Visual event-related potentials (ERPs) have been widely used during the past two decades to study cognition in developmental populations. One of the main challenges related to current infant ERP practices is the need to obtain good-quality ERPs that provide reliable results with which to examine research issues. There are two main aspects that play a role in obtaining a valid ERP trial: the infant’s attention to the stimuli and an EEG signal free of artefacts—primarily eye and body movement artefacts—. The aims of this thesis were to explore alternative methodological approaches that could potentially improve current ERP practice as well as to assess the validity of the current ERP editing methods. Specifically, the first aim was to study the use of eye tracking techniques to improve current ERP data collection methods and ERP final data quality. In Chapter 2, eye tracking technology was used as a mean to study how advanced ERP components are negatively affected by the fixation location distance to the stimulus. Chapter 3 evaluated the changes in attention and visual behaviour of a gaze-contingent paradigm used during ERP data collection in 10-month-old infants. The second aim was to investigate to what extent the current infant ERP editing methods are reliable and to assess the need for their further standardization within the field. Chapter 3 focused on understanding the current editing methods used in ERP infant research for selecting ERP trials free of artefacts. The study analyzed the agreement between three experienced infant ERP human editors and one algorithm for selecting artefact-free ERP trials as well as the EEG signal characteristics that influenced the selection of valid trials. The implications of the results of the three studies presented in this thesis for current infant ERP practices are discussed in Chapter 5. Overall, this thesis highlights the importance and need of the field to work towards the improvement of current methodologies for collecting and analyzing infant ERP data.