Implicit Learning of Temporal Discriminations in Perception

We measured the temporal accuracy of signal transfer in the brain by means of periodic pulse stimuli in various sensory modalities using an adaptive threshold algorithm. Trains of supra-threshold signal pairs in 0° or 180° phase shifts appeared, and the subject indicated whether the signals of each pair in a train were simultaneous or not at various nominal frequencies of the pairs. The signals were spatially separate flashes of light, clicks, tactile pulses, or combinations thereof in intermodal comparisons. Temporal discrimination thresholds involving one signal in central photopic vision and the other in audition or tactile presentation to the finger tips did not exceed frequencies from 3 to 8 Hz, and in vision alone the average synchronism thresholds were about 10 Hz (SOA 0.05 s with 0.008 s pulses). The practice derived from 9 experimental sessions during 6 weeks improved temporal accuracy by factors ranging from 1.2 to 2.0 on the average in a sample of 33 naive subjects (university students), although no explicit feedback was given. The practice effect was lasting, for the average performance decrement was only 9% in 7 months. Thus, a considerable temporal modifiability must exist in the brain because a large learning effect was found in the simple temporal synchronism discrimination tasks.

Temporal Integration and Contrast Sensitivity in Foveal and Peripheral Vision

Spatial contrast sensitivity functions and temporal integration functions for gratings with dark surrounds were measured at various eccentricities in photopic vision. Contrast sensitivity decreased with increasing eccentricity at all exposure durations and spatial frequencies tested. The decrease was faster at high than at low spatial frequencies, but similar at different exposure durations. When cortically similar stimulus conditions were produced at different eccentricities by M-scaling, contrast sensitivity became independent of visual field location at all exposure durations tested. The results support the view that in photopic vision spatiotemporal information processing is qualitatively similar across the visual field, and that quantitative differences result from retinotopical differences in ganglion cell sampling. For gratings of constant retinal area temporal integration (improvement of contrast sensitivity with increasing exposure duration) was more extensive at high than at low retinal spatial frequencies but independent of cortical spatial frequency and eccentricity. For M-scaled gratings temporal integration was more extensive at high than at low cortical spatial frequencies but independent of retinal spatial frequency and eccentricity. The results suggest that the primary determinant of temporal integration is not spatial frequency but grating value that is calculated as AF2 square cycles (cycle2), where A is grating area and F spatial frequency.

Serum lipid fatty acids and temporal processing acuity in children with oral clefts

We investigated the relation between a biological factor (fatty acids, FA) and a cognitive processing speed factor (temporal processing acuity, TPA) that are both suggested to relate to neuronal and cognitive functioning. Blood samples of 49 ten-year-old children with oral clefts were collected for FA analysis in serum triglycerides, cholesteryl esters, and phospholipids on the same day as they performed behavioral TPA tasks (simultaneity/nonsimultaneity judgments) in several perceptual modalities (visual, auditory, tactile, audiotactile, visuotactile, and audiovisual). This population has larger than expected variation in the relevant cognitive measures (TPA, learning ability, and intelligence). Sequential regression analyses (adjusted for age, gender, and cleft type) showed that saturated FAs were not generally associated with TPA. Monounsaturated erucic and nervonic acids were inversely related with TPA. Of the polyunsaturated fatty acids, eicosapentaenoic and docosahexaenoic acids were positively associated with TPA, whereas gamma-linolenic acid was inversely related to TPA. In summary, we found significant relations between a biological (certain FAs) and a cognitive factor (TPA).

Serum lipid fatty acids, phonological processing, and reading in children with oral clefts

Reading skill is suggested to be related to phonological processing ability and polyunsaturated fatty acids (PUFAs). Here we investigated whether fatty acids (FAs) are related to phonological processing, whether the relations between PUFAs and reading generalize to other FAs, whether these relations are mediated by phonological processing, and whether relations of FAs are specific for language-related functions. Blood samples of 49 ten-year-old children with oral clefts were collected for FA proportion analysis in serum cholesteryl esters and phospholipids. On the same day, they performed tasks of phonological processing, reading, and both verbal and nonverbal intelligence. Sequential regression analyses (adjusted for age, gender, and cleft type) showed that phonological processing was inversely related to myristic acid in phospholipids and positively related to eicosapentaenoic acid in cholesteryl esters. Reading was inversely related to palmitoleic and gammalinolenic acids in phospholipids. The relations between FAs and reading were not mediated by phonological processing and FAs related only to language-related functions.

