Primary motor cortex activation during action observation revealed by wavelet analysis of the EEG
Introduction
The premotor and anterior inferior parietal cortices of monkeys contain neurons that fire during both the execution and observation of object-directed actions (di Pellegrino et al., 1992, Gallese et al., 1996, Gallese et al., 2002, Kohler et al., 2002). Based on these properties it has been argued that these ‘mirror’ neurons form the basis of an action–representation system whereby actions are understood by directly mapping the actions of a conspecific animal onto the motor structures that the observer uses to execute those actions (Rizzolatti et al., 2001, Umilta et al., 2001).
Various imaging modalities have now shown that a similar system exists in humans (see Rizzolatti et al., 2001, for a review). In particular, human neurophysiological recordings have shown that the rhythmical activities of primary sensorimotor areas, specifically the beta (∼20 Hz) and mu (∼10 Hz) rhythms, exhibit modulations consistent with the hypothesis that the neural substrates that generate these rhythms play a functional role in the human mirror neuron system (Hari et al., 1998, Jarvelainen et al., 2001, Muthukumaraswamy and Johnson, 2004). Various studies have located the main generators of the beta rhythm in primary motor cortex, whereas mu generators are located more posteriorly in somatosensory cortex (Salmelin and Hari, 1994, Hari et al., 1998, Rossi et al., 2002, Cheyne et al., 2003).
A number of magnetoencephalographic studies have shown that following median nerve stimulation the amplitude of the beta rhythm is transiently attenuated and then approximately 400 ms after stimulation rebounds to a level exceeding its original baseline (Salmelin and Hari, 1994, Salenius et al., 1997). The rebound typically lasts until approximately 1 s after nerve stimulation. This pattern of beta band power decrease followed by beta power increase is similar to what occurs during voluntary movement, described by both the MEG (Salmelin and Hari, 1994, Taniguchi et al., 2000) and EEG (Pfurtscheller, 1992, Pfurtscheller and Lopes da Silva, 1999). The decrease in power in the beta band is usually referred to as event-related desynchronisation (ERD) while the increase in power is referred to as event-related synchronisation (ERS) (Pfurtscheller and Lopes da Silva, 1999).
It has been shown on several occasions that the rebound of the beta rhythm following median nerve stimulation is significantly diminished when subjects perform a movement (Salenius et al., 1997) and is also diminished when subjects observe an experimenter manipulating an object (Hari et al., 1998, Jarvelainen et al., 2001). This effect is consistent with the hypothesis that the beta rhythm may be a neurophysiological index of the activity of a human mirror neuron system. More recently it has been shown that this ∼20 Hz modulation occurs also during the observation of movement sequences (no object-interaction) when subjects are required to later recall the sequence but not when the subjects are asked to perform arithmetic calculations, a cognitive/attentive task (Rossi et al., 2002). This suggests that beta rhythm modulations are not simply due to generalised attentional differences between baseline and experimental conditions. Modulation of EEG beta rhythm during the observation of aimless movements has been shown in a naturalistic setting (Babiloni et al., 2002), but an equivalent experiment coupled to median nerve stimulation has not been reported. Quantification of task-related changes with concurrent nerve stimulation can be useful because these induced events are frequently more stable than spontaneous rhythms (Salenius et al., 1997); however, only a few studies have examined EEG beta-rhythmicity coupled to median nerve-stimulation. Pfurtscheller (1981) was the first to report increased beta oscillatory activity following median nerve stimulation and this finding has been replicated in several studies since (Pfurtscheller et al., 2002, Muller et al., 2003, Stancak et al., 2003). Only one study (Pfurtscheller et al., 2002), limited to 4 recording electrodes, has examined beta rhythm during movement performance, and to our knowledge, no previous EEG studies have examined beta rebound effects during action observation.
Typically, experimental changes in EEG rhythms are extracted by narrow bandpass filtering of the recorded signal and then averaging across trials (see Pfurtscheller and Lopes da Silva, 1999, for several variations on this approach). This technique is perhaps not ideal because the bandpass limits are set before the analysis is conducted, although a number of techniques have been suggested in order to define these bands prior to analysis (Pfurtscheller and Lopes da Silva, 1999). An alternative approach to measurement of time-varying energy in a signal at a given frequency band is by convolution of the EEG signal with a family of wavelets (Grossmann et al., 1989, Tallon-Baudry et al., 1997). Wavelet analysis does not require a priori selection of a narrow frequency band, and provides a better compromise between temporal and frequency resolution than time–frequency techniques based on short-time Fourier transforms (Sinkkonen et al., 1995, Misiti et al., 1997).
