Author Correction: Moderate acute alcohol intoxication increases visual motion repulsion

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

www.nature.com/scientificreports www.nature.com/scientificreports/ At the end of the Methods and Materials section, under the subheading "Data Analysis", the following sentence should be included: "Repeated measures ANOVA was applied on the extracted threshold and log10-transformed lapse rates. " Results shown in Figure 3 of this Article are corrected following the re-analysis. The corrected Figure 3 is shown as Figure 1.
In the Results section, under the subheading "Increased motion repulsion after alcohol administration", the following paragraph: "Repeated measures ANOVA with the surround direction factor (0°, ±30°, ±60° and ±90°) and conditional factor (sober, placebo and alcohol) revealed that there were significant main effects of center-surround direction differences on MR (F (3, 81) = 198.26, p < 0.001, ehat = 0.763) and the condition (F (2, 54) = 19.36, p < 0.001, ehat = 0.992), as well as a significant interaction between them (F (6, 162) = 2.76, p = 0.023, ehat = 0.789). This interaction effect was driven by a significant difference in the MR between the placebo and intoxication states with surround directions of 30° and 60°. Tests were then conducted under the placebo and intoxication conditions to identify significant differences at each test direction. We found a significant difference between both the surround direction (F (3, 81) = 181.97, p < 0.001, ehat = 0.808) and alcohol condition (F (1, 27) = 11.65, p = 0.002), as well as a significant interaction between them (F (3, 81) = 4.43, p = 0.011, ehat = 0.821). Compared to the placebo condition, the amplitudes measured in the intoxication condition were significantly higher, with a surround direction of ±60° (p < 0.01), but not ±0° (p = 0.74), ±30° (p = 0.37) or ±90° (p = 0.41). " should read: "Repeated measures ANOVA with the surround direction factor (0°, ±30°, ±60° and ±90°) and conditional factor (sober, placebo and alcohol) revealed that there were significant main effects of center-surround direction differences on MR (F(3,75) = 169.7, p < 0.001, ehat = 0.764) and the condition (F(2,50) = 12.81, p < 0.001, ehat = 0.991), as well as a significant interaction between them (F(6,150) = 4.58, p = 0.0007, ehat = 0.832). This interaction effect was driven by a significant MR increase in intoxication state at surround direction of 60°. Tests were then conducted under the placebo and intoxication conditions to identify significant differences at each test direction. Compared to the placebo condition, the amplitudes measured in the intoxication condition were significantly higher at surround direction of 60° (t(25) = −5.27, p < 0.001) and 90° (t(25) = −2.70, p = 0.0121), but not at 0° (t(25) = −2.33, p = 0.028) and 30° (t(25) = −1.43, p = 0.16) (Bonferroni correction of significance level to 0.05/4 = 0.0125). " In the Results section, under the subheading "Discrimination performance", the following paragraph: "The direction discrimination thresholds in the intoxicated state were larger than those in the sober and placebo conditions (F (2, 54 = 29.63, p < 0.001, ehat = 0.775), and the thresholds were modulated by the surround direction (F (3, 81) = 12.13, p < 0.001, ehat = 0.812). Importantly, there was no interaction between the condition and surround direction (F (6, 162) = 0.53, p = 0.697, ehat = 0.603), showing a similar trend as that of the threshold variation for different surround directions. " should read: "The direction discrimination thresholds in the intoxicated state were larger than those in the sober and placebo conditions (F(2,50) = 12.21, p = 0.0005, ehat = 0.684), and the thresholds were modulated by the surround "If this was the case, then one should expect to see a correlation between the thresholds and biases for a given condition at a fixed surround of 30° or 60°. Pearson's correlation was conducted between the perceptual bias and discrimination threshold. No correlation was found in the ± 30° (Pearson r = 0.15, p = 0.24) or the ± 60° (Pearson r = 0.15, p = 0.46) surround direction in the intoxication condition. These results demonstrate that changes in bias may not be related to "higher cognitive" effects. " should read: "We note that if this was the case, the main effect of such a task change would be to shift the psychometric functions (midpoints) in opposite directions from the repulsion effects. This would in fact decrease the repulsion effect in intoxicated condition when compared to control and placebo conditions, an effect that our data did not show. On the other hand, intoxicated condition might simultaneously influence bias and thresholds due to either low-sensory effects or high-cognitive effects, thus showing correlated changes in both measures across subjects. We computed the Pearson's correlations between the thresholds and biases for each condition and surround. Among the 12 correlations (3 conditions x 4 surrounds), two showed significant correlations (Alcohol at 60°: r = 0.55, p = 0.0039; placebo at 90°: r = 0.55, p = 0.0032) after adjustment for multiple tests (Bonferroni adjustment to 0.05/12 = 0.0042) (all others p > 0.01). These results suggest that changes in bias and threshold may not be related to "higher cognitive" effects in intoxicated condition only. " www.nature.com/scientificreports www.nature.com/scientificreports/ Additionally, we re-analyzed the data for all orders of Alcohol measure and we found that the main claims of the study are still valid, irrespective of which Alcohol measure is considered -1 st , 2 nd or 3 rd . Below, we provide results of the analysis as applied in the main article for measures number 1 and number 3 (measure number 2 is presented in Figure 1 of this Correction).
From the 28 persons with at least one Alcohol measure, the Bayesian fitting procedure of psychometric function parameters extraction gave in one measure A1 (Alcohol measure 1) two lapse rates above 0.25. Inspecting the staircases of the subject showed that the person responded to the surround direction instead of the center target for about half of the trials. No other such case was present among the remaining 27 subjects' staircases. Thus, for alcohol measure 1 there were 27 data sets, for alcohol measure 2 there were 26 data sets and alcohol measure 3 there were 21 data sets. Figure 2 presents the results of each analysis.
The ANOVA analyses are presented in Table 1. We note that (1) for Bias measures Condition always showed significant effect with Alcohol measure giving stronger repulsion, (2) Thresholds were always higher in Alcohol measure across all three measures, (3) the lapse rates were higher in alcohol conditions for all three measures and did not interact with surround direction.
The Authors apologize for the errors, and any inconvenience caused.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.