拓扑不变性质在运动诱发视盲现象中的作用

知觉物体的定义是认知科学领域富有争议的中心问题之一．本研究通过运动诱发视盲(MIB)现象的研究发现：a．两个图形当它们是连通的,或位于同一个洞中,或是具有相同的洞的个数,更倾向于整体共同消失或重现;b．前景图形和背景图形的拓扑性质差异,较之非拓扑性质差异,显著地削弱MIB现象;c．相对于其他各种几何性质的变化,拓扑性质的变化能更快地被知觉到返回意识状态．这些结果揭示了拓扑性质定义的知觉物体是MIB过程操作的基本单元,提示了拓扑性质改变产生新知觉物体的表征可能是意识下的过程,进一步支持了知觉物体的拓扑学定义．     

视觉拥挤效应的神经机制

当出现在边缘视野的一个物体被周围其他物体包围时,视觉系统对它的识别会很困难,这种现象叫做视觉拥挤效应.研究拥挤效应既有利于理解人类进行客体识别的过程,也对治疗黄斑变性、弱视和阅读障碍等视觉病变有显著的临床意义.自拥挤效应被提出以来,对拥挤效应的特性、神经机制和影响因素等都做了深入地研究.本文将系统地综述拥挤效应的研究进展,包括其特性、现有的理论假设、计算模型、可能涉及的大脑区域以及近年来利用知觉学习消除拥挤效应的一些工作,最后对该领域的未来发展给出建议.尽管在这个领域已经获得了丰富的成果,但在许多问题上仍有争议,未来还需要更为巧妙的设计和精确的技术进一步解决这些问题.     

连续闪现抑制下对洞的无意识知觉

大量收敛一致的实验证据表明,图形的大范围性质可以由拓扑性质来描述,并且其检测发生在视觉过程的最早期,这些证据几乎全部来自对意识上知觉的研究,而拓扑性质的意识下加工机制尚有待发掘.意识下知觉是人的感官系统客观上接受刺激呈现但主观上没有察觉的知觉,其机制和应用长期以来一直是知觉研究的热点.本文采用了一种双眼竞争的变式——连续闪现抑制,将待检测的刺激图形掩蔽,使之处于意识下状态,考察意识下知觉中拓扑性质的加工."洞的个数"作为一种拓扑性质是本文的研究对象.通过量度被抑制图形从发生变化到被知觉的被抑制时间,或者被试对被抑制的变化图形的正确检测率,我们发现,相比于不变或者各种非拓扑性质变化,意识下知觉中的拓扑性质(洞的个数)的变化会使图形更快、更容易被检测到.本研究揭示了拓扑性质(洞的个数)在意识下知觉中的优先性,将拓扑知觉理论从意识上知觉领域拓展到了意识下知觉领域,为拓扑性质加工的早期性提供了有力的证据;另一方面,本研究也提示了拓扑性质经由皮层下视通路加工处理的可能性.     

拓扑差异物体的构型处理：一项脑电研究

根据陈霖的拓扑知觉理论,在视觉感知中,拓扑差异的检测要快于局部特征的检测.本研究组认为,这一现象可能是对有＂洞＂图形和无＂洞＂图形的感知差异所造成的.本文旨在检验这两类图形所存在的知觉差异究竟是什么.利用脑电（ERP）技术,测量＂洞＂和＂非洞＂图形所引起的视觉诱发电位（VEP）,探究＂洞＂和＂非洞＂的区别.两类图形引起的VEP区别体现在N170上,＂洞＂图形的反转使得N170的潜伏期显著变化,而＂非洞＂图形的反转不会引起N170潜伏期的变化.人脸与非人脸物体的反转引起的N170变化与本结果相似.本研究认为,对＂洞＂的感知更偏向于整体加工,而对于＂非洞＂则更偏向于特征加工.     

