A Systematic Review of Visual Competition Effects in the Visual World Paradigm

The visual world paradigm (VWP) is a commonly-used method for studying speech processing. The VWP involves participants looking at an array of objects and tracking their eye movements as they listen to speech. Visual competition effects—a specific finding identified using the VWP—describe the tendency for participants to fixate on visually similar objects to those mentioned in the audio. There has been little metanalytic work examining the methodology of VWP studies, despite recent evidence that common differences in study design and analysis can greatly affect outcomes. The current paper assessed visual similarity coding, preview time selection and analysis time window selection and found that methodological decisions of the literature had an overall lack of consistency and several instances of practices that might impede proper interpretation of results. Future VWP research on visual competitors should consider using validated visual stimuli, alternative methods of statistical analysis that don’t require specific time windows and provide additional justification for methodological decisions. Future research should also attempt to resolve the lack of clarity regarding the theoretical background behind the VWP.