视错觉产生的神经机制

传统认知心理学对视觉错觉的解释存在局限性。认知神经科学在移动错觉、轮廓错觉、颜色错觉研究中的一批新发现,可以综合并概括为一种视错觉发生的神经关联重叠机制。该重叠性表现为错觉与相应的真实知觉具有本质上相同的神经活动,即错觉与真实知觉享有共同的神经关联,没有只对错觉反应而不对真实知觉反应的特定神经关联物,而错觉神经活动的特别之处是在真实知觉发生的相同脑区有更强烈的激活     

Structure-from-motion: dissociating perception, neural persistence, and sensory memory of illusory depth and illusory rotation

In the structure-from-motion paradigm, physical motion on a screen produces the vivid illusion of an object rotating in depth. Here, we show how to dissociate illusory depth and illusory rotation in a structure-from-motion stimulus using a rotationally asymmetric shape and reversals of physical motion. Reversals of physical motion create a conflict between the original illusory states and the new physical motion: Either illusory depth remains constant and illusory rotation reverses, or illusory rotation stays the same and illusory depth reverses. When physical motion reverses after the interruption in presentation, we find that illusory rotation tends to remain constant for long blank durations (T _blank ≥ 0.5 s), but illusory depth is stabilized if interruptions are short (T _blank ≤ 0.1 s). The stability of illusory depth over brief interruptions is consistent with the effect of neural persistence. When this is curtailed using a mask, stability of ambiguous vision (for either illusory depth or illusory rotation) is disrupted. We also examined the selectivity of the neural persistence of illusory depth. We found that it relies on a static representation of an interpolated illusory object, since changes to low-level display properties had little detrimental effect. We discuss our findings with respect to other types of history dependence in multistable displays (sensory stabilization memory, neural fatigue, etc.). Our results suggest that when brief interruptions are used during the presentation of multistable displays, switches in perception are likely to rely on the same neural mechanisms as spontaneous switches, rather than switches due to the initial percept choice at the stimulus onset.     

Apparent motion of subjective surfaces

Apparent motion of an illusory surface was produced by presenting two spatially separated illusory squares in an appropriately timed sequence. Control experiments showed that the effect arose from the illusory contours themselves and not from motion of the cut sectors on the discs. When a template of this movie was superimposed on 'wallpaper' composed of a regular matrix of spots, the spots appeared to move with the illusory surface even though they were physically stationary. This effect ('motion capture') suggests that the motion of certain salient features in the visual field gets spontaneously attributed to even static elements in the vicinity.     

Figure-background segregation and the formation of illusory figures

Three experiments were carried out to test the relationship between figure-background segregation and illusory contours. Illusory figures are believed to arise as byproducts of figure-background segregation. When, in a scene, part of what should be the background becomes an illusory figure, a mechanism of contour attribution favoring the area in which the illusory figure appears takes place. This mechanism is prevented from operating when the attribution of the contour is inhibited by the presence of "groupable" (connectable) contours. Spatial proximity is one of the factors affecting such grouping: the closer the connectable contours, the more likely is their grouping in a single unit and the less likely is the emergence of an illusory figure. Experimental results showed that the illusory effect was established when contours were prevented from being connected. This outcome is interpreted as evidence that a mechanism of contour attribution is effective in the formation of illusory figures.     

Directional harmonic theory: a computational Gestalt model to account for illusory contour and vertex formation

Visual illusions and perceptual grouping phenomena offer an invaluable tool for probing the computational mechanism of low-level visual processing. Some illusions, like the Kanizsa figure, reveal illusory contours that form edges collinear with the inducing stimulus. This kind of illusory contour has been modeled by neural network models by way of cells equipped with elongated spatial receptive fields designed to detect and complete the collinear alignment. There are, however, other illusory groupings which are not so easy to account for in neural network terms. The Ehrenstein illusion exhibits an illusory contour that forms a contour orthogonal to the stimulus instead of collinear with it. Other perceptual grouping effects reveal illusory contours that exhibit a sharp corner or vertex, and still others take the form of vertices defined by the intersection of three, four, or more illusory contours that meet at a point. A direct extension of the collinear completion models to account for these phenomena tends towards a combinatorial explosion, because it would suggest cells with specialized receptive fields configured to perform each of those completion types, each of which would have to be replicated at every location and every orientation across the visual field. These phenomena therefore challenge the adequacy of the neural network approach to account for these diverse perceptual phenomena. I have proposed elsewhere an alternative paradigm of neurocomputation in the harmonic resonance theory (Lehar 1999, see website), whereby pattern recognition and completion are performed by spatial standing waves across the neural substrate. The standing waves perform a computational function analogous to that of the spatial receptive fields of the neural network approach, except that, unlike that paradigm, a single resonance mechanism performs a function equivalent to a whole array of spatial receptive fields of different spatial configurations and of different orientations, and thereby avoids the combinatorial explosion inherent in the older paradigm. The present paper presents the directional harmonic model, a more specific development of the harmonic resonance theory, designed to account for specific perceptual grouping phenomena. Computer simulations of the directional harmonic model show that it can account for collinear contours as observed in the Kanizsa figure, orthogonal contours as seen in the Ehrenstein illusion, and a number of illusory vertex percepts composed of two, three, or more illusory contours that meet in a variety of configurations.     

