Donkin Chris, Little Daniel R, Houpt Joseph W
School of Psychology, University of New South Wales.
School of Psychological Sciences, University of Melbourne.
J Exp Psychol Hum Percept Perform. 2014 Jun;40(3):1183-202. doi: 10.1037/a0035947. Epub 2014 Mar 17.
The ability to trade accuracy for speed is fundamental to human decision making. The speed-accuracy trade-off (SAT) effect has received decades of study, and is well understood in relatively simple decisions: collecting more evidence before making a decision allows one to be more accurate but also slower. The SAT in more complex paradigms has been given less attention, largely due to limits in the models and statistics that can be applied to such tasks. Here, we have conducted the first analysis of the SAT in multiple signal processing, using recently developed technologies for measuring capacity that take into account both response time and choice probability. We show that the primary influence of caution in our redundant-target experiments is on the threshold amount of evidence required to trigger a response. However, in a departure from the usual SAT effect, we found that participants strategically ignored redundant information when they were forced to respond quickly, but only when the additional stimulus was reliably redundant. Interestingly, because the capacity of the system was severely limited on redundant-target trials, ignoring additional targets meant that processing was more efficient when making fast decisions than when making slow and accurate decisions, where participants' limited resources had to be divided between the 2 stimuli.
用准确性换取速度的能力是人类决策的基础。速度 - 准确性权衡（SAT）效应已经经过了数十年的研究，并且在相对简单的决策中已被充分理解：在做出决策前收集更多证据能让人更准确，但速度会变慢。在更复杂的范式中的SAT受到的关注较少，主要是由于可应用于此类任务的模型和统计方法存在局限性。在这里，我们利用最近开发的同时考虑反应时间和选择概率来测量能力的技术，对多信号处理中的SAT进行了首次分析。我们表明，在我们的冗余目标实验中，谨慎的主要影响在于触发反应所需的证据阈值量。然而，与通常的SAT效应不同，我们发现当参与者被迫快速反应时，他们会策略性地忽略冗余信息，但仅当额外刺激确实冗余时才会如此。有趣的是，由于在冗余目标试验中系统的能力受到严重限制，忽略额外目标意味着在快速决策时比在做出缓慢而准确的决策时处理效率更高，因为在后者中参与者有限的资源必须在两种刺激之间分配。
