Active manual control of object views facilitates visual recognition.

Active exploration of large-scale environments leads to better learning of spatial layout than does passive observation [1] [2] [3]. But active exploration might also help us to remember the appearance of individual objects in a scene. In fact, when we encounter new objects, we often manipulate them so that they can be seen from a variety of perspectives. We present here the first evidence that active control of the visual input in this way facilitates later recognition of objects. Observers who actively rotated novel, three-dimensional objects on a computer screen later showed more efficient visual recognition than observers who passively viewed the exact same sequence of images of these virtual objects. During active exploration, the observers focused mainly on the 'side' or 'front' views of the objects (see also [4] [5] [6]). The results demonstrate that how an object is represented for later recognition is influenced by whether or not one controls the presentation of visual input during learning.

The McCollough effect reveals orientation discrimination in a case of cortical blindness.

BACKGROUND: The McCollough effect is a colour after-effect that is contingent on the orientation of the patterns used to induce it. To produce the effect, two differently oriented grating patterns--such as a red-and-black vertical grating and a green-and-black horizontal grating--are viewed alternatively for a few minutes. After this period of adaptation, if the black-and-white test gratings are viewed in the same orientation as the adaptation patterns, the white sections of the vertical grating will appear pale green and the white sections of the horizontal grating will appear pink. The McCollough effect indicates that colour- and orientation-coding mechanisms interact at some point during visual processing; but the question remains as to whether this interaction occurs at an early or later stage in the cortical visual pathways. In an attempt to answer this question, we studied a patient who had suffered extensive damage to extrastriate visual areas of the brain, which had left him able to see colour but little else. RESULTS: Neuropsychological and perceptual tests demonstrated that the patient, P.B., has a profound impairment in form perception and is even unable to discriminate between 90 degrees differences in the orientation of grating stimuli. He is also unable to use orientation information to control his reaching or grasping. Nevertheless, P.B. can name and discriminate different colours reliably, including those used to induce the McCollough effect. After adaptation with red-and-green gratings, P.B. appropriately reported the orientation-contingent aftereffect colours, even though he continued to be unable to discriminate the orientations of the test patterns. CONCLUSIONS: These results indicate that at some level in P.B.'s visual system orientation is being coded, but it is at a level that he is unable to use in making orientation judgements or in visuomotor control. Given the massive insult to the extrastriate cortex in P.B., it is likely that the anatomical locus of the mechanisms underlying the McCollough effect is within primary visual cortex or even earlier in the visual pathway.

The effects of visual object priming on brain activation before and after recognition.

BACKGROUND: Recognizing an object is improved by recent experience with that object even if one cannot recall seeing the object. This perceptual facilitation as a result of previous experience is called priming. In neuroimaging studies, priming is often associated with a decrease in activation in brain regions involved in object recognition. It is thought that this occurs because priming causes a sharpening of object representations which leads to more efficient processing and, consequently, a reduction in neural activity. Recent evidence has suggested, however, that the apparent effect of priming on brain activation may vary as a function of whether the neural activity is measured before or after recognition has taken place. RESULTS: Using a gradual 'unmasking' technique, we presented primed and non-primed objects to subjects, and measured activation time courses using high-field functional magnetic resonance imaging (fMRI). As the objects were slowly revealed, but before recognition had occurred, activation increased from baseline level to a peak that corresponded in time to the subjects' behavioural recognition responses. The activation peak for primed objects occurred sooner than the peak for non-primed objects, and subjects responded sooner when presented with a primed object than with a non-primed object. During this pre-recognition phase, primed objects produced more activation than non-primed objects. After recognition, activation declined rapidly for both primed and non-primed objects, but now activation was lower for the primed objects. CONCLUSIONS: Priming did not produce a general decrease in activation in the brain regions involved in object recognition but, instead, produced a shift in the time of peak activation that corresponded to the shift in time seen in the subjects' behavioural recognition performance.

Differences in perceived shape from shading correlate with activity in early visual areas.

The perception of shape from shading depends on the orientation of the shading gradient [1] [2] [3] [4]. Displays composed of elements with vertically oriented shading gradients of opposite polarity produce a strong and stable percept of 'concave' and 'convex' elements. If the shading gradients are rotated 90 degrees , the depth percept is reduced and appears much more ambiguous. Results from psychophysical [1] [2] [3] [4] [5] [6], neuropsychological [7] and computational studies [8] [9] suggest that the perception of shape from shading engages specific mechanisms in early cortical visual areas. In a three-dimensional functional magnetic resonance imaging (fMRI) study at 1.5 Tesla using a three-dimensional, interleaved-echoplanar imaging technique and a surface radio frequency (RF) coil placed under the visual cortex, we investigated the activity in these early visual areas associated with viewing shape from shading displays at two different orientations. We found significantly greater activation in area V1 and neighbouring low-level visual areas of cortex when subjects viewed displays that led to weak and unstable depth percepts than when they viewed displays that led to strong and stable depth percepts.

