Twin mechanisms: Rapid scene recognition involves both feedforward and feedback processing

Spatial frequency-based information plays an important role in visual perception. By combining behavioral and electroencephalogram (EEG) measurements, we investigated the mechanisms of the interaction and information integration between different spatial frequency bands. The observers performed a scene categorization task on hybrid images that were generated by combining the low spatial frequency (LSF) component of one image with the high spatial frequency (HSF) component of another image. The results showed that the recognition of the HSF component was interfered by the non-attended LSF component at semantic level. The strength of the semantic interference was modulated by the physical similarity between the LSF and HSF components. Analyses of the EEG data revealed an early anterior N1 component (122ms from stimulus onset) that was related to the observed interaction of the semantic and physical information between the LSF and HSF components. These findings demonstrate that the semantic information from different spatial frequency bands can be integrated at early stage of the perceptual processing. This early integration is likely to occur at frontal areas in order to initiate top-down facilitation.

ABSTRACTCurrent models of visual perception suggest that, during scene categorization, low spatial frequencies (LSF) are rapidly processed and activate plausible interpretations of visual input. This coarse analysis would be used to guide subsequent processing of high spatial frequencies (HSF). The present study aimed to further examine how information from LSF and HSF interact and influence each other during scene categorization. In a first experimental session, participants had to categorize LSF and HSF filtered scenes belonging to two different semantic categories (artificial vs. natural). In a second experimental session, we used hybrid scenes as stimuli made by combining LSF and HSF from two different scenes which were semantically similar or dissimilar. Half of the participants categorized LSF scenes in hybrids, and the other half categorized HSF scenes in hybrids. Stimuli were presented for 30 or 100 ms. Session 1 results showed better performance for LSF than HSF scene categorization. Session 2 scene categorization was faster when participants attended and categorized LSF than HSF scene in hybrids. The semantic interference of a semantically dissimilar HSF scene on LSF scene categorization was greater than the semantic interference of a semantically dissimilar LSF scene on HSF scene categorization, irrespective of exposure duration. These results suggest a LSF advantage for scene categorization, and highlight the prominent role of HSF information when there is uncertainty about the visual stimulus, in order to disentangle between alternative interpretations.

Glare visual acuity and contrast sensitivity are important indicators of visual quality. Studies have shown that the glare visual acuity and contrast sensitivity in dry eye patients tend to degenerate, further affecting their quality of life. The objective of this study was to investigate the effect of notch filters on glare VA and contrast sensitivity in patients with dry eye or with dry eye syndrome. 36 subjects in the 20‒65 age group were diagnosed as having dry eye disease or perceived dry eye syndromes themselves who were included after the initial screening with the OSDI questionnaire, and one was subsequently excluded as they had undergone retinal detachment surgery. Finally, 35 subjects (14 male and 21 female) with a mean age of 40.66 ± 15.62years participated in this study. All subjects wore their habitual prescriptions and four different filter lenses (namely 480, 620, dual 480 & 620 notch filter, and FL-41 tinted lens), and measured the parameters of glare visual acuity and contrast sensitivity using CSV-1000 and sine wave contrast test (SWCT), respectively. Student t-test and Repeated measurement analysis (R-ANOVA) were utilized by using SPSS 26.0 software. A dual-wavelength 480 & 620nm optical notch filter had a significant anti-glare effect decreasing glare disabilities or discomfort, and leading to better visual quality, the same effect was also shown on a 480nm notch filter lens. All participants showed a significant difference among the baseline, three notch filters (480nm, 620nm, dual-wavelength 480 & 620nm), and FL-41 tinted lens were used on SWCT_A (1.5cpd, F = 3.054, p = 0.019) and SWCT_E (18cpd, F = 2.840, p = 0.049); but did not show statistical different on SWCT_B (3cpd, F = 0.333, p = 0.771), SWCT_C (6cpd, F = 1.779, p = 0.159), and SWCT_D (12cpd, F = 1.447, p = 0.228). The baseline showed the best visual performance on CS at a low spatial frequency (SWCT_A, 1.5cpd), any filter might reduce the contrast sensitivity at low spatial frequencies in the clinical trial, whereas 480nm notch filter showed the best effectiveness on CS at a high spatial frequency (SWCT_E, 18cpd), the FL-41 lens that also filters out the 480nm band does not achieve the same effect. Moreover, patients with dry eye or those older than 40years old preferred optical multilayer notch filters to FL-41 tinted lenses. The 480- & 620-nm dual-wavelength and 480-nm single-wavelength notch filters have the best effect on the glare visual acuity and contrast sensitivity (CS) at high spatial frequencies in dry eye patients. The 620-nm notch filter performs better in CS at low and mid-low spatial frequencies; the FL-41 tinted lens performs poorly for glare VA and CS spatial frequencies examination. Patients with glare disabilities or CS disturbance at high spatial frequencies may choose a 480-nm notch filter lens, and patients who have CS disturbance at low spatial frequencies may consider a 620-nm notch filter for the prescription.

