Neural mechanisms of visual associative processing

This is a review of our work on multiple microelectrode recordings from the visual cortex of monkeys and subdural recordings from humans--related to the potential underlying neural mechanisms. The former hypothesis of object representation by synchronization in visual cortex (or more generally: of flexible associative processing) has been supported by our recent experiments in monkeys. They demonstrated local synchrony among rhythmic or stochastic gamma-activities (30-90 Hz) and perceptual modulation, according to the rules of figure-ground segregation. However, gamma-synchrony in primary visual cortex is restricted to few millimeters, challenging the synchronization hypothesis for larger cortical object representations. We found that the spatial restriction is due to gamma-waves, traveling in random directions across the object representations. It will be argued that phase continuity of these waves can support the coding of object continuity. Based on models with spiking neurons, potentially underlying neural mechanisms are proposed: (i) Fast inhibitory feedback loops can generate locally synchronized gamma-activities; (ii) Hebbian learning of lateral and feed forward connections with distance-dependent delays can explain the stabilization of cortical retinotopy, the limited size of synchronization, the occurrence of gamma-waves, and the larger receptive fields at successive levels; (iii) slow inhibitory feedback can support figure-ground segregation; (iv) temporal dispersion in far projections destroys coherence of fast signals but preserves slow amplitude modulations. In conclusion, it is proposed that the hypothesis of flexible associative processing by gamma-synchronization, including coherent representations of visual objects, has to be extended to more general forms of signal coupling.     

Cortical representation of space around the blind spot

The neural mechanism that mediates perceptual filling-in of the blind spot is still under discussion. One hypothesis proposes that the cortical representation of the blind spot is activated only under conditions that elicit perceptual filling-in and requires congruent stimulation on both sides of the blind spot. Alternatively, the passive remapping hypothesis proposes that inputs from regions surrounding the blind spot infiltrate the representation of the blind spot in cortex. This theory predicts that independent stimuli presented to the left and right of the blind spot should lead to neighboring/overlapping activations in visual cortex when the blind-spot eye is stimulated but separated activations when the fellow eye is stimulated. Using functional MRI, we directly tested the remapping hypothesis by presenting flickering checkerboard wedges to the left or right of the spatial location of the blind spot, either to the blind-spot eye or to the fellow eye. Irrespective of which eye was stimulated, we found separate activations corresponding to the left and right wedges. We identified the centroid of the activations on a cortical flat map and measured the distance between activations. Distance measures of the cortical gap across the blind spot were accurate and reliable (mean distance: 6-8 mm across subjects, SD approximately 1 mm within subjects). Contrary to the predictions of the remapping hypothesis, cortical distances between activations to the two wedges were equally large for the blind-spot eye and fellow eye in areas V1 and V2/V3. Remapping therefore appears unlikely to account for perceptual filling-in at an early cortical level.     

Interhemispheric transfer of phosphenes generated by occipital versus parietal transcranial magnetic stimulation

Phosphenes represent a perceptual effect of transcranial magnetic stimulation (TMS) or electric stimulation of visual cortical areas. One likely neural basis for the generation of static phosphenes is the primary visual cortex (V1) although evidence is controversial. A peculiar feature of V1 is that it has sparse callosal connections with the exception of a central portion of visual field representation. In contrast, visually responsive cortical areas in the parietal lobe have widespread callosal connections. Thus, interhemispheric transfer (IT) time of off-centre phosphenes should be slower when generated by V1 than by visual parietal areas. To verify this possibility, in Exp. 1 we measured IT of phosphenes generated by TMS applied to V1 and in Exp. 2 we measured IT of phosphenes obtained by TMS applied to posterior parietal cortex. In both experiments, we obtained static bright circular phosphenes appearing in the contralateral hemifield. We measured IT time behaviorally by comparing unimanual simple reaction time to the onset of a phosphene under crossed or uncrossed hemifield-hand condition (Poffenberger paradigm). In keeping with our prediction, we found a substantially longer IT time for V1 than for parietal phosphenes. Additionally, an IT similar to that obtained with V1 stimulation was found when participants were asked to imagine the phosphenes previously experienced during TMS. In conclusion, the present results suggest that IT of phosphenes either generated by V1 TMS or imagined is subserved by slower callosal channels than those of real visual stimuli or parietal phosphenes.     

