User:Psyc4600/Group8

Neural Correlates of Visual Motion Detection
Motion detection in the cortically blind is a phenomenon that is manifested as the ability to detect motion by patients suffering from damage to the primary visual cortex that has resulted in cortical blindness. Generally, it involves the detection of the presence of a moving stimulus, with the ability to discriminate the direction of movement being more difficult. The primary visual cortex, also known as striate cortex or area V1, contains a topographic map of the visual field, so damage to V1 generally results in blindness in the corresponding part of the visual field. Some patients with damage to V1, though clinically blind due to a scotoma covering their entire visual field, have shown the ability to detect and even discriminate between visual stimuli presented to them. This phenomenon is generally known as blindsight. There are two known types of blindsight: Type 1 blindsight and type 2 blindsight. Patients suffering from type 1 blindsight are not consciously aware of any stimuli being presented to them, but yet are able to predict different aspects of presented visual stimuli at levels that are significantly above chance. Type 2 blindsight involves patients having some awareness of a visual stimuli, but still lacking any sort of visual perception. Blindsight can also be observed in subjects with damage to other areas of the visual system such as visual area V4 or V5. Preserved abilities for the perception and action of and towards an object can be explained by the dissociable pathways of the dorsal and ventral streams, respectively.

Primary Visual Cortex (V1)
Axons of the eye’s retinal ganglion cells exit the retina via the optic nerve, crossing at the optic chiasm. At this point axons from the nasal retina cross over to the other side of the brain while axons from the temporal retina remain on the same side. After passing the optic chiasm, the axons of the retinal ganglion cells become collectively known as the optic tract. The majority of axons of the optic tract then terminate in the lateral geniculate nucleus (LGN), while others project to the superior colliculi, the suprachiasmatic nucleus (SCN), and the pretectum. The LGN serves as the primary relay for visual processing in the cerebral cortex, mostly projecting axons into the visual areas of the occipital cortex. From here, the primary visual cortex (V1), also known as the striate cortex projects axons to other areas of the cerebral cortex (extrastriate) in two major pathways know as the dorsal and ventral stream. The dorsal stream (known as the “where pathway”) includes the middle temporal area and leads from the striate cortex into the parietal lobe. It is thought to be responsible for spatial aspects of vision such as positional awareness and guidance of actions. The ventral stream (known as the “what pathway”) extends from the striate cortex into the inferior part of the temporal lobe. It is strongly connected to the dorsal stream and both the medial temporal lobe and the limbic system. It is believed to be responsible for object identification.

Area V5
Area V5, or middle temporal lobe(MT), is located bilaterally at the occipito-temporal junction of the extrastriate cortex. This area contains cells that are sensitive to motion, and that 90% of them respond preferentially to a particular direction of motion and will not respond at all to the opposite direction of motion. None of which are colour sensitive. V5 is organized in columns: one set defined by movement directionality rather than orientation, and a second set whose cells are sensitive to binocular disparity.

Lesion Study
V5 is an area of choice for examining the effects of lesions within the dorsal pathway, primarily because of its high incidence of directionally selective neurons. Rudolph and Pasternak (1999) used a match-to-sample task to examine in greater detail the MT lesion-induced inability to extract motion from directional noise. They compared the effects of unilateral MT/MST lesions on the ability of monkeys to discriminate the direction of motion of random-dots or of drifting gratings masked by noise. On each trial, the monkeys indicated whether the two sequentially presented stimuli, sample and test, moved in the same or in different directions. The delay separating the two comparison stimuli was very brief (200 milliseconds) thus, the task imposed only minimal requirements on the ability to remember the preceding stimulus. Deficits in direction discrimination were found with both types of stimuli, but no permanent deficits were found in contrast thresholds for discriminating the direction of drifting gratings measured in an identical task. This selectively increased susceptibility to noise was specific to the domain of motion perception, since the same monkeys showed no deficit in discrimination the orientation, and were not found when the monkeys were required to discriminate stimulus direction, even when not found when the monkeys were required to discriminate stimulus direction, even when the motion stimulus was masked by noise. These selective results demonstrated the importance of MT neurons for the ability to discriminate motion direction in the presence of noise.

Dorsal and Ventral Stream Distinction


From early visual processing in the occipital cortex, two important pathways can be distinguished that may be specialized for different types of information. First there is the dorsal route (or the 'where' pathway), which includes the middle temporal area and leads from the primary visual cortex into the parietal lobe. It is thought to be responsible for spatial aspects of vision such as positional awareness and guidance of actions. There is also the ventral stream (or the 'what' pathway), which extends from the primary visual cortex into the inferior part of the temporal lobe. It is strongly connected to the dorsal stream and both the medial temporal lobe and the limbic system. It is believed to be responsible for object identification. The following contains distinctions between both routes in its visual processing centres.

