2001) examined this notion by asking participants to imagine either themselves manipulating an object or the experimenter manipulating an object. Their results showed a dramatic increase in activation in the right inferior parietal lobule at the junction with the temporal cortex when participants imagined the experimenter performing the manipulation compared to when they imagined performing it themselves. When participants imagined themselves as agents from a first-person perspective, there was stronger activation of the inferior parietal cortex in the left hemisphere. Evidently, humans require differential physiological substrates to avoid confusion between self and other (Decety & Sommerville, 2003). Recent neuroimaging studies have also discussed further brain areas associated with self-versus other-related processes. Forms of self and other mental state reasoning – such as coordinating representations of self and representations of other – particularly involve medial prefrontal areas (Decety and Sommerville, 2003, Gallagher and Frith, 2003).
Against this background, the present study examined whether observing actions that encompass different levels of motor familiarity, that is, either self- or other-generated, differentially activates the brain areas involved in action observation as well as areas related to self-other processes. We consider motor familiarity to influence an essential part of the complex process of effect prediction that enhances the recruitment of internal models. One idea of the processes significant in sensorimotor integration proposes that input–output and output–input transformations between the motor system and external events are based on internal models. These internal models are either neural mechanisms that predict sensory consequences from efference copies of motor commands (i.e., the forward model) or that calculate the inverse, the necessary motor commands for motor control (for example Kawato, 1999, Wolpert and Kawato, 1998). These models can be combined, when predictions of a forward model are based on the output of a prior inverse model. If the output of an inverse model is stored in memory, motor familiarity is a factor on the recruitment of models. Higher familiarity leads to more memory retrieval, while lower familiarity leads to more spatial calculation of action effects. We assume that the processing of visual kinematics includes a stage of comparison between the observed kinematics and the stored internal models (action repertoire). Early stages in this processing stream are expected to elicit neural activation regardless of the result of this comparison. The next stages of the processing stream are recruited differentially depending on the outcome of the comparison between the observed kinematics and the own motor repertoire. We focused on these processing steps in this study. We examined participants in an fMRI scanner while watching point-light displays (PLDs) of own movements and those of others. More precisely, they watched degraded PLD presentations of their own and others’ table tennis strokes under the instruction to deduce whether the subsequent track of the ball would be down the line or cross-court. Note that self-recognition was not part of the task. Participants were all novices who were familiar with basic table tennis, but had no experience of observing PLD presentations of their own and others’ table tennis strokes.
Assuming that motor familiarity (i.e., self-generated actions) influences performance on the present action observation task our first hypothesis is that the detection rate for ball-flight direction will increase when participants observe their own strokes. Our second hypothesis is on the neural level and states that activation within areas associated with own motor representations and self-related processes such as parts of the IPL or the vmPFC will be increased when observing own table tennis strokes compared to those of others.
Section snippets
Participants
Twenty-one persons participated in this experiment for either course credit points or financial compensation: 20 students at Giessen University and one trainee at the Institute of Sport Science. One participant's data were dropped from all analyses, because performance on “own videos” was extremely low at 22% correct answers—more than two standard deviations below the mean (M=67.71%, SD=16.72).
Thirteen participants were female, 7 were male, and their mean age was 20.9 years (SD=1.80 years). All
Behavioral data
Each participant was asked to anticipate ball flight (down the line or cross-court) in 128 different videos. Reaction times revealed no statistically significant differences (one-way ANOVA, F[2, 59]=0.07, p=.93) between the three conditions “own videos” (M=976 ms, SD=305), “other videos” (M=988 ms, SD=206), and “rotated videos” (M=1005 ms, SD=219).
The percentage of correct responses was above chance level (random probability=50%; own: M=70.00%; SD=13.35; t=6.70, p<.01; other: M=63.71%; SD=8.99; t
Discussion
This study used fMRI to investigate the neural correlates of motor familiarity in the perception of goal-directed actions and the prediction of nonvisible action effects. Visual information on the movements – table-tennis strokes – was degraded to the kinematics of PLDs. By presenting PLDs of one's own actions or PLDs of actions generated by others different degrees of motor familiarity were applied.
We were able to replicate a behavioral effect showing the benefit of motor familiarity when
Acknowledgments
This research was supported by the Deutsche Forschungsgemeinschaft (DFG, Research Group 560 “Action and Perception”). The authors thank Jonathan Harrow for native speaker advice.
References (56)
The default self: feeling good or being right?
Trends in Cognitive Sciences
(2007)- et al.
Utilizing the ventriloquism-effect to investigate audio-visual binding
Neuropsychologia
(2007) - et al.
Self-awareness and action
Current Opinion in Neurobiology
(2003) - et al.
Abnormalities in the awareness of action
Trends in Cognitive Sciences
(2002) - et al.
Body and movement: consciousness in the parietal lobes
Neuropsychologia
(2010) - et al.
Shared representations between self and other: a social cognitive neuroscience view
Trends in Cognitive Sciences
(2003) - et al.
Experiencing oneself vs. another person as being the cause of an action: the neural correlates of the experience of agency
Neuroimage
(2002) - et al.
Implicit and explicit processes in social cognition
Neuron
(2008) - et al.
Functional imaging of ‘theory of mind
Trends in Cognitive Sciences
(2003) Internal models for motor control and trajectory planning
Current Opinion in Neurobiology
(1999)
Pianists duett better when they play with themselves: on the possible role of action simulation in synchronization
Consciousness and Cognition
Selective responses in right inferior frontal and supramarginal gyri differentiate between observed movements of oneself vs. another
Neuropsychologia
The assessment and analysis of handedness: the Edinburgh inventory
Neuropsychologia
Perceptual resonance: action-induced modulation of perception
Trends in Cognitive Sciences
Dissociation between ventral and dorsal fMRI activation during object and action recognition
Neuron
Effect anticipation modulates deviance processing in the brain
Brain Research
Multiple paired forward and inverse models for motor control
Neural Networks
Internal models in the cerebellum
Trends in Cognitive Sciences
Simulation during observation of human actions—theories, empirical studies, applications
Vision Research
Meeting of minds: the medial frontal cortex and social cognition
Nature Reviews Neuroscience
Attenuated human auditory middle latency response and evoked 40-Hz response to self-initiated sounds
European Journal of Neurosciences
From the perception of action to the understanding of intention
Nature Reviews Neuroscience
Prediction, cognition and the brain
Frontiers in Human Neuroscience
Recognizing and predicting movement effects— identifying critical movement features
Experimental Psychology
Leader or follower? Involvement of the inferior parietal lobule in agency
Neuroreport
Sensitivity of the action observation network to physical and observational learning
Cerebral Cortex
Recognizing one's one clapping: the role of temporal cues
Psychological Research
In search of the emotional self: an fMRI study using positive and negative emotional words
The American Journal of Psychiatry
Cited by (0)
- ☆
Study carried out at the Bender Institute of Neuroimaging.