Looming sensitive cortical regions without V1 input: evidence from a patient with bilateral cortical blindness

Human subcortical pathways automatically detect collision trajectory without attention and awareness

Detecting imminent collisions is essential for survival. Here, we used high-resolution fMRI at 7 Tesla to investigate the role of attention and consciousness for detecting collision trajectory in human subcortical pathways. Healthy participants can precisely discriminate collision from near-miss trajectory of an approaching object, with pupil size change reflecting collision sensitivity. Subcortical pathways from the superior colliculus (SC) to the ventromedial pulvinar (vmPul) and ventral tegmental area (VTA) exhibited collision-sensitive responses even when participants were not paying attention to the looming stimuli. For hemianopic patients with unilateral lesions of the geniculostriate pathway, the ipsilesional SC and VTA showed significant activation to collision stimuli in their scotoma. Furthermore, stronger SC responses predicted better behavioral performance in collision detection even in the absence of awareness. Therefore, human tectofugal pathways could automatically detect collision trajectories without the observers' attention to and awareness of looming stimuli, supporting "blindsight" detection of impending visual threats.

Subjective time dilation as a result of entropy

Subjective time dilation is an effect discovered using the oddball paradigm, where expanding visual stimuli, but not shrinking ones, induce a prolongation of subjective temporal perception compared to static stimuli. This disparity is often seen as another demonstration of humans' evolved reaction to approaching threats, since visual expansion resembles approaching potential threats and warrants extra attention. In this study, we show that by manipulating the relative sizes of stimuli, both expanding and shrinking stimuli can induce prolongation of subjective time in an oddball paradigm. We propose an alternative explanation based on information theory, linking subjective temporal perception to entropy, an objective property of stimulus that is information uncertainty. Temporal function as a logistic function thus serves as the framework inside which content functions like information processing are able to operate coherently.

Convolutional neural networks for vision neuroscience: significance, developments, and outstanding issues

Convolutional Neural Networks (CNN) are a class of machine learning models predominately used in computer vision tasks and can achieve human-like performance through learning from experience. Their striking similarities to the structural and functional principles of the primate visual system allow for comparisons between these artificial networks and their biological counterparts, enabling exploration of how visual functions and neural representations may emerge in the real brain from a limited set of computational principles. After considering the basic features of CNNs, we discuss the opportunities and challenges of endorsing CNNs as in silico models of the primate visual system. Specifically, we highlight several emerging notions about the anatomical and physiological properties of the visual system that still need to be systematically integrated into current CNN models. These tenets include the implementation of parallel processing pathways from the early stages of retinal input and the reconsideration of several assumptions concerning the serial progression of information flow. We suggest design choices and architectural constraints that could facilitate a closer alignment with biology provide causal evidence of the predictive link between the artificial and biological visual systems. Adopting this principled perspective could potentially lead to new research questions and applications of CNNs beyond modeling object recognition.

Exogenous Attention to Emotional Stimuli Presenting Realistic (3D) Looming Motion

Previous research shows that dynamic stimuli, on the one hand, and emotional stimuli, on the other, capture exogenous attention due to their biological relevance. Through neural (ERPs) and behavioral measures (reaction times and errors), the present study explored the combined effect of looming motion and emotional content on attentional capture. To this end, 3D-recreated static and dynamic animals assessed as emotional (positive or negative) or neutral were presented as distractors while 71 volunteers performed a line orientation task. We observed a two-phase effect: firstly (before 300 ms), early components of ERPs (P1p and N2po) showed enhanced exogenous attentional capture by looming positive distractors and static threatening animals. Thereafter, dynamic and static threatening distractors received enhanced endogenous attention as revealed by both late ERP activity (LPC) and behavioral (errors) responses. These effects are likely explained by both the emotional valence and the distance of the stimulus at each moment.

