Probing Unconscious Visual Processing with the McCollough ......transfer, the McCollough effect is...

CONSCIOUSNESS AND COGNITION 7, 494–519 (1998)ARTICLE NO. CC980369

Probing Unconscious Visual Processingwith the McCollough Effect1

G. Keith Humphrey and Melvyn A. Goodale

Department of Psychology, The University of Western Ontario, London, Ontario, Canada N6A 5C2.

The McCollough effect, an orientation-contingent color aftereffect, has been known forover 30 years and, like other aftereffects, has been taken as a means of probing the brain’soperations psychophysically. In this paper, we review psychophysical, neuropsychological,and neuroimaging studies of the McCollough effect. Much of the evidence suggests thatthe McCollough effect depends on neural mechanisms that are located early in the corticalvisual pathways, probably in V1. We also review evidence showing that the aftereffectcan be induced without conscious perception of the induction patterns. Based on these twolines of evidence, it is argued that our conscious visual experience of the world arises inthe cortical visual system beyond V1. 1998 Academic Press

. . . even when the external object of perception has departed, the impressionsit has made persist, and themselves are the objects of perception(Aristotle in McKeon, 1941, p. 621)

INTRODUCTION

Visual aftereffects are illusions or distortions of perception that are experiencedafter prolonged viewing of a visual stimulus. One of the classic examples of suchillusory perceptions was described over 2000 years ago by Aristotle in his treatiseon Dreams—De Somniis in Parva Naturalia. Aristotle notes that ‘‘. . . when personsturn away from looking at objects in motion, e.g., rivers, and especially those whichflow very rapidly, they find that the visual stimulations still present themselves, forthe things really at rest are then seen as moving . . .’’ (McKeon, 1941, p. 621). Suchillusory movement, which is in the opposite direction to that of the original stimulus,is now usually referred to as the motion aftereffect (e.g., Wade, 1994; Anstis, Verstra-ten, & Mather, 1998). Other aftereffects, sometimes called figural aftereffects, arebased on properties of static patterns, such as their spatial frequency, orientation, orcurvature. For example, after viewing a high contrast grating that is tilted slightly tothe right of vertical, a grating that is oriented vertically will appear to be tilted slightlyto the left of vertical (e.g., Blakemore, 1973). This aftereffect is called the tilt afteref-fect.

Besides being a visual amusement, visual aftereffects have been used to revealprinciples of visual functioning. In his book entitled Seeing: Illusion, Brain and Mind,John Frisby referred to aftereffects as the ‘‘psychologist’s microelectrode.’’ Frisbywas drawing attention to the parallel between the use of microelectrodes by neuro-

1 Address reprint requests to G. Keith Humphrey, Department of Psychology, The University of West-ern Ontario, London, Ontario, Canada N6A 5C2. E-mail: [email protected].

4941053-8100/98 $25.00Copyright 1998 by Academic PressAll rights of reproduction in any form reserved.

THE MCCOLLOUGH EFFECT 495

physiologists and the use of aftereffects by psychologists; these methods, althoughdifferent in many ways, are both used to explore the functional organization of thevisual system. Neurophysiologists use microelectrodes to investigate directly the re-sponse properties of single neurons in the monkey brain, while perception researchersuse aftereffects to infer the coding properties of what are assumed to be the corre-sponding neurons in the human brain. In the following paper, we will describe howwe and others have used visual aftereffects to probe unconscious visual processing.

THE MCCOLLOUGH EFFECT

The aftereffects mentioned above involve only a single stimulus dimension. In1965, Celeste McCollough discovered an aftereffect that involves two stimulus di-mensions—color and contour orientation. To produce the aftereffect, subjects wouldfirst be asked to look at two differently colored patterns, for example, black-and-redvertical gratings alternating with black-and-green horizontal gratings. After viewingthese patterns for a few minutes, subjects would then be shown test patterns composedof different orientations of black and white gratings. When shown these patterns,subjects typically report seeing desaturated colors on the white sections of verticaland horizontal black and white test gratings. The colors of the aftereffects are approxi-mately complementary to the color in which the pattern was presented during induc-tion. Vertical black and white gratings appear greenish, while horizontal black andwhite gratings look pink. If, however, the black and white gratings are oriented at45°, no color aftereffect is seen (see Fig. 1).

Subsequent research has shown that contingent aftereffects can be established be-tween a number of visual dimensions, such as, for example, spatial frequency andcolor, direction of movement and color, texture density and orientation, and texturedensity and color of the surround. There is a large literature that is concerned notonly with the properties of such aftereffects, but also with what they may tell us, ata theoretical level, about the mechanisms of vision (for various proposals and re-views, see Barlow, 1990; Broerse & O’Shea, 1995; Dodwell & Humphrey, 1990;Durgin, 1996; Durgin & Proffitt, 1996; Harris, 1980; Humphrey, 1998; Siegel &Allan, 1993; Skowbo, Timney, Gentry, & Morant, 1975; Stromeyer, 1978). In thispaper, most of our discussion will be devoted to the McCollough effect, the mostcommonly studied contingent aftereffect. We will present evidence that the McCol-lough effect depends on neural mechanisms that are located early in the cortical visualpathways. We will also review evidence showing that the aftereffect can be inducedwithout conscious perception of the induction patterns. These two lines of evidencewill be used to argue that our conscious visual experience of the world must ariselater in the cortical visual system than the site at which the McCollough effect isgenerated.

The Retinotopic Nature of the McCollough Effect

Considerable evidence suggests that many visual aftereffects, including the McCol-lough effect, reflect changes in mechanisms at an early or low level in the primaryvisual pathway. While there is no universal agreement on what is meant by the term‘‘low level’’ in this context, most researchers would accept that structures up to and

496 HUMPHREY AND GOODALE


including the striate cortex (V1) are low-level. In fact, much of the relevant researchsuggests that the McCollough effect is a low-level phenomenon in this anatomicalor physiological sense. As first noted by McCollough (1965), the hallmark of theeffect is its dependence on contour orientation; that is, the dependence of the per-ceived color on the orientations of the test patterns (and their relation to the orienta-tions of the patterns used during the induction period). Further, this orientation contin-gency depends on orientation with respect to the retina; there is no ‘‘correction’’ forhead tilt (e.g., Bedford & Reinke, 1993; Ellis, 1976). Because neurons in V1 of thevisual cortex of cat and monkey are the first to show orientation selectivity, the effectlikely involves cortical structures. But, because the effect is not subject to vestibularand/or perceptual correction for head tilt, it is likely that the effect is subserved byorientation sensitive mechanisms at an early stage of visual processing, perhaps asearly as V1. It is interesting to note, in this regard, that most orientation-sensitiveneurons in V1 appear to be tuned to oriented contours in retinal rather than real-world coordinates.2

The McCollough effect is also fairly specific to retinal location and to the spatialfrequency of the subtended retinal image. If the retinal location or spatial frequencyis changed between induction and testing, the aftereffect is considerably weakened orabsent, depending upon the magnitude of the change (e.g., Harris, 1980; Humphrey,Herbert, Symons, & Kara, 1994; Humphrey, Skowbo, Symons, Herbert, & Grant,1994; Lovegrove & Over, 1972; Stromeyer, 1972a,b). In short, the McCollough effectdoes not show translational invariance or size constancy. Because cells in V1 havesmaller receptive fields than those at subsequent stages of processing and, as a conse-quence, code more locally than cells further on in the visual system (e.g., Buser and

FIG. 1. An illustration of typical induction and testing conditions for the McCollough effect. Sub-jects would view the vertical red-and-black grating alternating with the horizontal green-and-black grat-ing for a few minutes. After such induction, black and white vertical gratings would appear green andhorizontal gratings would appear pink (as in the simulated aftereffect shown in the lower left of thefigure). Black and white oblique test gratings, as in the lower right of the figure, would not appearcolored.

FIG. 2. The left panel shows an example of oblique discrimination trials in the four-alternativediscrimination task used by Humphrey, Gurnsey, and Fekete (1991) and Humphrey, Goodale andGurnsey (1991). The right panel shows a simulated aftereffect to illustrate how simple the discriminationis after McCollough effect induction.

FIG. 3. An example of the displays in the visual search task used by Humphrey, Gurnsey, andFekete (1991). The task was to search for a ‘‘target’’ item among varying numbers of ‘‘distractors.’’In this case, the target was composed of an outer square that contained a black and white, right obliquegrating and an inner square containing a left oblique grating. The distractor items had the left obliquegrating on the outside and the right oblique grating on the inside. Before McCollough induction, thetask was difficult (search slopes for target present trials were, on average, 107 ms/item). After McCol-lough induction, the task became much easier (average search slopes were 4 ms/item for trials in whichthe target was present), as is evident in the simulated McCollough effect on the right of the figure.

2 There have been studies that have examined the effects of body tilt on the receptive field characteris-tics of cortical cells in cats. Although a few cells have been found to be influenced by changes in bodytilt, the majority of cortical cells have not (e.g., Denney & Adorjani, 1972; Horn, Stechler, & Hill, 1972).


Imbert, 1992), these cells are prime candidates for the neuronal substrate of thisaftereffect. In sum, the spatial characteristics of the McCollough effect reflect theprojection of the stimulus on the retina and not on how the stimulus might be per-ceived in the real world.

Research has also shown that the wavelength of the induction patterns, and nottheir perceived color, determines the aftereffect hue (Thompson & Latchford, 1986).Again, some research suggests that color sensitive cells in V1 respond to the wave-length characteristics of a stimulus, while cells further along in the visual system,such as many of those in V4, respond to more complex aspects of a color displayand appear to be associated with color constancies (Zeki, 1993).

