Postsaccadic processing of the retinal image during picture scanning

Eye movements were monitored while observers inspected photographs of natural scenes. At the end of each saccade (i.e., at the beginning of each period of steady fixation), the stimulus was replaced for a certain period of time by a uniform field (Experiment 1) or a blurred version of the stimulus scene (Experiment 2). Total fixation duration was measured as a function of the duration of the initial uniform field or the blurred image that followed the saccade. It was found that fixation duration increased proportionally with the duration of the initial replacement field, even for durations as short as 25 msec. These results suggest that the visual system uses information on the retina right after each saccade is completed and that the blurred, low-resolution information used in Experiment 2 (cutoff frequency of 0.8 cpd) is not sufficient for the requirements of picture processing in this task.     

Simple rotary motion is integrated across fixations

Participants were shown a line rotating at a constant angular velocity and were asked to judge whether motion was continuous or whether the line jumped (i.e., moved either forward or backward in the rotary cycle). In two experiments, the participants were significantly better than chance in detecting these jumps in simple rotary motion even when the jumps occurred during a saccade. Moreover, in Experiment 2, when the jump occurred during a saccade followed by a masking flash, perception of the jump was at least as good as when it occurred during a fixation followed by a masking flash. These results complement and extend the findings of Verfaillie and co-workers, who found that perceptions of some changes in biological motion were as good across fixations as when they occurred during a fixation. These findings are in contrast to the common finding, in the reading literature, that people are consciously "blind" to many changes in the text and to those, in the scene perception literature, in which detection of static changes across fixations is above chance but plausibly well below the level that would be expected if the change occurred during a fixation.     

Limits on integrating motion information across saccades

In two experiments, we investigated whether people could detect changes in the rotary motion of a cube. A rendering of a cube rotating at a constant angular velocity was presented on a video monitor and, at a key point in the trial, a cross was presented to one side of the cube as a cue for a saccade. On some trials, a change in the rotation occurred either about 100 msec before the saccade or during the saccade; on other trials, there was no change. The change consisted of moving the cube to a new position in the "rotation sequence," after which it continued to rotate at the same angular velocity as before. There was also a control on all trials to ensure that change detection was not due to the detection of low-level motion. Although detection of the change was well above chance when it occurred during the fixation, it was at chance when it occurred during the saccade, except in the case of one participant (who was in both experiments). This chance performance also occurred in Experiment 2 for (1) a slower rotation speed and (2) an axis of rotation that made the rotation planar. The participant who had above chance performance (and as good as that when the change occurred during a fixation) reported using a "strategy" that did not track the path of the cube. It thus appears that there is no natural way in which the visualsystem tracks this rotary motion, and that detection of change requires some sort of recoding. This finding raises the question of whether good performance in other, apparently similar, motion-detection tasks is a result of similar recoding.     

An adaptive algorithm for fixation,saccade, and glissade detection in eyetracking data

Event detection is used to classify recorded gaze points into periods of fixation, saccade, smooth pursuit, blink, and noise. Although there is an overall consensus that current algorithms for event detection have serious flaws and that a de facto standard for event detection does not exist, surprisingly little work has been done to remedy this problem. We suggest a new velocity-based algorithm that takes several of the previously known limitations into account. Most important, the new algorithm identifies so-called glissades, a wobbling movement at the end of many saccades, as a separate class of eye movements. Part of the solution involves designing an adaptive velocity threshold that makes the event detection less sensitive to variations in noise level and the algorithm settings-free for the user. We demonstrate the performance of the new algorithm on eye movements recorded during reading and scene perception and compare it with two of the most commonly used algorithms today. Results show that, unlike the currently used algorithms, fixations, saccades, and glissades are robustly identified by the new algorithm. Using this algorithm, we found that glissades occur in about half of the saccades, during both reading and scene perception, and that they have an average duration close to 24 msec. Due to the high prevalence and long durations of glissades, we argue that researchers must actively choose whether to assign the glissades to saccades or fixations; the choice affects dependent variables such as fixation and saccade duration significantly. Current algorithms do not offer this choice, and their assignments of each glissade are largely arbitrary.     

Saccadic eye movements and localization of visual stimuli

Visual localization phenomena were studied before, during, and after a saccade. Light flashes of 5 and 9 msec duration presented before and during the eye movement were mislocated in the saccade direction, the localization error being a time function. When the 9-msec duration stimulus and saccade did not overlap in time, a stripe was reported, when they did not, the stimulus was perceived as a point. If a long-duration stimulus moved perpendicularly to the saccade direction with the same “sigmoidal” velocity, a curvilinear trace was perceived, regardless of the linear trace of the image on the retina. A stimulus with stabilized retinal image was perceived as a stationary point during the saccade. A possible theory to deal with the data was suggested by modifying the algebra of outflow-inflow theories.     

