Faculty of Psychology, Warsaw School of Social Sciences and Humanities, ul. Grunwaldzka 98, Wroclaw, Poland.
Conscious Cogn. 2010 Mar;19(1):426-31. doi: 10.1016/j.concog.2009.12.011. Epub 2010 Feb 8.
Post-decision wagering (Persaud, McLeod, & Cowey, 2007) has been proposed as a method of demonstrating that perception can occur without conscious awareness. When wagering is independent from above-chance performance there is evidence of a lack of awareness of the correctness of the first-order discriminations. However, there are reasons to believe that the contingency analysis conducted by Persaud and colleagues failed to measure "the zero accuracy-wagering criterion". The author shows that a Pearson chi-square test employed by Persaud and colleagues is unable to accommodate the hypothesis of partial independence between accuracy and advantageous wagering. On the contrary, the problem of "the zero accuracy-wagering criterion" is best conducted with log-linear models. Moreover, log-linear analysis suggests that absence of advantageous wagering is not sufficient evidence for a lack of conscious awareness. In the case of the blindsight study, there was compelling evidence that the patient GY was partially aware of having information about the correctness of his decisions.
事后打赌（Persaud、McLeod 和 Cowey，2007）被提议作为一种证明感知可以在没有意识的情况下发生的方法。当打赌与高于机会的表现无关时，就有证据表明对第一级辨别正确性的意识不足。然而，有理由认为，Persaud 及其同事进行的协变分析未能衡量“零准确性-打赌标准”。作者表明，Persaud 及其同事使用的 Pearson 卡方检验无法适应准确性和有利打赌之间部分独立的假设。相反，“零准确性-打赌标准”的问题最好通过对数线性模型来处理。此外，对数线性分析表明，缺乏有利打赌并不是缺乏意识的充分证据。在盲视研究的情况下，有令人信服的证据表明患者 GY 对自己决策正确性的信息有部分意识。