小鼠对不同拓扑性质图形的识别

视觉感知的一系列研究都支持大范围拓扑感知的理论．拓扑性质作为整体性质,是视觉感知的基础．视觉对图形拓扑特征差异的感知要优先于对局部特征差异的感知．采用Y迷宫研究了小鼠对不同拓扑性质图形的识别．训练小鼠学习识别圆环和实心矩形这一对拓扑性质不同的图形,之后用拓扑特征相同或不同的其他图形测试小鼠,这些图形包括空心矩形、实心圆、缺口的圆环、缺口的空心矩形．实验结果表明,学会识别圆环(奖励)和实心矩形(无奖励)的小鼠无法区分实心圆和实心矩形以及圆环和空心矩形,但是能够分别从缺口圆环、缺口的空心矩形、实心圆与空心矩形组成的图形对中识别出空心矩形．因此证实了小鼠的视觉系统能够感知拓扑特征的差异并且具有对拓扑性质的概括能力．结果为拓扑知觉对视觉系统来说是基本的这一假设提供了证据．     

整体知觉依赖于双侧视皮层的协同工作: TMS研究

以往研究表明, 右侧和左侧大脑半球分别主要负责复合图形的整体和局部知觉. 本研究使用经颅磁刺激(TMS)研究对复合刺激的整体知觉是否需要双侧皮层的协同工作. 在实验一中, 被试辨别复合字母的整体或者局部属性, 在施加单脉冲TMS或者伪TMS的同时记录反应时. 尽管观测到了整体优势效应(即对整体目标的反应快于局部目标, 整体对于局部有更强的干扰), 但是TMS减弱了整体对局部加工的干扰并且增强局部对整体加工的干扰. 实验二排除了实验一中观测到的效应是知觉学习后果的可能性. 实验三利用复合图形刺激, 观测到了与实验一相似的TMS效应, 而且发现TMS延长对整体的反应时(RTs)并减少对局部的反应时. 最后, 实验一和实验三中观测到的TMS效应在左半球或者右半球没有显著差异. 这些结果支持了协同假设, 该假设认为复合刺激的整体知觉依赖于双侧视皮层的协同工作     

"大范围优先"对象形成的神经关联:前颞叶

知觉对象是什么?这事看起来很简单,却是认知科学许多领域中最核心的问题,目前为止还没有公认的解答."大范围优先"的拓扑学方法把知觉对象的定义严格地联系到拓扑变换下的不变量.知觉对象的直观概念-形状改变时其整体一致性得以保持-可以精确地被拓扑不变量,如连接组分和洞,所刻画.知觉对象的拓扑学定义已经被大量的行为学实验所验证.这些实验一致表明,虽然对象的一致性能够在非拓扑变换下得以保持,却会被拓扑变换所破坏,从而一个新的对象被感知到.特别是进一步的功能磁共振成像实验揭示,前颞页区参与拓扑知觉和知觉对象的形成,而这一脑区本来是形式视觉通路的终点.行为学上"大范围优先"的结果与视觉通路神经解剖学结果的悖逆,提示我们应该注意对象表征形成的问题和更广泛的意义上,知觉到底在何处发生的基本问题.     

双稳态知觉的脑机制

当对视觉输入的信息有多种解释时,人们的知觉状态会在这些解释之间随机切换.目前,这种多稳态或双稳态知觉的神经机制仍处于争论之中.本研究分别以鲁宾花瓶和内克尔立方体两种双稳态图形为对象来研究双稳态知觉在大脑结构上的机制.首先,通过计算两种双稳态知觉的切换频率,发现两者切换频率之间有正相关关系.在此基础上,通过计算两种知觉切换频率与大脑灰质体积的相关性,发现鲁宾花瓶和内克尔立方体图形的切换频率均和右侧额下回的灰质体积存在显著的负相关.本研究表明,不同双稳态知觉之间具有共同的神经基础,这一共同的基础位于右侧额下回,支持了自上而下的加工在双稳态知觉中具有重要的作用.     

