Retina, MEG

Spatial weighting of visual motion coding in manual following response and MEG

Hiroaki Gomi, Kaoru Amano and Toshitaka Kimura

Visual motion information is not only used for perception of external world, but also instantaneously used for motor behaviors. It was found that a large-field visual motion induces a short latency manual response during reaching, termed Manual Following Response (MFR). To understand its generation mechanisms, we examined the spatial integration characteristics of MFR and related MEG (magnetoencephalography) response. As stimulus size increased, the MFR amplitude markedly increased for the lower-SF stimuli (0.02, 0.05 c/deg), but not much for the higher-SF stimuli (0.2, 0.8 c/deg). The identical stimuli were utilized to characterize the visually evoked MEG responses. MEG responses, mainly originating from the temporo-occipital cortex, were highly correlated with the MFR as shown previously (Amano et al., 2009, J Neurophysiol.,101, 888-897), but some difference was found in their tunings for stimulus size. Sensitivity functions fitted by Gaussian indicate that spatial integration is wider for the MFR than for the MEG with lower-SF stimuli while it is narrower for the MFR than for the MEG with higher-SF stimuli. These results suggest that the motion signal integration for the MFR relatively weights the neural activities sensitive to the lower-SF stimuli on the visual periphery while MEG reflects more parafoveal neural activities.

MEG evidence for early sensitivity to spatial phase congruency in human primary visual cortex

Linda Henriksson, Aapo Hyvärinen and Simo Vanni

Salient features in images, such as edges and lines, have been shown to correspond to points where spatial frequency components across spatial scales are maximally in phase, i.e., at locations of maximal phase congruency. Recent functional magnetic resonance imaging (fMRI) studies suggest that human visual cortex, including the primary visual cortex (V1), is sensitive to the phase congruency information [Henriksson et al., 2009, Journal of Neuroscience, 29(45):14342-51; Perna et al., 2008, Journal of Vision, 8(10):15, 1-15]. However, given the limited temporal resolution of fMRI, it has remained unclear whether this sensitivity originates in V1 or in a higher-level visual area. In this magnetoencephalography (MEG) study, we compared responses for compound grating stimuli with congruent and random phase alignments. We observed greater activity for the congruent stimuli first in V1, already within 100 ms after the stimulus presentation (Wilcoxon's signed rank test across subjects, N=10, p < 0.05), and only later in higher-level visual areas. These results imply an early origin for phase congruency sensitivity in V1, and support the hypothesis that this information is used by the human visual system to locate natural broadband edges.

Visual discrimination performance dependence on the individual peculiarities of the retina structures

Olga Vakhrameeva, Mikhail Sukhinin, Svetlana Muravieva, Dmitry Demidov and Yuri E. Shelepin

Recently growing interest to possible association of amblyopia with structural changes in retina makes it important to investigate the dependence of the human normal visual acuity on the peculiarities of macular retina structures [Repka et al, 2006, Am J Ophthalmol, 142(2): 247-251], [Cagini C. et al, 2009, Curr Eye Res 34(12):1036-41.], [Tsilimbaris et al, 2012, Ophthalmologica, Apr 4. Epub ahead of print]. In this work the correlation between human ability to discriminate small features of visual objects and individual characteristics of macular retina structures was investigated. During the discrimination task subjects had to define the orientation of the Landolt C elements with different sizes, noise levels and presentation modes: monocular and binocular. The probability of the correct answer was used as a measure of subjects'visual acuity. Optical coherent tomography method was used to measure the individual features of subjects'retina, such as ratio of the foveola and fovea diameters (ff-ratio), and retina thickness in different macular regions. First results show that subjects with thinner retina in the fovea area and ff-ratio closer to unity demonstrate better performance in the discrimination task. The results are discussed in terms of the monocular versus binocular vision model and the limiting factor of the internal multiplicative noise formed by the retinal structures.

Visual hyperacuity estimated with apparent motion

Alexander Garusev and Victor Doubrovski

To investigate visual acuity 2-dot stimuli were used with the angular sizes about 6 arcsec and 7 arcsec spacing produced by LEDs and plastic light guides. Visual acuity is considered as the minimum angular distance between two luminous dots for which the visual system reacts as on separated light sources. Rather than asking the subject whether one or two dots were presented, indirect judgments about visible displacement, an apparent motion and flicker were used. Detection of any of these phenomena indicates that the visual system responds on two dots, not on one. According to our preliminary findings, detection of flicker and displacement occurs at angular distances about 14 arcsec, and apparent motion detection at 16 - 20 arcsec, that agrees with vernier visual acuity thresholds, but less than angular cone size. The plausible explanation is that the retina should be considered as quasiperiodic 3D structure. Hence, even with very small stimuli and displacements (less than cone sizes) the light field has sophisticated distribution within the thickness of retina. Than the large set of photoreceptors is excited.