Shallow neural networks trained to detect collisions recover features of visual loom-selective neurons

Animals have evolved sophisticated visual circuits to solve a vital inference problem: detecting whether or not a visual signal corresponds to an object on a collision course. Such events are detected by specific circuits sensitive to visual looming, or objects increasing in size. Various computational models have been developed for these circuits, but how the collision-detection inference problem itself shapes the computational structures of these circuits remains unknown. Here, inspired by the distinctive structures of LPLC2 neurons in the visual system of Drosophila, we build anatomically-constrained shallow neural network models and train them to identify visual signals that correspond to impending collisions. Surprisingly, the optimization arrives at two distinct, opposing solutions, only one of which matches the actual dendritic weighting of LPLC2 neurons. Both solutions can solve the inference problem with high accuracy when the population size is large enough. The LPLC2-like solutions reproduces experimentally observed LPLC2 neuron responses for many stimuli, and reproduces canonical tuning of loom sensitive neurons, even though the models are never trained on neural data. Thus, LPLC2 neuron properties and tuning are predicted by optimizing an anatomically-constrained neural network to detect impending collisions. More generally, these results illustrate how optimizing inference tasks that are important for an animal's perceptual goals can reveal and explain computational properties of specific sensory neurons.

Neural Mechanism of Blindsight in a Macaque Model

Some patients with damage to the primary visual cortex (V1) exhibit visuomotor ability, despite loss of visual awareness, a phenomenon termed "blindsight". We review a series of studies conducted mainly in our laboratory on macaque monkeys with unilateral V1 lesioning to reveal the neural pathways underlying visuomotor transformation and the cognitive capabilities retained in blindsight. After lesioning, it takes several weeks for the recovery of visually guided saccades toward the lesion-affected visual field. In addition to the lateral geniculate nucleus, the pathway from the superior colliculus to the pulvinar participates in visuomotor processing in blindsight. At the cortical level, bilateral lateral intraparietal regions become critically involved in the saccade control. These results suggest that the visual circuits experience drastic changes while the monkey acquires blindsight. In these animals, analysis based on signal detection theory adapted to behavior in the "Yes-No" task indicates reduced sensitivity to visual targets, suggesting that visual awareness is impaired. Saccades become less accurate, decisions become less deliberate, and some forms of bottom-up attention are impaired. However, a variety of cognitive functions are retained such as saliency detection during free viewing, top-down attention, short-term spatial memory, and associative learning. These observations indicate that blindsight is not a low-level sensory-motor response, but the residual visual inputs can access these cognitive capabilities. Based on these results we suggest that the macaque model of blindsight replicates type II blindsight patients who experience some "feeling" of objects, which guides cognitive capabilities that we naïvely think are not possible without phenomenal consciousness.

Preserved extrastriate visual network in a monkey with substantial, naturally occurring damage to primary visual cortex

Lesions of primary visual cortex (V1) lead to loss of conscious visual perception with significant impact on human patients. Understanding the neural consequences of such damage may aid the development of rehabilitation methods. In this rare case of a Rhesus macaque (monkey S), likely born without V1, the animal's in-group behaviour was unremarkable, but visual task training was impaired. With multi-modal magnetic resonance imaging, visual structures outside of the lesion appeared normal. Visual stimulation under anaesthesia with checkerboards activated lateral geniculate nucleus of monkey S, while full-field moving dots activated pulvinar. Visual cortical activation was sparse but included face patches. Consistently across lesion and control monkeys, functional connectivity analysis revealed an intact network of bilateral dorsal visual areas temporally correlated with V5/MT activation, even without V1. Despite robust subcortical responses to visual stimulation, we found little evidence for strengthened subcortical input to V5/MT supporting residual visual function or blindsight-like phenomena.

Neuronal mechanisms of motion detection underlying blindsight assessed by functional magnetic resonance imaging (fMRI)

Brain imaging offers a valuable tool to observe functional brain plasticity by showing how sensory inputs reshape cortical activations after a visual impairment. Following a unilateral post-chiasmatic lesion affecting the visual cortex, patients may suffer a contralateral visual loss referred to homonymous hemianopia. Nevertheless, these patients preserve the ability to unconsciously detect, localize and discriminate visual stimuli presented in their impaired visual field. To investigate this paradox, known as blindsight, we conducted a study using functional magnetic resonance imaging (fMRI) to evaluate the structural and functional impact of such lesion in a 33-year old patient (ML), who suffers a complete right hemianopia without macular sparing and showing strong evidences of blindsight. We thus performed whole brain and sliced thalamic fMRI scan sequences during an event-related motion detection task. We provided evidence of the neuronal fingerprint of blindsight by acquiring and associating neural correlates, specific structures and functional networks of the midbrain during blindsight performances which may help to better understand this condition. Accurate performance demonstrated the presence of residual vision and the ability to unconsciously perceive motion presented in the blind hemifield, although her reaction time was significantly higher in her blind-field. When the normal hemifield was stimulated, we observed significant contralateral activations in primary and secondary visual areas as well as motion specific areas, such as the supramarginal gyrus and middle temporal area. We also demonstrated sub-thalamic activations within the superior colliculi (SC) and the pulvinar. These results suggest a role of secondary subcortical structures in normal spontaneous motion detection. In a similar way, when the lesioned hemifield was stimulated, we observed contralateral activity in extrastriate areas with no activation of the primary lesioned visual cortex. Moreover, we observed activations within the SC when the blind hemifield was stimulated. However, we observed unexpected ipsilateral activations within the same motion specific areas, as well as bilateral frontal activations. These results highlight the importance of abnormal secondary pathways bypassing the primary visual area (V1) in residual vision. This reorganization in the structure and function of the visual pathways correlates with behavioral changes, thus offering a plausible explanation for the blindsight phenomenon. Our results may potentially impact the development of rehabilitation strategies to target subcortical pathways.

