Representations In Visual Areas Research Articles

A role for visual areas in physics simulations

ABSTRACT To engage with the world, we must regularly make predictions about the outcomes of physical scenes. How do we make these predictions? Recent computational evidence points to simulation—the idea that we can introspectively manipulate rich, mental models of the world—as one explanation for how such predictions are accomplished. However, questions about the potential neural mechanisms of simulation remain. We hypothesized that the process of simulating physical events would evoke imagery-like representations in visual areas of those same events. Using functional magnetic resonance imaging, we find that when participants are asked to predict the likely trajectory of a falling ball, motion-sensitive brain regions are activated. We demonstrate that this activity, which occurs even though no motion is being sensed, resembles activity patterns that arise while participants perceive the ball's motion. This finding thus suggests that mental simulations recreate sensory depictions of how a physical scene is likely to unfold.

Mechanisms of target localization in visual change detection: An interplay of gating and filtering

Localizing visual information requires drawing on retinotopic activation maps located in visual cortex that are modulated both by stimulus saliency and a top-down bias toward relevant inputs. The present study investigated the role of this spatial stimulus representation for target localization by means of event-related potentials (ERPs) of the electroencephalogram (EEG). In different blocks, participants were instructed to localize a luminance change or simply detect its presence in a fast sequence further containing a randomly occurring orientation change distractor. Both in the localization and detection task, the lowest behavioral performances were shown when target and distractor were presented contralateral to each other (spatial conflict condition). On ERP level, higher posterior asymmetries in favor of the relevant luminance change in the N1 and N2 time windows were observed for the localization compared to the detection task and suggested an increased top-down bias on spatial target representation in visual areas when the response to a relevant stimulus had a spatial component. The increased top-down bias included both a stronger gating of relevant information in the N1 time window for unilateral feature changes and a subsequent increased filtering process with an onset that was delayed as a function of the distraction caused by irrelevant signals in the display. These mechanisms were impaired when participants failed to localize the luminance change, indicating that the top-down amplification of spatial target representation in visual areas relative to the irrelevant surrounding is a substantial component of localizing visual feature changes.

Visual areas become less engaged in associative recall following memory stabilization

Numerous studies have focused on changes in the activity in the hippocampus and higher association areas with consolidation and memory stabilization. Even though perceptual areas are engaged in memory recall, little is known about how memory stabilization is reflected in those areas. Using magnetoencephalography (MEG) we investigated changes in visual areas with memory stabilization. Subjects were trained on associating a face to one of eight locations. The first set of associations (‘stabilized’) was learned in three sessions distributed over a week. The second set (‘labile’) was learned in one session just prior to the MEG measurement. In the recall session only the face was presented and subjects had to indicate the correct location using a joystick. The MEG data revealed robust gamma activity during recall, which started in early visual cortex and propagated to higher visual and parietal brain areas. The occipital gamma power was higher for the labile than the stabilized condition (time=0.65–0.9 s). Also the event-related field strength was higher during recall of labile than stabilized associations (time=0.59–1.5 s). We propose that recall of the spatial associations prior to memory stabilization involves a top-down process relying on reconstructing learned representations in visual areas. This process is reflected in gamma band activity consistent with the notion that neuronal synchronization in the gamma band is required for visual representations. More direct synaptic connections are formed with memory stabilization, thus decreasing the dependence on visual areas.

Learning to Attend: Modeling the Shaping of Selectivity in Infero-temporal Cortex in a Categorization Task

Recent experiments on behaving monkeys have shown that learning a visual categorization task makes the neurons in infero-temporal cortex (ITC) more selective to the task-relevant features of the stimuli (Sigala and Logothetis in Nature 415 318-320, 2002). We hypothesize that such a selectivity modulation emerges from the interaction between ITC and other cortical area, presumably the prefrontal cortex (PFC), where the previously learned stimulus categories are encoded. We propose a biologically inspired model of excitatory and inhibitory spiking neurons with plastic synapses, modified according to a reward based Hebbian learning rule, to explain the experimental results and test the validity of our hypothesis. We assume that the ITC neurons, receiving feature selective inputs, form stronger connections with the category specific neurons to which they are consistently associated in rewarded trials. After learning, the top-down influence of PFC neurons enhances the selectivity of the ITC neurons encoding the behaviorally relevant features of the stimuli, as observed in the experiments. We conclude that the perceptual representation in visual areas like ITC can be strongly affected by the interaction with other areas which are devoted to higher cognitive functions.

The role of the corpus callosum and extra striate visual areas in stereoacuity in Macaque monkeys

Stereoacuity was measured in six normal male rhesus monkeys by requiring them to indicate the closer of two vertical line targets. Their stereoacuity ranged from 13–23 arc sec and was as good as that of human observers tested in the same way. Sectioning the splenium of the corpus callosum, either before or after removing the foveal representation in visual area 2, had no effect on stereoacuity thresholds, indicating that the splenium is not essential for the detection of small retinal disparities at the visual mid-line. In contrast, removal of the foveal representation in V2 permanently and markedly elevated stereoacuity threshold, almost to the level observed in two monkeys after removal of the representation in striate cortex of the central 5° of the retina. Posterior infero-temporal ablation also impaired stereoacuity, though less strikingly. In these two monkeys, the effect of subsequent damage to the rostral superior colliculi was examined. One animal was unimpaired. In the other, in which the pre-tectum was damaged, the stereoacuity threshold was substantially raised, most plausibly as a consequence of imperfect binocular fixation.

The organization of binocular cortex in the primary visual area of the rabbit.

AbstractA conscious rabbit which crouches in the ‘freeze’ position has an unequivocal 24° wide binocular field formed by the overlap of the two 12° sectors of uniocular optical field which extend nasal of its midline. Although this investigation reveals that the horizontal representation of the uniocular visual field in cortical visual area I extends more nasal than in earlier reports, it is found to terminate at the midline when the eyes are set in the standard freeze position. The 12° sectors of uniocular field nasal of the midline were not represented in spite of being served by retina. The binocular field of the rabbit in the freeze posture thus appears to have no binocular representation in visual area I.Nevertheless, the presence of a binocular region was confirmed in rabbit visual area I but the projections to it, via the contralateral and ipsilateral eyes, deal with the nonoverlapping sectors of monocular field when the eyes are in the freeze position. The maps of the visual field obtained in one hemisphere via the contralateral and ipsilateral projections were subject to a horizontal divergent disparity of some 18°. The presence of binocular single units in these regions was also confirmed but their two receptive fields were necessarily located in different visual hemispheres and again subject to an 18° mean horizontal divergent disparity. They could not be simultaneously stimulated by any localized feature of an object and are thus precluded from involvement in binocular single vision during the freeze position.The systematic, rather than reportedly random, topography of the ipsilateral projection to visual area I ensured that it could be well fused with the similarly organized contralateral projection by means of an 18° vergence of the eyes from the freeze position. The horizontal receptive field disparity of binocular single units is then brought to a zero mean value which enables their receptive fields through each eye to be simultaneously stimulated by the same part of an image. Tested units then evinced response summation; a minimum requirement for binocular vision. In this equivalent primary position of the eyes, the entire binocular visual field is completely represented in visual area one of both hemispheres.The rabbit thus appears to employ a two‐state cortical system for forward vision. In the freeze position the visual field attains a cyclopean extent of 360° but the receptive fields of cortical binocular units are widely divergent. The classic binocular optical field may then project to some other region of brain which subserves binocular vision. Upon examination of a frontal object the animal must verge its eyes in order to obtain binocular single vision in visual area I; it then temporarily surrenders part of its rear visual field. An explanation in terms of ocular image quality is suggested.