Temporal order and processing acuity of visual, auditory, and tactile perception in developmentally dyslexic young adults

We studied the temporal acuity of 16 developmentally dyslexic young adults in three perceptual modalities. The control group consisted of 16 age- and IQ-matched normal readers. Two methods were used. In the temporal order judgment (TOJ) method, the stimuli were spatially separate fingertip indentations in the tactile system, tone bursts of different pitches in audition, and light flashes in vision. Participants indicated which one of two stimuli appeared first. To test temporal processing acuity (TPA), the same 8-msec nonspeech stimuli were presented as two parallel sequences of three stimulus pulses. Participants indicated, without order judgments, whether the pulses of the two sequences were simultaneous or nonsimultaneous. The dyslexic readers were somewhat inferior to the normal readers in all six temporal acuity tasks on average. Thus, our results agreed with the existence of a pansensory temporal processing deficit associated with dyslexia in a language with shallow orthography (Finnish) and in well-educated adults. The dyslexic and normal readers' temporal acuities overlapped so much, however, that acuity deficits alone would not allow dyslexia diagnoses. It was irrelevant whether or not the acuity task required order judgments. The groups did not differ in the nontemporal aspects of our experiments. Correlations between temporal acuity and reading-related tasks suggested that temporal acuity is associated with phonological awareness.

Rate of information segregation in developmentally dyslexic children

Slowed processing of sequential perceptual information is related to developmental dyslexia. We investigated this unimodally and crossmodally in developmentally dyslexic children and controls ages 8-12 years. The participants judged whether two spatially separate trains of brief stimuli, presented at various stimulus onset asynchronies (SOA) in one or two senses, were synchronous or not. The stimulus trains consisted of light flashes in vision, clicks in audition, and indentations of the skin in the tactile sense. The dyslexic readers required longer SOAs than controls for successful performance in all six comparisons. The crossmodal spatiotemporal resolution of the groups differed more than unimodal performance. The dyslexic readers' segregation performance was also less differentiated than that of the controls. Our results show that not only sensory but also polysensory nonverbal information processing is temporally impaired in dyslexic children.

Mu rhythm modulation during changes of visual percepts

Cooperation between vision and somatomotor behavior, such as manual exploration of objects, suggests close functional coupling between the visual and sensorimotor systems. We observed this type of interaction in human volunteers during binocular rivalry while following the level of sensorimotor mu rhythm with a whole-scalp neuromagnetometer. The observers viewed a weak vertical grating in the lower visual field of one eye and a strong horizontal grating in the same spatial window of the other eye. When stationary, the weak grating was permanently invisible because of its low contrast and spatial frequency. A sudden brief drifting movement of the weak grating wiped out the dominant grating, and the weak grating became visible for less than the 3-s interval between the movements. The postcentral 8- to 15-Hz mu rhythm was found in six of nine observers, and its level increased transiently by 10-15%, starting about 450 ms after the beginning of the movement. The mu level was also enhanced by the actual disappearance of the stronger stimulus, when it occurred in random order with the rivalry stimuli. Identical visual motion, when not accompanied by a perceptual dominance change, produced only minor effects on the mu rhythm. Our results show that a change in visual percept, even with no real or imagined motor response, is associated with modified activity of the postcentral gyrus. This modification may reflect visuohaptic interactions and/or activity of the distributed cortical network implementing visually guided movements.

Stronger occipital cortical activation to lower than upper visual field stimuli. Neuromagnetic recordings

We recorded whole-scalp magnetoencephalographic (MEG) responses to black-and-white checkerboards to study whether the human cortical responses are quantitatively similar to stimulation of the lower and upper visual field at small, 0-6 degrees, eccentricities. All stimuli evoked strong occipital responses peaking at 50-100 ms (mean 75 ms). The activation was modeled with a single equivalent current dipole in the contralateral occipital cortex, close to the calcarine fissure, agreeing with an activation of the V1/V2 cortex. The dipole was, on average, twice as strong to lower than to upper field stimuli. Responses to hemifield stimuli that extended to both lower and upper fields resembled the responses to lower field stimuli in source current direction and strength. These results agree with psychophysical data, which indicate lower visual field advantage in complex visual processing. Parieto-occipital responses in the putative V6 complex were similar to lower and upper field stimuli.