Visual Cortex: The Eccentric Area Prostriata in the Human Brain

Human area Prostriata is a small, unstudied portion of the visual brain set deep in the calcarine sulcus, next to V1. A recent neuroimaging study in humans indicates that this area is specialized to respond to rapidly moving stimuli in the far periphery, consistent with single-unit responses in other mammals.

Autism Pathogenesis: The Superior Colliculus

After been exposed to the visual input, in the first year of life, the brain experiences subtle but massive changes apparently crucial for communicative/emotional and social human development. Its lack could be the explanation of the very high prevalence of autism in children with total congenital blindness. The present theory postulates that the superior colliculus is the key structure for such changes for several reasons: it dominates visual behavior during the first months of life; it is ready at birth for complex visual tasks; it has a significant influence on several hemispheric regions; it is the main brain hub that permanently integrates visual and non-visual, external and internal information (bottom-up and top-down respectively); and it owns the enigmatic ability to take non-conscious decisions about where to focus attention. It is also a sentinel that triggers the subcortical mechanisms which drive social motivation to follow faces from birth and to react automatically to emotional stimuli. Through indirect connections it also activates simultaneously several cortical structures necessary to develop social cognition and to accomplish the multiattentional task required for conscious social interaction in real life settings. Genetic or non-genetic prenatal or early postnatal factors could disrupt the SC functions resulting in autism. The timing of postnatal biological disruption matches the timing of clinical autism manifestations. Astonishing coincidences between etiologies, clinical manifestations, cognitive and pathogenic autism theories on one side and SC functions on the other are disclosed in this review. Although the visual system dependent of the SC is usually considered as accessory of the LGN canonical pathway, its imprinting gives the brain a qualitatively specific functions not supplied by any other brain structure.

Functional reorganization of population receptive fields in a hemispherectomy patient with blindsight

Blindsight refers to the ability of some patients with destruction of the primary visual cortex (V1) to respond to stimuli presented in their clinically blind visual field despite lack of visual awareness. Here we tested a rare and well-known patient with blindsight following hemispherectomy, DR, who has had the entire cortex in the right hemisphere removed, and in whom the right superior colliculus is the only post-chiasmatic visual structure remaining intact. Compared to more traditional cases of blindsight after damage confined to V1, the study of blindsight in hemispherectomy has offered the invaluable opportunity to examine directly two outstanding questions: the contribution of the intact hemisphere to visual processing without awareness, and the nature of plastic and compensatory changes in these remaining contralesional visual areas. Population receptive field (pRF) mapping was used to define retinotopic maps, delineate the boundaries between the visual areas, examine changes in the sizes of receptive field centres within each visual area, and their variability as a function of eccentricity. Aside from the dorsal visual areas showing blurred borders between V2d and V3d, not otherwise detected with perimetric mapping, the retinotopic maps of DR did not differ substantially from those of three matched healthy controls. Interestingly, those dorsal compartments showed a significant increase in the RF sizes toward values typical of higher-order processing cortices, while no differences were observed in the corresponding ventral visual areas. Findings showed that whereas receptive field sizes at foveal and parafoveal eccentricities (≤ 4°) were not measurably altered, the pRF size increased by ~ 270% at 4-6° of eccentricity, and the size difference reached ~ 300% between 8° and 10°. We interpret these findings to suggest that an increase in pRF sizes could be indicative of cerebral plasticity involving the retinotopic reorganization of the dorsal visual areas.

