Perception

The true receptive field of more than 90% of neurons in the middle temporal visual area (MT) extends well beyond the classical receptive field (crf), as mapped with conventional bar or spot stimuli, and includes a surrounding region that is 50 to 100 times the area of the crf. These extensive surrounds are demonstrated by simultaneously stimulating the crf and the surround with moving stimuli. The surrounds commonly have directional and velocity-selective influences that are antagonistic to the response from the crf. The crfs of MT neurons are organized in a topographic representation of the visual field. Thus MT neurons are embedded in an orderly visuotopic array, but are capable of integrating local stimulus conditions within a global context. The extensive surrounds of MT neurons may be involved in figure–ground discrimination, preattentive vision, perceptual constancies, and depth perception through motion cues.

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

Is motion perception based on a local piecemeal analysis of the image or do ‘global’ effects also play an important role? Use was made of bistable apparent-motion displays in trying to answer this question. Two spots were flashed simultaneously on diagonally opposite corners of a 1 deg wide square and then switched off and replaced by two spots appearing on the other two corners. One can either see vertical or horizontal oscillation and the display is bistable just as a Necker cube is. If several such bistable figures are randomly scattered on the screen and presented simultaneously, then one usually sees the same motion axis in all of them, suggesting the presence of field-like effects for resolving ambiguity in apparent motion.

While viewing a single figure observers experience hysteresis: they tend to adhere to one motion axis or the other and can switch the axis only by looking away and looking back after 10–30 s have elapsed. The figure can be switched off and made to reappear at some other random location on the screen and it is then always found to retain its motion axis. Several such demonstrations are presented to show that spatial induction effects in metastable motion displays may provide a particularly valuable probe for studying ‘laws’ of perceptual organization.

A small square and a large triangle below it were presented in the first frame. These were switched off and replaced by a triangle alone in the second frame, shifted horizontally and upwards. The triangle appeared to move obliquely, as expected, but most observers also saw the square moving horizontally and hiding behind the triangle, although there was no stimulus corresponding to it in the second frame. The visual system invokes the occlusion ‘hypothesis’ in order to explain the otherwise mysterious disappearance of the square. The experiment suggests that apparently intelligent solutions can be rapidly computed by the visual system.

Two fields of random dots that were identical except for a slight shift in a central square region were presented in rapid alternation. This produced a vivid impression of a square moving back and forth above the background. When the kinematogram is presented in equiluminous red/green, the motion of the central region can still be seen, although over a narrower range of alternation rates, interstimulus intervals, and displacements than for black/white presentation. The perception of motion for equiluminous stimuli indicates that colour and motion can be analyzed conjointly by the visual system. However, as originally reported by Ramachandran and Gregory, the segregation of the oscillating central square from the background is lost at equiluminance. This segregation process therefore appears to be colour-blind.

A novel display is described which stimulates both the long-range and the short-range motion detecting processes simultaneously, but with opposing directions of movement. The direction in which the stimulus appears to move depends on retinal eccentricity and element size, but adaptation to the display always produces a motion aftereffect (MAE) direction opposite to the direction of the short-range component. The display may offer insights into the properties of the two-process motion detecting system.

A study is reported of the role of luminance and contrast in resolving ambiguous apparent motion (AM). Different results were obtained for the short-range (SR) and the long-range (LR) motion-detecting processes. For short-range jumps (7.5 min arc), the direction of ambiguous AM depended on brightness polarity, with AM only from white to white and from black to black. But for larger jumps, or when an interstimulus interval (ISI) was introduced, AM was less dependent on polarity, with white often jumping to black and black jumping to white.

Two potential AMs were pitted against each other, one carried by a light stimulus and the other by a dark stimulus. The stimulus whose luminance differed most from the uniform surround captured the AM. Visual response to luminance was linear, not logarithmic. When the stimulus was modified to give continuous AM in one direction it was followed by a negative aftereffect of motion only when the spatial displacement was 1 min arc. A larger displacement (10 min arc) gave good AM but no motion aftereffect. Thus only short-range motion adapts motion-sensitive channels.

A study is reported of the perception of random-dot two-frame apparent motion in which the durations of each exposure and the interstimulus interval between them were varied. The results are largely consistent with the rule that, for optimal motion detection, a portion of each exposure must fall within the same time interval of about 40 ms. In addition, motion perception is separably dependent on the displacement from one exposure to the next and on the time interval between those exposures, rather than on the ‘velocity implied by their ratio.

Null measurements given by cancelling forces, voltages, or whatever, are used in physics for gaining ‘objectivity’—by avoiding ‘subjective’ perceptions; but, somewhat paradoxically, null methods can be useful for studying perception itself. Here we consider cancelling opposed movements for photometry with coloured lights, and some recent experiments, carried out with John Harris, on nulling ‘real’ against opposed ‘apparent’ motion for teasing out some neural movement channels.

Not only translations and rotations, but also intriguing ‘plastic deformations’ are observed in apparent motion. What kinds of invariants does the visual system depend on during these transformations to determine that two figures of different shapes nevertheless represent the same object? Experiments are reported in which seven pairs of stimuli with topological differences were used. The evidence suggests that topological invariants may be used in the perception of apparent motion. In spite of variations in other factors, such as brightness, spatial frequency, terminators, etc, subjects displayed a strong preference for motion from a central figure to a figure with the same topological invariants. The results emphasize the importance of topological structure in figure perception.

The detection of spatiotemporal correlation in visual displays has been studied with stroboscopically presented random-noise patterns and with a signal-to-noise ratio paradigm in which the moving pattern was masked with spatiotemporal white noise. These methods reveal the ability of the visual system to detect correlation of spatiotemporal structures, rather than luminance contrast. The effects of stroboscopic rate, exposure duration, target size, and the extent of discrete spatial shifts were studied in both the central and the peripheral visual field. Evidence for orientation-selective and speed-selective mechanisms was found, as well as for extensive spatiotemporal integration. Bounds on parameters of spatial and temporal correlation and integration were obtained. The results are similar to those reported earlier, and also extend them. Their relation to results obtained through other paradigms (eg the motion aftereffect) is explored.

The sensitivity of the visual system to motion of differentially moving random dots was measured. Two kinds of one-dimensional motion were compared: standing-wave patterns where dot movement amplitude varied as a sinusoidal function of position along the axis of dot movement (longitudinal or compressional waves) and patterns of motion where dot movement amplitude varied as a sinusoidal function orthogonal to the axis of motion (transverse or shearing waves). Spatial frequency, temporal frequency, and orientation of the motion were varied. The major finding was a much larger threshold rise for shear than for compression when motion spatial frequency increased beyond 1 cycle deg⁻¹. Control experiments ruled out the extraneous cues of local luminance or local dot density. No conspicuous low spatial-frequency rise in thresholds for any type of differential motion was seen at the lowest spatial frequencies tested, and no difference was seen between horizontal and vertical motion. The results suggest that at the motion threshold spatial integration is greatest in a direction orthogonal to the direction of motion, a view consistent with elongated receptive fields most sensitive to motion orthogonal to their major axis.