A Method for Rehabilitating Vision after Stroke

We developed a method for rehabilitating eye movements and binocular fusion, and tested the method in one patient. An infarct of the pons caused the paresis of the lateral rectus muscle of the left eye. Beginning from the third week in hospital, the patient was trained in eye movements and binocular fusion. Fusion was made possible by means of prisms that moved the images of targets in central vision to the threshold of fusion in the primary eye position. During practice sessions lasting 0.5 to 2 h daily the patient kept the images fused by making a horizontal head movement when necessary. Several eye-movement sessions were held daily, consisting of voluntary saccades and fixations as far to the left as possible. The strength of the prismatic correction required for fusion decreased, and four months after the onset of stroke the patient could fuse without prisms in the primary position. His binocular vision became practically normal in one year. The plasticity of the visual system can be utilised in rehabilitation by a practice that uses minimal remedial means necessary for correct function at each level of performance.

Activation of human V5 complex and rolandic regions in association with moving visual stimuli

We recorded magnetoencephalographic responses from seven healthy humans during the presentation of stationary and rotating radial gratings. Rotations lasting 1 s evoked movement-specific sustained activity in the parieto-occipitotemporal border area, in agreement with the activation of the V5 complex specialized for the analysis of movement. The source areas of the movement-specific sustained fields were transiently active 100-130 ms after the onsets of both rotating and stationary stimuli, suggesting that movement-related cortical areas respond to any transient changes in the visual environment. Transients were evoked also in other brain areas 60-200 ms after onsets of both stimuli. Four subjects displayed additional motion-related sustained activity in the rolandic region. Sustained activity continued after the stimulus movement in several subjects during perception of the movement aftereffect. The transient activity may evoke visual attention while sustained activity of the V5 complex may be related to the conscious perception of movement.

Cortical magnification, scale invariance and visual ecology

The visual world of an organism can be idealized as a sphere. Locomotion towards the pole causes translation of retinal images that is proportional to the sine of eccentricity of each object. In order to estimate the human striate cortical magnification factor M, we assumed that the cortical translations, caused by retinal translations due to the locomotion, were independent of eccentricity. This estimate of M agrees with previous data on magnifications, visual thresholds and acuities across the visual field. It also results in scale invariance in which the resolution of objects anywhere in the visual field outside the fixated point is about the same for any viewing distance. Locomotion seems to be a possible determinant in the evolution of the visual system and the brain.

Illusory perception of gratings stimulating a small number of neurones

We studied pattern perceptions caused by drifting gratings presented monocularly in the nasal and temporal visual fields at various suprathreshold contrasts. The grating and its surround and background were matched in luminance. Small grating produced illusions and reduced perceptions. When grating area or contrast increased from a subthreshold value, the gratings were first seen as mere flashes. Then each grating was sometimes perceived as a single small bright spot or point. Next each grating was seen as a single dark or bright line. Finally the stimuli were perceived as gratings consisting of several bars. Orientation or direction of movement were perceived correctly, but velocity, colour and number of bars were often perceived as illusions. Thus, in spite of the illusions, some features of the stimuli could have allowed correct discriminations. The area and contrast limits of illusory perception depended on eccentricity. Irrespective of retinal size, the stimuli were not perceived correctly as gratings at any eccentricities when the gratings were smaller than about 1 x 1 mm in their calculated cortical area and stimulated a small constant number of retinal ganglion cells. Relations between the results and retinal aliasing, cortical columns and phase locking of neuronal oscillations are discussed.

Analysis of spatial structure in eccentric vision

The analysis of spatial structure, ie, the encoding of relative positions between pattern elements, was studied in central and eccentric vision. In a two-alternative forced-choice task the observer had to discriminate between two patterns consisting of short line segments. At each trial the two patterns were flashed for 140 msec and the observer indicated whether the patterns were identical or mirror symmetric. Psychometric functions were measured by changing pattern size at each eccentricity in order to find the threshold size allowing 75% of correct responses. The scaling factor, required for discriminating between mirror symmetric and identical patterns independent of eccentricity, was found to be similar to the size-scaling proposed by Levi et al (Vision Res 25:963, 1985) for vernier acuity tasks.