McCollough (1965) and many others (e.g., Murch, 1972; Humphrey, Gurnsey, &Bryden, 1994; Savoy, 1984) have reported that the effect does not transfer from oneeye to the other under typical conditions for assessing interocular transfer. By typicalconditions we mean that the adapting stimuli are presented to one eye while theother eye is covered or is presented with unpatterned input. Under these inductionconditions, the aftereffect is seen when test patterns are presented to the adapted eyebut not when they are presented to the unadapted eye. In not showing interoculartransfer, the McCollough effect is unusual. Figural and motion aftereffects typicallyshow substantial interocular transfer (e.g., Blake, Overton, & Lema-Stern, 1981,Moulden, 1980). The presence of interocular transfer is often used as evidence that themechanisms underlying the aftereffect are cortical in origin, since neurons sensitive toinput from the two eyes are first seen in V1. The lack of interocular transfer of theMcCollough effect suggests several possibilities. It could be the case that the effectarises at a site that precedes the emergence of binocular units, such as the first stagesof processing in V1, before binocular cells are encountered, or even earlier. Or, alter-natively, it could be the case that the effect is mediated by mechanisms in V1 orbeyond where many cells are binocular but it fails to show interocular transfer be-cause one component of the effect, that involving color, may be largely monocular(Savoy, 1984).

Despite the absence of interocular transfer, a ‘‘binocular’’ McCollough effect canbe demonstrated under certain induction conditions (Seaber & Lockhead, 1989; Vid-yasagar, 1976). For example, subjects might be induced with red-and-black verticaland green-and-black horizontal gratings with both eyes open. Alternating with thisbinocular adaptation, each eye is also separately induced with the opposite colorand orientation contingency—green-and-black vertical and red-and-black horizontalgratings. After this complicated induction regime, subjects report a ‘‘monocular’’McCollough effect when they view test patterns with a single eye; for example, avertical black and white grating appears pink. But when they view the same testpattern with both eyes open they report a binocular aftereffect, in which the verticaltest pattern now appears greenish.

One would expect that this purely binocular effect would depend on the presenceof binocular cells. An experiment by Seaber and Lockhead (1989) provides someevidence for this. They studied both normal and strabismic3 subjects using the same

3 Strabismus refers to the misalignment of the axes of the eyes. If strabismus occurs early in life, oneof its possible consequences is the lack of normal binocular vision. Research with animals suggests that


induction regime described above and found that, although normal subjects reportedboth the monocular and binocular aftereffects, strabismic subjects reported only themonocular aftereffects. This result suggests then that the binocular McCollough effectdepends on binocular neurons, presumably drastically reduced in number in the stra-bismic subjects

Collectively these results suggest that the McCollough effect depends on corticalmechanisms at an early stage in the primary visual pathways. The fact that the effectseems to depend on the wavelength characteristics of the input suggests that theadaptation involves early color coding mechanisms. In addition, the retinotopic natureof the effect suggests that it occurs at an early site in visual processing where theneurons have small receptive fields. Although these results point to an early site,other results suggest that this site is cortical. In particular, the orientation specificityof the aftereffect and the fact that one can induce binocular McCollough effects bothimplicate cortical mechanisms. In sum, the evidence supports the view that theMcCollough effect reflects an adaptation of mechanisms at an early stage of corticalvisual processing, perhaps as early as V1.

Visual Form Agnosia and the McCollough Effect

Another way to attack the question of the neural substrates of the McCollougheffect is to study individuals who have sustained damage to particular visual path-ways. Moreover, by studying the perceptual reports of such people, we can gaininsight into the role of conscious perception in the mediation of this aftereffect. Onemight suspect, given the suggestion that early, rather than late, visual mechanisms areinvolved, that conscious perception might not be necessary to induce the McCollougheffect. To this end, we examined the McCollough effect in several individuals whohave profound deficits in form and orientation perception. All of these people sufferedbrain damage from an anoxic episode (lack of oxygen to the brain). Such damageoften leads to what some neurologists have called ‘‘apperceptive agnosia,’’ in whichelementary visual functions and general cognitive ability are quite intact, but visualform discrimination and recognition are dramatically impaired (Lissauer, 1890). Thisdeficit is sometimes called ‘‘visual form agnosia,’’ a term which emphasizes theselective loss of visual form perception (Benson & Greenberg, 1969; for review, seeFarah, 1990). In fact, in many cases, but perhaps particularly with individuals whoseanoxic episode occurred as a result of carbon monoxide poisoning, there is substantialsparing of color vision relative to form perception (e.g., Adler, 1950; Campion, 1987;Efron, 1968; Milner & Heywood, 1989; Vecera & Gilds, 1997). The pattern of braindamage in these cases is diffuse but tends to be greatest in the occipital lobe, butwith relatively less damage in V1 than in extrastriate areas.

The first person that we studied, D.F., showed the typical pattern of diffuse braindamage from anoxia as a result of carbon monoxide poisoning. Nevertheless, thedamage was most evident in the posterior regions of the brain, although evenhere, it was confined mainly to the ventrolateral regions of the occipital cortex, spar-

this loss of binocular function is related to a reduction in the number of cells in the cortex that can beactivated by both eyes (e.g., Hubel & Wiesel, 1965).


ing V1. D.F. has a profound impairment in visual form and orientation perception(Goodale, Milner, Jakobson, & Carey, 1991; Humphrey, Goodale, Jakobson, & Ser-vos, 1994; Milner, Perrett, Johnston, Benson, Jordan, Heeley, Bettucci, Mortara,Mutani, Terazzi, & Davidson, 1991; Servos, Goodale, & Humphrey, 1993). She isunable to discriminate between such simple geometric forms as a triangle and a circle,let alone recognize the faces of her relatives and friends or the visual shapes of com-mon objects. D.F. has no difficulty identifying people from their voices and she hasno problem identifying objects placed in her hands. Her perceptual problems areexclusively visual and are largely restricted to the form of objects. Her color visionis close to normal (Milner & Heywood, 1989). Moreover, she can use color and othersurface features to identify objects (Humphrey, Goodale, Jakobson, & Servos, 1994;Servos et al., 1993) and, to some extent, she can even use shape from shading (Hum-phrey, Symons, Herbert, & Goodale, 1996) which may rely on low-level mechanisms(Humphrey, Goodale, Bowen, Gati, Vilis, Rutt, Menon, 1997). What she seems un-able to perceive are the contours of objects—no matter how the contours are defined(Milner at al., 1991).

D.F.’s performance on various clinical and psychophysical tests has demonstratedthat the profound deficit in her form perception cannot be explained by disturbancesin low-level sensory processing (Milner et al., 1991). But the most compelling reasonto doubt that her perceptual deficit is due to some sort of low-level disturbance inprocessing is the fact that, in another domain, visuomotor control, she remains exqui-sitely sensitive to the form of objects. For example, when presented with a large slotwhich could be placed in one of a number of different orientations, D.F. showed greatdifficulty in indicating the orientation of the slot either verbally or even manually byrotating a handheld card. Nevertheless, when she was asked simply to reach out andinsert a card into the slot, she performed as well as normal subjects, rotating herhand in the appropriate direction as soon as she began the movement (Milner et al.,1991; Goodale et al., 1991). In other tests, even though D.F. could not discriminatebetween target objects that differed in outline shape, she could nevertheless pick upsuch objects successfully, placing her index finger and thumb on stable grasp points(Goodale, Meenan, Bulthoff, Nicolle, Murphy, & Racicot, 1994).

D.F. has a severe problem, then, in her perceptual experience of form; indeed, sheappears to have lost all conscious perception of this visual attribute. Nevertheless,she is able to use the orientation and shape of objects to guide skilled movements,such as grasping. This result, together with a wealth of supporting data in the monkey,has been used by Goodale, Milner, and their colleagues (e.g., Goodale & Milner,1992; Milner & Goodale, 1995; Goodale & Humphrey, 1998) to develop an accountof the division of labor between the well-known dorsal and ventral visual projectionsfrom V1. According to their account, the ventral stream of projections to inferotemp-oral cortex mediates the visual perception of objects, providing a foundation for visualcognition. It is the ventral stream that delivers our conscious visual experience ofthe world. In contrast, the dorsal stream of projections to posterior parietal cortexmediates the visual control of skilled actions directed at objects. This stream convertsincoming visual information into the required coordinates for action—coordinatesthat are specific for the particular effector that is used. The visual processing in thisstream is more ‘‘automatic’’ and is less accessible to consciousness. In D.F.’s case,


the processing of form and shape by the ventral ‘‘perception’’ stream has been se-verely compromised, while the processing of similar information by the dorsal ‘‘ac-tion’’ stream remains intact.

The lack of visual awareness of contour orientation, but relatively preserved colorvision, in D.F. makes for an interesting experiment on the McCollough effect. If theMcCollough effect depends on the adaptation of neural mechanisms that directlysupport ‘‘seeing’’ the orientation of the inducing stimuli, then one might not expectD.F. to experience the aftereffect. On the other hand, it is possible that the neuralmechanisms mediating the effect—possibly in V1—are not available to consciousperception and that, despite D.F.’s inability to see different contour orientations, shestill would experience the McCollough effect. To test for the McCollough effect inD.F., we used a task that we have used in studies of normal subjects as well (Hum-phrey, Gurnsey, & Fekete, 1991). In this task, which is illustrated in Fig. 2, subjectswere required to discriminate a briefly presented monochromatic patch of square-wave grating oriented at 45° from three others oriented at 135° (and vice versa). Onother trials, subjects also were asked to discriminate between horizontal and verticalgratings. Subjects performed the task before and after McCollough effect induction.Before induction, normal subjects found it especially difficult to discriminate betweenthe obliques, particularly at short exposure durations. This result would be expected,given the many results showing difficulties in the perception of oblique orientations(e.g., Appelle, 1972; Essock, 1980). After induction with colored oblique gratings,however, performance was strikingly better, suggesting that the color aftereffect func-tioned as a physically present color difference even at exposure durations as shortas 67 ms. In other words, the color aftereffect permitted the stimulus that differedfrom the three others to ‘‘pop out’’ of the display even at these brief exposure dura-tions (see Fig. 2).