The drawing of objects by a visual form agnosic: contribution of surface properties and memorial representations.

Although edge-based representations of objects are thought to play a central role in object identification, it is clear that real objects convey more information about their form than line drawings. Patients with visual form agnosia, for example, are able to identify real objects more easily than the corresponding line drawings of those objects, even if exactly the same projection planes are used [Goodale et al. Object versus picture identification in a patient with visual form agnosia. Paper presented at the annual meeting of the Association for Research in Vision and Ophthalmology (April, 1991). Sarasota, FL, 1991]. To compare these two modes of representation in another way, we asked a patient (D.F.) with profound visual form agnosia to make line drawings of a series of common objects, either from long-term memory, from the real objects themselves, or from line drawings of those objects. When four independent judges rated the drawings as to how well they represented the target objects, drawings from memory received higher ratings than drawings of real objects which in turn received higher ratings than those based on line drawings. These results complement those of Goodale et al. (1991) and suggest that cues derived from surface properties and depth can assist in the demarcation of the critical features necessary for the accurate portrayal of objects. They also suggest that despite D.F.'s perceptual deficits, her long-term representation of objects is relatively intact.

A functional theory of the McCollough effect.

Pattern-contingent color aftereffects, or McCollough effects (MEs), are used to probe the visual brain's operations psychophysically. Their neural substrate is unknown, however, and theories about them are weak. Our theory proposes a strong functional role for MEs and a neuropsychological basis that accounts for "top-down" (global) constraints ignored by other theories. The functional aspect of the theory is based on the concept of contingent adaptation level, following Helson (1964), and on the "error-correcting device" of Andrews (1964), which tracks and adjusts internal representation to external-world contingencies. The neuropsychological part of the theory postulates that global properties are the result of MEs being generated not at the individual detector level but in vectorfields of which the detectors are elements. It is an implementation of Lie transformation group theory (Hoffman, 1966). Evidence for this model is assessed.

The objects of action and perception.

Two major functions of the visual system are discussed and contrasted. One function of vision is the creation of an internal model or percept of the external world. Most research in object perception has concentrated on this aspect of vision. Vision also guides the control of object-directed action. In the latter case, vision directs our actions with respect to the world by transforming visual inputs into appropriate motor outputs. We argue that separate, but interactive, visual systems have evolved for the perception of objects on the one hand and the control of actions directed at those objects on the other. This 'duplex' approach to high-level vision suggests that Marrian or 'reconstructive' approaches and Gibsonian or 'purposive-animate-behaviorist' approaches need not be seen as mutually exclusive, but rather as complementary in their emphases on different aspects of visual function.

Repetition priming and the time course of object recognition: an fMRI study.

We investigated the effects of repetition priming on the time course of recognition in several visual areas of the brain using fMRI. We slowed down recognition by gradually revealing the stimuli, in order to prolong the pre-recognition phase. Activation was lower for primed than for non-primed objects overall in both the occipitotemporal region (OTR) and the intraparietal region (IPR). A difference was found between primed and non-primed objects in the rate of increase of OTR activation. We concluded that the IPR, in addition to the OTR, was affected by repetition priming, and that this effect was different from that seen in the OTR.

A neurological dissociation between shape from shading and shape from edges.

We studied the ability of a neurological patient, who has deficits in various aspects of form perception, to perform region segregation tasks requiring discriminations based on several image properties that are related to the three-dimensional structure of objects. The patient could discriminate the apparent three-dimensional structure and orientation of shapes defined by shading gradients, but could not make such discriminations for shapes in which edges were depicted as lines or as luminance discontinuities. These results suggest that the neural pathways that compute shape from shading gradients may be independent of those that compute shape based on edges, and, based on the patient's pattern of brain damage, they also indicate a relatively early functional separation in the requisite inputs.

Eye position sense contributes to the judgement of slant.

We measured monocular judgements of the slant of a cube face while varying eye position in the absence of stereoscopic and external lighting cues. Errors were found to be small, only 10% on average of the cube's eccentricity. Two factors appear to have contributed approximately equally to this error: an underestimate of cube slant as seen by the eye and an underestimate of eye position. When prism adaptation altered the sensed eye position, the pattern of slant judgements changed to reflect the altered sense of eye position.