Mesopic spectral responses and the Purkinje shift of macaque lateral geniculate nucleus cells

Spectral responsiveness of different classes of macaque LGN cells at eccentricities smaller than 12 degrees was studied at low light intensities. Cone thresholds of cells varied from 1 to 10 td. Rod inputs were found in all classes of cells, including inhibitory inputs in some cells. Rod inputs were not apparent above 10-40 td, giving a total mesopic range of about 1-40 td. Strong rod-mediated responses could be evoked in broadband phasic cells and in spectrally opponent cells excited by short wavelengths. Only weak if any excitatory responses could be evoked by short wavelengths at scotopic levels in spectrally opponent long-wavelength excited cells. Hence, rod inputs do not confound the spectral responsiveness of cells because no spectrally opponent cell excited by long-wavelength stimuli at photopic levels became significantly responsive to short wavelength stimuli at mesopic or scotopic intensities. The so-called "rod color" may be blue. An increase in the dominance of wide-band cell responses that may explain the Bezold-Brücke hue shift was observed at higher stimulus radiances at wavelengths near 450 and 650 nm.

Light adaptation in cells of macaque lateral geniculate nucleus and its relation to human light adaptation

Responses of macaque lateral geniculate nucleus (LGN) cells to stimuli of different incremental intensities and wavelength compositions were studied at different levels of light adaptation from scotopic to low photopic levels. Stimuli were large in comparison with receptive-field size. Human increment thresholds were measured for comparison. The strength of responses grew in many cells from threshold up to a saturation level as a logarithmic function of incremental intensity. More complex intensity-response functions were also obtained, particularly from parvocellular layer (PCL) cells, but the shape and slope of the intensity-response function changed as a function of adaptation level only with chromatic backgrounds. As a function of adaptation level, the intensity-response functions shifted along the logarithmic abscissa but not sufficiently for a complete contrast constancy. Thus responses to any constant contrast became smaller when adaptation level decreased. The change from cone to rod responses, when possible, took place without noticeable change in shape of intensity-response functions, and much of the adaptive shift of the functions could be attributed to the change-over between rods and cones. Differences between different cells in light adaptation and dark-adapted sensitivity were large, mostly because of variation in the strength of rod input. The strongest excitatory rod inputs were found in PCL cells activated by short-wavelength light, so that the highest sensitivity at low levels of illumination occurred in blue- and blue-green-sensitive cells. The lowest increment thresholds based on cones matched the thresholds of macaque cone late receptor potentials recorded by Boynton and Whitten (3). They were also similar to human cone thresholds measured psychophysically but only for small stimulus sizes that may approximate the size of the receptive-field centers. Human sensitivity was much higher when measured with large stimulus sizes, indicating integration at post-geniculate neural levels. Light adaptation, as evaluated with respect to contrast constancy and Weber law behavior, was similarly incomplete in monkey single cells and human perception. A few cat LGN cells were studied in a control experiment; results agreed with previous findings. The light adaptation of cat cels was more complete and sensitivity higher than those observed under comparable conditions in macaque single cells and human. The maintained activity level of cells was little affected by the intensity of steady backgrounds. Thus, the steady-state hyper-polarisation of receptors was not transmitted to LGN cells.

Cell responses in dorsal layers of macaque lateral geniculate nucleus as a function of intensity and wavelength

We studied the relationship between light intensity and cell response to various wavelengths and wavelength combinations in the dorsal, parvocellular layers of the macaque lateral geniculate nucleus. When response is plotted as a function of the logarithm of stimulus intensity, the slope and shape of curves depends on wavelength. For wavelengths near the crossover point between excitatory and suppressive responses, nonmonotonic curves are common. Consequently, the form of spectral-response functions depends on stimulus intensity. Responses to combined stimuli made up of wavelengths close together near one spectral extreme are approximately additive. If one wavelength is near the crossover point, responses are nonadditive so that a midspectral wavelength, only producing a weak excitatory response, is able to occlude more vigorous responses to wavelengths near the spectral ends. Responses of parvocellular layer cells are consistent with their being a result of linear interaction of opponent cone mechanisms, the response of each of which follows a modified hyperbolic tangent function (22). Responses to all wavelength combinations, even those showing strikingly nonadditive effects, could be predicted from the additive opponent model described above.