We examined D.F.’s performance on the task described above, although at longerexposure durations than used in the research with normal subjects (Humphrey et al.,1991). Before induction, not only did she perform at a chance level in discriminatingbetween 45° and 135° oblique gratings, but she also failed to discriminate horizontaland vertical gratings, a task that was trivial for normal subjects at these durations.After induction with one oblique grating in red and the other in green, her perfor-mance improved markedly, but only for the oblique gratings. She now performedthe oblique discrimination trials by relying on the difference in aftereffect colors. Incontrast, her performance on the horizontal–vertical discriminations remained at achance level.

D.F. then experienced an orientation-contingent color aftereffect—the McCol-lough effect—even though she had no conscious awareness of the orientation of theinduction or test patterns. Given the pattern of brain damage in D.F., this result isconsistent with the hypothesis that the McCollough effect reflects an adaptation ofearly visual mechanisms, maybe as early as V1, which are relatively spared in D.F.There are other possibilities, however. It must be remembered that, despite D.F.’sprofound deficit in form and orientation perception, she is able to orient her handappropriately in anticipation of grasping the same objects that she cannot discriminateperceptually (Goodale et al., 1991; Milner & Goodale, 1995). This suggests that infor-mation about orientation and form, while inaccessible to perceptual awareness, can


be used for the control of skilled actions. It is possible, then, that the coding oforientation for the McCollough effect is carried out by the same cortical mechanismsin the dorsal stream that are presumably mediating her spared visuomotor control.What is needed is a demonstration that the McCollough effect can occur in an individ-ual with extrastriate damage that prevents any processing of form that might be leftin V1 from reaching those higher cortical areas mediating the visual control of action.

To address these issues, we studied the McCollough effect in a second individual(Humphrey, Goodale, Corbetta, & Aglioti, 1995). This person, P.B., suffered exten-sive brain damage as a result of cardiac arrest due to an accidental electrocution.MRI scans show a diffuse brain atrophy, particularly in the parieto-occipital cortex.Like D.F., P.B. shows a profound insensitivity to visual form and is unable to recog-nize the simplest of geometrical shapes. P.B., also like D.F., shows remarkably intactcolor perception. But P.B. is different from D.F. in one important way; he showsnone of the spared visuomotor abilities that are so evident in D.F. In other words,P.B.’s deficit in form processing extends to both perceptual judgments and visuomo-tor control. Thus, we were able to use P.B. to test the possibility that the McCollougheffect seen in D.F. was mediated by the apparently intact dorsal ‘‘action’’ stream.Indeed, if P.B. were to show the McCollough effect, then this would be additionalsupport for the proposal that the requisite interaction between color and form takesplace in early vision, possibly in V1.

Despite the fact that P.B. is essentially blind for form, he nevertheless showedevidence for an orientation-contingent color aftereffect. Although he required a rela-tively long induction period to show the McCollough effect, the fact that he showedit at all suggests that, whatever underlies this phenomenon, it does not require anyconscious perception of form—or even form processing for action.

In addition to D.F. and P.B., we have also examined the McCollough effect intwo other individuals with visual form agnosia. Both of these people, like D.F. andP.B., developed profound impairments in form and orientation perception followingan anoxic event but, again, like D.F. and P.B., still perceive color. Although we didnot study their performance as extensively as we did that of D.F. and P.B., we wereable to demonstrate that both individuals could experience the McCollough effect.

In sum, the performance of the individuals with visual form agnosia shows clearlythat the conscious perception of contour orientation is not necessary for inductionof the McCollough effect. Whatever the population of neurons might be that isadapted in these people, that population by itself does not support the perception ofform. This population, however, must in some way code orientation. Given the rela-tive sparing of V1 in these individuals, and given that V1 is the first place whereorientation-sensitive cells can be recorded in the retinocortical pathway, the resultsare consistent with the proposal that V1 is the site which normally mediates theMcCollough effect.

The results of a study by Savoy and Gabrieli (1991) of patients with Alzheimer’sdisease are also consistent with an early locus for the generation of the McCollougheffect. The patients showed a normal McCollough effect after 5 minutes of inductionand retained the effect for at least one hour. Anatomical studies of patients withAlzheimer’s disease have suggested that V1 of the visual cortex is spared relativeto extrastriate visual areas (Lewis, Campbell, Terry, & Morrison, 1987). Savoy and


Gabrieli speculate that the patients they tested may have shown this same pattern ofdegeneration and propose that V1 may be the critical locus for the McCollough effect.

Binocular Rivalry and the McCollough Effect

The research with the individuals with visual form agnosia shows that consciousperception of form is not required to induce or, indeed, to experience the McCollougheffect. But what about normal subjects? Can the McCollough effect be induced evenwhen subjects are unaware of the inducing stimuli? One way to address this issueis to take advantage of a phenomenon called binocular rivalry. Binocular rivalry isproduced when two very different images are presented to the two eyes and the twoimages cannot be fused into a single percept. In this situation, one usually sees theimage presented to one eye alternating with the image presented to the other eye. Itappears that the input from one eye is ‘‘suppressed’’ when the input to the otheris perceived. It is the suppression that occurs in binocular rivalry that provides theopportunity to examine whether or not awareness of the inducing stimuli is necessaryfor the induction of the McCollough effect.

White, Petry, Riggs, and Miller (1978; Experiment 1) examined whether the mag-nitude of the McCollough effect was dependent on the length of time the inducingpatterns were actually present on the retina or on the amount of time that the inductionpatterns were perceived. In one condition, White and his colleagues presented theinduction patterns to one eye and a black and white checkerboard-like pattern to theother eye. In this ‘‘rivalrous’’ condition, the induction patterns were not seen, at leastfor a large part of the induction period, because of suppression by the highly salientcheckerboard-like pattern presented to the other eye. In their nonrivalrous condition,the induction patterns were almost always perceived because a nonsalient homoge-neous stimulus was presented to the other eye. Nevertheless, the McCollough effectsin the rivalrous condition were of the same strength as those obtained in nonrivalrousconditions. Such a result suggests that much of the adaptation was taking place with-out the subject’s conscious awareness of the adapting patterns. This lack of depen-dence on perceptual awareness also characterizes many other aftereffects, as well.Thus, binocular rivalry does not interfere with the production of simple aftereffects,such as the linear motion aftereffect (Lehmkuhle & Fox, 1975; O’Shea & Crassini,1981),4 the tilt aftereffect (Wade & Wenderoth, 1978), and spatial frequency adapta-tion (Blake & Fox, 1974). In fact, for all of these aftereffects, it is the amount oftime the inducing stimuli are present on the retina and not the amount of time theyare perceived that determines the strength of the aftereffect.

The fact that the strength of the McCollough effect and other aftereffects is notdiminished during binocular rivalry means that the mechanisms underlying these af-tereffects reside at a site in the visual pathway that may precede, or is at least indepen-dent of the site of binocular rivalry. Although some older research on binocular ri-

4 There is evidence, however, that suppression during binocular rivalry does interfere with some motionaftereffects (Lack, 1978; Wiesenfelder & Blake, 1990). These studies involved spiral and rotationalmotion, so it may be that different levels of the visual system are adapted, depending on the type ofmotion (see Wiesenfelder & Blake, 1990, for more discussion).


valry suggested that it occurred at a relatively early locus in the retinocorticalpathway, recent research suggests otherwise (Andrews & Purves, 1997; Kovacs,Papathomas, Yang, & Feher, 1996; Leopold & Logothetis, 1996; Logothetis, Leo-pold, & Sheinberg, 1996; Logothetis & Schall, 1989; Sheinberg & Logothetis, 1997;for discussion, see Crick, 1996; Sengpeil, 1997). In a particularly elegant set of exper-iments, Logothetis and his colleagues have correlated the activity of neurons in thecortical visual pathway with monkeys’ perception of stimuli in a binocular rivalryparadigm. In their experiments, the monkeys were presented with different imagesto the two eyes and were trained to indicate which of the two images was dominantat any given moment. Using this behavioral method and simultaneously recordingfrom different visual areas in cortex, Logothetis and his colleagues found that theactivity of fewer than 20% of the cells in early visual areas such as V1 and V2 wascorrelated with the monkeys’ reports of their percepts. In higher-order visual areas,the number of cells whose activity correlated with the monkeys’ reports increasedto around 40% in V4 and to about 90% in inferotemporal cortex. These results, cou-pled with the fact that binocular rivalry does not affect the strength of the McCollougheffect, provide additional support for the suggestion that the mechanisms that mediatethis aftereffect occurs earlier, rather than later, in the cortical visual pathway. Again,the evidence suggests that V1 might be the critical site.

Attention and the McCollough Effect

Some researchers divide visual processes operationally into ‘‘high-level’’ and‘‘low-level’’ stages—based, to some extent, on the role of focal attention in taskswhere these two different processes are thought to operate (for review and criticaldiscussion, see Wolfe, in press). In particular, it has been suggested that ‘‘featuresearch’’ tasks are those which can be carried out in parallel and do not require focalattention (e.g., Treisman, 1988) or that can operate at very brief presentations (e.g.,Julesz, 1984). Detecting a green disc among red discs would be an example of afeature search task, requiring only low-level processing. ‘‘Conjunction search’’ tasksinvolve searching for targets defined by a conjunction of properties. Searching for agreen square among red squares, red discs, and green discs is an example of a con-junction search task. Such search is slower and serial in nature and requires focalattention [but see Mack & Rock (1998) and Nakayama & Joseph (in press) for differ-ent interpretations of such findings].

The low-level processes, frequently investigated in psychophysical studies, are of-ten thought to reflect the activity of early visual areas. In contrast, high-level pro-cesses have been associated with activity in extrastriate regions, such as V4 andbeyond. Certainly, many neurophysiological studies with primates have shown thatresponse properties of cells in V1 are not generally modulated by attentional manipu-lations, while those at higher levels, such as in V4 and inferotemporal cortex, canbe readily modulated by attentional demands (e.g., Desimone & Duncan, 1995). Ifthe McCollough effect results from adaptation of low-level mechanisms, one mightexpect that attentional manipulations would have little or no effect on it.