Effects of Ketamine on perceptual grouping in rats

Ketamine is a selective NMDA glutamate receptor antagonist that disrupts cognitive and behavioral function. Evidence exists that NMDA receptors play a role in lateral cortical connections, suggesting involvement in integrating information across the cortex. To investigate NMDA receptors' role in cortical integration at a perceptual level, psychophysical measures were made of perceptual grouping, which requires global analysis of neural representations of stimulus elements. Rats were trained to discriminate solid lines as well as patterns of dots that could be perceptually grouped into vertical or horizontal stripes. Psychophysical measures determined thresholds of perceptual grouping capacities. Rats receiving maximum subanesthetic doses of Ketamine discriminated solid patterns normally, but were impaired on dot pattern discrimination when greater demands were placed on perceptual grouping. These results demonstrate a selective disruption by Ketamine of visual discrimination that requires perceptual grouping of stimulus patterns. These results also provide evidence associating NMDA receptor-dependent neural mechanisms with context-dependent perceptual function.     

Manifestation of scotomas created by transcranial magnetic stimulation of human visual cortex.

Reduced visual performance under transcranial magnetic stimulation (TMS) of human visual cortex demonstrates suppression whose spatial extent is not directly visible. We created an artificial scotoma (region missing from a visual pattern) to directly visualize the location, size and shape of the TMS-induced suppression by following a large-field, patterned, visual stimulus with a magnetic pulse. The scotoma shifted with coil position according to known topography of visual cortex. Visual suppression resulted in pattern-dependent distortion, and the scotoma was filled in with temporally adjacent stimuli, suggesting spatial and temporal completion mechanisms. Thus, perceptual measurements of TMS-induced suppression may provide information about cortical processing via neuronal connections and temporal interactions of neural signals.     

Hierarchy of cortical responses underlying binocular rivalry

During binocular rivalry, physical stimulation is dissociated from conscious visual awareness. Human brain imaging reveals a tight linkage between the neural events in human primary visual cortex (V1) and the dynamics of perceptual waves during transitions in dominance during binocular rivalry. Here, we report results from experiments in which observers' attention was diverted from the rival stimuli, implying that: competition between two rival stimuli involves neural circuits in V1, and attention is crucial for the consequences of this neural competition to advance to higher visual areas and promote perceptual waves.     

Neural responses in the macaque v1 to bar stimuli with various lengths presented on the blind spot

Although there is no retinal input within the blind spot, it is filled with the same visual attributes as its surround. Earlier studies showed that neural responses are evoked at the retinotopic representation of the blind spot in the primary visual cortex (V1) when perceptual filling-in of a surface or completion of a bar occurs. To determine whether these neural responses correlate with perception, we recorded from V1 neurons whose receptive fields overlapped the blind spot. Bar stimuli of various lengths were presented at the blind spots of monkeys while they performed a fixation task. One end of the bar was fixed at a position outside the blind spot, and the position of the other end was varied. Perceived bar length was measured using a similar set of bar stimuli in human subjects. As long as one end of the bar was inside the blind spot, the perceived bar length remained constant, and when the bar exceeded the blind spot, perceptual completion occurred, and the perceived bar length increased substantially. Some V1 neurons of the monkey exhibited a significant increase in their activity when the bar exceeded the blind spot, even though the amount of the retinal stimulation increased only slightly. These response increases coincided with perceptual completion observed in human subjects and were much larger than would be expected from simple spatial summation and could not be explained by contextual modulation. We conclude that the completed bar appearing on the part of the receptive field embedded within the blind spot gave rise to the observed increase in neuronal activity.     

A computational model of perceptual fill-in following retinal degeneration

The ablation of afferent input results in the reorganization of sensory and motor cortices. In the primary visual cortex (V1), binocular retinal lesions deprive a corresponding cortical region [lesion projection zone (LPZ)] of visual input. Nevertheless, neurons in the LPZ regain responsiveness by shifting their receptive fields (RFs) outside the retinal lesions; this re-emergence of neural activity is paralleled by the perceptual completion of disrupted visual input in human subjects with retinal damage. To determine whether V1 reorganization can account for perceptual fill-in, we developed a neural network model that simulates the cortical remapping in V1. The model shows that RF shifts mediated by the plexus of spatial- and orientation-dependent horizontal connections in V1 can engender filling-in that is both robust and consistent with psychophysical reports of perceptual completion. Our model suggests that V1 reorganization may underlie perceptual fill-in, and it predicts spatial relationships between the original and remapped RFs that can be tested experimentally. More generally, it provides a general explanation for adaptive functional changes following CNS lesions, based on the recruitment of existing cortical connections that are involved in normal integrative mechanisms.     

Electrophysiological correlates of visually processing subject's own name

The study aimed to test the modulation induced by 1 Hz repetitive Transcranial Magnetic Stimulation (rTMS) of the occipital cortex on the alpha phase synchronization under repetitive flash stimuli in 15 migraine without aura patients compared to 10 controls. The EEG was recorded by 7 channels, while flash stimuli were delivered at 9, 18, 21 and 24 Hz in basal, rTMS (15 min of 1 Hz stimulation of the occipital cortex) and sham conditions. Migraine patients displayed increased alpha-band phase synchronization under visual stimulation, while an overall desynchronizing effect was evident in controls. The rTMS resulted in a slight increase of synchronization index in migraine patients, which did not cause significant differences in respect to the basal and sham conditions. The synchronizing-desynchronizing changes of alpha rhythm under repetitive flash stimulation, seem independent from the state of occipital cortex excitability. Other mechanisms beyond cortical excitability may contribute to explain migraine pathogenesis.     

