Original article Infant ERP amplitudes change over the course of an experimental session: Implications for cognitive processes and methodology Manuela Stets ⇑ , Vincent M. Reid Department of Psychology, Durham University, UK Received 11 June 2010; received in revised form 5 October 2010; accepted 10 October 2010 Abstract Event-related Potentials (ERP) studies are a widely used methodology to investigate the early cognitive development in infants of all age ranges. We examined changes in amplitude contribution in a set of previously published data (see [4]) in relation to the Neg- ative component as a function of number of trials contributing to each individual average and with time as a co-variate to that con- tribution. With only seven trials contributing to each individual’s average the Nc for the object-directed condition was significantly more negative than the one for the averted-gaze condition, which is the opposite of the effect reported in Hoehl et al. [4]. The analysis including time as a co-variate revealed that the effect did not alter according to the amount of time taken during the experiment to collect each trial, suggesting that data of the quality needed to contribute to the ERP is itself a measure of time varying components of the experimental session. We conclude that infants initially allocate more attentional resources to object-directed gaze. This sug- gests that the temporal dynamics of infant ERP violates assumptions present in adult ERP, and that differences between conditions in relatively few trials inform us of cognitive capacities during infancy. Ó 2010 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved. Keywords: Infant; ERP; Signal-to-noise ratio; ERP methodologies; Eye gaze; Attention 1. Introduction Event-related Potentials (ERPs) are used to investi- gate the cognitive and neural architecture of many phenomena in the field of psychology. As it is a non- invasive way of investigating cognitive processes related to the perception and processing of a certain stimulus or a kind of stimuli, it is well-suited for research with infants and young children (e.g., [1,2]). This is particu- larly the case with infant populations as behavioral responses that can be measured in older infants are often not clearly present at younger ages (see [3]). For exam- ple, an infant may have the capacity to react to certain stimuli and differently process multiple streams of stim- uli, but no evidence for this would be manifested at the behavioral level despite differences in processing seen at the level of neural correlates (see [4–6]). One important characteristic of ERP-studies con- ducted to date with infants is that the methodologies applied and the ways of interpreting the data are derived from cognitive neuroscience with adult populations. These two populations need to be treated differently with respect to how ERP data are recorded and ana- lyzed (see also [7]). Should these elements be standard- ized, there would still remain differences between infants and adults concerning several physical character- istics such as the thickness of the skull or myelination, leading to different morphologies of the resultant ERPs (see [2,7]). Also, the cognitive processes triggered by the 0387-7604/$ - see front matter Ó 2010 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.braindev.2010.10.008 ⇑ Corresponding author. Address: Department of Psychology, Dur- ham University, Science Site, South Road, Durham DH1 3LE, UK. Tel.: +44 191 33 40808. E-mail address: Manuela.stets@durham.ac.uk (M. Stets). www.elsevier.com/locate/braindev Brain & Development 33 (2011) 558–568