Houck and Hoffman (1986) examined whether attentional manipulations duringinduction affected the strength of the McCollough effect. They showed that, evenwhen attention was devoted to a task that would shift attention away from the induc-


ing patterns during the induction period, the aftereffects were similar in strength tothose found when attention was focused on the inducing patterns.5

The results of Houck and Hoffman (1986) demonstrate that attention to the adapt-ing patterns is not necessary for induction of the McCollough effect. It has also beensuggested that attention is not required for the perception of the McCollough either.Thompson and Travis (1989), following earlier studies by Mayhew and Frisby(1978), showed that it took subjects several seconds to discriminate two textures,each composed of overlaid gratings with different orientations. If, however, the sub-jects were first exposed to an induction period in which each of the textures wascolored red or green, respectively, then, after such induction, the two textures couldbe discriminated much more quickly. Hence, it could be argued that the McCollougheffect emerges preattentively, prior to the (attentive) detection of the involved orienta-tions, since to see the aftereffect colors the orientation of the oriented gratings wouldhave to have been detected.

Humphrey et al. (1991) found results similar to those of Thompson and Travis(1989) using visual search tasks. They investigated whether aftereffect colors couldact like a simple ‘‘feature’’ in a visual search task. As mentioned earlier, in thesetasks, the time it takes to detect a ‘‘target’’ item is independent of the number ofbackground or ‘‘distractor’’ items (e.g., Treisman, 1988). This result suggests thatsimple features can be detected in parallel without focal attention. The targets anddistractors consisted of two concentric square regions in which, for the target, theouter square contained a black and white, right oblique grating and the inner squarecontained a left oblique grating (see Fig. 3). The distractor items had the left obliquegrating on the outside and the right oblique grating on the inside. Like the stimuliused by Thompson and Travis (1989), these displays were very difficult to discrimi-nate before McCollough effect induction. Without McCollough induction, the timetaken to detect a target among distractors increased substantially as the number ofdistractors increased. After induction, however, detection time was essentially inde-pendent of the number of distractors because the color difference between the targetand distractors ‘‘popped out’’ (see Fig. 3). The subjects had rapid and effortlessaccess to the aftereffect colors and did not have to attend to any particular region oritem in the displays to ‘‘bind’’ the colors to the patterns.

In summary, the studies on attention, and those described earlier that used binocu-lar rivalry, are consistent with the hypothesis that the McCollough effect results fromadaptation of low-level visual mechanisms. Attending to the patterns is not necessaryfor induction or perception of the McCollough effect. In short, one need have anyconscious perception of the stimuli that subsequently evoke a striking visual afteref-fect.

V1 May Be Necessary but Not Sufficient for the McCollough Effect

Up to this point, we have been making the case that the McCollough effect ismediated by mechanisms in V1. Evidence from a number of different lines of research

5 It should be noted that attention may modulate some aftereffects, such as those involving certainkinds of visual motion (Chaudhuri, 1990; Rees, Frith, & Lavie, 1997; Takeuchi & Kita, 1994). For thisreason, one cannot make blanket statements about the lack of attentional modulation of aftereffects.


all point to an early site for generation of the McCollough effect. Nevertheless, eventhough modification of the mechanisms in V1 may be the critical factor in the produc-tion of the McCollough effect, other structures may be essential for the consciousperception of the colors invoked by the oriented contours.

We recently had a chance to explore the role of structures beyond V1 in the genera-tion of McCollough effect in an individual (B.K.), who suffered a right occipitalstroke that spared V1 but damaged adjacent extrastriate areas. B.K., an eminent neu-ropsychologist, has described his own case in a fascinating paper (Kolb, 1990). Peri-metric testing shows that B.K. has a scotoma of about 6° along the vertical midlineand about 15° lateral along the horizontal midline in the upper left visual field. Thearea beyond this zone of ‘‘blindness’’ in the upper left quadrant has reduced vision,in the sense that there is still an inability to perceive dim lights in this area. Recentmagnetic resonance imaging of B.K.’s brain shows that most of his damage appearsto be in areas V2/VP (where VP refers to ventral area V3) and that he sufferedrelatively little or no damage to V1. Also, because his scotoma shows a sharp cutat the horizontal meridian, it is likely to be restricted to areas V2/VP (Horton &Hoyt, 1991; Tootell, Hadjikhani, Mendola, Marrett, & Dale, 1998).

Despite this damage, B.K. can still perceive motion and color in his scotoma, butcannot see form within the scotoma (for further discussion, see Kolb, 1990). Hismotion perception could be mediated by pathways from V1 to MT/V5 and/or fromsubcortical structures to MT/V5 (ffytche, Guy, & Zeki, 1995; Gross, 1991). Account-ing for B.K.’s color phenomenology in his scotoma is perhaps more problematic.There could be input from V1 directly to extrastriate areas involved in color pro-cessing. There is evidence in the macaque monkey for a V4 input from the centralrepresentation in V1 (Zeki, 1993; see also Felleman & Van Essen, 1991), althoughthe most prominent input to V4 is from V2.

B.K., then, offered a unique opportunity to test the McCollough effect within hisscotoma and to compare the results with similar testing in a comparable area withinhis normal visual field. It is interesting to note that, when we first discussed thisexperiment with B.K., he thought he would experience the McCollough effect in hisscotoma because he perceives color there and because, after staring at colored patchesthat fall in his scotoma, he experiences appropriately colored afterimages. We alsotested D.F., one of the people with visual form agnosia, using the same stimulusconditions that we used with B.K.

Both D.F. and B.K. first performed several pretest tasks of color and orientationdiscrimination. In one color identification task, desaturated colors were presented onan oblique grating. Desaturated colors were used in an attempt to mimic the desatu-rated colors of the McCollough effect. The gratings were either pink-and-black orgreen-and-black. The gratings were presented briefly one at a time and the task wasto identify the color of each grating. There were also orientation identification tasks.For example, D.F. and B.K. were asked to identify the orientations of left and rightoblique grating patches that were presented briefly in the same location in the leftor the right visual field. The stimuli were constructed to fall in B.K.’s scotoma inthe left field and at a mirror-symmetric location in the right visual field. These samelocations were used for testing D.F.

As shown in Fig. 4, both D.F. and B.K. were excellent at identifying colors in


FIG. 4. Pretest and post McCollough induction results for B.K. and D.F. The left panel shows thepretest results for identifying the colors of oblique pink-and-black and green-and-black gratings. BothB.K. and D.F. performed this task much better than chance (where chance is 50% correct). When theyhad to identify the orientation of black and white oblique gratings (right or left oblique), D.F. performedat a chance level in both visual fields, as did B.K. when the gratings were presented in the upper leftvisual field and fell within his scotoma. In the right visual field, B.K. performed well above chance.After McCollough induction with oblique gratings in red and green, B.K. and D.F. were presented withthe black and white oblique gratings and reported the aftereffect colors. As shown in the right panel,D.F. reported appropriate aftereffects in both visual fields, while B.K. reported aftereffects only in hisright visual field. In the left visual field, within his scotoma, B.K. did not experience the McCollougheffect and thus his reporting of aftereffect colors was at chance.

both visual fields. D.F. could not identify the orientation of oblique gratings at betterthan a chance level in either visual field. B.K. could identify the obliques in his rightvisual field, but was at chance when the patterns were presented within his scotoma.

The pretesting, then, showed that D.F. and B.K. could clearly identify the colorsof oblique gratings, even when that color was desaturated—as would be the case forthe colors normally experienced in the McCollough effect. This was true even whenthe colored test stimuli were presented in B.K.’s scotoma. Nevertheless, neither D.F.nor B.K., when the stimuli were presented in his scotoma, could identify the orienta-tion of the oblique gratings. After completing these pretests, D.F. and B.K. wereadapted for 25 minutes to alternating oblique gratings in red and green.6 The induction

6 We tested B.K. on two occasions, separated by about 8 months, for the McCollough effect. On bothoccasions, we used a 25-minute induction period. The results for each test session were essentially thesame and he did not experience the McCollough effect in his scotoma during either session. The resultspresented in Fig. 4 are from the first test session.


patterns extended well beyond B.K.’s scotoma into his upper, lower, left, and rightfields. After induction, black and white oblique gratings were presented again, as inthe pretest orientation identification task. Now, however, this task was actually theMcCollough task and, unlike the pretest, the subjects were instructed to report theperceived color of the two patterns. That is, they were asked to indicate whether thegratings, which normally would evoke the McCollough effect, looked pink or green.On this task, D.F. performed well in both fields, as she had in the earlier studies,and reported appropriate aftereffect colors (see Fig. 4). As expected, B.K. reportedappropriate aftereffect colors reliably in his right undamaged field. In his scotoma,however, he did not see any aftereffect colors and so his reporting of such colorswas at chance (see Fig. 4).

We began this section by proposing that, even though modification of the mecha-nisms in V1 may be the critical factor in the production of the McCollough effect,other structures may be essential for the conscious perception of the colors invokedby the oriented contours. Our study with B.K., who has little or no damage to V1but significant damage to adjacent extrastriate areas, suggests that this may be true.He showed no evidence of experiencing the McCollough effect in his scotoma. Thereis a puzzle remaining to be answered, however. Despite the fact that B.K. does notexperience the McCollough effect in his scotoma, he can see colors in this regionof his visual field and even reports experiencing simple color afterimages. This para-dox may be resolved as we learn more about the different projections carrying wave-length information from the eye to the cortex.

