Power for detecting the presence of set size differences in the contralateral delay activity
Date:
Abstract
The contralateral delay activity, an event-related potential that tracks closely with memory load, has been a very useful and popular tool in learning about the visual working memory system. However, with recent concerns around the reproducibility of science, ensuring studies are high powered to ensure the reliability of the results is critical. I present downsampling analyses estimating the power to detect the presence of the contralateral delay activity, and to detect set-size differences in the contralateral delay activity. The results are surprising - typical numbers of subjects and trials appear to be very underpowered at detecting set-size differences.