Show simple item record

dc.contributor.authorFriedman, Lee ( )
dc.date.accessioned2019-09-30T13:50:43Z
dc.date.available2019-09-30T13:50:43Z
dc.date.issued2020-04
dc.identifier.citationFriedman L. Brief communication: Three errors and two problems in a recent paper: gazeNet: End-to-end eye-movement event detection with deep neural networks (Zemblys, Niehorster, and Holmqvist, 2019). Behav Res Methods. 2020 Apr 13. doi: 10.3758/s13428-019-01342-x.en_US
dc.identifier.urihttps://digital.library.txstate.edu/handle/10877/8649
dc.description.abstractA final version of this research is now published as of April 13, 2020 in Behavior Research Methods. Access to this article is through this link: Brief communication: Three errors and two problems in a recent paper: gazeNet: End-to-end eye-movement event detection with deep neural networks (Zemblys, Niehorster, and Holmqvist, 2019) https://rdcu.be/b3z4n
dc.description.abstractZemblys et al. (Behavior Research Methods, 51(2), 840–864, 2019) reported on a method for the classification of eyemovements (“gazeNet”). I have found three errors and two problems with that paper that are explained herein. Error 1: The gazeNet classification method was built assuming that a hand-scored dataset from Lund University was all collected at 500 Hz, but in fact, six of the 34 recording files were actually collected at 200 Hz. Of the six datasets that were used as the training set for the gazeNet algorithm, two were actually collected at 200 Hz. Problem 1 has to do with the fact that even among the 500 Hz data, the inter-timestamp intervals varied widely. Problem 2 is that there are many unusual discontinuities in the saccade trajectories from the Lund University dataset that make it a very poor choice for the construction of an automatic classification method. Error 2 The gazeNet algorithm was trained on the Lund dataset, and then compared to other methods, not trained on this dataset, in terms of performance on this dataset. This is an inherently unfair comparison, and yet nowhere in the gazeNet paper is this unfairness mentioned. Error 3 arises out of the novel event-related agreement analysis employed by the gazeNet authors. Although the authors intended to classify unmatched events as either false positives or false negatives, many are actually being classified as true negatives. True negatives are not errors, and any unmatched event misclassified as a true negative is actually driving kappa higher, whereas unmatched events should be driving kappa lower.
dc.formatText
dc.format.extent33 pages
dc.format.medium1 file (.pdf)
dc.format.medium1 file (.zip)
dc.format.medium1 file (.txt)
dc.language.isoen_USen_US
dc.relationCorrected Eye-movement Event Matching Procedure and Computation of the Event Level Cohen's Kappa https://digital.library.txstate.edu/handle/10877/8650
dc.subjectEye movement classificationen_US
dc.subjectgazeNeten_US
dc.subjectDeep neural networksen_US
dc.subjectArticle commentary
dc.titleBrief Communication: Three Errors and Two Problems in a Recent Paper: gazeNet : End-to-end Eye-movement Event Detection with Deep Neural Networks (Zemblys, Niehorster, and Holmqvist, 2019)en_US
txstate.documenttypePaper
txstate.departmentComputer Science


Download

Thumbnail
Icon

This item appears in the following Collection(s)

Show simple item record