Drew Trafton, Williams Lauren H
Department of Psychology, University of Utah, 380 S 1530 E Beh S 1003, Salt Lake City, UT 84112 USA.
Cogn Res Princ Implic. 2017;2(1):44. doi: 10.1186/s41235-017-0082-3. Epub 2017 Nov 22.
Searching for targets in the visual world, or visual search, is something we all do every day. We frequently make 'false-negative' errors, wherein we erroneously conclude a target was absent when one was, in fact, present. These sorts of errors can have tremendous costs, as when signs of cancers are missed in diagnostic radiology. Prior research has characterized the cause of many of these errors as being due to failure to completely search the area where targets may be present; indeed, roughly one-third of chest nodules missed in lung cancer screening are never fixated (Drew, Võ, Olwal, Jacobson, Seltzer and Wolfe, Journal of Vision 13:3, 2013). This suggests that observers do not have a good representation of what areas have and have not been searched prior to declaring an area target free. Therefore, in six experiments, we sought to examine the utility of reducing the uncertainty with respect to what areas had been examined via online eye-tracking feedback. We hypothesized that providing information about what areas had or had not been examined would lead to lower rates of false negatives or more efficient search, namely faster response times with no cost on target detection accuracy. Neither of these predictions held true. Over six experiments, online eye-tracking feedback did not yield any reliable performance benefits.
在视觉世界中寻找目标，即视觉搜索，是我们每天都会做的事情。我们经常会犯“假阴性”错误，即当目标实际上存在时，我们却错误地得出目标不存在的结论。这类错误可能会造成巨大代价，比如在放射诊断中漏诊癌症迹象。先前的研究将许多这类错误的原因归结为未能彻底搜索目标可能存在的区域；事实上，肺癌筛查中漏诊的胸部结节约有三分之一从未被注视过（德鲁、武、奥尔瓦尔、雅各布森、塞尔策和沃尔夫，《视觉杂志》13:3，2013年）。这表明观察者在宣布某个区域没有目标之前，对哪些区域已经被搜索过、哪些区域没有被搜索过并没有清晰的认知。因此，在六个实验中，我们试图通过在线眼动追踪反馈来检验减少关于哪些区域已被检查的不确定性的效用。我们假设，提供关于哪些区域已被检查或未被检查的信息会降低假阴性率或提高搜索效率，即反应时间更快，同时不影响目标检测准确率。但这两个预测都不成立。在六个实验中，在线眼动追踪反馈并未产生任何可靠的性能提升。
