Links Between Action Perception and Action Production in 10-Week-Old Infants

1. Links Between Action Perception and Action Production in 10-Week-Old Infants Vincent M. Reid1 and Katharina Kaduk 1Department of Psychology, Durham University, UK Correspondence to: Vincent Reid (vincent.reid@durham.ac.uk) Introduction It has been supposed that infants first must acquire a specific level of reasoning about their own actions in order to be able to understand the actions of others. Further, they do so by analogy to self-produced actions (e.g. Barresi & Moore, 1996; Meltzoff, 1995). Despite these claims, only one study has investigated such issues in infants younger than 6 months of age. Sommerville et al. (2005) provided experience of object contact for 3-month-old infants by providing them with ‘sticky mittens,’ allowing object engagement far earlier than would naturally occur. Only those infants who were given object contact experience were able to focus on the relation between an experimenter and her object-directed goal during a subsequent habituation task. Control infants did not process the difference in stimuli when the observed object directed goal was modified. However, the issue of influence of action production on action perception can be addressed without requiring the infant to discriminate others’ goals. Intentional action is a highly complex form of movement. By investigating the role of action experience in the perception of biological motion (BM), we can examine this issue at an earlier age in development. We provided infants with the experience of walking via the walking reflex. As this reflex disappears at 12 weeks, we tested earlier than this date. We then compared the processing of BM between a group of infants who had experience of walking and a control group which had no experience of walking. We predicted that walking experience would alter BM processing but only for BM movements which participants had experienced. Figure 2. Experience group: Upright and inverted walking. Black=upright, Grey = inverted BM Figure 3. Experience group: Highlight of electrode P3, displaying the effect from 350ms-550ms, with increased positivity for upright vs inverted walking. Method Participants 25 infants aged 9-11 weeks (mean age 10.6 weeks) participated in the study. Following data editing we had nine participants in the experience and 10 in the no experience groups due to too few trials for the formation of adequate individual average ERPs from 6 participants. Procedure Thirteen 10-week-old participants were induced to produce the stepping reflex for three minutes. This reflex is enhanced at this age by submerging the legs in water in order to reduce leg mass (Thelen & Fisher, 1983); we therefore provided a warm bath for this purpose. Twelve 10-week-old infants served as controls by having the same bath experience but without the step reflex. These infants were moved within the bath in the same way as the stepping infants, but their legs did not touch the ground; they did not produce the step reflex (see Figure. 1). The number of steps was calculated for each infant, with “steps” defined as per Thelen and Fisher (1983). A rater who was not aware of the aims of the study assessed nine participants. Cohen’s Kappa for walking steps was 0.79. Once this pre-test phase was completed, two forms of biological motion depicting walking and crawling in upright and inverted positions were presented to infants in a standard ERP paradigm. EEG Editing and Analysis EEG was recorded continuously with Ag–AgCl electrodes from 32 scalp locations of the 10–20 system, referenced to the vertex (Cz). Data was amplified via a Neuroscan 32-channel amplifier. Horizontal and vertical electro-oculargram were recorded bipolarly. Sampling rate was set at 250 Hz. EEG data was re-referenced offline to the linked mastoids. The EEG recordings were segmented into epochs of waveform that comprised a 100 ms baseline featuring a triangular central fixation object and 1000 ms of upright or inverted biological motion comprising walking or crawling. For the elimination of electrical artifacts caused by eye and body movements, EEG data was rejected offline by visual editing. No channel interpolation took place due to too few electrodes in the montage. Figure 4. No experience group grand average for upright and inverted walking. Black=upright, Grey = inverted BM. No significant differences were found at any electrode. Figure 5. Experience group upright and inverted crawling stimuli. Black=upright, Grey = inverted BM. No significant differences were found at any electrode. Figure 6. No experience group upright and inverted crawling stimuli. Black=upright, Grey = inverted BM. No significant differences were found at any electrode. Discussion This study suggests that parietal regions are associated with the perception of biological motion even at 9-11 weeks. This further suggests that the gradual development of BM processing skills begins earlier in development than has previously been suggested by electrophysiological evidence. Crucially, the positivity in parietal regions that is associated with a unified percept is present for those infants with experience only. This strongly suggests that experience refines the perception of biological motion. We conclude that even at 10 weeks the link between action perception and action production is tightly woven. Further, it is likely that this link is a fundamental component of the development of social information processing mechanisms. It is therefore suggested that BM processing is the foundation of social-cognitive development upon which other elements of social cognition are built (see Reid & Striano, 2007, for a proposed model of how this may work). Figure 1. Examples of Experience group bath walking (left), no experience group bath (central) and ERP phase (right). Results Visual inspection suggested no effects other than in parietal electrodes. We analyzed these data in a 2 (experience group) X 3 location: left P7, P3, central Pz, right P4, P8) repeated measures ANOVA for upright and inverted walking and crawling stimuli. We found that those infants who had the experience of walking produced parietal region activity from 350ms-550ms after stimulus onset that was more positive in mean amplitude for the upright walking when contrasted with the inverted walking stimuli, F(1,8) = 6.767, p = 0.025. This is indicative of perceiving the stimuli as containing BM as suggested by the results of Reid et al., (2006) and Marshall et al., (2009). Those infants who had not had the experience of walking did not differentiate the conditions within this epoch as measured by ERP amplitudes. No other significant effects were found for any stimulus type at any timepoint. Acknowledgements We thank Manuela Stets for testing some of the infants for this study. We also thank the infants and their parents for their participation in this study. This study was funded by an ESRC Small Grant (project number RES-000-22-3277) References Barresi,J. & Moore,C. (1996). Behavioral and Brain Sciences, 19(1), 107-122.Marshall, P., et al. (2009). Developmental Neuropsychology, 34, 368-377. Meltzoff, A.N. (1995). Developmental Psychology, 31, 838-850. Reid, V.M. et al. (2006). Neuroscience Letters, 95, 211-214. Reid, V.M., & Striano, T. (2007). European J. Dev. Psychology, 4(1), 100-110. Sommerville, J.A., et al. (2005). Cognition, 96, B1-B11. Thelen, E., & Fisher, D.M. (1983). Journal of Motor Behavior, 18, 353-377.