Risto Näsänen

Cortical magnification factor predicts the photopic contrast sensitivity of peripheral vision

THE area of visual cortex devoted to the analysis of a con-slant-size region in the visual field diminishes progressively for more peripheral locations. The change is described quantitatively by the cortical magnification factor, which indicates the linear extent of cortex in mm corresponding to one degree of visual angle at various eccentricities (angular distances from the middle fovea). The human cortical magnification factor has been estimated by Cowey and Rolls1 from the data of Brindley and Lewin2, who mapped the phosphenes (sensations of light) caused in the lower nasal visual field by electrical stimulation of the human visual cortex. Building on these results, we have studied the spatial contrast sensitivity functions in man at various eccentricities. We used two methods: in one the retinal image sizes of the test gratings were kept similar at different eccentricities and in the other, the calculated sizes of cortical projections of the test gratings were made similar at different locations. Our results indicate that peripheral contrast sensitivity and acuity are inferior to foveal performance, because of the diminished cortical projection area.

Spatial frequency bandwidth used in the recognition of facial images

The purpose of the study was to find out what spatial frequency information human observers use in the recognition of face images. Signal-to-noise ratio thresholds for the recognition of facial images were measured as a function of the centre spatial frequency of narrow-band additive spatial noise. The relative sensitivity of recognition to different spatial frequencies was derived from these results. The maximum sensitivity was found at 8-13 c/face width and the bandwidth was just under two octaves. Qualitatively similar results were obtained with stimuli in which Fourier phase was randomised within a narrow band of different centre spatial frequencies. This resulted in a considerable increase of energy threshold around 8 c/face width and less elsewhere. Further, contrast energy thresholds were measured as a function of the centre spatial frequency of band-pass filtered face images. As a function of object spatial frequency (c/face width), energy threshold first decreased and then increased. The lowest energy thresholds found around 10 c/face width were lower than the energy threshold for unfiltered images. This is what one would expect if face recognition is narrow-band, since band-pass filtered images of optimal centre spatial frequency do not contain unused contrast energy at low and high spatial frequencies. In conclusion, the results suggest that the recognition of facial images is tuned to a relatively narrow band (< 2 octaves) of mid object spatial frequencies.

NEW VISUAL ACUITY TEST FOR PRE-SCHOOL CHILDREN

A new test chart was developed for the measurement of visual acuity of pre‐school children. The symbols of the test are circle, square, apple and house. These were so designed that each symbol measures visual acuity similarly. This feature of the test was verified experimentally. The visual acuity values measured by the individual symbols correlated highly with the visual acuity values measured with the whole test (0.82–0.86). The correlation between the visual acuity values measured repeatedly, the reliability of the new test, was found to be 0.94 for adult subjects. The new visual acuity test thus fulfils the statistical criteria of a good visual acuity test. Because both children and nurses seem to like the new test, it may be useful in the assessment of visual acuity in pre‐school children.

Cortical magnification and peripheral vision

In a generalized form, the cortical magnification theory of peripheral vision predicts that the thresholds of any visual stimuli are similar across the whole visual field if the cortical stimulus representations calculated by means of the cortical magnification factor are similar independently of eccentricity. Failures of the theory in spatial vision were analyzed, and the theory was tested with five visual acuity tasks and two hyperacuity tasks. Almost all increases in thresholds with eccentricity were explained by the theory in five of these tasks, which included the two-dot vernier hyperacuity test, the measurement of visual acuities with gratings, the Snellen E test, and two acuity tests that required either separation between dots or discrimination between two mirror-symmetric forms. The two-dot vernier thresholds could be explained as a special case of orientation discrimination, and orientation discrimination at different eccentricities was in agreement with the cortical magnification theory. The increase of thresholds in peripheral vision was larger than predicted by the theory in the Landolt visual acuity and bisection hyperacuity tests, possibly because of retinal undersampling.

