Role of synchrony in contour binding: some transient doubts sustained

被引:29
作者
Dakin, SC [1 ]
Bex, PJ [1 ]
机构
[1] UCL, Inst Ophthalmol, London EC1V 9EL, England
关键词
D O I
10.1364/JOSAA.19.000678
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
The temporal correlation hypothesis proposes that neurons signal mutual inclusion in complex features, such as extended contours, by phase-locking their firing [C. M. Gray and W. Singer, Proc. Natl. Acad. Sci, USA 86, 1698 (1989)]. Although this hypothesis remains controversial, a number of recent psychophysical studies have suggested that temporal correlation among features can indeed promote perceptual grouping. In particular, subjects are better at detecting extended visual contours embedded within a field of distractor elements when a small delay is present between a cycling presentation of the contour and the background [Nature 394, 179 (1988)]. We have replicated this finding and examined three potentially confounding factors. First, we controlled local density and used more curved contours composed of bandpass elements to confirm that the effect was associated with contour integration and not with the operation of coarse-scale spatial filters. Second, we minimized the effects of saccadic eye movements (which could combine with the flicker of the asynchronous display to introduce motion cues at the contour location) both by using a fixation marker that was visible only when observers made a saccade (allowing them to reject these trials) and by retinally stabilizing the stimulus. We report that eye movements contribute to the effect, Third, we asked if either visible persistence or transients at the onset and the offset of the asynchronous stimuli might contribute to the effect. We report that the effect is largely abolished by the inclusion of prestimulus and poststimulus masks and is entirely abolished by ramping the contrast of the stimulus on and off. Neither ramping, masking, nor stabilization should specifically disrupt a contour-binding scheme based on temporal synchrony, and we conclude that it is the transient component at the onset and the offset of these stimuli that is responsible for the reported advantage for asynchronous presentation. (C) 2002 Optical Society of America.
引用
收藏
页码:678 / 686
页数:9
相关论文
共 21 条
[11]   Cortical oscillatory responses do not affect visual segmentation [J].
Kiper, DC ;
Gegenfurtner, KR ;
Movshon, JA .
VISION RESEARCH, 1996, 36 (04) :539-544
[12]   Visual form created solely from temporal structure [J].
Lee, SH ;
Blake, R .
SCIENCE, 1999, 284 (5417) :1165-1168
[13]   The influence of temporal phase differences on texture segmentation [J].
Leonards, U ;
Singer, W ;
Fahle, M .
VISION RESEARCH, 1996, 36 (17) :2689-2697
[14]   MODEL FOR VISUAL SHAPE RECOGNITION [J].
MILNER, PM .
PSYCHOLOGICAL REVIEW, 1974, 81 (06) :521-535
[15]   The VideoToolbox software for visual psychophysics: Transforming numbers into movies [J].
Pelli, DG .
SPATIAL VISION, 1997, 10 (04) :437-442
[16]  
ROGERSRAMACHANDRAN D, 1991, INVEST OPHTH VIS SCI, V32, P1034
[17]   Synchrony unbound: A critical evaluation of the temporal binding hypothesis [J].
Shadlen, MN ;
Movshon, JA .
NEURON, 1999, 24 (01) :67-77
[18]  
SINGER W, 1995, ANNU REV NEUROSCI, V18, P555, DOI 10.1146/annurev.ne.18.030195.003011
[19]   Visual synchrony affects binding and segmentation in perception [J].
Usher, M ;
Donnelly, N .
NATURE, 1998, 394 (6689) :179-182
[20]  
VONDERMALSBURG C, 1981, MODELS NEURAL NETWOR, V2