A Neuroimaging Study of the McCollough Effect

To help localize the cortical visual structures that may be directly involved in theMcCollough effect, we recently conducted a functional magnetic resonance imaging(fMRI) study of the McCollough effect (James, Humphrey, Gati, Menon, & Goodale,1998). Although the evidence reviewed earlier is consistent with the proposal thatadaptation of mechanisms in V1 may underlie the McCollough effect, the researchwith B.K. suggests that the experience of the aftereffect likely involves extrastriateareas. Indeed, the perception of color in humans has been associated with activityin extrastriate regions as well as in V1. Both neuropsychological (Kennard, Law-den, Morland, & Ruddock, 1995; Meadows, 1974; Zeki, 1990) and neuroimagingstudies (Guylas, Heywood, Popplewell, Roland, & Cowey, 1994; Kleinschmidt,Lee, Requart, & Frahm, 1996; McKeefry & Zeki, 1997; Sakai, Watanabe, Onodera,Uchida, Kato, Yamamoto, Koizumi, & Miyashita, 1995; Wandell, Baseler, Poirson,Boynton, & Engel, in press; Zeki, Watson, Lueck, Friston, Kennard, & Frackowiak,1991) have suggested that neural circuitry in the lingual and fusiform gyri in theventromedial occipital cortex is involved in the perception of color in humans. Also,some neuroimaging studies have shown activity in areas V1/V2 (Engel, Zhang, &Wandell, 1997; Kleinschmidt et al., 1996; McKeefry & Zeki, 1997; Wandell et al.,in press), in addition to extrastriate regions, when subjects view colored stimuli. Per-haps more important, a recent fMRI study has shown that the perception of simplecolored afterimages (where no contingencies were involved) was associated withactivation of the posterior fusiform gyrus (Sakai et al., 1995). To examine the relation-


ship between the McCollough effect and cortical activity, we used high resolutionfMRI to measure the blood oxygen level-dependent or BOLD response (Ogawa,Tank, Menon, Ellermann, Kim, Merkle, & Ugurbil, 1992) after McCollough adapta-tion. To set up our experimental and control conditions, we took advantage of thefact that the McCollough effect is tuned to the orientation of the inducing stimuli,such that test stimuli oriented 45° from the inducing stimuli do not evoke the afteref-fect. In various conditions, subjects were presented with ‘‘congruent’’ test patternsthat had the same orientations as the inducing stimuli and thus would evoke theMcCollough effect, and with ‘‘noncongruent’’ test patterns that were oriented 45°from the inducing stimuli and would not evoke the effect.

Our results showed that, although there was some evidence for activation in areasV1/V2, the most consistent activation was in the lingual and fusiform areas. Thisresult is in agreement with earlier research in which color afterimages activated re-gions of the posterior fusiform gyrus (Sakai et al., 1995), as well as other reportsimplicating lingual and fusiform gyri in color perception in humans. It appears thatregions of the lingual and fusiform gyri are activated during the perception of color,whether that perception is due to the presentation of colored stimuli, afterimages ofcolored stimuli, or orientation-contingent color-aftereffects, as described here. Theactivation observed in lingual and fusiform gyri also parallels other research demon-strating that the fMRI signal in extrastriate cortex correlates with the perception ofa visual aftereffect. In these experiments, perception of the motion aftereffect wascorrelated with activity in the V5/MT complex, an extrastriate area that has beenimplicated in the perception of motion (Tootell, Reppas, Dale, Look, Sereno, Malach,Brady, & Rosen, 1995).

Although we found activation in those extrastriate areas that have been implicatedin the perception of color, we found little evidence for activation correlated with theMcCollough effect in areas V1/V2. It is true that in one subject we did see reliableactivation associated with the McCollough effect in these regions, but overall thechanges in activation we observed were much less reliable than those observed inthe lingual/fusiform areas. The absence of a strong effect in areas V1/V2 was a bitsurprising. The bulk of the evidence from psychophysical and neuropsychologicalstudies would lead one to expect that these early visual areas would be activatedwhen subjects experienced the McCollough effect. On reflection, however, perhapsthe lack of reliable activation is not so surprising, after all. There are a number ofreasons to think so. For one thing, the mechanisms in early visual areas that arecrucial for producing the McCollough effect are probably operating on a spatial scalethat is not well measured by fMRI, at least under the imaging conditions that weused. All the stimuli in our study, both the congruent and the noncongruent, wereoriented gratings—patterns that are already powerful stimuli for driving activity inV1. Thus, any differences due to the McCollough effect would have to be superim-posed on what are already high levels of activation. Another possibility that shouldbe considered is that the adaptation that leads to the perception of the aftereffect mayinvolve neural mechanisms that are not likely to produce differences in blood flow—which, of course, is what is being measured indirectly by fMRI. As we will discussin the next section, it is likely that the process of adaptation produces only subtlechanges in the profile of the activity of the neural elements coding color and orienta-


tion. We hope to address these issues in further research, including a study in whichwe will examine activation in V1, not only during the perception of the McCollougheffect, but also during its induction. Nevertheless, this initial fMRI study on theMcCollough effect has revealed a strong relationship between perception of the after-effect and activation in the lingual and fusiform gyri—and has provided a hint, atleast, that changes in activation are also occurring in areas V1/V2.

What Is ‘‘Adapted’’ during Induction of the McCollough Effect?

Although a great deal of evidence suggests that many aftereffects, including theMcCollough effect, result from ‘‘adaptation’’ of low-level mechanisms in the corticalvisual pathways, we have not described in any detail what such adaptation may meanat the level of neural circuitry. As discussed above, the nature of these changes willhave important implications for studies using functional neuroimaging. To outlineperhaps the most common explanation for aftereffects, suggested by Sutherland(1961), consider the tilt aftereffect discussed above. After viewing a grating tilted tothe right of vertical for a few minutes, subjects typically report that a vertical gratingappears tilted slightly to the left (see also Blakemore, 1973). It is proposed that fora particular sensory dimension, such as contour orientation, values on that dimensionare represented by a group of neurons in the visual cortex that are ‘‘tuned’’ for differ-ent orientations. The tuning of these neurons is assumed to be broad, so that thetuning function of a particular neuron overlaps with those of other neurons respondingmaximally to similar orientations. It is further proposed that the joint pattern of activa-tion of these neurons determines the subjective quality of the percept, the perceivedorientation in this case. Finally, it is assumed that prolonged presentation of a particu-lar stimulus, as happens in a typical aftereffect induction regime (i.e., staring at atilted grating), differentially adapts this ensemble of neurons. Although the detailednature of this adaptation is not well understood, it has often been suggested thatadaptation produces a depression of sensitivity or a ‘‘fatigue’’ in those neurons thatare most closely tuned to the induction stimulus (see, for example, Movshon & Len-nie, 1979). In the case of the tilt aftereffect, neurons tuned to orientations near andslightly to the right of vertical would be fatigued during the induction period. Thisdepression in activity produces a shift in the distribution of activity in the populationof neurons that code orientation and causes a vertical grating to be perceived as tiltedto the left.

By analogy, it has been proposed that the McCollough effect is caused by thefatiguing of ‘‘double duty’’ units that code simultaneously for orientation and color(e.g., Houck & Hoffman, 1986; Sharpe, Harris, Fach, & Braun, 1991). There is someevidence for such neurons in monkey striate cortex (e.g., Hubel & Wiesel, 1968;Leventhal, Thompson, Liu, Zhou, & Ault, 1995; Michael, 1978). Such units wouldnormally respond best to a particular combination of color and orientation.

Explanations of the McCollough effect based on fatiguing of simple neural ele-ments have been criticized on several grounds. First, McCollough effects last farlonger than could be expected from what is known of the physiology of single cellsin the visual system. McCollough effects can last for hours, days, or weeks withoutcomplete dissipation (Riggs, White, & Eimas, 1974; Jones & Holding, 1975), as can


many ‘‘simple’’ aftereffects (for review, see Humphrey, 1998). This is generallyconsidered to be much longer than the fatiguing and recovery times of simple neuralprocesses. A second criticism is that the rate of decay depends on the conditions ofstimulation. The McCollough effect is preserved in the absence of stimulation(Jones & Holding, 1975; MacKay & MacKay, 1975) and the form of decay can bemodified to some extent by pre- and postinduction exposure to achromatic gratings(Skowbo, 1988). A third criticism is that, because there are many different types ofcontingent aftereffects, the number of such double-duty detectors would have to beprohibitively large (Harris, 1980; for review, see Durgin, 1996; Humphrey, 1998).A fourth criticism stems from functional considerations. It is simply not clear whatfunction neural fatigue of this kind would serve.

If the fatigue of double-duty units cannot account for the McCollough effect andother contingent aftereffects, then what other explanations are there? One obviouspossibility is that the induction period leads to a modification in connection strengthbetween mechanisms that code different stimulus attributes (e.g., Barlow, 1990, 1997;Humphrey, 1998; Savoy, 1984). In the case of the McCollough effect, mechanismsthat code for color and separate achromatic systems that code for orientation wouldbe involved. On the basis of present understanding, however, it is unclear just wherethe color and form mechanisms underlying the McCollough effect reside. We haveargued that the McCollough effect reflects changes at a low level of the visual systemand have suggested that perhaps V1 has the requisite properties. It could be that theso-called ‘‘blobs’’ of V1 which, according to some research, contain a high propor-tion of color selective cells that are monocular and contain unoriented receptive fields(Livingstone & Hubel, 1984; Ts’o & Gilbert, 1988; but see Leventhal et al., 1995)serve as the color component, while other cells in V1 that are orientation and spatialfrequency sensitive, but show a broadband response to wavelength, subserve the formcomponent (Savoy & Gabrieli, 1991). This would again place the locus for the pro-duction of the effect in V1. There is some recent research using single cell recordingtechniques, showing that V1 cells in macaque monkeys can adapt to at least somecontingencies (Carandini, Barlow, O’Keefe, Poirson, & Movshon, 1997). While thecontingencies were between two different orientations, and not between color andorientation, the results are consistent with the proposal that V1 can subserve contin-gent aftereffects like the McCollough effect.

