Sally S, Gurnsey R
Department of Psychology, Concordia University, Montreal, Quebec, Canada.
Spat Vis. 2001 Jun;14(2):217-34. doi: 10.1163/156856801300202940.
Humans are extremely sensitive to symmetry when it is foveated but sensitivity drops as a symmetrical region of a fixed size is moved into the periphery. A psychophysical study was undertaken to determine if eccentricity dependent sensitivity loss could be overcome by magnifying stimuli at each eccentricity (E) by a factor F = 1 + E/E2, where E2 indicates the eccentricity at which the size of a stimulus must be doubled, relative to a foveal standard, to achieve equivalent performance. The psychophysical task required subjects to decide on each trial in which of two intervals a symmetrical stimulus had been presented. Stimuli were presented at a range of sizes and eccentricities (0 to 8 degrees) and the probability of a correct discrimination was computed for each condition. In Experiment 1, thresholds were measured with stimuli set to maximum available contrast and, in Experiment 2, stimuli were presented at a constant multiple of contrast detection threshold. In both experiments, a single scaling function removed most of the eccentricity dependent variation from the data. However, the E2 value recovered for one subject tested in both experiments was larger by about 65% when stimuli were not equated for visibility. We conclude that symmetry detection can be equated across a range of eccentricities by scaling stimuli with an E2 in the range of 0.88 to 1.38 degrees. Failure to equate for visibility across all viewing conditions may result in an inflated estimate of E2.
当对称区域位于中央凹时,人类对其极为敏感,但当一个固定大小的对称区域移至周边时,敏感度会下降。开展了一项心理物理学研究,以确定是否可以通过将每个离心率(E)处的刺激放大F = 1 + E/E2倍来克服离心率依赖性敏感度损失,其中E2表示相对于中央凹标准,刺激大小必须翻倍才能达到同等表现的离心率。心理物理学任务要求受试者在每次试验中判定对称刺激出现在两个时间间隔中的哪一个。刺激以一系列大小和离心率(0至8度)呈现,并针对每种情况计算正确辨别的概率。在实验1中,将刺激设置为最大可用对比度来测量阈值,在实验2中,以对比度检测阈值的恒定倍数呈现刺激。在两个实验中,一个单一的缩放函数消除了数据中大部分与离心率相关的变化。然而,在两个实验中都进行测试的一名受试者,当刺激的可见度未达到均等时,其恢复的E2值大约大65%。我们得出结论,通过用0.88至1.38度范围内的E2缩放刺激,可以在一系列离心率上使对称检测达到均等。未能在所有观看条件下使可见度达到均等可能会导致对E2的估计偏高。