场景知觉中的眼动分析指标:基于注视和眼跳视角

眼动范式是场景知觉研究的重要方法之一,可以通过对场景知觉过程中眼动变化的实时记录,真实地反映场景信息加工的内在心理机制;然而,人们的心理活动是极其复杂的,与之相应的眼动指标也具有多样性和复杂性。对眼动指标的分析,有不同的分析维度,比如整体和局部、时间和空间等。本文回顾了已有的眼动指标分类,并尝试从注视和眼跳的视角对场景知觉过程中的眼动指标进行分类。在此基础上,从界定标准、心理学意义、研究应用等方面对相应的眼动指标进行分析和介绍,最后提出了眼动指标分析和应用可能存在的问题,以及未来研究可能拓展的领域。     

视觉表象眼动的变化：知识还是技能表征的差异？

以表象看到一个运动员完成三级跳远项目为实验任务,对表象任务的信息通达水平、眼动注视点的活动位置和被试对三级跳远项目的知识水平和技能水平进行系统的操纵,通过2个实验探讨了视觉表象眼动的变化是基于知识学习表征差异还是技能训练表征差异的问题。实验1以没有三级跳远运动专业技能知识且对该运动的认知水平也较低的大学生为被试,结果表明,在完成高信息通达水平的表象任务时,注视点需要较短的持续时间,但眼跳距离会增大,眼跳频率会变低;实验2对表象任务的知识学习表征水平和技能训练表征水平进行操纵,分别以对实验任务进行过知识学习和专业技能训练的人为被试,结果表明,随着被试知识习得水平和技能水平表征能力的提高,不同表象任务信息通达水平间的眼动差异将消失,但知识学习和技能表征的差异在平均眼跳时间上有差异,技能训练型的被试其平均眼跳时间要短于知识学习型被试,达到临界水平显著,注视点平均持续时间和平均眼跳距离等均没有差异。     

Emotional attention: effects of emotion and gaze direction on overt orienting of visual attention

In the present study we considered the two factors that have been advocated for playing a role in emotional attention: perception of gaze direction and facial expression of emotions. Participants performed an oculomotor task in which they had to make a saccade towards one of the two lateral targets, depending on the colour of the fixation dot which appeared at the centre of the computer screen. At different time intervals (stimulus onset asynchronies, SOAs: 50,100,150 ms) following the onset of the dot, a picture of a human face (gazing either to the right or to the left) was presented at the centre of the screen. The gaze direction of the face could be congruent or incongruent with respect to the location of the target, and the expression could be neutral or angry. In Experiment 1 the facial expressions were presented randomly in a single block, whereas in Experiment 2 they were shown in separate blocks. Latencies for correct saccades and percentage of errors (saccade direction errors) were considered in the analyses. Results showed that incongruent trials determined a significantly higher percentage of saccade direction errors with respect to congruent trials, thus confirming that gaze direction, even when task-irrelevant, interferes with the accuracy of the observer’s oculomotor behaviour. The angry expression was found to hold attention for a longer time with respect to the neutral one, producing delayed saccade latencies. This was particularly evident at 100 ms SOA and for incongruent trials. Emotional faces may then exert a modulatory effect on overt attention mechanisms.     

The ability of the saccadic system to change motor plans in scanning letter strings

Evidence from recent studies bolsters the idea of preestablished motor plans in scanning isolated items. Thus, refixation saccades are preplanned at the same time as the primary saccade directed to a peripheral item and are completed with fixed amplitudes irrespective of the first fixation position in the item. In order to examine the saccadic system's ability to correct the motor plan during its execution on the basis of new visual information, an experiment was conducted in which 11-letter strings were changed to two 5-letter strings at different times after the primary saccade was directed to the stimulus. The results demonstrate that the saccadic system is able to cancel the preplanned refixation saccade and plan a saccade directed to the next item only when the new visual information is available at least 220 msec before the execution of the saccade.     

Eye movement control during reading: Evidence for direct control

In three experiments, subjects read text as their eye movements were monitored and display changes in the text were made contingent upon the eye movements. In one experiment, a window of text moved in synchrony with the eyes. In one condition, the size of the window was constant from fixation to fixation, while in the other condition the size of the window varied from fixation to fixation. In the other experiments, a visual mask was presented at the end of each saccade which delayed the onset of the text, and the length of the delay was varied. The pattern of eye movements was influenced by both the size of the window and the delay of the onset of the text, even when the window size or text delay was varying from fixation to fixation. However, there was also evidence that saccade length was affected by the size of the window on the prior fixation and that certain decisions to move the eye are programmed either before the fixation begins or are programmed during the fixation but without regard to the text fixated on. The results thus provide support for a mixed control model of eye movements in reading, in which decisions about when and where to move the eyes are based on information from the current fixation, the prior fixations, and possibly, other sources as well.     