Realtime vs Memory
There is a distinction between the relative timing of influence of the dorsal and ventral stream. A study by Striemer et al. (2009) studied the patient CB, a 75 year old man with a left field hemianopia caused by a right occipital stroke. This study tested the hypothesis that visuomotor systems in the dorsal stream work optimally in real-time, however it is the ventral stream that is engaged when action is driven by memory They employed a reaching task in which CB was required to reach from a start button to target strip as fast as possible with the right hand, while avoiding obstacles located either in or out from the midline, in his sighted (right) or blind (left) field, or in both. CB was required to wear PLATO goggles which were open for 500 ms at the beginning of each trial. CB performed similar to controls when objects were placed in his right (sighted) field. He was also sensitive to the position of obstacles in his blind visual fields, shown by his reaches being significantly right when obstacles were placed close to the midline. The reach trajectories showed a clear separation, with significantly different endpoints of the reach. However, when a 2 s delay was inserted after the PLATO goggles closed (CB was instructed not to reach until he heard an auditory “go” signal from his headphones) CB was no longer sensitive to the position of obstacles in his blind field, although in his sighted field he retained this ability. This provides support that the ventral stream is involved in retrieving visual information from memory, as the damage to this stream in CB’s case impairs his ability to perceive objects in his blind field after a delay. However, his dorsal stream is intact and thus on this real-time basis, in which direct input is received from the retina, CB remains sensitive to objects in his blind field.

Vision for Action versus Vision for Perception
Both streams process information about the structure of objects and about their spatial locations, and both are subject to the modulatory influences of attention. However there is a difference between the two when it comes to visual experiences. The ventral stream transforms visual input into perceptual representations that embody the characteristics of objects and their spatial relations, whereas the dorsal stream tends to use the visual input to mediate visual control of skilled actions, such as reaching and grasping, directed at objects in the world. From this description the terms ‘vision for action’ and ‘vision for perception’ were coined to the ventral and dorsal stream, respectively. Milner and Goodale (2008) have further explored this notion as follows.

Perception is deemed the conscious visual experience that we have about the current stimulus that can be translated into a subjective report. However, the concept needs to be extended to include ‘unconscious’ or ‘preconscious’ perception of objects and events. This refers to mental representations that potentially could reach conscious awareness. Perception represents our visual experience of the world, but not that it provides the direct foundation for action. Action is considered the identification of possible and actual goal objects, and the selection of an appropriate course of action to deal with those objects, within the ventral stream. It provides visual information to enable the identification of a goal object such as a baseball, and to enable other cognitive systems to plan the action of picking up that ball. Vision for action includes the use of visual information in the detailed programming and real-time control at the level of elementary movements. It also works only in real time and is not normally engaged unless the target object is visible during the programming phase, which is when bottom-up visual information is being converted into the appropriate motor commands. The dorsal stream, vision of perception, uses this bottom-up information to specify the required movement parameters such as the trajectory of the reach and the required grip needed to grasp the target object.

Allocentric versus Egocentric
The dorsal pathway projects from the striate to the posterior parietal cortex. Several subcortical structures are associated with this stream and are critical for online visuomotor control or goal directed action including the superior colliculus and the cerebellum. The ventral stream, involved in visuospatial perception involves areas within the occipito-temporal cortex and is associated with the medial temporal lobe which is involved in long term storage. Carey et al. (2006), dicuss the distinctions between the two streams in the way objects are encoded. In the dorsal stream, objects are encoded egocentrically meaning that a target object is encoded relative to the observer. This co-ordinate based system is specific to the dorsal steam. On the other hand, allocentric processing of visual information is associated with the ventral stream.

A study by Cary et al. (2006), examined the participant DF, a right handed woman who suffered from brain damage due to carbon monoxide poisoning. This caused her to have a condition called visual agnosia in which she was impaired in her percetual judgements of objects including their size, shape and orientation. However, DF's color perception as well as visuomotor tasks remained intact. This study by Carey et al. (2006), examined DF's ability to use allocentric information for guidance of grasping movements. More specifically, they looked at her ability to adjust her grasp to take hold of objects through holes drilled in them (two or three), as opposed to holding them at the edges. When three holes were used, DF was completely insensitive to the orientation of fingers, however, when there were two holes DF was able to adjust hand orientation appropriately, however was unable to alter grip to distance between the two holes. This study found that DF was impaired in the use of information about the distance between two wholes in guiding her grasping distance, however was able to use the information between herself and the wholes in order to correctly orient her hands. This suggests that she has an impairment using allocentric cues, but egocentric spatial cues are still intact. This makes sense due to the fact that DF has an intact dorsal stream and thus egocentric processing is assumed to be intact. In allocentric tasks, her visuospatial perceptual deficit can be seen. Both DF and SB (also has visual agnosia) unconsciously takes account of the locations of the objects when making a reaching movement, similar to the controls. They both showed an impairments in bisecting the space between two objects. Also, a patient with bilateral optic ataxia consisting of damage the the dorsal stream shows an opposite pattern to DF, that being- impaired implicit object avoidance with intact bisection between two objects.