Manipulating and recognizing virtual objects: where the action is.

In an earlier report (Harman, Humphrey, & Goodale, 1999), we demonstrated that observers who actively rotated three-dimensional novel objects on a computer screen later showed faster visual recognition of these objects than did observers who had passively viewed exactly the same sequence of images of these virtual objects. In Experiment 1 of the present study we showed that compared to passive viewing, active exploration of three-dimensional object structure led to faster performance on a "mental rotation" task involving the studied objects. In addition, we examined how much time observers concentrated on particular views during active exploration. As we found in the previous report, they spent most of their time looking at the "side" and "front" views ("plan" views) of the objects, rather than the three-quarter or intermediate views. This strong preference for the plan views of an object led us to examine the possibility in Experiment 2 that restricting the studied views in active exploration to either the plan views or the intermediate views would result in differential learning. We found that recognition of objects was faster after active exploration limited to plan views than after active exploration of intermediate views. Taken together, these experiments demonstrate (1) that active exploration facilitates learning of the three-dimensional structure of objects, and (2) that the superior performance following active exploration may be a direct result of the opportunity to spend more time on plan views of the object.

Codes and operations in picture matching.

Ellis and Allport (1986; see also Ellis, Allport, Humphreys & Collis, 1989) proposed a model of object perception wherein successively more abstract descriptions are generated as a function of processing time. The aspect of their model that is examined here is the proposal that viewer-centred representations of objects decay rapidly whereas object-centred or semantic-level representations do not. To test the model, a picture-matching task was used in which subjects decided whether successively presented pictures rotated in the frontal plane had the same name. The pictures were either identical pictures, pictures of different objects with the same name, or pictures of objects with different names. The two successive pictures could be in the same orientation or in a different orientation. In Experiment 1, two orientations (0 degrees upright and 120 degrees) and two ISIs were examined (100 ms and 2 s). In Experiment 2, two orientation (0 degrees and 60 degrees) and three ISIs were examined (100 ms, 2 s, and 5 s). In neither experiment was there any evidence that viewpoint-specific representations disappeared at longer ISIs. These results, although consistent with other research on the perception of rotated objects, did not replicate the results of Ellis and Allport (1986) and are inconsistent with their model.

An examination of the effects of axis foreshortening, monocular depth cues, and visual field on object identification.

Four experiments are reported on the identification of line drawings of common objects. In each experiment, performance on "unconventional" views of the objects, in which the major axis of the object was foreshortened, was compared to performance on more "conventional" views without appreciable foreshortening. In each experiment, except Experiment 2, where performance on the two views was experimentally equated, the foreshortened views were more difficult to identify than were the conventional views. Experiments 1 and 2 showed that if the foreshortened views were presented on a background with strong monocular depth cues, object identification was improved. This result suggests that part of the difficulty in identifying objects depicted from such a view stems from an improper depth interpretation of the object depictions. Experiments 3 and 4 examined visual field differences in the identification of the two types of object view. Results reported in the neuropsychological literature have shown that people with right-hemisphere damage have particular difficulty with the identification of unconventional views of objects that foreshorten major axes. Accordingly, it was expected that there would be a left visual field advantage for the foreshortened views. Neither experiment yielded any visual field effects consistent with this expectation. Possible reasons for the lack of a field effect are discussed.

Bilateral symmetry detection: testing a 'callosal' hypothesis.

At the end of the 19th century Mach observed that vertical symmetry is more easily perceived than is symmetry at other orientations, and proposed this resulted from bilateral symmetry in the visual system. Numerous studies of symmetry detection have been conducted, but none has been concentrated on Mach's proposal. Recent interpretations of Mach's hypothesis suggest the corpus callosum mediates the vertical-symmetry advantage. In this 'callosal' hypothesis it is suggested that the detectability of symmetry should be narrowly tuned around vertical, and that presentation of patterns away from fixation should disrupt the vertical advantage. We found that the vertical advantage was disrupted by presentation of patterns 1.2 deg from fixation, while detection of symmetry at other orientations was not disrupted. At fixation the orientation tuning was at least within +/- 10 degrees of vertical. The detection of vertical symmetry at fixation was found to be anomalous in two subjects born without a corpus callosum as compared with controls, but relatively normal for presentation off fixation. The three experiments reported are in agreement with some of the predictions derived from the callosal hypothesis. It appears that the callosal hypothesis may account for the relative advantage of vertical symmetry at fixation, but other mechanisms must operate to detect symmetry at other orientations and positions.