Of course, mechanisms that code for color and orientation exist beyond V1 andthere is evidence that they maintain separation in extrastriate regions, as well (Deyoe,Felleman, Van Essen, & McClendon, 1994). Van Essen and DeYoe (1995) havereferred to the two streams that make up the occipitotemporal pathway as the ‘‘blob-dominated stream’’ and the ‘‘interblob-dominated stream.’’ They propose that theblob-dominated stream might be specialized to represent surface characteristics ofobjects, such as texture and color information, while the interblob-dominated streammight be more specialized for dealing with form information (see also Allman &Zucker, 1993; Grossberg, 1987). Induction of the McCollough effect would not onlyaffect cells in V1, but the changes in visual processing produced by such adaptationwould lead to modification of the signals sent to structures further downstream inthe occipitotemporal pathways. It is presumably some of this activity in higher ordervisual areas that we are detecting in our fMRI study.


Whatever the mechanisms underlying contingent aftereffects might be, it seemslikely that they serve some useful function. It has been proposed by a number ofresearchers that such changes in perception reflect a recalibration of the relationshipbetween different stimulus dimensions (e.g., Barlow, 1990; Dodwell & Humphrey,1990; Durgin & Profitt, 1996; Humphrey, 1998). The recalibration mechanisms,which are thought to operate early in the cortical visual pathways, adjust the signalsfrom neural elements coding independent stimulus dimensions when particular valueson those dimensions are correlated more than they would (or should) be in the realworld. Such ‘‘spuriously’’ high correlations might arise from time to time, as is thecase during induction of the McCollough effect, and, for this reason, there wouldneed to be mechanisms that can recalibrate the system to cope with these inflatedcorrelations. Richard Gregory (1998, p. 149) put it well: ‘‘The McCollough effectis a dramatic and experimentally useful phenomenon—illustrating how vision scalesand adapts itself generally avoiding its own errors, to make sense of the world.’’

CONSCIOUSNESS, V1, AND THE MCCOLLOUGH EFFECT

We have been marshaling evidence to suggest that the critical mechanisms for theinduction of the McCollough effect are located early in the cortical visual pathway,specifically in V1. We have also suggested however, that the perception of theMcCollough effect depends on mechanisms beyond V1. To use the terminology ofVan Essen and DeYoe (1995), the conscious perception of the color aftereffect proba-bly depends on extrastriate mechanisms in the blob-dominated stream. In fact, ourfMRI study indicates that such extrastriate activation may involve parts of the lingualand fusiform gyri.

The suggestion that even though McCollough induction modifies processes in V1,the perception of the aftereffect depends on activity in extrastriate regions is consis-tent with Crick and Koch’s hypothesis (1995a,b; 1998) that V1 activity by itself isnot sufficient for perception—although it is clearly necessary. Crick and Koch reviewa variety of evidence, including results of aftereffect studies, that is consistent withtheir hypothesis. For example, recent research by He and his colleagues shows thatadaptation to contour orientation can also occur without awareness of the orien-tation of the adaptation gratings (He, Cavanagh, & Intriligator, 1996, 1997; He,Smallman, & McLeod, 1995). Because these aftereffects are thought to be mediatedby orientation-sensitive neurons in V1, the results imply that the neurons mediatingvisual awareness must be located past this stage. Some recent research also suggeststhat V1 neurons may respond to high temporal frequencies that we can’t see (Gur &Snodderly, 1997).

As we saw earlier, even though people with visual form agnosia cannot see orienta-tion, they nonetheless experience the McCollough effect. This demonstrates that ori-entation selectivity is still operative in these individuals, but at a level that is inacces-sible to conscious perception. There is evidence in at least some of the individualsthat V1 is relatively intact; what appears to be damaged is the output to the temporallobe that codes for orientation and form. To use the terminology of Van Essen andDeYoe (1995) once again, these individuals appear to have damage to the interblobdominated stream. Taken together, these observations are also consistent with the


Crick and Koch hypothesis and suggest that, even though V1 can support the induc-tion of an orientation-contingent aftereffect, it cannot by itself deliver the consciousperception of orientation.

The case of B.K. suggests, however, that although V1 neurons are likely to beadapted during McCollough induction, the actual ‘‘read-out,’’ or the neural activitythat directly underlies the perception of the aftereffect, occurs at higher cortical levels,a conclusion that is consistent with the results of our fMRI study. It appears thatsome crucial circuits involved in perceiving the colors associated with the McCol-lough effect are damaged in B.K., since he fails to report a McCollough effect in hisscotoma. B.K.’s lesion in areas V2/VP may have disrupted the connections betweenV1 and higher cortical areas, perhaps located in the fusiform and/or lingual gyri, thatmediate the perception of color. All of this, of course, is highly speculative. As wementioned earlier, B.K.’s case is a puzzling one, in that he still shows some sensitivityto color in his scotoma, suggesting that at least some color-coded input from the eyeis reaching higher cortical areas.

In summary, the evidence we have reviewed suggests that the McCollough effectand other orientation-specific aftereffects reflect changes in the activity of cells at anearly stage of the visual cortical pathways, probably in V1. But even though themechanisms in V1 are necessary for the conscious perception of the McCollougheffect, they are not sufficient—a proposal that we have argued is consistent withCrick and Koch’s (1995a,b; 1998) hypothesis that the activity in primary visual cortexis not directly accessible to consciousness. The psychologist’s microelectrode, itseems, is not only of use in the study of the principles of sensory coding, but may,like the neurophysiologist’s microelectrode, be an important tool in the search forthe neural correlates of consciousness.

ACKNOWLEDGEMENTS

Preparation of this paper and of the research of the authors that is described was supported by grantsfrom the Natural Sciences and Engineering Research Council of Canada to G.K.H. and by the MedicalResearch Council of Canada to M.A.G.

REFERENCES

Adler, A. (1950). Course and outcome of visual agnosia. Journal of Nervous and Mental Disease, 111,41–51.

Allman, J., & Zucker, S. W. (1993). On cytochrome oxidase blobs in visual cortex. In L. Harris & M.Jenkin (Eds.), Spatial vision in humans and robots (pp. 33–38). Cambridge, UK: Cambridge Univ.Press.

Andrews, T. J., & Purves, D. (1997). Similarities in normal and binocularly rivalrous viewing. Proceed-ings of the National Academy of Sciences USA, 94, 9905–9908.

Anstis, S., Verstraten, F. A. J., & Mather, G. (1998). The motion aftereffect. Trends in Cognitive Sciences,2, 111–117.

Appelle, S. (1972). Perception and discrimination as a function of stimulus orientation: The ‘‘obliqueeffect’’ in man and animals. Psychological Bulletin, 78, 266–278.

Barlow, H. B. (1990). A theory about the functional role and synaptic mechanism of visual after-effects.In C. Blakemore (Ed.), Vision: Coding and efficiency (pp. 363–375). Cambridge, UK: CambridgeUniv. Press.


Barlow, H. B. (1997). The knowledge used in vision and where it comes from. Philosophical Transac-tions of the Royal Society of London B, 352, 1141–1147.

Bedford, F. L., & Reinke, K. S. (1993). The McCollough effect: Dissociating retinal from spatial coordi-nates. Perception & Psychophysics, 54, 515–526.

Benson, D. F., & Greenberg, J. P. (1969). Visual form agnosia: A specific defect in visual discrimination.Archives of Neurology, 20, 82–89.

Blake, R., & Fox, R. (1974). Adaptation to invisible gratings and the site of binocular rivalry suppression.Nature, 249, 488–490.

Blake, R., Overton, R., & Lema-Stern, S. (1981). Interocular transfer of visual aftereffects. Journal ofExperimental Psychology: Human Perception and Performance, 7, 367–381.

Blakemore, C. (1973). The confounded brain. In R. L. Gregory & E. H. Gombrich (Eds.), Illusion innature and art (pp. 8–47). London, UK: Duckworth.

Broerse, J., & O’Shea, R. P. (1995). Local and global factors in spatially-contingent coloured aftereffects.Vision Research, 35, 207–226.

Buser, P., & Imbert, M. (1992). Vision (R. H. Kay, Trans.). MIT Press. (Original work published 1987)

Campion, J. (1987). Apperceptive agnosia: The specification and description of constructs. In G. W.Humphreys & M. J. Riddoch (Eds.), Visual object processing: A cognitive neuropsychological ap-proach (pp. 197–232). London, UK: Erlbaum.

Carandini, M., Barlow, H. B., O’Keefe, L. P., Poirson, A. B., & Movshon, J. A. (1997). Adaptation tocontingencies in macaque primary visual cortex. Philosophical Transactions of the Royal Societyof London B, 352, 1149–1154.

Chaudhuri, A. (1990). Modulation of the motion aftereffect by selective attention. Nature, 344, 60–62.

Crick, F. (1996). Visual perception: Rivalry and consciousness. Nature, 379, 485–486.

Crick, F., & Koch, C. (1995a). Are we aware of neural activity in primary visual cortex? Nature, 375,121–123.

Crick, F., & Koch, C. (1995b). Cortical areas in visual awareness—Reply. Nature, 377, 294–295.

Crick, F., & Koch, C. (1998). Consciousness and neuroscience. Cerebral Cortex, 8, 97–107.

Denney, D., & Adorjani, C. (1972). Orientation specificity of visual cortical neurons after head tilt.Experimental Brain Research, 14, 312–317.

Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual attention. Annual Reviewof Neuroscience, 18, 193–222.

DeYoe, E. A., Felleman, D. J., Van Essen, D. C., & McClendon, E. (1994). Multiple processing streamsin occipitotemporal visual cortex. Nature, 371, 151–154.

Dodwell, P. C., & Humphrey, G. K. (1990). A functional theory of the McCollough effect. PsychologicalReview, 97, 78–89.

Durgin, F. H. (1996). Visual aftereffect of texture density contingent on color of frame. Perception &Psychophysics, 58, 207–223.

Durgin, F. H., & Profitt, D. R. (1996). Visual learning in the perception of texture: Simple and contingentaftereffects of texture density. Spatial Vision, 9, 423–474.

