Perception

In order to secure useful information the perceptual system must combine information present in the retinal counterpart of the to-be-discriminated distal variable with information about other variables that affect the state of the retinal counterpart. This combinatorial process may be called—following Woodworth's lead—the process of ‘taking-into-account’.

Seven applications of the ‘taking-into-account’ hypothesis are described. Salient features of the hypothesis are examined and the empirical status of the hypothesis is summarized and evaluated. Finally a model of the information-processing sequence suggested by the hypothesis is presented.

Kinaesthesis, the sensing of body movement, which is essential for controlling activity, depends on registering the changes which accompany body movement. While there are two basic types of change—mechanical (articular, cutaneous, and vestibular) and visual—and so two potential sources of kinaesthetic information, the mechanical changes have traditionally been considered the basis of kinaesthesis, vision being considered a purely exteroceptive sense. J.J. Gibson, on the other hand, has argued that vision is a powerful kinaesthetic sense. To test this idea visual–mechanical kinaesthetic conflicts were created by moving the visible surroundings linearly forward and backward around a passively or actively moving subject. In most cases vision dominated. Therefore vision is not a purely exteroceptive sense, nor is visual kinaesthesis simply an adjunct to mechanical kinaesthesis. Vision is an autonomous kinaesthetic sense.

With the growing realisation by psychologists of the need to take account of response biases in perception and memory, the lack of a simple and realistic model to enable ‘sensitivity’ and ‘bias’ to be measured separately in complex tasks is becoming acute. Though signal-detection theory (TSD) has proved very useful in two-alternative tasks, it is virtually intractable in multiple-choice tasks; for this reason experimenters have tended to resort to Luce's Choice Model (LCM) as an approximation to it. The relationship between TSD and LCM, however, though close in the two-alternative case, becomes increasingly tenuous as the number of alternatives is increased. Nevertheless the substitution may still be made, provided certain cautions are observed: computer simulations have been used here to study both forced- and unforced-choice tasks, and establish guidelines for experimenters who wish to apply a two-factor analysis to data from complex tasks.

Infants aged 5–12 weeks were shown a silent colour film whose clarity/focus was contingent on their sucking on a dummy nipple. In the ‘suck-for-clear’ condition the mean rate of sucking increased significantly over baseline level, and decreased when the contingency shifted to ‘suck-for-blur’. When the initial condition was suck-for-blur, sucking rate remained close to baseline level (even after the shift to suck-for-clear). Time spent looking at the clear film increased in both conditions, but there was little change in looking at the blurred pictures. With the introduction of the contingency conditions patterns of looking at the clearing and cleared pictures changed, and looking at the cleared picture increased in the suck-for-clear but not the suck-for-blur condition. Asymmetry of the results indicates that infants are better able to use an active response for instrumental means than to inhibit a response to achieve instrumental control. Implications for the development of voluntary control of action are explored.

Infants 6 to 16 weeks old exhibited shape constancy for plain solid objects. Habituation of visual fixation occurred with respect to repeated viewing of a cube in the same orientation and of a cube in different orientations, but not when viewing photographs of a cube in different orientations. Since the photographs reproduced the outline shape and the luminance patterns of the solid object, it is suggested that motion parallax mediates the shape constancy. Infants responded to the photographs as if the latter were patterns rather than as if they were representations of solid objects.

Principles of the perception of a special kind of partially rigid motion—bending motion—are explored. The proximal stimuli consisted of an outline quadrangle with various combinations of changes of length and direction of the four sides. The result demonstrated a perceptual preference for rotary motion over bending motion, and of bending motion over two-dimensional form change (stretching). The principle of minimum object change can predict this result, implying that rotation, bending, and stretching are perceptually preferred in this order.

The site of adaptation resulting from exposure to spatially discordant visual and auditory cues was determined by comparing, for both sensory modalities, shifts in pointings with shifts in estimations of ‘straight ahead’. The exposure task consisted of monitoring a synchronous series of flashes and noises separated 15° laterally by prisms. The shifts it produced for both modalities were of nearly equal magnitude in the two tests.