Surface cues reduce the latency to name rotated images of objects.

Jolicoeur (1985, Memory & Cognition 13 289-303) found a linear increase in the latency to name line drawings of objects rotated (0 degrees to 120 degrees) from the upright (0 degrees) in the initial trial block. This effect was much shallower in later blocks. He proposed that the initial effect may indicate that mental rotation is the default process for recognising rotated objects, and that the decrease in this effect, seen with practice, may reflect the increased use of learned orientation-invariant features. Initially, we were interested in whether object-colour associations that may be learned during the initial block, could account for the reduced latency to name rotated objects, seen in later blocks. In experiment 1 we used full-cue colour images of objects that depicted colour and other surface cues. Surprisingly, given that Jolicoeur's findings were replicated several times with line drawings, we found that even the initial linear trend in naming latency was shallow. We replicated this result in follow-up experiments. In contrast, when we used less-realistic depictions of the same objects that had fewer visual cues (ie line drawings, coloured drawings, greyscale images), the results were comparable to those of Jolicoeur. Also, the initial linear trends were steeper for these depictions than for full-cue colour images. The results suggest that, when multiple surface cues are available in the image, mental rotation may not be the default recognition process.

Probing unconscious visual processing with the McCollough effect.

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

Encoding 'regular' and 'random' sequences of views of novel three-dimensional objects.

When we look at an object as we move or the object moves, our visual system is presented with a sequence of different views of the object. It has been suggested that such regular temporal sequences of views of objects contain information that can aid in the process of representing and recognising objects. We examined whether seeing a series of perspective views of objects in sequence led to more efficient recognition than seeing the same views of objects but presented in a random order. Participants studied images of 20 novel three-dimensional objects rotating in depth under one of two study conditions. In one study condition, participants viewed an ordered sequence of views of objects that was assumed to mimic important aspects of how we normally encounter objects. In the other study condition, participants were presented the same object views, but in a random order. It was expected that studying a regular sequence of views would lead to more efficient recognition than studying a random presentation of object views. Although subsequent recognition accuracy was equal for the two groups, differences in reaction time between the two study groups resulted. Specifically, the random study group responded reliably faster than the sequence study group. Some possible encoding differences between the two groups are discussed.

The role of structure in infant visual pattern perception.

The research of the authors and others on the role of structure in infant visual pattern perception is reviewed. Two new experiments are reported on preference for symmetry in infants. It is shown that infants reliably prefer patterns with multiple axes of bilateral symmetry relative to asymmetrical patterns. Also, the results demonstrate that vertically oriented single axis bilateral symmetry is more salient than horizontally oriented symmetry in infants as it is in adults. It is argued that sensitivity to various kinds of pattern structure reflects fundamental operations of the visual system.

Recognizing novel views of three-dimensional objects.

The purpose of the experiments reported was to examine how novel, three-dimensional shapes are represented in long-term memory and how this might be differentially affected by monocular and binocular viewing. Three experiments were conducted. The first experiment established that slide projections of the novel objects could be recognized readily if seen in the same orientation as seen during learning. The second and third experiments examined generalization to novel depth rotations of the objects. The second experiment used slide projections of the objects. The results indicated that the representation of the objects seen during training was quite viewpoint-specific as recognition of objects in novel orientations was relatively poor. In the third experiment subjects were shown the real objects under monocular or binocular viewing. Overall, the results are consistent with a growing body of recent research showing that, at least under certain conditions, the visual system stores viewpoint-specific representations of objects.

Pattern and space perception in young infants.

Research is reviewed which reveals the surprisingly advanced perceptual skills of very young infants and some changes in these capacities which occur early in life; possible mechanisms which may underlie these changes are discussed. Newborns readily turn toward visual, auditory, and tactual stimulation, indicating that primitive localization systems operate at birth. However, their pattern perception appears to be more limited, with the notable exception of certain facial configurations which may have a privileged status. During the period from 1 to 3 months of life, auditory localization responses decrease substantially from neonatal levels while interest in visual patterns increases; indeed, during this period infants seem to become 'captured' by visual stimuli. By 4 months of age, infants turn rapidly and accurately towards off-centered sounds again, as they begin to reach for visible and invisible sounding objects. Between 3 and 4 months of age, they become sensitive to various types of static pattern regularities such as symmetry and other global configurational properties, and to dynamic aspects of faces (e.g. changes in facial expressions). Major structural maturation of the visual cortex at this age may underlie these new levels of auditory-visual spatial integration and pattern analysis abilities.