Simon T J, Vaishnavi S
School of Psychology, Georgia Institute of Technology, Atlanta 30332-0170, USA.
Percept Psychophys. 1996 Aug;58(6):915-26. doi: 10.3758/bf03205493.
Two experiments showed that, when selective eye movements were disabled by the presentation of stimuli in the form of afterimages, increased inspection time and facilitative stimulus configurations failed to increase the subitizing limit of 4 objects. Afterimages of two to eight dots induced by a photographic flashgun were shown to 3 adult subjects. For more than 4 objects, enumeration errors occurred at a rate of 20%-30%. Enumeration was effectively perfect for 2-4 linearly configured dots, with occasional errors surprisingly occurring in that range when dots appeared in groups of up to 3 items. No errors occurred in nonafterimage control conditions. Enumeration errors were attributed to failures of individuating dots to be counted due to the deactivation of selective eye movements in afterimages. A third experiment supported this interpretation by disabling eye movements with briefly presented stimuli and producing results much like those of the afterimage conditions.
两项实验表明，当以残像形式呈现刺激使选择性眼动被禁用时，检查时间的增加和促进性刺激配置未能提高4个物体的即时数数极限。用摄影闪光灯产生的2至8个点的残像展示给3名成年受试者。对于超过4个物体，计数错误发生率为20%-30%。对于2-4个呈线性排列的点，计数实际上是完美的，不过当点以最多3个一组出现时，在该范围内偶尔也会惊人地出现错误。在非残像对照条件下未出现错误。计数错误归因于残像中选择性眼动的失活导致要计数的单个点无法区分。第三个实验通过用短暂呈现的刺激使眼动失效并产生与残像条件非常相似的结果，支持了这一解释。