Linear signal transmission from prepotentials to cells in the macaque lateral geniculate nucleus

Prepotentials preceding a neuronal action potential were recorded extracellularly in the lateral geniculate nucleus (LGN) of the macaque. Although prepotentials are found less frequently in the macaque than in the cat LGN, their electrical characteristics are similar, suggesting that they represent the arrival of impulses in a retinal afferent, as in the cat. The visual response properties of prepotentials and associated cells were similar under a variety of conditions, indicating that, apart from some response attenuation, little signal processing takes place in macaque LGN. A constant fraction of prepotentials above a threshold frequency gave rise to neuronal action potentials independent of the stimuli used, so that the frequency of cell action potentials was linearly related to the frequency of prepotentials. Since the maintained discharge rates of a cell and its prepotential always fell on the linear relation, the net responses of a cell and its prepotential to visual stimuli were approximately proportional to one another.

Resolution of gratings oriented along and across meridians in peripheral vision

Grating resolution was measured at various locations of the visual field for four grating orientations. As an instance of the oblique effect, vertical and horizontal gratings produced the highest resolution values in the central area. At eccentricities larger than about 20 deg, the oblique effect was replaced by a meridional resolution effect, in which resolution was systematically best for meridionally oriented grating bars and worst for grating bars perpendicular to the visual-field meridians. The origin of the effect seems to be neural because it was not caused by peripheral refractive errors or optical distortion of the peripheral retinal image.

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.

An estimation and application of the human cortical magnification factor

Comparisons of the published data on the density D of receptive fields of retinal ganglion cells and on the cortical magnification factor M indicated that M2 is directly proportional to D in primates. Therefore, the human M can be estimated for the principal meridians of the visual field from the density-distribution of retinal ganglion cells and from the density of the centralmost cones. Using the previously published empirical data, we estimated the values of the human M and express the values in four simple equations that can be used for finding the value of M for any location of the visual field. The monocular values of M are not radially symmetric. These analytically expressed values of M make it possible to predict contrast sensitivity and resolution for any location of the visual field. We measured contrast sensitivity functions at 25 different locations and found that the functions could be made similar by scaling the retinal dimensions of test gratings by the inverse values of M. Visual acuity and resolution could be predicted accurately for all retinal locations by means of a single constant multiplier of the estimated M. The results indicate that the functional and structural properties of the visual system are very closely and similarly related across the whole retina. Visual acuity, e.g., bears the same optimal relation to the density of sampling executed by retinal ganglion cells at all locations of the visual fields.

Visual resolution, contrast sensitivity, and the cortical magnification factor

This study shows that photopic contrast sensitivity and resolution can be predicted by means of simple functions derived by using the cortical magnification factor M as a scale factor of mapping from the visual field into the striate cortex. We measured the minimum contrast required for discriminating the direction of movement or orientation of sinusoidal gratings, or for detecting them in central and peripheral vision. No qualitative differences were found between central and peripheral vision, and almost all quantitative differences observed could be removed by means of a size compensation derived from M. The results indicated specifically that (1) visual patterns can be made equally visible if they are scaled so that their calculated cortical representations become equivalent; (2) contrast sensitivity follows the same power function of the cortical area stimulated by a grating at any eccentricity; (3) area and squared spatial frequency are reciprocally related as determinants of contrast sensitivity; and (4) acuity and resolution are directly proportional to M, and the minimum angle of resolution is directly proportional to M-1. The power law of spatial summation expressed in (2) and (3) suggests the existence of a central integrator that pools the activity of cortical neurons. This summation mechanism makes the number of potentially activated visual cells the most important determinant of visibility and contrast sensitivity. The functional homogeneity of image processing across the visual field observed here agrees with the assumed anatomical and physiological uniformity of the visual cortex.

