Object words modulate the activity of the mirror neuron system during action imitation

Abstract

Background: Although research has demonstrated that the mirror neuron system (MNS) plays a crucial role in both action imitation and action-related semantic processing, whether action-related words can inversely modulate the MNS activity remains unclear.

Methods: Here, three types of task-irrelevant words (body parts, verbs, and manufactured objects) were presented to examine the modulation effect of these words on the MNS activity during action observation and imitation. Twenty-two participants were recruited for the fMRI scanning and remaining data from 19 subjects were reported here.

Results: Brain activity results showed that word types elicited different modulation effects over nodes of the MNS (i.e., the right inferior frontal gyrus, premotor cortex, inferior parietal lobule, and STS), especially during the imitation stage. Compared with other word conditions, action imitation following manufactured objects words induced stronger activation in these brain regions during the imitation stage. These results were consistent in both task-dependent and -independent ROI analysis.

Conclusion: Our findings thus provide evidence for the unique effect of object words on the MNS during imitation of action, which may also confirm the key role of goal inference in action imitation.

Keywords: action words; fMRI; imitation; mirror neuron; objects.

Figure 1

Sample stimuli from the experimental stimuli set and the procedure of each trial: Each trial included three stimuli stages that were displayed for 2500 ms, spaced by three interstage intervals which varied between 500–6500 ms. Participants were asked to read the word silently during the word stage, to view the action video clips in the action observation stage, and to imitate the action when the green dot was presented in the action imitation stage. Duration of the events: Word = 2.5 s; Jitter 1 = 0.5–6.5 s; Observation = 2.5 s; Jitter 2 = 0.5–6.5 s; imitation=2.5 s, Jitter 3 = 0.5–6.5 s

Figure 2

Overall activation in three stages and the overlapping brain regions for action observation and imitation. (a) Activation for word stage, (b) Activation in the stage of action observation, and (c) Activation for imitation stage. The overlapping activation between (b) and (c) are shown in the conjunction analysis (d). Images are shown with a statistical threshold of voxel‐wise uncorrected p < .001. Coordinates and statistics are provided in Table 2

Figure 3

Mean parameter estimates in the four ROIs identified from the ANOVA analyses. The object word condition induced stronger activation in the left IPL, right IFG, and right PostG. The word effect was significant in the left PrecG. * p < .05, ** p < .01. Error bars are standard error of the mean

Figure 4

Mean parameter estimates in the six independent ROIs. The object word condition induced the strongest activity over the bilateral vPM, aIPS, and STS only in the imitation stage. The Word × Stage effect was significant in the vPM, aIPS, and STS, indicating that the word effect occured in the imitation stage and object words induced strongest activity than other words and checkerboard. Error bars are standard error of the mean. † p < .1, * p < .05 ** p < .01