Low spatial frequency (LSF) image content has been proposed to play a superior functional role in emotional content extraction via the magnocellular pathway biasing attentional resources toward emotional content in visual cortex. We investigated whether emotionally unpleasant complex images that were presented either unfiltered or with LSF content only in the background while subjects performed a foreground task will withdraw more attentional resources from the task compared to unemotional, neutral images (distraction paradigm). We measured steady-state visual evoked potentials (SSVEPs) driven by flickering stimuli of a foreground task. Unfiltered unpleasant images resulted in a significant reduction of SSVEP amplitude compared to neutral images. No statistically significant differences were found with LSF background images. In a behavioral control experiment, we found no significant differences for complexity ratings between unfiltered and LSF pictures. Content identification was possible for unfiltered and LSF picture (correct responses > 74%). An additional EEG study examined typical emotion-related components for complex images presented either as unfiltered, LSF, or high spatial frequency (HSF, as an additional control) filtered, unpleasant, and neutral images. We found a significant main effect of emotional valence in the early posterior negativity. Late positive potential differences were only found for unfiltered and HSF images. Results suggest that, while LSF content is sufficient to allow for content and emotional cue extraction when images were presented alone, LSF content is not salient enough to serve as emotional distractor that withdraws attentional resources from a foreground task in early visual cortex.

Motion detection and the coincidence of structure at high and low spatial frequencies

Cerebral regions and hemispheric specialization for processing spatial frequencies during natural scene recognition. An event-related fMRI study

In this paper, a novel criterion for evaluating the beam quality of high energy laser is proposed, which is called “power outside the large bucket”, P OLB for short. The novel criterion does not demonstrate the divergence of the beam but focuses on the high spatial frequency wavefront aberration of the beam. The P OLB values of the laser beams with various aberrations are calculated. It shows that the more high spatial frequency components in the aberration the larger P OLB value is. Moreover, it is theoretically analyzed that the laser beams with various aberrations are corrected by ideal adaptive optics (AO) systems with different deformation mirror (DM) actuator numbers. It is shown that the residual error of the corrected wavefront aberration with many high spatial frequency components is quite large. Finally, the dependence of the residual wavefront error on the P OLB value is investigated. Only if the P OLB is smaller than 3.8/3.4/2.5/1.7 the residual error of the wavefront may be smaller than λ/10, which is corrected by an ideal AO system with 127/61/37/19 actuators. It is necessary to employ a complicated AO system for improving the laser beam of which the P OLB value is large. The novel evaluating criterion P OLB is able to demonstrate the amount of high spatial frequency aberration and the residual wavefront error corrected by AO system. It is an accessible and useful criterion for evaluating the beam quality of high energy lasers.