[Inhibition and space-frequency characteristics of the complex receptive fields of the cat visual cortex]

Responses of complex receptive fields of the cat straitum to moving sinusoidal grating were studied. Stimulation of the receptive field with some spatial frequencies suppresses spontaneous discharges. Responses of the receptive field corroborate previously made predictions that the spatial--frequency characteristics of the receptive field should have the main and the secondary maximums and negative areas in case the complex fields perform piece-wise Fourier--transformation of image. The changes of impulse frequency in field's response are predicted by comparing the changes of instantaneous spectrum of grating entering the field with spatial frequency characteristic of the field. The data evidence that the complex field is rather a spatial--frequency filter than a detector. Some complex fields reveal a lateral inhibitory area behind the field's nucleus in direction of stimulus movement. The complex fields with no lateral inhibitory areas seem to serve for piece Fourier--description of image, those with lateral areas--for picking out the countour between textures.

Retinal mechanisms of visual adaptation and afterimages

Recent results obtained from recordings of isolated photoreceptor activity and from correlations of this activity with time-dependent changes in the responses of other retinal cells in several vertebrates have made a thorough revision of former theories of visual adaptation necessary. The present paper reviews the current state of research and relates the new discoveries with psychophysical findings in an attempt to explain human light and dark adaptation from the novel starting point. The former conceptions of adaptation have to be replaced with a three-level process consisting of photochemical receptor neural and network adaptation. Several adaptive mechanisms can be discerned at each level. Depending on adaptation conditions, any level of the three can play a dominating role and can also produce afterimages that display the behaviour of the mechanisms working at each level. The total achievement of visual adaptation is an optimized end product of the actions of all the various mechanisms.

Psycholphysical 'measurement' of cortical colour mechanisms: reply of Meyer

In a paper recently published in this journal, Meyer criticised our study on relationships between channels for colour and spatial frequence 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.

Dark adaptation and receptive field organisation of cells in the cat lateral geniculate nucleus

The receptive fields of LGN cells were investigated with stationary light and dark spot and annulus stimuli. Stimulus size and background intensity were varied while stimulus/background contrast was kept constant. The speed of dark adaptation vaired considerably from cell to cell. Dark adaptation made responses more sustained in all neurones and eliminated the oscillatory on-responses evoked under some conditions in the light-adapted cells. Dark adaptation led also to a disappearance of early phasic inhibition in on-responses, and increased response rise time and latency. The power of surround responses to inhibit centre responses decreased slightly at low levels of light adaptation in LGN cells but much less than in retinal ganglion cells. Some other traces of changing retinal surround effects also appeared inthe LGN on dark adaptation. For example, the functional size of receptive fields increased at low levels of illuminance as has been observed in retinal ganglion cells and the receptive fields as estimated from response peaks were larger than those estimated from sustained components.

Responses of cells in the cat lateral geniculate nucleus to moving stimuli at various levels of light and dark adaptation

The responses of neurones in laminae A and A1 of the cat lateral geniculate nucleus to moving stimuli were investigated at different background luminances. Moving bright slits, dark bars and edges were employed; the contrast of stimuli against the background was held constant. Background intensities varied from 10(-3) to 10(2) td. Responses as stimuli passed across the centres of LGN receptive fields became stronger with increasing levels of light adaptation up to 10(-1)-10(1) td and then remained constant. Responses as stimuli passed through surround regions altered qualitatively with adaptation level, generally increasing in strength and complexity with background luminance. As a bright slit for on-centre cells or dark bar for off-centre cells left the surround, in almost all units a strong secondary peak could be elicited by an appropriate selection of the adaptation conditions. Many features of the responses to moving stimuli could not be predicted from the responses to stationary stimuli under different adaptation conditions described in the previous paper.

Central inhibitory interactions in human vision

Contrast threshold and perceived orientation of a line segment were measured when another line segment was simultaneously presented either to the same or the other eye; the angle between the two line segments was varied. The presence of the masking line elevated the contrast threshold under both conditions and the threshold increased similarly both in monoptic and dichoptic masking when the masking angle was made smaller. The presence of the masking line affected also the perceived orientation of the test line. The effect was similar both in monoptic and dichoptic masking; the largest change of perceived orientation occurred at about 15 degrees masking angle and the change was smaller at other angles. The effect disappeared within a short distance when the masking line was removed farther from the test line. The similarity of the monoptic and dichoptic threshold elevations demonstrates that there are lateral inhibitory interactions between central neural units in the human visual system. It is likely that the interacting units mediate the perception of contour orientation, for the threshold elevation functions were consistent with concurrent changes of perceived orientation. The results are evidence for the hypothesis that inhibition between orientation detectors is a factor in the perceptual expansion of acute angles.