The results suggest that auditory–visual conflict is resolved through recalibrations at the level of sound–head and of eye–head articulations.

Experimentation was conducted to compare tactile pattern recognition on three body loci—back, abdomen, and inner thigh. Five blind subjects, experienced users of a tactile vision substitution system which displays images picked up by a television camera through patterns of vibratory stimulation on the skin, were asked to identify letter images displayed on three body regions. Though the previous experience had been limited to stimulation on the back, recognition accuracy and latency were significantly better on the abdomen. The results indicate that immediate transfer of previous learning occurs when body locus of stimulation is shifted, and further suggest that neural networks for tactile pattern recognition do not include specific peripheral neural receptors.

A subject wore for six days a microphone on each hand, connected to stereo headphones. This effectively placed his ears on his hands. Hand movements, with eyes closed, produced apparent movements of sound sources, and crossing the hands over appeared to reverse the sound field. No perceptual adaptation to this auditory rearrangement was found.

The effect on dynamic visual acuity (DVA) of target angular velocity ω and target exposure time t_e was examined. The correlation between DVA and static visual acuity (SVA), the ability to discriminate small spatial separations in a stationary target, was also determined.

DVA was measured for pairings of four values of ω (25, 30, 40, and 45 deg s⁻¹) with four values of t_e (0·4, 0·6, 0·8, and 1·0 s). The twenty-four subjects viewed Landolt-ring targets that were presented as slide projections on a rear-projection screen. The average thresholds for each condition were determined and compared. SVA was measured and correlated with the DVA scores for each condition.

The results indicated that: (i) DVA deteriorated for increased ω or decreased t, with the latter having the most significant effect, (ii) DVA and SVA were not significantly correlated, (iii) the experimental apparatus and procedure were highly reliable, and (iv) DVA improvement with testing was significant for some conditions.

When a narrow slit moves over a line figure (or the equivalent stimulus is simulated by a short line segment appropriately displacing within a moving slit), observers often perceive an extended figure although the stimulus information is ambiguous. In several experiments it is shown that the perception of a figure tends to occur, provided the stimulus information is compatible with the perceptual ‘solution’ of a figure revealed by a moving aperture: the visible segment of the figure must completely fill the aperture, the surround of the aperture must appear to be opaque and extend an adequate distance on both sides, the aperture must be perceived as an opening rather than as a figure, and, if the slope of the segment is visible, it must change appropriately from moment to moment. Thus the outcome can be thought of as an intelligent, elegant solution to the problem posed by the transforming proximal stimulus.

Inexperienced observers show a delay before experiencing the stereoscopic percept from a random-dot stereo pair. This perception time is progressively reduced with repeated exposures of the stereogram. We have investigated the specificity of this perceptual learning effect, using stereograms made up of short oblique line elements. Learning with a stereogram consisting of 45° line elements transferred completely to an uncorrelated pattern with the same element orientation, but there was a marked failure of transfer to a pattern whose elements had the opposed oblique orientation. Thus the stereoscopic skill that has been acquired may be specific to those orientation analysers that were stimulated during the training period.

The somatosensory evoked response to vibrotactile stimuli was studied in 20 subjects. This type of response to stimuli of 1000 ms duration displays ‘on’ and ‘off’ components. The latter consists of a surface negative deflection of about 100–120 ms peak latency, followed by a slower positive wave. As in the averaged visual evoked response an afteractivity of frequency 8–10 Hz may follow the ‘on’ component. This activity may either mask the ‘off’ component, be initiated by the termination of the stimulus, or its rhythm can be interrupted by the stimulus end.

In view of the similarity between the averaged evoked response to visual and vibrotactile stimuli, it is suggested that visual and somatosensory information may be processed according to a common pattern.