Rapid learning in cortical coding of visual scenes

Experience-dependent plasticity in adult visual cortex is believed to have important roles in visual coding and perceptual learning. Here we show that repeated stimulation with movies of natural scenes induces a rapid improvement in response reliability in cat visual cortex, whereas stimulation with white noise or flashed bar stimuli does not. The improved reliability can be accounted for by a selective increase in spiking evoked by preferred stimuli, and the magnitude of improvement depends on the sparseness of the response. The increase in reliability persists for at least several minutes in the absence of further movie stimulation. During this period, spontaneous spiking activity shows detectable reverberation of the movie-evoked responses. Thus, repeated exposure to natural stimuli not only induces a rapid improvement in cortical response reliability, but also leaves a 'memory trace' in subsequent spontaneous activity.     

Attention modulates responses in the human lateral geniculate nucleus

Attentional mechanisms are important for selecting relevant information and filtering out irrelevant information from cluttered visual scenes. Selective attention has previously been shown to affect neural activity in both extrastriate and striate visual cortex. Here, evidence from functional brain imaging shows that attentional response modulation is not confined to cortical processing, but can occur as early as the thalamic level. We found that attention modulated neural activity in the human lateral geniculate nucleus (LGN) in several ways: it enhanced neural responses to attended stimuli, attenuated responses to ignored stimuli and increased baseline activity in the absence of visual stimulation. The LGN, traditionally viewed as the gateway to visual cortex, may also serve as a 'gatekeeper' in controlling attentional response gain.     

Perceptual load modulates visual cortex excitability to magnetic stimulation

Much recent research has shown that the level of perceptual load in a task determines the perception of task-irrelevant stimuli and associated neural activity, but the mediating neural mechanisms remain unclear. Here we show that increasing the level of perceptual load in a static letter search task results in an increase in the intensity of transcranial magnetic stimulation over V5/MT required to elicit the perception of a moving phosphene. These findings suggest that the neural mechanisms mediating the effects of perceptual load involve reduced visual cortex excitability in task-unrelated areas.     

Tracking the mismatch information in visual short term memory: An event-related potential study

Although the mechanisms of visual short term memory (VSTM) are extensively investigated in the recent decade, how we compare the representation stored in VSTM to the perceptual input to detect and process the mismatch information remains largely unclear. The current study explored whether there is an ERP component tracking the mismatch process in VSTM by adopting a delayed matching task. To exclude non-memory factors, for instance, using perceptual representation which dominated in previous studies using this paradigm, we lengthened the blank interval between the two sequentially displayed stimuli to 4 s to ensure the first stimuli is stored in VSTM. In order to test the sensitivity of this potential neural index and its functional relation to VSTM comparison process, colored shapes were adopted as materials while both the target feature color and the irrelevant feature shape could be changed. We found both the target feature change and the irrelevant feature change elicited a more negative component N270 (or N2-enhancement) around the anterior areas, with their neural sources located at frontal lobe. These results suggest that the N270 can sensitively reflect the mismatch information between the representation in VSTM and the perceptual input. Moreover, it may reflect the limited-capacity process in the VSTM-perception comparison, in which a deliberative comparison was conducted after an unlimited-capacity comparison process.     

Early induced beta/gamma activity during illusory contour perception

The temporal binding hypothesis proposes that visual feature binding is achieved by neuronal synchronization. Nevertheless, the existing human neurophysiological evidence for the neuronal synchronization in visual feature binding—the oscillatory induced beta/gamma activity (IB/GA) is under suspicion. The previously observed IB/GA occurs at a later stage (after 200 ms), thus leading to the objection that IB/GA may be related to some later top-down processes rather than the early perceptual processing. However, the present EEG study identified an IB/GA as early as 90 ms after stimulus onset, which was stronger for a Kanizsa-type illusory contour (IC, a classic example of visual feature binding) than for a control stimulus. This finding provides new human evidence for the temporal binding hypothesis that neuronal synchronization occurs at the early stage of visual feature binding.     