An Investigation of Attention Allocation During Sequential Eye Movement Tasks

Allocation of visuo-spatial attention during dynamic viewing was investigated with a dual task. Primary tasks (reading, scanning, searching) all required sequential left-to-right eye movements. An additional speeded manual response was made to a visual probe that appeared early or late after the onset of a randomly determined fixation (25 or 170 msec probe delay). The probe appeared to the left, directly above, or to the right of the currently fixated character (-10, -5, 0, +5, or +10 characters probe eccentricity). Faster probe detection near the location of the forthcoming eye fixation was found in the search task, but not during reading or scanning. Fixation times increased and saccade lengths decreased as a consequence of probing in all three tasks. Fixations were, however, less prolonged when the probe appeared in the right than when it appeared in the left hemifield, and saccades were greatest when the probe appeared at +10 characters. The results extend the notion of goal-directed attention shifts to free viewing and highlight the impact of task-specific processing requirements.     

Integrating pictorial information across eye movements

Six experiments are reported dealing with the types of information integrated across eye movements in picture perception. A line drawing of an object was presented in peripheral vision, and subjects made an eye movement to it. During the saccade, the initially presented picture was replaced by another picture that the subject was instructed to name as quickly as possible. The relation between the stimulus on the first fixation and the stimulus on the second fixation was varied. Across the six experiments, there was about 100-130 ms facilitation when the pictures were identical compared with a control condition in which only the target location was specified on the first fixation. This finding clearly implies that information about the first picture facilitated naming the second picture. Changing the size of the picture from one fixation to the next had little effect on naming time. This result is consistent with work on reading and low-level visual processes in indicating that pictorial information is not integrated in a point-by-point manner in an integrated visual buffer. Moreover, only about 50 ms of the facilitation for identical pictures could be attributed to the pictures having the same name. When the pictures represented the same concept (e.g., two different pictures of a horse), there was a 90-ms facilitation effect that could have been the result of either the visual or conceptual similarity of the pictures. However, when the pictures had different names, only visual similarity produced facilitation. Moreover, when the pictures had different names, there appeared to be inhibition from the competing names. The results of all six experiments are consistent with a model in which the activation of both the visual features and the name of the picture seen on the first fixation survive the saccade and combine with the information extracted on the second fixation to produce identification and naming of the second picture.     

Time course of information processing during scene perception: The relationship between saccade amplitude and fixation duration

The present study focuses on two aspects of the time course of visual information processing during the perception of natural scenes. The first aspect is the change of fixation duration and saccade amplitude during the first couple of seconds of the inspection period, as has been described by Buswell (), among others. This common effect suggests that the saccade amplitude and fixation duration are in some way controlled by the same mechanism. A simple exponential model containing two parameters can describe the phenomena quite satisfactorily. The parameters of the model show that saccade amplitude and fixation duration may be controlled by a common mechanism. The second aspect under scrutiny is the apparent lack of correlation between saccade amplitude and fixation duration (Viviani, ). The present study shows that a strong but nonlinear relationship between saccade amplitude and fixation duration does exist in picture viewing. A model, based on notions laid out by Findlay and Walker's () model of saccade generation and on the idea of two modes of visual processing (Trevarthen, ), was developed to explain this relationship. The model both fits the data quite accurately and can explain a number of related phenomena.     

视觉表象操作加工的眼动实验研究

本研究通过视觉表象旋转和扫描的眼动实验探讨表象的心理表征方式。实验一结果表明,眼动指标具有与反应时相类似的旋转角度效应。实验二结果显示,表象扫描的反应时和眼动指标都具有与知觉扫描加工一样的距离效应。由此可以认为,表象眼动与知觉眼动模式具有相似性;表象具有相对独立的心理表征方式并有其特殊的加工过程;表象的心理表征可以是形象表征,而非一定是抽象的命题或符号表征     

What are human express saccades?

