Kahn M W, Fox H, Rhode R
Department of Psychology, University of Arizona, Tucson 85721.
J Pers Assess. 1988 Fall;52(3):516-23. doi: 10.1207/s15327752jpa5203_14.
In a previous study of the ability of expert Rorschach interpreters to detect faking that used true and malingered protocols, the experts faired very poorly. In this study, 50% of these same protocols were scored by the Exner system and analyzed by Exner's Semantic Computer Interpretation program. The program indicated invalidity of protocols only on the basis of low R and designated the faked protocols high on psychotic descriptors, barely indicating psychosis for the true schizophrenic protocols. Unlike the judges, however, the computer gave no psychotic designation to the normal protocols. The scoring-computer analysis method was as susceptible to faking as were the clinical judgments.
在之前一项关于罗夏墨迹测验专家解读员辨别伪装能力的研究中,使用了真实和伪装的测验记录,专家们的表现非常糟糕。在这项研究中,同样的这些测验记录中有50%由艾克纳系统评分,并通过艾克纳的语义计算机解读程序进行分析。该程序仅基于低R值指出测验记录无效,并将伪装的测验记录判定为在精神病性描述方面得分高,而对于真正的精神分裂症测验记录几乎未显示出精神病特征。然而,与评判人员不同的是,计算机没有给正常的测验记录判定为精神病性。评分计算机分析方法与临床判断一样容易受到伪装的影响。