In the present study, we wished to characterise the spectral response of the EEG to median nerve stimulation using wavelet analysis, and to map these effects topographically using a high-density recording montage. We wished to determine if the ‘rebound’ rhythms extracted by this technique are diminished during overt movements, as has been found using narrow-band frequency analysis of the EEG/MEG. We also examined several different conditions of action observation to determine (a) if they exert effects comparable to those reported in the MEG literature; and (b) if so, the relative magnitude of their effects on the EEG rebound rhythms. We also tested some novel conditions including observation of somatosensory stimulation and observation of aimless thumb movement to test what modulatory effects these conditions might have.
Section snippets
Subjects
Five males and three females (mean age 23; range 21–26) participated in the experiment. All subjects were right-handed as assessed by the Edinburgh Handedness Inventory (Oldfield, 1971) and gave informed consent to participate. The University of Auckland Human Subjects Ethics Committee approved all experimental procedures.
Procedure
Subjects' right median nerves were stimulated using a Digitimer stimulation unit (Digitimer Ltd) such that painless contractions of the thumb muscle were elicited. Stimulus
Results
Fig. 1 shows the grand-averaged spectrogram for the 8 subjects in the baseline condition. The most striking feature of the spectrogram is the increase in beta energy that occurs from 500 to 1000 ms that peaks at ∼18 Hz. Prior to this energy increase there is a decrease from the prestimulus level that occurs from 50 to 300 ms. This pattern is also evident in the mu rhythm (∼10 Hz) but the onset of the mu rhythm rebound is later than that of beta, occurring from 750 to 1300 ms. In Fig. 2, the
Discussion
The present results show that wavelet-based time–frequency analysis extracts EEG features that are consistent with results from conventional analyses based on bandpass filtering of the MEG and EEG (Hari et al., 1998, Pfurtscheller et al., 2002, Rossi et al., 2002). The mean frequency band for the mu rhythm was found to be 10–12 Hz whereas for the beta rhythm it was 15–22 Hz. The mu rhythm tended to rebound slightly later than the beta rhythm (500 versus 750 ms) and shared a similar topography,
Acknowledgements
This work was supported by Royal Society of New Zealand Marsden Grants UOA813 and Royal Society of New Zealand International Science and Technology Linkages Fund Grant 3603503. The authors thank Douglas Cheyne and William Gaetz for discussions of the wavelet analysis, and three anonymous reviewers for helpful comments on a previous version of this manuscript.
References (37)
- et al.
Impaired mirror-image imitation in Asperger and high-functioning autistic subjects
Curr Biol
(2003) - et al.
Human cortical electroencephalography (EEG) rhythms during the observation of simple aimless movements: a high-resolution EEG study
NeuroImage
(2002) - et al.
Neuromagnetic imaging of cortical oscillations accompanying tactile stimulation
Cogn Brain Res
(2003) - et al.
EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis
J Neurosci Methods
(2004) - et al.
Scalp electrode impedance, infection risk, and EEG data quality
Clin Neurophysiol
(2001) - et al.
Timing of human cortical functions during cognition: role of MEG
Trends Cogn Sci
(2000) - et al.
Event-related beta EEG changes during wrist movements induced by functional electrical stimulation of forearm muscles in man
Neurosci Lett
(2003) The assessment and analysis of handedness: the Edinburgh inventory
Neuropsychologia
(1971)Central beta rhythm during sensorimotor activities in man
Electroencephalogr Clin Neurophysiol
(1981)Event-related synchronization (ERS): an electrophysiological correlate of cortical areas at rest
Electroencephalogr Clin Neurophysiol
(1992)
Event-related EEG/MEG synchronisation and desynchronisation: basic principles
Clin Neurophysiol
Contrasting behavior of beta event-related synchronization and somatosensory evoked potential after median nerve stimulation during finger manipulation in man
Neurosci Lett
Somatosensory processing during movement observation in humans
Clin Neurophysiol
Modulation of human cortical rolandic rhythms during natural sensorimotor tasks
NeuroImage
Spatiotemporal characteristics of sensorimotor neuromagnetic rhythms related to thumb movement
Neuroscience
Functional segregation of movement-related rhythmic activity in the human brain
NeuroImage
Gabor filters—an informative way for analyzing event-related brain activity
J Neurosci Methods
Desynchronization of cortical rhythms following cutaneous stimulation: effects of stimulus repetition and intensity, and of the size of corpus callosurn
Clin Neurophysiol
Cited by (169)
Mu-ERD reflects action understanding, but the effect is small
2024, Brain ResearchAcetaminophen changes Mu rhythm power related to pain empathy
2023, NeuropsychologiaImproving performance in motor imagery BCI-based control applications via virtually embodied feedback
2020, Computers in Biology and Medicine