It is argued in [1] that [2] was able to classify EEG responses to visual\nstimuli solely because of the temporal correlation that exists in all EEG data\nand the use of a block design. We here show that the main claim in [1] is\ndrastically overstated and their other analyses are seriously flawed by wrong\nmethodological choices. To validate our counter-claims, we evaluate the\nperformance of state-of-the-art methods on the dataset in [2] reaching about\n50% classification accuracy over 40 classes, lower than in [2], but still\nsignificant. We then investigate the influence of EEG temporal correlation on\nclassification accuracy by testing the same models in two additional\nexperimental settings: one that replicates [1]'s rapid-design experiment, and\nanother one that examines the data between blocks while subjects are shown a\nblank screen. In both cases, classification accuracy is at or near chance, in\ncontrast to what [1] reports, indicating a negligible contribution of temporal\ncorrelation to classification accuracy. We, instead, are able to replicate the\nresults in [1] only when intentionally contaminating our data by inducing a\ntemporal correlation. This suggests that what Li et al. [1] demonstrate is that\ntheir data are strongly contaminated by temporal correlation and low\nsignal-to-noise ratio. We argue that the reason why Li et al. [1] observe such\nhigh correlation in EEG data is their unconventional experimental design and\nsettings that violate the basic cognitive neuroscience design recommendations,\nfirst and foremost the one of limiting the experiments' duration, as instead\ndone in [2]. Our analyses in this paper refute the claims of the "perils and\npitfalls of block-design" in [1]. Finally, we conclude the paper by examining a\nnumber of other oversimplistic statements, inconsistencies, misinterpretation\nof machine learning concepts, speculations and misleading claims in [1].\n