Human auditory evoked responses during hangover

Auditory evoked responses (AER) to trains of 6 click stimuli (1 click/sec) were studied in 9 subjects under hangover, tired control, and normal control conditions in order to find out whether the symptoms of hyperexcitability during hangover have a correlate in the characteristics of the AER. In addition, the audiograms were measured. AERs to the first click in a stimulus train were markedly smaller during hangover than in the other 2 states. The amplitude levels of the AERs during the repetition of the click stimulus were, however, similar under all three conditions. The audiograms obtained in the three states were similar except for a very slight decrease of auditory threshold sensitivity during hangover as compared with the tired control condition. The results show that the effects of hangover on AERs resemble those of alcohol intoxication. The symptoms of hyperexcitability during hangover cannot be explained in terms of increased peripheral sensitivity.

Dark adaptation and short-wavelength backgrounds decrease perceived size

The effects of background luminance, contrast, and background wavelength on the perceived size of small line figures were studied at mesopic levels of light adaptation. Perceived size diminished at low levels of background luminance. The effect disappeared at high levels of luminance. Perceived size of luminous circles increased as a logarithmic function of background luminance when the background intensity did not exceed 25 td(1). The strength of the size effect decreased as a function of circle diameter from 0-125 to 2 deg of visual angle(2). Perceived size of small luminous circles, subtending less than 0-5 deg, also increased as a function of contrast at low values of contrast but at very high values of contrast there was a decrease in perceived size. Background luminance had the same effect on the perceived size of circles as on the perceived size of spatial cycles in gratings. Control experiments led to the conclusion that dark adaptation is the primary source of the size effects. The main evidence for this conclusion was obtained from a demonstration that the same background luminance produced either an increase or a decrease in perceived size, depending on the adaptational state of the eye. It was also found that a shift from cone vision to rod vision contributes to the effects, for a stimulus looked smaller on a short-wavelength background than on a long-wavelength background. The size effects can be predicted from the changes of receptive-field properties of single neurones under corresponding conditions of stimulation, if it is assumed that the perception of size is mediated by size-specific channels formed of single neurones. Stimulation that leads to an activation of small receptive fields appears to indicate to the brain the presence of small retinal images. If small receptive fields are experimentally made responsive to larger retinal images, an underestimation of size results.

Diphasic and polyphasic temporal modulations multiply apparent spatial frequency

The effects of diphasic and polyphasic flicker on apparent spatial frequency were studied in several experiments through simultaneous spatial-frequency matches. In diphasic flicker the spatial phase of a sinusoidal grating alternated between two values in a counterphase fashion, and in polyphasic flicker the spatial phases of gratings were varied discretely in time in a variable number of steps. Both forms of flicker increased the apparent spatial frequency at low temporal frequencies, in the same manner as low-frequency monophasic flicker has been found to do. At high temporal frequencies, diphasic flicker doubled the apparent spatial frequency, as reported by Kelly (1966). We found that through high-frequency polyphasic flicker the spatial effect that Kelly mentions can be generalised to spatial frequency multiplication: polyphasic flicker produces not only the apparent second harmonic but also the third and the fourth harmonic, depending on the phase parameters. A numerical analysis showed that the spatial high-frequency effects can be explained through temporal integration of nonlinearly filtered input signals if a value of 200 td(1) is assumed for the nonlinearity constant in

[Formula: see text]

where B( I) is the brightness, I is the retinal illuminance, K is a scale constant, and I½ is the constant of nonlinearity. A minimum value of 60 ms had to be estimated for integration time. An investigation of the integration time with diphasic flicker indicated that spatial integration time decreases when the level of light adaptation increases, and that the integration time for spatial effects is longer than for flicker fusion. The spatial effects of low-frequency and high-frequency flicker differ in so many respects that different neural processes have to be postulated for their explanation.