Temporal contrast sensitivity and cortical magnification

We measured temporal and spatial contrast sensitivity functions of foveal and peripheral photopic vision at various locations in the nasal visual field. Sensitivity decreased monotonically with increasing eccentricity when it was measured by using the same test gratings at different eccentricities. When the gratings were normalized in area, spatial frequency, and translation velocity by means of the cortical magnification factor M so that the calculated cortical representations of the gratings became equivalent at different eccentricities, the temporal contrast sensitivity functions became similar at all eccentricities. The normalization was effective under all experimental conditions that included various kinds of temporal modulation from 0 to 25 Hz (movement, counterphase flicker and on-off flicker) and different threshold tasks (detection, orientation discrimination, and discrimination of movement direction), independently of the subjective appearances of the gratings at threshold. We conclude that central and peripheral vision are qualitatively similar in spatiotemporal visual performance. The quantitative differences observed without normalization seem to be caused by the spatial sampling properties of retinal ganglion cells that are directly related to the values of M used in the normalization.

Modelling the dependence of contrast sensitivity on grating area and spatial frequency

We modelled the human foveal visual system in a detection task as a simple image processor comprising (i) low-pass filtering due to the optical transfer function of the eye, (ii) high-pass filtering of neural origin, (iii) addition of internal neural noise, and (iv) detection by a local matched filter. Its detection efficiency for gratings was constant up to a critical area but then decreased with increasing area. To test the model we measured Michelson contrast sensitivity as a function of grating area at spatial frequencies of 0.125-32 c/deg for simple vertical and circular cosine gratings. In circular gratings luminance was sinusoidally modulated as a function of the radius of the grating field. In agreement with the model, contrast sensitivity at all spatial frequencies increased in proportion to the square-root of grating area at small areas. When grating area exceeded critical area, the increase saturated and contrast sensitivity became independent of area at large grating areas. Spatial integration thus obeyed Pipers law at small grating areas. The critical area of spatial integration, marking the cessation of Pipers law, was constant in solid degrees at low spatial frequencies but inversely proportional to spatial frequency squared at medium and high spatial frequencies. At low spatial frequencies the maximum contrast sensitivity obtainable by spatial integration increased in proportion to spatial frequency but at high spatial frequencies it decreased in proportion to the cube of the increasing spatial frequency. The increase was due to high-pass filtering of neural origin (lateral inhibition) and the decrease was mainly due to the optical transfer function of the eye. Our model explained 95% of the total variance of the contrast sensitivity data.

Modelling contrast sensitivity as a function of retinal illuminance and grating area

We extended the contrast detection model of human vision [Rovamo, Luntinen & Näsänen (1993b) Vision Research, 33, 2773-2788] to low light levels by taking into account the effect of light-dependent quantal noise. The extended model comprises (i) low-pass filtering due to the optical modulation transfer function of the eye, (ii) addition of light-dependent noise at the event of quantal absorption, (iii) high-pass filtering of neural origin (lateral inhibition), (iv) addition of internal neural noise, and (v) detection by a local matched filter whose efficiency decreases with increasing grating area. To test the model we measured foveal contrast sensitivity as a function of retinal illuminance and grating area at spatial frequencies of 0.125-32 c/deg. In agreement with the model, monocular contrast sensitivity at all grating areas increased in proportion to I when retinal illuminance (I) was smaller than critical illuminance. Thereafter the increase saturated and contrast sensitivity became independent of retinal illuminance. Similarly, at all levels of retinal illuminance contrast sensitivity increased in proportion to A when grating area (A) was smaller than critical area. Thereafter the increase saturated and contrast sensitivity became independent of area. Critical level of retinal illuminance increased in proportion to the spatial frequency squared. Critical area marking the saturation of spatial integration was constant at low spatial frequencies but decreased in inverse proportion to spatial frequency squared at medium and high spatial frequencies. The maximum contrast sensitivity obtainable by spatial integration in bright light increased at low spatial frequencies in proportion to spatial frequency, was constant at medium spatial frequencies, and decreased in inverse proportion to spatial frequency cubed at high spatial frequencies. The increase was due to the neural modulation transfer function of the visual pathways whereas the decrease was due to the optical modulation transfer function of the eye. The model explained 91-99% of the total variance of our contrast sensitivity data at various spatial frequencies.