Efron, R. (1968). What is perception? In R. S. Cohen, & M. Wartofsky (Eds.), Boston studies in thephilosophy of science: Vol. 4 (pp. 137–173). Dordrecht, Netherlands: D. Reidel Publishing Com-pany.

Ellis, S. R. (1976). Orientation constancy of the McCollough effect. Perception & Psychophysics, 19,183–192.

Engel, S., Zhang, X., & Wandell, B. (1997). Colour tuning in human visual cortex measured with func-tional magnetic resonance imaging. Nature, 388, 68–71.

Essock, E. A. (1980). The oblique effect of stimulus identification considered with respect to two classesof oblique effects. Perception, 9, 37–46.

Farah, M. J. (1990). Visual agnosia. Cambridge, MA: MIT Press.


Felleman, D. J., & Van Essen, D. C. (1991). Distributed hierarchical processing in primate visual cortex.Cerebral Cortex, 1, 1–47.

ffytche, D. H., Guy, C. C., & Zeki, S. (1995). The parallel visual motion inputs into areas V1 and V5of human cerebral cortex. Brain, 118, 1375–1394.

Frisby, J. P. (1980). Seeing: Illusion, brain and mind. Oxford, UK: Oxford Univ. Press.

Goodale, M. A., & Humphrey, G. K. (1998). The objects of action and perception. Cognition, 57, 179–205.

Goodale, M. A., & Milner, A. D. (1992). Separate visual pathways for perception and action. Trendsin Neurosciences, 15, 20–25.

Goodale, M. A., Milner, A. D., Jakobson, L. S., & Carey, D. P. (1991). A neurological dissociationbetween perceiving objects and grasping them. Nature, 349, 154–155.

Goodale, M. A., Meenan, J. P., Bulthoff, H. H., Nicolle, D. A., Murphy, K. S., & Racicot, C. I. (1994).Separate neural pathways for the visual analysis of object shape in perception and prehension.Current Biology, 4, 604–610.

Gregory, R. L. (1998). Eye and brain (5th edition). Oxford, UK: Oxford Univ. Press.

Gross, C. G. (1991). Connections of striate cortex and the superior colliculus to visual function in areaMT, the superior temporal polysensory area, and inferior temporal cortex. Neuropsychologia, 29,497–515.

Grossberg, S. (1987). Cortical dynamics of three-dimensional form, color, and brightness perception:II. Binocular theory. Perception & Psychophysics, 41, 117–158.

Guylas, B., Heywood, C. A., Popplewell, D. A., Roland, P. E., & Cowey, A. (1994). Visual form discrimi-nation from color or motion cues: Functional anatomy by positron emission tomography. Proceed-ings of the National Academy of Sciences USA, 91, 9965–9969.

Gur, M., & Snodderly, D. M. (1997). A dissociation between brain activity and perception: Chromaticallyopponent cortical neurons signal chromatic flicker that is not perceived. Vision Research, 37, 377–382.

Harris, C. S. (1980). Insight or out of sight? Two examples of perceptual plasticity in the human adult.In C. S. Harris (Ed.), Visual coding and adaptability (pp. 95–149). Hillsdale, NJ: Erlbaum.

He, S., Cavanagh, P., & Intriligator, J. (1996). Attentional resolution and the locus of visual awareness.Nature, 383, 334–337.

He, S., Cavanagh, P., & Intriligator, J. (1997). Attentional resolution. Trends in Cognitive Sciences, 1,115–121.

He, S., Smallman, H. S., & MacLeod, D. I. A. (1995). Neural and cortical limits on visual resolution.Investigative Ophthalmology and Visual Science (Supplement), 36, 2010.

Horn, G., Stechler, G., & Hill, R. M. (1972). Receptive fields of units in the visual cortex of the cat inthe presence and absence of bodily tilt. Experimental Brain Research, 15, 113–132.

Horton, J. C., & Hoyt, W. F. (1991). Quadrantic visual field defects. Brain, 114, 1703–1718.

Houck, M. R., & Hoffman, J. E. (1986). Conjunction of color and form without attention: Evidencefrom an orientation-contingent color aftereffect. Journal of Experimental Psychology: Human Per-ception and Performance, 12, 186–199.

Hubel, D. H., & Wiesel, T. N. (1965). Binocular interaction in striate cortex of kittens reared withartificial strabismus. Journal of Neurophysiology, 28, 1041–1059.

Hubel, D. H., & Wiesel, T. N. (1968). Receptive fields and functional architecture of monkey striatecortex. Journal of Physiology, 195, 215–243.

Humphrey, G. K. (1998). The McCollough effect: Misperception and reality. In V. Walsh & J. Kulikowsi(Eds.), Perceptual constancy: Why things look as they do (pp. 31–68). Cambridge, UK: CambridgeUniv. Press.

Humphrey, G. K., Goodale, M. A., Bowen, C. V., Gati, J. S., Vilis, T., Rutt, B. K., and Menon, R. S.(1997). Differences in perceived shape from shading correlate with activity in early visual areas.Current Biology, 7, 144–147.


Humphrey, G. K., Goodale, M. A. G., Corbetta, M., & Aglioti, S. (1995). The McCollough effect revealsorientation discrimination in a case of cortical blindness. Current Biology, 5, 545–551.

Humphrey, G. K., Goodale, M. A., & Gurnsey, R. (1991). Orientation discrimination in a visual formagnosic: Evidence from the McCollough effect. Psychological Science, 5, 331–335.

Humphrey, G. K., Goodale, M. A., Jakobson, L. J., & Servos, P. (1994). The role of surface informationin object recognition: Studies of a visual form agnosic and normal subjects. Perception, 23, 1457–1481.

Humphrey, G. K., Gurnsey, R., & Bryden, P. J. (1994). An examination of colour-contingent patternaftereffects. Spatial Vision, 8, 77–94.

Humphrey, G. K., Herbert, A. M., Symons, L. A., & Kara, S. (1994). McCollough effect to ‘‘form’’:A local phenomenon. Journal of Experimental Psychology: General, 123, 86–90.

Humphrey, G. K., Skowbo, D., Symons, L. A., Herbert, A. M., & Grant, C. (1994). Text-contingentcolor-aftereffects: A re-examination. Perception & Psychophysics, 56, 405–413.

Humphrey, G. K., Symons, L. A., Herbert, A. M., & Goodale, M. A. (1996). A neurological dissociationbetween shape from shading and shape from edges. Behavioural Brain Research, 76, 117–125.

Humphrey, G. K., Gurnsey, R., & Fekete, E. (1991). Rapid discrimination of McCollough effects. Per-ception, 20, 467–482.

James, T. W., Humphrey, G. K., Gati, J. S., Menon, R. S., & Goodale, M. A. (1998). The McCollougheffect produces increased fMRI signal in the fusiform and lingual gyri. Presented at the 4th Interna-tional Conference on Functional Mapping of the Human Brain. Montreal, Quebec, Canada.

Jones, P. D., & Holding, D. H. (1975). Extremely long-term persistence of the McCollough effect.Journal of Experimental Psychology: Human Perception and Performance, 1, 323–327.

Julesz, B. (1984). Toward an axiomatic theory of preattentive vision. In G. Edelman, W. Einar, & W.Cowan (Eds.), Dynamic aspects of neocortical function (pp. 585–612). New York: Wiley.

Kennard, C., Lawden, M., Morland, A. B., & Ruddock, K. H. (1995). Colour identification and colourconstancy are impaired in a patients with incomplete achromatopsia associated with prestriate corti-cal lesions. Proceedings of the Royal Society of London B, 260, 169–175.

Kleinschmidt, A., Lee, B. B., Requardt, M., & Frahm, J. (1996). Functional mapping of color processingby magnetic resonance imaging of responses to selective P- and M-pathway stimulation. Experimen-tal Brain Research, 110, 279–288.

Kolb, B. (1990). Recovery from occipital stroke: A self-report and an inquiry into visual processes.Canadian Journal of Psychology, 44, 130–147.

Kovacs, I., Papathomas, T. V., Yang, M., & Feher, A. (1996). When the brain changes its mind: Interocu-lar grouping during binocular rivalry. Proceedings of the National Academy of Sciences USA, 93,15508–15511.

Lack, L. C. (1978). Selective attention and the control of binocular rivalry. New York: Mouton.

Lehmkuhle, S. W., & Fox, R. (1975). Effect of binocular rivalry suppression on the motion aftereffect.Vision Research, 15, 855–859.

Leopold, D. A., & Logothetis, N. K. (1996). Activity changes in early visual cortex reflect monkeys’percepts during binocular rivalry. Nature, 379, 549–553.

Leventhal, A. G., Thompson, K. G., Liu, D., Zhou, Y., & Ault, S. J. (1995). Concomitant sensitivityto orientation, direction, and color of cells in layers 2, 3, and 4 of monkey striate cortex. The Journalof Neuroscience, 15, 1808–1818.

Lewis, D. A., Campbell, M. J., Terry, R. D., & Morrison, J. H. (1987). Laminar and regional distributionsof neurofibrillary tangles and neuritic plaques in Alzheimer’s disease: A quantitative study of visualand auditory cortices. The Journal of Neuroscience, 7, 1799–1808.

Livingstone, M., & Hubel, D. (1984). Anatomy and physiology of a color system in the primate visualcortex. The Journal of Neuroscience, 4, 309–356.

Lissauer, H. (1890). Ein Fall von Seelenblindheit nebst einem Beitrage zur Theorie derselben. Archivfur Psychiatrie und Nervenkrankheiten, 21, 222–270.


Logothetis, N. K., Leopold, D. A., & Sheinberg, D. L. (1996). What is rivalling during binocular rivalry?Nature, 380, 621–624.

Logothetis, N. K., & Schall, J. D. (1989). Neuronal correlates of subjective visual perception. Science,245, 761–763.