Monophasic temporal modulation increases apparent spatial frequency

There are three different types of flicker—monophasic, diphasic, and polyphasic—and they have different spatial consequences with gratings. Monophasic flicker does not alter the spatial phase in time, but spatial phase changes are caused by the other two types. The effects of sinusoidal monophasic flicker on the apparent spatial frequency of sinusoidal gratings were studied in the present experiments by simultaneous spatial-frequency matches. Monophasic temporal modulation increased apparent spatial frequency. The stimulus conditions for producing a maximum effect were (a) a relatively low spatial frequency, (b) a relatively high level of light adaptation, (c) an intermediate temporal frequency (4–8 Hz), and (d) an intermediate contrast. The largest apparent increases exceeded 30%. The magnitude of the spatial effect was correlated with data on temporal resolution and contrast sensitivity collected under the same conditions. The spatial effect had a high correlation with the critical flicker-fusion frequency, and the contrast-sensitivity enhancements caused by the flicker followed functions similar to those of the spatial effect when the spatial frequency and the level of adaptation were varied. We interpret the spatial effect of low-frequency monophasic flicker as evidence that there are channels for spatial frequency in human vision whose spatial tuning depends on the spatial distribution of sensitivity in the receptive fields of single neurones. Flicker at intermediate temporal frequencies decreases the functional effectiveness of centre—surround antagonism, and makes channels otherwise responsive to high spatial frequencies responsive to low spatial frequencies. As the central decoding of channel outputs can be assumed invariant, an increase in the apparent spatial frequency follows from the change of tuning properties if the channels mediate the perception of spatial frequency. An analysis of the variability of spatial frequency matches indicated that spatial frequency discrimination, if considered in relative terms, is independent of the spatial frequency, the mean luminance, and the contrast of gratings within broad ranges of variation.

Relationships between channels for colour and spatial frequency in human vision

Five forms of relationships and four types of channels are possible between two systems of sensory channels. The relationships between channels for colour and spatial frequency were studied in three adaptation experiments. In the first, a new colour-specific spatial aftereffect was found, which indicates the existence of channels that are specific both to colour and to spatial frequency. The second showed that the spatial-frequency aftereffect of Blakemore and Sutton is not colour specific, which indicates that there are channels for spatial frequency that are not colour specific. The third demonstrated that coloured afterimages are not spatial-frequency specific immediately after adaptation, although they become so later. This indicates that there are channels for colour that are not spatial-frequency specific. The existence of these three types of channels implies that the channel systems for colour and spatial frequency overlap partially and mutually in the human visual system. This kind of organisation of channel systems, if it exists, may form the psychophysical structure that is required for the capacity of simultaneous integration and differentiation in the perception of colour and size of visual objects.

Perceived curvature of arcs and dot patterns as a function of curvature, arc length, and instructions

Arcs of circles, with six arc lengths and four radii of curvature, and an equivalent set of figures composed of three dots were used as stimuli. Subjects in Group I imagined the circle from which an arc or dot triplet was taken and indicated the centre of the circle. Group II subjects estimated the location of the point that was equidistant from the middle and ends of an arc, or equidistant from the three dots of a triplet. The results from arcs showed, in Group I, an underestimation of curvature that decreased as a function on the length of the arc. In Group II, however, overestimation of the curvature of most arcs occurred, indicating a strong influence of the difference in the perceptual task on the results. The effect of instructions was similar with the dot figures but, in general, more errors resembling overestimation of curvature occurred with these figures.

Contrast and confluxion as components in geometric illusions

Two experiments on illusions were planned to test the predictive power of explanations based on size contrast and confluxion. The predictions turned out to be correct. A modification of the Müller-Lyer figure and an illusion of divided distance were used as stimulus figures. In the latter the dividing distances were underestimated. It proved to be methodologically necessary to measure contrast effects as differences between results of two experimental procedures, because small intervals as such were over-estimated. It was not possible to explain all the results by means of the constancy theory in the form suggested by Gregory. Neither contrast nor constancy alone is sufficient to explain the geometric illusions. The development of an adequate theory by combining these two explanations seems possible.