Can subthreshold summation be observed with the Ehrenstein illusion?

Subthreshold summation between physical target lines and illusory contours induced by edges such as those produced in the Kanizsa illusion has been reported in previous studies. Here, we investigated the ability of line-induced illusory contours, using Ehrenstein figures, to produce similar subthreshold summation. In the first experiment, three stimulus conditions were presented. The target line was superimposed on the illusory contour of a four-arm Ehrenstein figure, or the target was presented between two dots (which replaced the arms of the Ehrenstein figure), or the target was presented on an otherwise blank screen (control). Detection of the target line was significantly worse when presented on the illusory contour (on the Ehrenstein figure) than when presented between two dots. This result was consistent for both curved and straight target lines, as well as for a 100 ms presentation duration and unlimited presentation duration. Performance was worst in the control condition. The results for the three stimulus conditions were replicated in a second experiment in which an eight-arm Ehrenstein figure was used to produce a stronger and less ambiguous illusory contour. In the third experiment, the target was either superimposed on the illusory contour, or was located across the central gap (illusory surface) of the Ehrenstein figure, collinear with two arms of the figure. As in the first two experiments, the target was either presented on the Ehrenstein figure, or between dots, or on a blank screen. Detection was better in the dot condition than in the Ehrenstein condition, regardless of whether the target was presented on the illusory contour or collinear with the arms of the Ehrenstein figure. These three experiments demonstrate the ability of reduced spatial uncertainty to facilitate the detection of a target line, but do not provide any evidence for subthreshold summation between a physical target line and the illusory contours produced by an Ehrenstein figure. The incongruence of these results with previous findings on Kanizsa figures is discussed.     

The perception of moving subjective contours by 4-month-old infants

We investigated whether 4-month-old infants are capable of perceiving illusory contours produced by the Kanizsa-square display, first introduced by Prazdny (1983, Perception & Psychophysics 34 403-404), which tests whether a viewer perceives the illusory contour in the absence of brightness contrast (illusory brightness). Because the illusory square appears to move across the computer screen and infants are attracted to motion, this display holds their interest. In experiment 1, 4-month-old infants were tested for their ability to distinguish between a continuously moving illusory square and a continuously moving control display in which the pacman elements were rotated so that the perception of subjective contours did not occur. Data analysis revealed a significant preference for the subjective contour display. In experiment 2, habituation-dishabituation was used with 4-month-old infants. They were tested for their ability to discriminate between the illusory Kanizsa square that continuously moved back and forth and an illusory square which changed positions randomly. Although the infants did not show differences in dishabituation as a function of the habituation display, they looked significantly longer at the continuously moving display.     

On the Reality of Illusory Conjunctions

The reality of illusory conjunctions in perception has been sometimes questioned, arguing that they can be explained by other mechanisms. Most relevant experiments are based on migrations along the space dimension. But the low rate of illusory conjunctions along space can easily hide them among other types of errors. As migrations over time are a more frequent phenomenon, illusory conjunctions can be disentangled from other errors. We report an experiment in which series of colored letters were presented in several spatial locations, allowing for migrations over both space and time. The distribution of frequencies were fit by several multinomial tree models based on alternative hypothesis about illusory conjunctions and the potential sources of free-floating features. The best-fit model acknowledges that most illusory conjunctions are migrations in the time domain. Migrations in space are probably present, but the rate is very low. Other conjunction errors, as those produced by guessing or miscategorizations of the to-be-reported feature, are also present in the experiment. The main conclusion is that illusory conjunctions do exist.     