“大范围优先”对象形成的神经关联：前颞叶

知觉对象是什么？这事看起来很简单,却是认知科学许多领域中最核心的问题,日前为止还没有公认的解答。“大范围优先”的拓扑学方法把知觉对象的定义严格地联系到拓扑变换下的不变量。知觉对象的直观概念—形状改变时其整体一致性得以保持一可以精确地被拓扑不变量,如连接组分和洞,所刻画。知觉对象的拓扑学定义已经被大量的行为学实验所验证。这些实验一致表明,虽然对象的一致性能够在非拓扑变换下得以保持,却会被拓扑变换所破坏,从而一个新的对象被感知到。特别是进一步的功能磁共振成像实验揭示,前颞页区参与拓扑知觉和知觉对象的形成,而这一脑区本来是形式视觉通路的终点。行为学上“大范围优先”的结果与视觉通路神经解剖学结果的悖逆,提示我们应该注意对象表征形成的问题和更广泛的意义上,知觉到底在何处发生的基本问题。     

视觉系统中的性别差异

大脑性别差异一直是神经科学领域中的一个热门话题.以前的性别差异研究往往关注与高级认知相关的脑区结构和功能差异,而对低级感知觉系统中的性别差异没有足够的重视.近年来,越来越多的研究结果表明,男性和女性在视知觉功能上也存在着明显的差异.本文首先回顾和梳理了视觉系统中存在性别差异的行为学和神经生物学证据,然后对视觉系统中性别差异的来源提出了两种可能的解释,接着讨论了视知觉功能存在性别差异的进化学意义,最后强调了在感知觉研究中把性别作为一个实验变量的的重要性,并对后续的性别差异研究提出了一些具体的建议.     

Crowding by invisible flankers

Background

Human object recognition degrades sharply as the target object moves from central vision into peripheral vision. In particular, one''s ability to recognize a peripheral target is severely impaired by the presence of flanking objects, a phenomenon known as visual crowding. Recent studies on how visual awareness of flanker existence influences crowding had shown mixed results. More importantly, it is not known whether conscious awareness of the existence of both the target and flankers are necessary for crowding to occur.

Methodology/Principal Findings

Here we show that crowding persists even when people are completely unaware of the flankers, which are rendered invisible through the continuous flash suppression technique. Contrast threshold for identifying the orientation of a grating pattern was elevated in the flanked condition, even when the subjects reported that they were unaware of the perceptually suppressed flankers. Moreover, we find that orientation-specific adaptation is attenuated by flankers even when both the target and flankers are invisible.

Conclusions

These findings complement the suggested correlation between crowding and visual awareness. What''s more, our results demonstrate that conscious awareness and attention are not prerequisite for crowding.     

Grouping and Crowding Affect Target Appearance over Different Spatial Scales

Crowding is the impairment of peripheral target perception by nearby flankers. A number of recent studies have shown that crowding shares many features with grouping. Here, we investigate whether effects of crowding and grouping on target perception are related by asking whether they operate over the same spatial scale. A target letter T had two sets of flanking Ts of varying orientations. The first set was presented close to the target, yielding strong crowding. The second set was either close enough to cause crowding on their own or too far to cause crowding on their own. The Ts of the second set had the same orientation that either matched the target’s orientation (Grouped condition) or not (Ungrouped condition). In Experiment 1, the Grouped flankers reduced crowding independently of their distance from the target, suggesting that grouping operated over larger distances than crowding. In Experiments 2 and 3 we found that grouping did not affect sensitivity but produced a strong bias to report that the grouped orientation was present at the target location whether or not it was. Finally, we investigated whether this bias was a response or perceptual bias, rejecting the former in favor of a perceptual grouping explanation. We suggest that the effect of grouping is to assimilate the target to the identity of surrounding flankers when they are all the same, and that this shape assimilation effect differs in its spatial scale from the integration effect of crowding.     

Effect of spatial crowding on aggressive behavior in a bonobo colony

According to a previously proposed coping model (De Waal [1989] Zoo Biol. 1(suppl):141–148), gregarious primates will react to crowding by adjusting their behavior in order to limit the amount of aggression displayed. Depending on the duration of the crowding, this can be achieved by the use of different strategies. Chimpanzees (Pan troglodytes) react to a medium‐term duration of spatial crowding by an active tension‐reduction mechanism whereby allogrooming markedly increases while aggression rises only slightly. Under comparable circumstances, a similar increase in grooming is found in bonobos (Pan paniscus). However, to test the coping model for bonobos, data on aggressive behavior during crowding are indispensable. This study provides such information by comparing the aggressive repertoire of a well‐established bonobo colony during a crowded period in winter with those that occurred during an uncrowded control period in summer. Given that during winter the aggression rate differed between food and nonfood contexts, we accounted for a potential effect of food context on aggression. During the crowded period, a significantly higher total frequency of aggression was found than during the control period. However, this increase was small in relation to the reduction in space when compared to similar experiments in other species. Together with observations of increased grooming under crowded conditions, our data confirm the hypothesis that bonobos cope with the increase in tension created by crowding (medium‐term duration) by applying a tension‐reduction strategy. Our data further show that although not all aggressive behaviors appear to be influenced by a high‐density condition, bonobos do not specifically limit their aggressive behaviors to mild, nonprovocative aggressions in a crowded environment. Zoo Biol 23:383–395, 2004. © 2004 Wiley‐Liss, Inc.     

