Naturalistic Stimulation in M/EEG: Promises and Pitfalls
- Phillip Alday
- Thursday 4 April 2019
2311 BD Leiden
Laboratory experiments on language often lack ecological validity. In addition to the intrusive laboratory equipment, the language used is often highly constrained in an attempt to control possible confounds. More recent research with naturalistic stimuli has been largely confined to fMRI, where the low temporal resolution helps to smooth over the uneven finer structure of natural language use. Nonetheless, it is possible to study both speech and language under naturalistic stimulation using temporally sensitive methods such as EEG and MEG. The key to this work is the use of modern statistical and computational methods, but these also have their own complications and limits. The full complexity of naturalistic language use cannot be captured by carefully controlled designs alone, so it is important that we are aware of both the promises and pitfalls offered to us by naturalistic stimulation in M/EEG.