Lovegrove, W. J., & Over, R. (1972). Color adaptation of spatial frequency detectors in the human visualsystem. Science, 176, 541–543.

Mack, A., & Rock, I. (1998). Inattentional blindness. Cambridge, MA: MIT Press.

MacKay, D. M., & MacKay, V. (1975). What causes decay of pattern-contingent chromatic aftereffects?Vision Research, 15, 462–464.

Mayhew, J. E. W., & Frisby, J. P. (1978). Texture discrimination and Fourier analysis in human vision.Nature, 275, 438–439.

McCollough, C. (1965). Color adaptation of edge-detectors in the human visual system. Science, 149,1115–1116.

McKeefry, D. J., & Zeki, S. (1997). The position and topography of the human colour centre as revealedby functional magnetic resonance imaging. Brain, 120, 2229–2242.

McKeon, R. P. (Editor) (1941). The basic works of Aristotle. New York: Random House.

Meadows, J. (1974). Disturbed perception of colours associated with localized cerebral lesions. Brain,97, 615–632.

Michael, C. R. (1978). Color vision mechanisms in monkey striate cortex: Simple cells with dual oppo-nent-color receptive fields. Journal of Neurophysiology, 41, 1233–1241.

Milner, A. D., & M. A. Goodale. (1995). The visual brain in action. Oxford, UK: Oxford Univ. Press.

Milner, A. D., & Heywood, C. A. (1989). A disorder of lightness discrimination in a case of visual formagnosia. Cortex, 25, 489–494.

Milner, A. D., Perrett, D. I., Johnston, R. S., Benson, P. J., Jordan, T. R., Heeley, D. W., Bettucci, D.,Mortara, F., Mutani, R., Terazzi, R., & Davidson, D. L. W. (1991). Perception and action in visualform agnosia. Brain, 114, 405–428.

Movshon, J. A., & Lennie, P. (1979). Pattern-selective adaptation in visual cortical neurones. Nature,278, 850–852.

Moulden, B. (1980). Aftereffects and the integration of neural activity within a channel. PhilosophicalTransactions of the Royal Society of London B, 290, 39–55.

Murch, G. M. (1972). Binocular relationships in a size and color orientation specific aftereffect. Journalof Experimental Psychology, 93, 30–34.

Nakayama, K., & Joseph, J. S. (in press). Attention, pattern recognition, and pop-out in visual search.In R. Parasuraman (Ed.), The attentive brain. Cambridge, MA: MIT Press.

Ogawa, S., Tank, D. W., Menon, R., Ellermann, J. M., Kim, S-G., Merkle, H., & Ugurbil, K. (1992).Proceedings of the National Academy of Sciences, 89, 5951–5955.

O’Shea, R., & Crassini, B. (1981). Interocular transfer of the motion aftereffect is not reduced by binocu-lar rivalry. Vision Research, 21, 801–804.

Rees, G., Frith, C. D., & Lavie, N. (1997). Modulating irrelevant motion perception by varying attentionalload in an unrelated task. Science, 278, 1616–1619.

Riggs, L. A., White, K. D., & Eimas, P. D. (1974). Establishment and decay of orientation-contingentaftereffects of color. Perception & Psychophysics, 16, 535–542.

Sakai, K. K., Watanabe, E., Onodera, Y., Uchida, I., Kato, H., Yamamoto, E., Koizumi, H., & Miyashita,Y. (1995). Functional mapping of the human colour centre with echo-planar magnetic resonanceimaging. Proceedings of the Royal Society of London B, 261, 89–98.

Savoy, R. L. (1984). ‘‘Extinction’’ of the McCollough effect does not transfer interocularly. Perception &Psychophysics, 36, 571–576.

Savoy, R. L., & Gabrielli, J. D. E. (1991). Normal McCollough effect in Alzheimer’s disease and globalamnesia. Perception & Psychophysics, 49, 448–455.


Seaber, J. H., & Lockhead, G. R. (1989). McCollough aftereffects in strabismus and amblyopia. VisionResearch, 29, 609–617.

Sengpeil, F. (1997). Binocular rivalry: Ambiguity resolved. Current Biology, 7, 443–450.

Servos, P., Goodale, M. A., & Humphrey, G. K. (1993). The drawing of objects by a visual form agnosic:Contribution of surface properties and memorial representations. Neuropsychologia, 31, 251–259.

Sharpe, L. T., Harris, J. P., Fach, C. C., & Braun, D. I. (1991). Contingent aftereffects: Lateral interactionsbetween color and motion. Perception and Psychophysics, 49, 434–447.

Sheinberg, D. L., & Logothetis, N. K. (1997). The role of temporal cortical areas in perceptual organiza-tion. Proceedings of the National Academy of Sciences USA, 94, 3408–3413.

Siegel, S., & Allan, L. G. (1993). Pairings in learning and perception: Pavlovian conditioning and contin-gent aftereffects. In D. Medin (Ed.), The psychology of learning and motivation, Vol. 28 (pp. 127–160). New York: Academic Press.

Skowbo, D. (1988). Interference with McCollough effects via pre- and postinduction exposure to achro-matic gratings: Time course and magnitude of aftereffect decrement. Perception and Psychophysics,44, 295–303.

Skowbo, D., Timney, B. N., Gentry, T. A., & Morant, R. B. (1975). McCollough effects: Experimentalfindings and theoretical accounts. Psychological Bulletin, 82, 497–510.

Stromeyer, C. F., III. (1972a). Contour contingent color aftereffects: Retinal area specificity. AmericanJournal of Psychology, 85, 227–235.

Stromeyer, C. F., III. (1972b). Edge-contingent aftereffects in human vision: Spatial frequency specific-ity. Vision Research, 12, 717–733.

Stromeyer, C. F. (1978). Form-color aftereffects in human vision. In R. Held, H. W. Leibowitz, &H. L. Teuber (Eds.), Perception: Handbook of sensory physiology, Vol. VIII (pp. 97–142). NewYork: Springer-Verlag.

Sutherland, N. S. (1961). Figural after-effects and apparent size. Quarterly Journal of ExperimentalPsychology, 13, 222–228.

Takeuchi, T., & Kita, S. (1994). Attentional modulation in motion aftereffect. Japanese PsychologicalResearch, 36, 94–107.

Thompson, P., & Latchford, G. (1986). Colour-contingent after-effects are really wavelength-contingent.Nature, 320, 525–526.

Thompson, P., & Travis, D. (1989). Making Mayhew and Frisby effortlessly discriminable. Perception,18, 231–235.

Tootell, R. B. H., Hadjikhani, N. K., Mendola, J. D., Marrett, S., & Dale, A. M. (1998). From retinotopyto recognition: fMRI in human visual cortex. Trends in Cognitive Sciences, 2, 174–183.

Tootell, R. B. H., Reppas, J. B., Dale, A. M., Look, R. B., Sereno, M. I., Malach, R., Brady, T. J., &Rosen, B. R. (1995). Visual motion aftereffect in human cortical area MT revealed by functionalmagnetic resonance imaging. Nature, 375, 139–141.

Treisman, A. (1988). Features and objects: The fourteenth Bartlett Memorial Lecture. Quarterly Journalof Experimental Psychology, 40A, 201–237.

Ts’o, D. Y., & Gilbert, C. D. (1988). The organization of chromatic and spatial interactions in the primatestriate cortex. The Journal of Neuroscience, 8, 1712–1727.

Van Essen, D. C., & DeYoe, E. A. (1995). Concurrent processing in the primate visual cortex. In M.Gazzaniga (Ed.), The cognitive neurosciences (pp. 382–400). Cambridge, MA: MIT Press.

Vecera, S. P., & Gilds, K. S. (1997). What is it like to be a patient with apperceptive agnosia? Conscious-ness and Cognition, 6, 237–266.

Vidyasagar, T. R. (1976). Orientation specific colour adaptation at a binocular site. Nature, 261, 39–40.

Wade, N. J. (1994). A selective history of the study of visual motion aftereffects. Perception, 23, 1111–1134.


Wade, N. J., & Wenderoth, P. (1978). The influence of color and contrast rivalry on the magnitude ofthe tilt aftereffect. Vision Research, 18, 827–835.

Wandell, B. A., Baseler, H., Poirson, A. B., Boynton, G. M., & Engel, S. (in press). Computationalneuroimaging: Color tuning in two human cortical areas using fMRI. In K. Gegenfurtner & L. T.Sharpe (Eds.), Color vision: From molecular genetics to perception. Cambridge, UK: CambridgeUniv. Press.

White, K. D., Petry, H. M., Riggs, L. A., & Miller, J. (1978). Binocular interactions during establishmentof McCollough effects. Vision Research, 18, 1201–1215.

Wiesenfelder, H., & Blake, R. (1990). The neural site of binocular rivalry relative to the analysis ofmotion in the human visual system. The Journal of Neuroscience, 10, 3880–3888.

Wolfe, J. M. (in press). Visual search: A review. In H. Pashler (Ed.), Attention. London, UK: UniversityCollege London Press.

Zeki, S. (1990). A century of cerebral achromatopsia. Brain, 113, 1721–1777.

Zeki, S. (1993). A vision of the brain. Oxford, UK: Blackwell Scientific Publications.Zeki, S., Watson, J. D. G., Lueck, C. J., Friston, K. J., Kennard, C., & Frackowiak, R. S. J. (1991). A direct

demonstration of functional specialization in human visual cortex. The Journal of Neuroscience, 11,641–649.

Received May 27, 1998

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具

http://www.xuebalib.com/cloud/

http://www.xuebalib.com/

http://www.xuebalib.com/cloud/


http://www.xuebalib.com/vip.html

http://www.xuebalib.com/db.php

http://www.xuebalib.com/zixun/2014-08-15/44.html


Probing Unconscious Visual Processing with the McCollough ......transfer, the McCollough effect is...

Documents

Transcript of Probing Unconscious Visual Processing with the McCollough ......transfer, the McCollough effect is...