Shrinking Bouma’s window: How to model crowding in dense displays

In crowding, perception of a target deteriorates in the presence of nearby flankers. Traditionally, it is thought that visual crowding obeys Bouma’s law, i.e., all elements within a certain distance interfere with the target, and that adding more elements always leads to stronger crowding. Crowding is predominantly studied using sparse displays (a target surrounded by a few flankers). However, many studies have shown that this approach leads to wrong conclusions about human vision. Van der Burg and colleagues proposed a paradigm to measure crowding in dense displays using genetic algorithms. Displays were selected and combined over several generations to maximize human performance. In contrast to Bouma’s law, only the target’s nearest neighbours affected performance. Here, we tested various models to explain these results. We used the same genetic algorithm, but instead of selecting displays based on human performance we selected displays based on the model’s outputs. We found that all models based on the traditional feedforward pooling framework of vision were unable to reproduce human behaviour. In contrast, all models involving a dedicated grouping stage explained the results successfully. We show how traditional models can be improved by adding a grouping stage.     

The neural correlates of crowding-induced changes in appearance

Object recognition in the peripheral visual field is limited by crowding: the disruptive influence of nearby clutter. Despite its severity, little is known about the cortical locus of crowding. Here, we examined the neural correlates of crowding by combining event-related fMRI adaptation with a change-detection paradigm. Crowding can change the appearance of objects, such that items become perceptually matched to surrounding objects; we used this change in appearance as a signature of crowding and measured brain activity that correlated with the crowded percept. Observers adapted to a peripheral patch of noise surrounded by four Gabor flankers. When crowded, the noise appears oriented and perceptually indistinguishable from the flankers. Consequently, substitution of the noise for a Gabor identical to the flankers ("change-same") is rarely detected, whereas substitution for an orthogonal Gabor ("change-different") is rarely missed. We predicted that brain areas representing the crowded percept would show repetition suppression in change-same trials but release from adaptation in change-different trials. This predicted pattern was observed throughout cortical visual areas V1-V4, increasing in strength from early to late visual areas. These results depict crowding as a multistage process, involving even the earliest cortical visual areas, with perceptual consequences that are increasingly influenced by later visual areas.     

Thresholds vary between spatial and temporal forced-choice paradigms: the case of lateral interactions in peripheral vision

Psychophysicists use spatial or temporal two-alternative forced-choice (2AFC) paradigms interchangeably. Thus, experiments with the same general goal are carried out using one or the other paradigm by distinct or even the same research groups. For example, this situation has occurred both in studies on visual sensitivity in dyslexia and in studies on lateral interactions in peripheral vision. Conflicting results in either field (e.g. whether or not dyslexics have a visual deficit and whether or not peripheral detection is facilitated by the presence of flankers) appear to be resolved on the surmise that spatial and temporal 2AFC paradigms indeed produce different results. We designed experiments in which peripheral detection thresholds for Gabor patches (in the presence or absence of suprathreshold flankers) could be measured using completely equivalent spatial and temporal 2AFC paradigms so that any resultant difference can be unequivocally attributed to the effect of the paradigms themselves. The results showed that spatial 2AFC renders significantly lower sensitivity than temporal 2AFC when the target is presented along with suprathreshold flankers, but about the same sensitivity as temporal 2AFC when the target is presented alone. In the end, this resulted in statistically significant facilitation in peripheral vision only when measured with temporal 2AFC. Separate experiments at each of several peripheral locations revealed that the presence and magnitude of this effect varies not only with psychophysical paradigm but also with retinal locus.     