Psychophysical and neuroimaging responses to moving stimuli in a patient with the Riddoch phenomenon due to bilateral visual cortex lesions

Patients with injury to early visual cortex or its inputs can display the Riddoch phenomenon: preserved awareness for moving but not stationary stimuli. We provide a detailed case report of a patient with the Riddoch phenomenon, MC. MC has extensive bilateral lesions to occipitotemporal cortex that include most early visual cortex and complete blindness in visual field perimetry testing with static targets. Nevertheless, she shows a remarkably robust preserved ability to perceive motion, enabling her to navigate through cluttered environments and perform actions like catching moving balls. Comparisons of MC's structural magnetic resonance imaging (MRI) data to a probabilistic atlas based on controls reveals that MC's lesions encompass the posterior, lateral, and ventral early visual cortex bilaterally (V1, V2, V3A/B, LO1/2, TO1/2, hV4 and VO1 in both hemispheres) as well as more extensive damage to right parietal (inferior parietal lobule) and left ventral occipitotemporal cortex (VO1, PHC1/2). She shows some sparing of anterior occipital cortex, which may account for her ability to see moving targets beyond ~15 degrees eccentricity during perimetry. Most strikingly, functional and structural MRI revealed robust and reliable spared functionality of the middle temporal motion complex (MT+) bilaterally. Moreover, consistent with her preserved ability to discriminate motion direction in psychophysical testing, MC also shows direction-selective adaptation in MT+. A variety of tests did not enable us to discern whether input to MT+ was driven by her spared anterior occipital cortex or subcortical inputs. Nevertheless, MC shows rich motion perception despite profoundly impaired static and form vision, combined with clear preservation of activation in MT+, thus supporting the role of MT+ in the Riddoch phenomenon.

Subcortical pathways to extrastriate visual cortex underlie residual vision following bilateral damage to V1

Residual vision, or blindsight, following damage to the primary visual cortex (V1) has been investigated for almost half a century. While there have been many studies of patients with unilateral damage to V1, far fewer have examined bilateral damage, mainly due to the rarity of such patients. Here we re-examine the residual visual function and underlying pathways of previously studied patient SBR who, as a young adult, suffered bilateral damage restricted to V1 which rendered him cortically blind. While earlier work compared his visual cortex to healthy, sighted participants, here we consider how his visual responses and connections compare to patients with unilateral damage to V1 in addition to sighted participants.

Detection of drifting Gabor patches of different contrasts (1%, 5%, 10%, 50% and 100%) was tested in SBR and a group of eight patients with unilateral damage to V1. Performance was compared to the neural activation in motion area hMT+ measured using functional magnetic resonance imaging. Diffusion tractography was also used to determine the white matter microstructure of the visual pathways in all participants.

Like the patients with unilateral damage, patient SBR showed increased % BOLD signal change to the high contrast stimuli that he could detect compared to the lower contrast stimuli that were not detectable. Diffusion tractography suggests this information is conveyed by a direct pathway between the lateral geniculate nucleus (LGN) and hMT+ since this pathway had microstructure that was comparable to the healthy control group. In contrast, the pathway between LGN and V1 had reduced integrity compared to controls. A further finding of note was that, unlike control participants, SBR showed similar patterns of contralateral and ipsilateral activity in hMT+, in addition to healthy white matter microstructure in the tract connecting hMT+ between the two hemispheres. This raises the possibility of increased connectivity between the two hemispheres in the absence of V1 input.

In conclusion, the pattern of visual function and anatomy in bilateral cortical damage is comparable to that seen in a group of patients with unilateral damage. Thus, while the intact hemisphere may play a role in residual vision in patients with unilateral damage, its influence is not evident with the methodology employed here.

•

Bilaterally hemianopic patient SBR has neural patterns like unilateral patients.

•

hMT+ activity increases with stimulus contrast and better stimulus detection.

•

Like in unilateral patients, the pathway between LGN and hMT+ is intact in SBR.