Mismaking memories: neural precursors of memory illusions in electrical brain activity

Memory illusions--vivid experiences of events that never occurred--could result from inaccuracies either in retrieving memories or in initially storing them. In two experiments, people studied lists of associated words that either did or did not induce later illusory (false) memories of associated but nonpresented lure words. The amplitude of the electrical brain activity during study of words (approximately 500-1,300 ms) that were themselves later correctly remembered reliably distinguished list words that led to such illusory memories from those that did not. This encoding difference associated with subsequent illusory memory (referred to as a DIM)--presumably reflecting item-specific encoding differences--is a neural precursor of memory illusions.     

Analysis of the perception of motion concomitant with a lateral motion of the head

This study is concerned with two questions regarding the illusory motion of objects that occurs concomitantly with motion of the head. One is whether this illusory concomitant motion, unlike the perception of real motion, is paradoxical in the sense that, although the object appears to move, it does not appear to go anywhere. The second question is whether illusory concomitant motion can be explained by errors in convergence produced by a tendency for the convergence of the eyes to displace in the direction of the resting state of convergence. Both questions receive a negative answer. In Experiment 1, it is shown that the illusory motion perceptually can add to or subtract from apparent motion resulting from real motion. In Experiment 2, it is shown that, for a binocularly viewed object at a near distance, the error in convergence (fixation disparity) is far too small to be an explanation for the illusory object motion associated with a moving head. The results of both experiments support an interpretation of illusory concomitant motion in terms of errors in the apparent distance of the stimulus object and the veridical perception of its direction.     

Absence of compensation and reasoning-like processes in the perception of orientation in depth

When errors are present in the perceived depth between the parts of a physically stationary object, the object appears to rotate as the head is moved laterally (Gogel, 1980). This illusory rotation has been attributed either to compensation (Wallach, 1985, 1987) or to inferential-like processes (Rock, 1983). Alternatively, the perceived distances of and directions to the parts of the object are sufficient to explain the illusory perceived orientations and perceived rotations of the stimulus. This was examined in three experiments. In Experiment 1, a perceived illusory orientation of a stimulus object extended in depth was produced by misleading binocular disparity and was measured at two different lateral positions of the head under two conditions. In the static condition, the head was stationary at different times at each of the two measurement positions of the head. In the dynamic condition, continuous motion of the head occurred between these two positions. In Experiment 2, static and dynamic conditions of illusory stimulus orientation were observed with the head stationary. In Experiment 3, a perspective illusion instead of binocular disparity produced the errors in perceived depth. In no experiment did the perceived orientation of the object differ for the static and dynamic conditions. In the absence of head motion, neither compensatory nor inferential-like processes were available. It is concluded that these processes are not needed to explain either illusory or nonillusory perceptions of the orientation or rotation of stimuli viewed with a laterally moving head.     

Absence of compensation and reasoning-like processes in the perception of orientation in depth.

When errors are present in the perceived depth between the parts of a physically stationary object, the object appears to rotate as the head is moved laterally (Gogel, 1980). This illusory rotation has been attributed either to compensation (Wallach, 1985, 1987) or to inferential-like processes (Rock, 1983). Alternatively, the perceived distances of and directions to the parts of the object are sufficient to explain the illusory perceived orientations and perceived rotations of the stimulus. This was examined in three experiments. In Experiment 1, a perceived illusory orientation of a stimulus object extended in depth was produced by misleading binocular disparity and was measured at two different lateral positions of the head under two conditions. In the static condition, the head was stationary at different times at each of the two measurement positions of the head. In the dynamic condition, continuous motion of the head occurred between these two positions. In Experiment 2, static and dynamic conditions of illusory stimulus orientation were observed with the head stationary. In Experiment 3, a perspective illusion instead of binocular disparity produced the errors in perceived depth. In no experiment did the perceived orientation of the object differ for the static and dynamic conditions. In the absence of head motion, neither compensatory nor inferential-like processes were available. It is concluded that these processes are not needed to explain either illusory or nonillusory perceptions of the orientation or rotation of stimuli viewed with a laterally moving head.     

Illusory continuous motion from oscillating positive-negative patterns: implications for motion perception

A black and white (positive) grating pattern was superimposed in exact register on its own photographic negative. Four operations were repetitively applied to this positive pattern so that it moved fractionally to the right, grew dimmer, moved back to the left, and grew brighter again. This sequence produced a strong illusion of continuous apparent motion to the right for as long as the cycle was repeated. The small relative motion between the two patterns generated two new illusory effects: enhanced real movement (ERM) and reversed real movement (RRM). The dimming and brightening phases gave rise to reversed apparent movement (RAM). All three effects are attributed to spatial filtering by neural mechanisms, which shifts the effective position of the positive-negative contours.     

