Exploration of Methodological and Participant-Related Influences on the Number of Artifacts in ERP Data
Stephanie M. Shields, Caitlin E. Morse, and Dr. David F. Nichols
- Examined individual differences in participants and quality of EEG recordings, hoping to find ways for EEG researchers to minimize the occurence of artifacts in their data.
- Data collected from Fall 2014-Fall 2015.
- Poster presented at Synapse 2015.
- Manuscript published in IMPULSE.
Event related potential (ERP) data has low signal-to-noise ratio, requiring the conduction of a large number of trials in order to collect sufficient amounts of data. As a consequence, collecting ERP data can be time consuming and can result in a great deal of wasted efforts. Therefore, it would highly beneficial if researchers could minimize the number of artifacts that occur in the data, therein minimizing the number of discarded trials and the number of trials needed. Consequently, this study examined both methodological and individual difference variables in order to discern which factors could be adjusted in an attempt to minimize the amount of data that must be discarded. We used electroencephalography (EEG) to measure the responses of over 70 Roanoke College students during a task that had previously been shown to result in some of the aforementioned issues. The participants were asked to look at a computer screen as 80 images of faces were displayed. In two of the four runs, participants were asked not to blink for the duration of the run (approximately 2.5 minutes), but in the other two runs, images of fingerprints were also displayed to give participants a chance to blink. The task participants started with was randomly assigned. Subjects also completed a survey, which examined possible individual difference variables. For example, participants were asked about their reason(s) for participation, whether they knew the researchers prior to the study, and their emotional state both at the beginning and end of the conduction of the EEG. Results will be presented in relation to these factors to determine if they have an impact on the number of artifacts observed. We hope at the end of this study to be able to make suggestions to researchers about how to minimize the number of artifacts in EEG data.