Gamma oscillations in the superior colliculus and pulvinar in response to faces support discrimination performance in monkeys

The subcortical visual pathway including the superior colliculus (SC), pulvinar, and amygdala has been implicated in unconscious visual processing of faces, eyes, and gaze direction in blindsight. Our previous studies reported that monkey SC and pulvinar neurons responded preferentially to images of faces while performing a delayed non-matching to sample (DNMS) task to discriminate different visual stimuli (Nguyen et al., 2013, 2014). However, the contribution of SC and pulvinar neurons to the discrimination of the facial images and subsequent behavioral performance remains unknown. Since gamma oscillations have been implicated in sensory and cognitive processes as well as behavioral execution, we hypothesized that gamma oscillations during neuronal responses might contribute to achieving the appropriate behavioral performance (i.e., a correct response). In the present study, we re-analyzed those neuronal responses in the monkey SC and pulvinar to investigate possible relationships between gamma oscillations in these neurons and behavioral performance (correct response ratios) during the DNMS task. Gamma oscillations of SC and pulvinar neuronal activity were analyzed in three phases around the stimulus onset [inter-trial interval (ITI): 1000ms before trial onset; Early: 0-200ms after stimulus onset; and Late: 300-500ms after stimulus onset]. We found that human facial images elicited stronger gamma oscillations in the early phase than the ITI and late phase in both the SC and pulvinar neurons. Furthermore, there was a significant correlation between strengths of gamma oscillations in the early phase and behavioral performance in both the SC and pulvinar. The results suggest that gamma oscillatory activity in the SC and pulvinar contributes to successful behavioral performance during unconscious perceptual and behavioral processes.

Activation of the Human MT Complex by Motion in Depth Induced by a Moving Cast Shadow

A moving cast shadow is a powerful monocular depth cue for motion perception in depth. For example, when a cast shadow moves away from or toward an object in a two-dimensional plane, the object appears to move toward or away from the observer in depth, respectively, whereas the size and position of the object are constant. Although the cortical mechanisms underlying motion perception in depth by cast shadow are unknown, the human MT complex (hMT+) is likely involved in the process, as it is sensitive to motion in depth represented by binocular depth cues. In the present study, we examined this possibility by using a functional magnetic resonance imaging (fMRI) technique. First, we identified the cortical regions sensitive to the motion of a square in depth represented via binocular disparity. Consistent with previous studies, we observed significant activation in the bilateral hMT+, and defined functional regions of interest (ROIs) there. We then investigated the activity of the ROIs during observation of the following stimuli: 1) a central square that appeared to move back and forth via a moving cast shadow (mCS); 2) a segmented and scrambled cast shadow presented beside the square (sCS); and 3) no cast shadow (nCS). Participants perceived motion of the square in depth in the mCS condition only. The activity of the hMT+ was significantly higher in the mCS compared with the sCS and nCS conditions. Moreover, the hMT+ was activated equally in both hemispheres in the mCS condition, despite presentation of the cast shadow in the bottom-right quadrant of the stimulus. Perception of the square moving in depth across visual hemifields may be reflected in the bilateral activation of the hMT+. We concluded that the hMT+ is involved in motion perception in depth induced by moving cast shadow and by binocular disparity.

The superior colliculus is sensitive to gestalt-like stimulus configuration in hemispherectomy patients

Patients with cortical blindness following a lesion to the primary visual cortex (V1) may retain nonconscious visual abilities (blindsight). One intriguing, though largely unexplored question, is whether nonconscious vision in the blind hemifield of hemianopic patients can be sensitive to higher-order perceptual organization, and which V1-independent structure underlies such effect. To answer this question, we tested two rare hemianopic patients who had undergone hemispherectomy, and in whom the only post-chiasmatic visual structure left intact in the same side of the otherwise damaged hemisphere was the superior colliculus (SC). By using a variant of the redundant target effect (RTE), we presented single dots, patterns composed by the same dots organized in quadruple gestalt-like configurations, or patterns of four dots arranged in random configurations, either singly to the intact visual hemifield or bilaterally to both hemifields. As reported in a number of prior studies on blindsight patients, we found that bilateral stimulation yielded faster reaction times (RTs) than single stimulation of the intact field for all conditions (i.e., there was an implicit RTE). In addition to this effect, both patients showed a further speeding up of RTs when the gestalt-like, but not the random shape, quadruple patterns were projected to their blind hemifield during bilateral stimulation. Because other retino-recipient subcortical and cortical structures in the damaged hemisphere are absent, the SC on the lesioned side seems solely responsible for such an effect. The present results provide initial support to the notion that nonconscious vision might be sensitive to perceptual organization and stimulus configuration through the pivotal contribution of the SC, which can enhance the processing of gestalt-like or structured stimuli over meaningless or randomly assembled ones and translate them into facilitatory motor outputs.