Time and duration: A persistent illusion

Attention is drawn to a persistent illusory correlation between the words time and duration. This illusory correlation led Ward (1991) to conclude that neural adaptation and informational adaptation are asynchronous. But when the illusion is dispelled so that like is compared to like, it then appears possible that informational adaptation and neural adaptation may be closely linked.     

Sharp-edged vs. diffuse illusory circles: The effects of varying luminance

As had been found previously, one experiment demonstrated that reducing the luminance of a pattern that induced an abrupt-edged illusory figure increased mean ratings of that illusory effect. More importantly, the same result had not been found for a pattern that induced a diffuse illusory figure; in fact, a second experiment with such patterns produced a reliable tendency in the opposite direction. These results are at variance with the suggestion that an abrupt-edged illusory effect involves merely a minor variation in the boundary definition of a diffuse illusory lightness effect.     

Intrusion patterns in rapid serial visual presentation tasks with two response dimensions.

A rapid serial visual presentation (RSVP) paradigm using both one and two response dimensions was used to test parallel processing models of stimulus dimensions. Fifty subjects were asked to report the identity and/or color of a target uppercase word inserted in a series of lowercase words. The results produced a predominance of posttarget intrusions for color responses and a predominance of pretarget intrusions for identity responses. The requirement of a response to a second dimension impaired hit rates but did not change the pattern of intrusions. An examination of the distributions of intrusions in each response dimension as a function of the response given to the other dimension showed an unexpectedly high percentage of simultaneous hits, a moderate covariation between both responses, and the same patterns of intrusions when compared with the general distributions. While these results seem to be compatible with parallel models of processing for stimulus dimensions, two modifications to this model are suggested. First, the processing of response dimension(s) needs some attentional resources. Second, provision for a mixed model is indicated, which would include trials where no illusory conjunctions are formed.     

Orientation unbound: Dissociation of identity and orientation under rapid serial visual presentation

Participants were shown rapid sequences of three letters, flanked by digits, each rotated 0 degree, 90 degrees, 180 degrees, or 270 degrees clockwise from upright. In Experiment 1, the participants tried to report the letter that matched the orientation of an arrow, presented either before (before task) or after (after task) the sequence. A third task (total task) required them to report all of the letters. Accuracy for individual letters was significantly better in the total task than in the before task, and better in the before task than in the after task, suggesting particular difficulty in binding orientation to identity. In Experiment 2, the participants were given letter probes and were asked to indicate the orientation of the probed letter. Although report was above chance, there were frequent illusory conjunctions. Since perception of orientation must depend on prior establishment of identity, our results suggest that orientation and identity may become unbound during processing and are held in parallel storage systems.     

Intrusion patterns in rapid serial visual presentation tasks with two response dimensions

A rapid serial visual presentation (RSVP) paradigm using both one and two response dimensions was used to test parallel processing models of stimulus dimensions. Fifty subjects were asked to report the identity and/or color of a target uppercase word inserted in a series of lowercase words. The results produced a predominance of posttarget intrusions for color responses and a predominance of pretarget intrusions for identity responses. The requirement of a response to a second dimension impaired hit rates but did not change the pattern of intrusions. An examination of the distributions of intrusions in each response dimension as a function of the response given to the other dimension showed an unexpectedly high percentage of simultaneous hits, a moderate covariation between both responses, and the same patterns of intrusions when compared with the general distributions. While these results seem to be compatible with parallel models of processing for stimulus dimensions, two modifications to this model are suggested. First, the processing of response dimension(s) needs some attentional resources. Second, provision for a mixed model is indicated, which would include trials where no illusory conjunctions are formed.     

The effect of support ratio on infants' perception of illusory contours

We used a preferential looking technique to investigate the effect of support ratio (a ratio of the physically specified contours to the total edge length) on the perception of Kanizsa illusory contours in infants aged 3-8 months. Previous work has shown that for adult observers the illusory-contour strength increases proportionally with the support ratio. When the support ratio was relatively high (66%), infants preferred illusory contours to non-illusory figures by 3-4 months of age (experiment 1). In contrast, only infants 7-8 months old showed this preference for illusory contours when the support ratio was reduced to 37% (experiment 3). Further, infants showed no preference for an outline version of the illusory-contour figure, which produced no illusory contours (experiment 2). This result confirms that the infants' preference reflects their perception of illusory contours. Our results show that (i) illusory-contour perception emerges at around 3-4 months of age, but (ii) that this ability is very limited until around 7-8 months of age.