The receptive field of a visual neuron is the particular region in which a stimulus will modify the firing of that neuron. Retinal ganglion cells have a coaxial-shaped structure formed by the central excitatory and outer inhibitory regions. These regions respond oppositely to light. This shape can be mathematically modeled by a balanced Difference of Gaussian (DOG) function. In digital image processing, the DOG filter is used for detecting objects' edges. Bacteriorhodopsin (bR) is a photosensitive protein which resembles the visual pigment rhodopsin. When bR is illuminated, it generates a positive transient photocurrent, and, when light is switched off, a negative transient photocurrent is produced. This peculiar behavior is similar to the response of ganglion receptive fields. In this study, we fabricate a two-dimensional binarized DOG (b-DOG) filter, in which the central part mimics the excitatory part of the ganglion cell receptive fields, whereas the outer ring reproduces the effect observed in their outer inhibitory regions. This filter consists of photosensitive protein bR films and electrolyte solution that are sandwiched between ITO electrodes. To analyze the spatial-temporal frequency sensitivity, we use moving sine wave gratings with different pitches with a controlled scanning speed. When the temporal frequency is kept constant, the spatial frequency sensitivity matched with the Fourier transform of the b-DOG function. On the other hand, when the spatial frequency does not vary, the temporal frequency sensitivity corresponded well with the Fourier transform of Difference of Gamma function. Difference of Gamma function is known as the impulse response of the visual nerve of animals. We separately analyze the independent spatial and temporal frequencies collected from the spatial-temporal characteristics of the filter. The spatial-temporal frequency characteristics of b-DOG filter are similar to those of the X-type retinal ganglion cells. However, unlike the retinal ganglion cell, the b-DOG filter is a perfect linear filter. The analog image processing using this b-DOG filter is performed by scanning a standard test image. It is found that an edge can be detected just by scanning the image and plotting the zero-crossing point. With increasing input image size, the spatial frequency peak detected by the b-DOG filter shifts towards higher frequencies. Since there are only sharp edges where high spatial frequency components are present, the detected results of the analog filter are similar to that of the digital filter. Consequently, the phase difference between the input and output image is approximately the same for all pixels, and it agrees with the digital edge detection results. Some illusion images are scanned on the b-DOG filter to verify the occurrence of the illusion similar to vision. The Hermann grid illusion generated with lower-order vision was observed. It is found that no luminance information is necessary for the Herman grid illusion, because this b-DOG filter does not extract image luminance. The visual function elements similar to X-type retinal ganglion cells prepared in this study are useful to constructively understand the visual information processing mechanism of the organism.

Interocular suppressive interactions in amblyopia depend on spatial frequency

Text contains a range of different spatial frequencies but the effectiveness of spatial frequencies for normal variations in skilled adult reading ability is unknown. Accordingly, young skilled adult readers showing fast or slow reading ability read sentences displayed as normal or filtered to contain only very low, low, medium, high, or very high spatial frequencies. Reading times and eye movement measures of fixations and saccades assessed the effectiveness of these displays for reading. Reading times showed that, for each reading ability, medium, high, and very high spatial frequencies were all more effective than lower spatial frequencies. Indeed, for each reading ability, reading times for normal text were maintained when text contained only medium, high, or very high spatial frequencies. However, reading times for normal text and for each spatial frequency were all substantially shorter for fast readers than for slow readers, and this advantage for fast readers was similar for normal, medium, high, and very high spatial frequencies but much larger for low and very low spatial frequencies. In addition, fast readers made fewer and shorter fixations, fewer and shorter regressions, and longer forward saccades, than slow readers, and these differences were generally similar in size for normal, medium, high, and very high spatial frequencies, but larger when spatial frequencies were lower. These findings suggest that fast and slow adult readers can each use a range of different spatial frequencies for reading but fast readers make more effective use of these spatial frequencies and especially those that are lower. (PsycINFO Database Record

In the present paper, we report two experiments out of series of studies designed to examine various aspects of visual search for icons of differing spatial frequencies. Specifically, the present experiments explore whether there exists a search asymmetry between high and low spatial frequency icons (A amongst B > B amongst A), and whether observers can limit their search to the relevant set of items in a display containing both types of icons. Our results show that a classic search asymmetry does not exist for spatial frequency; that, rather, both types of targets ‘pop out’; that search for a high spatial frequency target amongst high spatial frequency distractors is less efficient than search for a low spatial frequency target amongst low spatial frequency distractors; and that observers are partially able to limit their search to the relevant subset in mixed displays. Implications for the design of touch screen user interfaces are discussed.

PurposeThe spatial frequency (SF) characteristics of accommodation in children are not well understood. In this study, we measured accommodative responses to grating targets to investigate the SF dependence of accommodation in children.MethodsThe effects of SF and contrast on the accommodative system were evaluated in two groups of children, including 22 with emmetropia and 20 with myopia. The contrast detection thresholds at five SFs were measured using a near-contrast sensitivity function test. The accommodative responses to grating targets with low (1.5 cycles per degree [cpd]), medium (6 cpd), and high (18 cpd) SFs were measured with a Grand Seiko WAM-5500 in dynamic mode for 30 seconds under standard and detection threshold contrast conditions. The accommodative lag and accommodative microfluctuations (AMFs) were compared between the two groups.ResultsUnder standard contrast conditions, no significant difference was found in the accommodative lag across SFs (F = 2.03, P = 0.14) or between the two groups (F = 3.57, P = 0.07). The AMFs were lowest at 6 cpd in emmetropia group (F = 6.51, P = 0.003) and in total (F = 10.82, P < 0.001). Children in emmetropia group showed greater AMFs at high SFs under detection threshold contrast conditions than under standard contrast conditions (P < 0.05).ConclusionsThis study demonstrated that the instability of accommodation was SF dependent in children. The AMFs in children were smallest at the medium SF for standard contrast grating targets. Myopic children are less sensitive to the low-contrast-induced blur for high SFs than emmetropic children.Translational RelevanceThis study provides a possibility to stabilize accommodative response of children by transforming SF components of fixation targets.

