Corrigendum to “Microsaccade directions do not predict directionality of illusory brightness changes of overlapping transparent surfaces” [Vision Research 46 (2006) 3823–3830]

In the above-referenced article, we reported that microsaccade directions do not predict the directionality of illusory brightness changes of overlapping transparent surfaces. While our results and conclusions remain the same, there was a methodological error that requires correction. We used the SRresearch Eyelink2 algorithm to identify microsaccades, but because of an error that was not discovered until after publication, peri-blink eye movements were not removed before microsaccade detection as per standard protocol. As a result, peri-blink eye movements may have been identified as microsaccades when they were not. This could have increased the noise level, leading to the null result reported. We have repeated the study with a greater number of subjects (n = 10), removing any potential spurious microsaccades in the neighborhood of eyeblinks. We replicate our previous null finding using a more conservative analysis and microsaccade detection algorithm modified from Engbert and Kliegl [Engbert, R., & Kliegl, R. (2003). Microsaccades uncover the orientation of covert attention. Vision Research, 43, 1035–1045]. Our results confirm our previous finding that microsaccade directions do not predict directionality of illusory brightness changes of overlapping transparent surfaces, and vice versa. Details of the new analysis and data can be found online in the supplementary material in Appendix A. Although the results are the same, because this replication was carried out with more subjects and analyzed in a more conservative manner, the results in Appendix A effectively replace those in the previous report.