Perception

Many simple cells of the visual cortex have long narrow receptive fields, which are strongly excited by lines oriented along their long axis. In the present psychophysical analysis, we assume that there are similar independent ‘subunits’ which contribute, by probability summation, to the detection of a line. If a line segment is shorter than the subunit length, then extending the line length will increase the sensitivity of all the subunits affected by the line, and a relatively large increase in visual sensitivity will occur, corresponding to this ‘physiological summation’ within subunits. However, for a line segment which is much longer than the subunit length, the main effect of extending line length is to stimulate more subunits, resulting in a relatively small increase in sensitivity owing to probability summation. Thus a study of sensitivity (reciprocal threshold) as a function of line length may be used to test the subunit model and to estimate the subunit length. Here we use a quantitative model to demonstrate that sensitivity/line length data may be well fitted, assuming independent subunits having a constant length of 8·6′—in good agreement with the length of Andrews' ‘ff’ units.

Changes in the physical rate of a clicking or fluttering sound caused changes in the rate at which a simultaneously viewed light appears to flicker, even though the physical flicker rate remains constant. Perceived flicker rate increases in response to a rate of change of flicker frequency, and this auditory ‘driving’ does not depend on whether the auditory and visual sources have the same location. Visual evoked potentials do not correlate with ‘driving’. Thus, the effect of auditory flutter upon perceived visual frequency is not due to the properties of Morrell-type bimodal neurons, nor does it reflect the activities of neurons responsible for maintaining correspondence between perceived auditory space and perceived visual space. The effect is possibly due to modification of a subjective criterion rather than, as previously suggested, to the entrainment or time-locking of physiological signals in the visual pathway.

Reaching behavior in eighteen- to thirty-two-week-old infants was studied as a function of binocular convergence. The infant looked at the object to be reached for through prism arrangements which changed convergence only. The reaches obtained were nearly always directed at the virtual object defined by convergence. Corrections of the reaches, if any, were made rather late and often not before the hand arrived at the place of the virtual object.

Prism adaptation as measured by negative aftereffects (NA), proprioceptive shifts (PS), and visual shifts (VS) was assessed as a function of amount of exposure time and target specificity, whether an exposure and a test target background were the same or different, to determine the validity of Wilkinson's linear model (NA = PS+ VS). With few exceptions the model was found to hold well up to 40 min of prism viewing regardless of type of exposure background. In addition target specificity affected magnitude of the NA component of adaptation but not the PS and the VS components.

The contribution of afterimages to spatial-frequency adaptation was studied by comparing a number of different fixation paradigms designed to maximize or minimize afterimages. While it is clear that adaptation is not an afterimage artifact, nevertheless afterimages are produced at low spatial frequencies and can considerably distort the results of adaptation experiments unless steps are taken to eliminate them.

The existence of a tilted binocular disparity space was experimentally established. The results of the experiment showed that a texture surface which under binocular viewing appears to be oriented vertically and perpendicular to the line of sight in fact is tilted such that the portions of the texture surface above and below fixation incline toward and away from the observer, respectively. Under monocular viewing such a discrepancy between apparent and objective vertical tilt is not observed. This suggests that under binocular viewing the cortical representations of the upper and lower visual hemifields are biased toward uncrossed and crossed disparities, respectively. The tilted binocular disparity space is explained in terms of an experientially induced disparity gradient effected during development of the binocular system.

Contrast thresholds were measured for vertical and oblique grating patterns. As shown previously, at higher spatial frequencies sensitivity for vertical is much greater than that for oblique. Present results show that this difference in sensitivity is found only with low rates of temporal modulation; with higher temporal frequencies this orientation difference disappears. Moreover, when contrast thresholds are based on the perception of flicker, vertical and oblique sensitivities are essentially identical even at low flicker rates. These results indicate that the so-called ‘oblique effect’ is confined to mechanisms with poor temporal resolving power, probably the sustained channels which have been studied psychophysically and neurophysiologically by others.

A technique is described for measuring visual preferences in cats. The method involves a self-presentation paradigm, but provides strict control of the visual field and orientation of contours in the field. Two procedures are presented: a single-choice design and a multiple-choice design, both of which use duration of self-produced exposure as the dependent measure. Although both procedures may serve as indices of preference order, the multiple-choice method is preferable for measuring preference strength. The application of the technique as a behavioural index of the effects of early environmental manipulation is suggested. Preliminary data from two normal cats indicate that, although homogeneous illumination is reinforcing, the presence of features, especially oriented contours, appears to increase the ‘interest’ value of the presentation.

A behavioural method is described for the measurement of various visual spatial acuities in kittens as young as thirty days of age. Examples are given of applications of the technique to measurement of the visual acuity for gratings in normal kittens as well as to studies of the time course of behavioural recovery from the effects of early monocular visual deprivation.

The use of the term ‘global’ in the context of stereopsis is discussed. It is concluded that different meanings of this term need to be kept carefully distinguished at all times. The discussion centres around a series of demonstrations introduced by Ramachandran and Nelson, and interpretations are offered for these demonstrations in terms of spatial-frequency-tuned stereopsis channels.

McGonigle and Jones take exception to Dodwell's explanation of anomalous transfer (AT) as the outcome of relational discriminations among a set of stimulus patterns which vary along a single dimension of orientational salience, from ‘horizontal’ to ‘vertical’. In particular they do not think that the continuum is generated by units with Hubel-and-Wiesel type retinal receptive fields. Instead, they invoke Garner's notions about stimulus structure to explain Dodwell's results, as well as their own finding of two situations where AT fails to occur. It seems that McGonigle and Jones missed the point of the relational discrimination explanation of AT. In fact, it is shown that AT is not predicted by this model for the conditions in which they failed to obtain it.

The relational model makes definite predictions about conditions under which AT will occur; as this is not true of their invocation of Garner's ideas, the former is to be preferred. Whether or not outputs are coded by Hubel-and-Wiesel type units is not particularly relevant to the main point, that AT is a result of relational learning. In fact, the model is an instance of how structure, in Garner's sense, can be generated.

We repudiate Dodwell's claim that his relational model is predictively more accurate than our Garner-based treatment of AT. We claim instead, that his case is based on a fundamental misunderstanding of the essentials of Garner's complex thesis, together with a misreading of our own report. Rather than accept his notion of a single continuum of salience as providing for the genesis of structure, we argue instead that his ‘single continuum’ compounds and confounds at least two continua which may be understood only where the criteria of judgement are made explicit.

It is demonstrated by the use of drawings: (i) that two-point perspective drawings based on projective geometry possess systematic distortions, and (ii) that these distortions can be eliminated in drawings based on hyperbolic geometry.

A visual illusion consisting of transparent halos extending beyond the boundaries of rotating discs is reported. The effect can be obtained by rotating a variety of black-and-white discs at moderate speeds. It is not due solely to rods, as opposed to cones, and does not appear to be explainable in terms of intermittent stimulation of portions of visual fields of fixed visual angle.

A natural biological formation presents triangles forming Stars of David. Elements are readily grouped into first one, then another star.

In a paper recently published in this journal, Meyer criticised our study on relationships between channels for colour and spatial frequency for not being able to demonstrate a size aftereffect not specific to colour, a McCollough effect not specific to size, or the functions of cortical colour mechanisms. In fact, our study attempted none of these demonstrations in the sense suggested by Meyer because the first would have been impossible for empirical reasons, the second for conceptual reasons, and the third for methodological reasons. Instead, our study yielded evidence that at least three different types of perceptual channel underlie our capacity to perceive the size and colour of objects.