The subcortical visual pathway to the amygdala has long been considered a rapid and crude stream for processing emotionally salient information that is reliant on low spatial frequency (LSF) information. Recently, research has called this LSF dependency into question. To resolve this debate, we take advantage of an anatomical hemiretinal asymmetry, whereby the nasal hemiretina sends a higher proportion of information through the subcortical pathway than the temporal hemiretina. We recorded brain activity using electroencephalography (EEG) in human participants (N = 40) while they completed a monocular viewing paradigm. Pairs of faces (one fearful and one neutral, or both neutral) were projected simultaneously to the nasal and temporal hemiretina in three contrast-equated blocks; faces filtered to display only (i) LSF, (ii) high spatial frequency (HSF), or (iii) unfiltered information (broadband spatial frequency; BSF). BSF fearful faces were found to produce a greater naso-temporal asymmetry, with greater N170 amplitudes evoked by BSF faces in the nasal field, compared to HSF faces. Conversely, the naso-temporal asymmetry for LSF fearful faces did not differ between BSF and HSF. Collectively, these findings provide crucial evidence that the subcortical pathway carries combined spatial frequency visual signals, with a potential bias against HSF content.

Eye movements transform a spatial scene into luminance modulations on the retina. Recent work has shown that this transformation is highly structured: within human temporal sensitivity, saccades deliver power that increases in proportion to spatial frequency (SF) up to a critical frequency and remains constant beyond that. Importantly, the critical SF increases with decreasing amplitude. Therefore, at sufficiently low SFs, larger saccades effectively deliver stronger input signals to the retina. Here we tested whether this input reformatting has the predicted perceptual consequences, by examining how large and small saccades (6o & 1o) affect contrast sensitivity. We measured relative sensitivity at two SFs: a reference (0.5 cpd), equal to the critical SF for the small saccade, and a probe at either a lower or higher SF (0.1/2.5 cpd). We predicted that large saccades enhance visibility only when the probe has a lower SF than the reference. Subjects (N=7) made instructed saccades while presented with a plaid of overlapping orthogonal gratings at the two SFs and reported which grating was more visible. Results closely follow theoretical predictions: psychometric functions following small and large saccades only differed with the lower SF probe, in which case the larger saccade significantly enhanced visibility. In sum, saccades enable selectivity not only in the spatial domain, but also in the spatial-frequency domain.

The red-green axis of colour vision evolved recently in primate evolutionary history. Signals serving red-green colour vision travel together with signals serving spatial vision, in the parvocellular (PC) division of the subcortical visual pathway. However, the question of whether receptive fields of PC pathway cells are specialized to transmit red-green colour signals remains unresolved. We addressed this question in single-cell recordings from the lateral geniculate nucleus of anaesthetized marmosets. Marmosets show a high proportion of dichromatic (red-green colour-blind) individuals, allowing spatial and colour tuning properties of PC cells to be directly compared in dichromatic and trichromatic visual systems. We measured spatial frequency tuning for sine gratings that provided selective stimulation of individual photoreceptor types. We found that in trichromatic marmosets, the foveal visual field representation is dominated by red-green colour-selective PC cells. Colour selectivity of PC cells is reduced at greater eccentricities, but cone inputs to centre and surround are biased to create more selectivity than predicted by a purely 'random wiring' model. Thus, one-to-one connections in the fovea are sufficient, but not necessary, to create colour-selective responses. The distribution of spatial tuning properties for achromatic stimuli shows almost complete overlap between PC cells recorded in dichromatic and trichromatic marmosets. These data indicate that transmission of red-green colour signals has been enabled by centre-surround receptive fields of PC cells, and has not altered the capacity of PC cells to serve high-acuity vision at high stimulus contrast.