When a fixation point is removed 200 msec prior to target onset (the gap condition), human subjects are said to produce eye movements that have a short latency (80–120 msec), that form the early peak of a bimodal latency distribution, and that have been labeled “human express saccades” (see, e.g., Fischer, 1987; Fischer & Breitmeyer, 1987; Fischer & Ramsperger, 1984, 1986). In three experiments, we sought to obtain this express saccade diagnostic pattern in the gap condition, We orthogonally combined target location predictability with the presence versus absence of catch trials (Experiment 1). When target location was fixed and catch trials were not used, we found mostly anticipations. In the remaining conditions, where responses were under stimulus control, bimodality was not frequently observed, and, whether it was or not, latencies were not in the express saccade range. Using random target locations, we then varied stimulus luminance and the mode of stimulus presentation (LEDs vs. oscilloscope) in the gap and overlap (fixation is not removed) conditions (Experiment2). Bimodality was rarely observed, the gap effect (overlap minus gap reaction time) was additive with luminance, and only the brightest targets elicited saccades in the express range. When fixed locations and no catch trials were combined with latency feedback (Experiment 3), we observed many responses in the express saccade range and some evidence for bimodality, but the sudden introduction of catch trials revealed that many early responses were not under stimulus control. Humanscan make stimulus-controlled saccades that are initiated very rapidly (80–120 msec), but unless catch trials or choice reaction time is used, it is not possible to distinguish such saccades from anticipatory responses that are prepared in advance and timed to occur shortly after target onset. Because the express saccade diagnostic pattern is not a characteristic feature of human saccadic performance, we urge investigators to focus their attention on the robustgap effect     

Eye movement control during scene viewing: Immediate degradation and enhancement effects of spatial frequency filtering

What controls how long the eyes remain fixated during scene perception? We investigated whether fixation durations are under the immediate control of the quality of the current scene image. Subjects freely viewed photographs of scenes in preparation for a later memory test while their eye movements were recorded. Using the saccade-contingent display change method, scenes were degraded (Experiment 1) or enhanced (Experiment 2) via blurring (low-pass filtering) during predefined saccades. Results showed that fixation durations immediately after a display change were influenced by the degree of blur, with a monotonic relationship between degree of blur and fixation duration. The results also demonstrated that fixation durations can be both increased and decreased by changes in the degree of blur. The results suggest that fixation durations in scene viewing are influenced by the ease of processing of the image currently in view. The results are consistent with models of saccade generation in scenes in which moment-to-moment difficulty in visual and cognitive processing modulates fixation durations.     

Evidence for visual persistence during saccadic eye movements

Summary 1. The persistence of visual perception was investigated under conditions of visual fixation as well as eye movement. The Ss' task was to discriminate brief double light impulses; their responses were recorded as a function of the duration of the interstimulus interval. Based on these data the critical interstimulus interval was calculated, which yielded equal response frequencies for the perception of one or two stimuli upon presentation of double light pulses.2. In the condition of visual fixation the two stimuli could not be discriminated until the mean value of interstimulus interval exceeded 73 msec. In the condition with eye movements, when the first stimulus was presented in the parafoveal region of the retina before the beginning of the saccade and the second stimulus in the foveal region just after termination of the eye movement, this duration was shown to be statistically of the same magnitude (76 msec).3. Possible alternative interpretations of this latter result, e.g., that it could be explained in terms of masking or saccadic suppression rather than visual persistence was discussed; it was attempted to invalidate such explanations by means of three control experiments.4. The main result, the persistence of visual perception during voluntary eye movements, was discussed in relation to the problem of spatial and temporal stability of visual perception.I thank Prof. Dr. H.W. Wendt for support in correcting the English translation.     

Look away! Eyes and arrows engage oculomotor responses automatically

The present study investigates how people’s voluntary saccades are influenced by where another person is looking, even when this is counterpredictive of the intended saccade direction. The color of a fixation point instructed participants to make saccades either to the left or right. These saccade directions were either congruent or incongruent with the eye gaze of a centrally presented schematic face. Participants were asked to ignore the eyes, which were congruent only 20% of the time. At short gaze—fixation-cue stimulus onset asynchronies (SOAs; 0 and 100 msec), participants made more directional errors on incongruent than on congruent trials. At a longer SOA (900 msec), the pattern tended to reverse. We demonstrate that a perceived eye gaze results in an automatic saccade following the gaze and that the gaze cue cannot be ignored, even when attending to it is detrimental to the task. Similar results were found for centrally presented arrow cues, suggesting that this interference is not unique to gazes.     

Gaze step distributions reflect fixations and saccades: a comment on

In three experimental tasks Stephen and Mirman (2010) measured gaze steps, the distance in pixels between gaze positions on successive samples from an eyetracker. They argued that the distribution of gaze steps is best fit by the lognormal distribution, and based on this analysis they concluded that interactive cognitive processes underlie eye movement control in these tasks. The present comment argues that the gaze step distribution is predictable based on the fact that the eyes alternate between a fixation state in which gaze is steady and a saccade state in which gaze position changes rapidly. By fitting a simple mixture model to Stephen and Mirman's gaze step data we reveal a fixation distribution and a saccade distribution. This mixture model captures the shape of the gaze step distribution in detail, unlike the lognormal model, and provides a better quantitative fit to the data. We conclude that the gaze step distribution does not directly suggest processing interaction, and we emphasize some important limits on the utility of fitting theoretical distributions to data.