Training-induced cortical representation of a hemianopic hemifield

Background: Patients with homonymous hemianopia often have some residual sensitivity for visual stimuli in their blind hemifield. Previous imaging studies suggest an important role for extrastriate cortical areas in such residual vision, but results of training to improve vision in patients with hemianopia are conflicting. Objective: To show that intensive training with flicker stimulation in the chronic stage of stroke can reorganise visual cortices of an adult patient. Methods: A 61-year-old patient with homonymous hemianopia was trained with flicker stimulation, starting 22 months after stroke. Changes in functioning during training were documented with magnetoencephalography, and the cortical organisation after training was examined with functional magnetic resonance imaging (fMRI). Results: Both imaging methods showed that, after training, visual information from both hemifields was processed mainly in the intact hemisphere. The fMRI mapping results showed the representations of both the blind and the normal hemifield in the same set of cortical areas in the intact hemisphere, more specifically in the visual motion-sensitive area V5, in a region around the superior temporal sulcus and in retinotopic visual areas V1 (primary visual cortex), V2, V3 and V3a. Conclusions: Intensive training of a blind hemifield can induce cortical reorganisation in an adult patient, and this case shows an ipsilateral representation of the trained visual hemifield in several cortical areas, including the primary visual cortex.

Effect of stimulus contrast on performance and eye movements in visual search

According to the visual span control hypothesis, eye movements are controlled in relation to the size of visual span. In reading, the decrease of contrast reduces visual span, saccade sizes, and reading speed. The purpose of the present study is to determine how stimulus contrast affects the speed of two-dimensional visual search and how changes in eye movements and visual span could explain changes in performance. The task of the observer was to search for, and identify, an uppercase letter from a rectangular array of characters in which the other items were numerals. Threshold search time, i.e. the duration of stimulus presentation required for search that is successful with a given probability, was determined by using a multiple-alternative staircase method. Eye movements were recorded simultaneously by using a video eye tracker. Four different set sizes (the sizes of stimulus array) (3x3-10x10), and five different contrasts (0.0186-0.412) were used. At all set sizes, threshold search time decreased with increasing contrast. Also the average number of fixations per search decreased with increasing contrast. At the smallest set size (3x3), only one fixation was needed except at the lowest contrast. Average fixation duration decreased and saccade amplitudes increased slightly with increasing contrast. The reduction of the number of fixations with increasing contrast suggests that visual span, i.e. the area from which information can be collected at one fixation, increases with increasing contrast. The reduction of the number of fixations together with reduced fixation duration result in reduced search times when contrast increases.

Temporal sensitivity in a hemianopic visual field can be improved by long-term training using flicker stimulation

Background: Blindness of a visual half-field (hemianopia) is a common symptom after postchiasmatic cerebral lesions. Although hemianopia severely limits activities of daily life, current clinical practice comprises no training of visual functions in the blind hemifield. Objective: To find out whether flicker sensitivity in the blind hemifield can be improved with intensive training, and whether training with flicker stimulation can evoke changes in cortical responsiveness. Methods: Two men with homonymous hemianopia participated in the experiments. They trained with flicker stimuli at 30° or with flickering letters at 10° eccentricity twice a week for a year, and continued training with more peripheral stimuli thereafter. Neuromagnetic responses were registered at 1–2-month intervals, and the Goldmann perimetry was recorded before, during and after training. Results: Flicker sensitivity in the blind hemifield improved to the level of the intact hemifield within 30° eccentricity in one participant and 20° eccentricity in the other. Flickering letters were recognised equally at 10° eccentricity in the blind and intact hemifields. Improvement spread from the stimulated horizontal meridian to the whole hemianopic field within 30°. Before training, neuromagnetic recordings showed no signal above the noise level in the hemianopic side. During training, evoked fields emerged in both participants. No changes were found in the Goldmann perimetry. Discussion: Results show that sensitivity to flicker could be fully restored in the stimulated region, that improvement in sensitivity spreads to the surrounding neuronal networks, and that, during training, accompanying changes occurred in the neuromagnetic fields.