Visual features that vary together over time group together over space

The visual system perceives objects as coherent even when some parts are hidden or discontinuous. How this representation is constructed from local features of many nearby objects is termed the 'binding problem.' Here we manipulate contrast in several drifting gratings that can be perceived as either independent objects or parts of a single object. Contrast modulations that are correlated in time enhance perceptual coherence, whereas uncorrelated modulations impair coherence. Presumably, correlated contrast modulations produce correlated responses in cortical neurons. Therefore, our results are consistent with the hypothesis that temporal correlation of neural activity is important for feature binding.     

Remapping in human visual cortex

With each eye movement, stationary objects in the world change position on the retina, yet we perceive the world as stable. Spatial updating, or remapping, is one neural mechanism by which the brain compensates for shifts in the retinal image caused by voluntary eye movements. Remapping of a visual representation is believed to arise from a widespread neural circuit including parietal and frontal cortex. The current experiment tests the hypothesis that extrastriate visual areas in human cortex have access to remapped spatial information. We tested this hypothesis using functional magnetic resonance imaging (fMRI). We first identified the borders of several occipital lobe visual areas using standard retinotopic techniques. We then tested subjects while they performed a single-step saccade task analogous to the task used in neurophysiological studies in monkeys, and two conditions that control for visual and motor effects. We analyzed the fMRI time series data with a nonlinear, fully Bayesian hierarchical statistical model. We identified remapping as activity in the single-step task that could not be attributed to purely visual or oculomotor effects. The strength of remapping was roughly monotonic with position in the visual hierarchy: remapped responses were largest in areas V3A and hV4 and smallest in V1 and V2. These results demonstrate that updated visual representations are present in cortical areas that are directly linked to visual perception.     

Contralateral visual field representation in area 17 of the cerebral cortex of the agouti: a comparison between the cortical magnification factor and retinal ganglion cell distribution

The cortical representation of the contralateral visual field in area 17 of the agouti's brain was studied by multiunit recording. The borders of area 17 were determined by electrophysiological, cytoarchitectonic and myeloarchitectonic criteria. The results were plotted in flat, bidimensional representations of the cerebral cortex to minimize perspective distortions. The V1 map, a first order topological transformation of the visual field, shows a magnified representation of the horizontal meridian that corresponds to a retinal specialization, the visual streak. The visual field representation has asymmetries that are not directly related to the topography of the retinal ganglion cell density. Whereas the ganglion cell density shows a plateau along the visual streak, the areal cortical magnification factor is higher in the region that corresponds to the intersection of the horizontal and vertical meridians. This suggests functional specializations that are not obvious when one considers the distribution of the whole ganglion cell population but which might be related to the distribution of specific ganglion cell classes.     

Visually cued motor synchronization: modulation of fMRI activation patterns by baseline condition

A well-known issue in functional neuroimaging studies, regarding motor synchronization, is to design suitable control tasks able to discriminate between the brain structures involved in primary time-keeper functions and those related to other processes such as attentional effort. The aim of this work was to investigate how the predictability of stimulus onsets in the baseline condition modulates the activity in brain structures related to processes involved in time-keeper functions during the performance of a visually cued motor synchronization task (VM). The rational behind this choice derives from the notion that using different stimulus predictability can vary the subject's attention and the consequently neural activity. For this purpose, baseline levels of BOLD activity were obtained from 12 subjects during a conventional-baseline condition: maintained fixation of the visual rhythmic stimuli presented in the VM task, and a random-baseline condition: maintained fixation of visual stimuli occurring randomly. fMRI analysis demonstrated that while brain areas with a documented role in basic time processing are detected independent of the baseline condition (right cerebellum, bilateral putamen, left thalamus, left superior temporal gyrus, left sensorimotor cortex, left dorsal premotor cortex and supplementary motor area), the ventral premotor cortex, caudate nucleus, insula and inferior frontal gyrus exhibited a baseline-dependent activation. We conclude that maintained fixation of unpredictable visual stimuli can be employed in order to reduce or eliminate neural activity related to attentional components present in the synchronization task.     

A neural model of figure-ground organization

Psychophysical studies suggest that figure-ground organization is a largely autonomous process that guides--and thus precedes--allocation of attention and object recognition. The discovery of border-ownership representation in single neurons of early visual cortex has confirmed this view. Recent theoretical studies have demonstrated that border-ownership assignment can be modeled as a process of self-organization by lateral interactions within V2 cortex. However, the mechanism proposed relies on propagation of signals through horizontal fibers, which would result in increasing delays of the border-ownership signal with increasing size of the visual stimulus, in contradiction with experimental findings. It also remains unclear how the resulting border-ownership representation would interact with attention mechanisms to guide further processing. Here we present a model of border-ownership coding based on dedicated neural circuits for contour grouping that produce border-ownership assignment and also provide handles for mechanisms of selective attention. The results are consistent with neurophysiological and psychophysical findings. The model makes predictions about the hypothetical grouping circuits and the role of feedback between cortical areas.