Reducing crowding by weakening inhibitory lateral interactions in the periphery with perceptual learning

We investigated whether lateral masking in the near-periphery, due to inhibitory lateral interactions at an early level of central visual processing, could be weakened by perceptual learning and whether learning transferred to an untrained, higher-level lateral masking known as crowding. The trained task was contrast detection of a Gabor target presented in the near periphery (4°) in the presence of co-oriented and co-aligned high contrast Gabor flankers, which featured different target-to-flankers separations along the vertical axis that varied from 2λ to 8λ. We found both suppressive and facilitatory lateral interactions at target-to-flankers distances (2λ - 4λ and 8λ, respectively) that were larger than those found in the fovea. Training reduces suppression but does not increase facilitation. Most importantly, we found that learning reduces crowding and improves contrast sensitivity, but has no effect on visual acuity (VA). These results suggest a different pattern of connectivity in the periphery with respect to the fovea as well as a different modulation of this connectivity via perceptual learning that not only reduces low-level lateral masking but also reduces crowding. These results have important implications for the rehabilitation of low-vision patients who must use peripheral vision to perform tasks, such as reading and refined figure-ground segmentation, which normal sighted subjects perform in the fovea.     

Enhanced growth at low population density in Daphnia: the absence of crowding effects or relief from visual predation?

1. It has been suggested that chemical information from crowded populations of an animal such as Daphnia carries a cue indicating imminent food limitation, and we suggest that in the presence of fish kairomones, it may also convey a hint of the need to enhance antipredation defences. 2. We performed two‐factorial experiments with Daphnia grown in flow‐through plankton chambers in medium containing high levels of Scenedesmus food plus chemical information on either low or high population density levels and in the presence or absence of fish chemical cues (kairomones) and recorded (i) the effects on Daphnia growth rate and reproduction, and (ii) the effects on Daphnia depth selection. Further depth‐selection experiments were performed to test the reaction of Daphnia to crowding information at different Daphnia concentrations and to test its effect on daytime and night‐time depth selection by different Daphnia instars in the presence of kairomones. 3. The effects of crowding information alone (in the absence of kairomones) were weak and were not significantly strengthened by the addition of kairomones. The effects of kairomones alone (in the absence of crowding information) were much stronger and were increased by the presence of crowding chemicals: Daphnia selected greater depths in daylight (the later the instar and the larger its body size, the greater the depth), their body growth was slower and daily reproductive investment reduced, compared with Daphnia grown in the absence of crowding information. This suggested that crowding chemicals carry a cue indicating the need to invest more into antipredation defences. 4. The adaptive significance of these effects was confirmed by the differential vulnerability to predation of the Daphnia when offered as prey to live roach after being grown for 6 days either in the presence (higher vulnerability) or in the absence (lower vulnerability) of information on high density. 5. The strong interaction between crowding information and fish kairomones may be explained either as the reaction to a cue indicating impending food stress or as the reaction to a signal of increased predation risk. While the former scenario is already known from crowding studies, the latter is a novel idea that stems from the old concept of ‘low‐density anti‐predation refuge’. The two scenarios are not mutually exclusive: each stems from the need to invest in survival rather than in growth and reproduction [Corrections were made to this paragraph after first online publication on 4 April 2012].     

Visual crowding is correlated with awareness

Crowding by nearby features causes identification failures in the peripheral visual field. However, prominent visual features can sometimes fail to reach awareness, causing scenes to be incorrectly interpreted. Here we examine whether awareness of the flanking features is necessary for crowding to occur. Flankers that were physically present were rendered perceptually absent with adaptation-induced blindness. In a letter identification task, targets were presented unflanked or with up to four flanker letters. On each trial, observers reported both the number of letters they perceived and the identity of a target letter. This paradigm allowed trial-by-trial assessment of awareness and crowding and ensured that both targets and flankers were attended. Target-letter identification performance was correlated with the number of flanking letters that were perceived on a given trial, regardless of the number that were physically present. Our data demonstrate that crowding can be released when flanking elements at attended locations are suppressed from visual awareness.