In the bisection task by Carey et al. (2006), subjects were required to make allocentric judgements in relation to the midpoint between objects. In the implicit obstacle avoidance task, hand path is related to the positions of non-target objects in egocentric spatial co-ordinates. There are different kinds of visuospatial processing for perceptual as opposed to visuomotor tasks, again reiterating the differences between vision for perception and vision for action. This double dissociation between these patients provides indirect support for the link between the dorsal stream and egocentric spatial processing and the ventral stream and allocentric processing.

In the current study by Carey et al. (2006), they assessed DF's visuomotor responses based on allocentric spatial representations. Coloured tokens were used and participants were told to copy exact spatial positions of tokens presented in the test array. They were also required to count the number of tokens, and estimate which pairs of tokens were closest and farthest apart. In the pantomime condition of this experiment participants were required to point to locations on a blank sheet identical to locations of targets on an adjacent stimulus array. DF was impaired in her use of allocentric coordinate information but performed well on tasks which may require coordinate-level spatial processing, nearest and farther apart pair judgments. The data from this study confirm that DF has good coordinate spatial information which guides pointing movements towards individual visual stimulus in an array in front of her. This is possible due to her intact dorsal stream, allowing for processing of egocentric spatial cues. However, allocentric tasks reveal a visuospatial perceptual deficit.

Visual Impairments
Damage to the dorsal/ventral stream is often associated with visual impairments. These visual impairments detrimentally effect the patients ability to visually recognize objects, both implicitly and explicitly, and to perform a visual motor task. By assessing these impairments, we can get a better idea of specific instances and how they relate to motion detection.

Visual Agnosia
Visual agnosia is a failure of adequate response to visual stimulation in the absence of absolute scotomata. There is an impaired ability to recognize objects from vision, despite intact basic visual processes.

Case of DF
Milner and colleagues (1991) presented DF with a letter box in which the orientation of a slot can be rotated. DF had problems matching orientation of slot to visually presented alternatives but was able to reach towards slot and orient hand appropriately to actually post object. This suggests a dissociation between visual perception (based on the impaired ventral stream) and visual control of action (using the spared dorsal stream). When DF was given a more complex object, for example a T-shape with appropriate slot, she was still moderately accurate but did indeed miss a few times, which tended to be at 90 degrees. This suggests that action is driven by orientation of a single edge therefore the dorsal route cannot adequately integrate different edges into whole objects.

Optic Ataxia
Optic ataxia is the impairment of visually guided reaching. This is due to damage to the posterior-superior parietal lobes. It reflects failure to transform visual perceptual information into appropriate motor commands. It is mostly, but not always, restricted to a particular hand that is opposite of the lesion, or the particular hand when in a particular side of space. The latter situation is unlikely to be purely motoric or purely visual but due to failure to integrate both (bad hand in bad side of space).

Hemianopia
Hemianopia is the complete loss of one hemisphere that leads to complete blindness in the corresponding half of the visual field. It is often associated with problems of space exploration, with increased latencies for eye movements towards the blind hemifield, and an almost random rather than systematic visual search of the space contralateral to the lesion

Smaller lesions lead to quadrantanopia, the loss of vision in one quadrant of the visual field.

Blindsight
Blindsight is the ability to react in a goal-directed way to stimuli presented within this blind part of the visual field. The visual analysis succeeds (fails to fail) since the patient is still able to detect, localize, and analyse the stimulus, even if this analysis is not available to conscious insight. Patients usually are not pleased at all if asked to respond to stimuli that they do not subjectively see.

In a study by Stoerig and Fahle 1995, they presented two dots in the visual field and asked the patients, with scotomata, to discriminate between direction of motion of two dots; one below and one above the scotoma. The patient always saw two dots that were presented within the intact visual field and hence had no exceptional problem answering the question regarding motion direction between these two dots, even if the direction was not obvious. This was also done with three dots which yielded better results even though the additional third point was presented in the middle of scotoma and hence was never consciously perceived. There was better motion discrimination in three dots even in temporal field, with the third dot presented in scotoma.

Striemer et al. (2009) tested if inputs from the primary visual cortex (V1) to the dorsal stream are needed for obstacle avoidance. They presented CB with visual targets simultaneously in his sighted and blind fields, and found he was faster to respond to a response button when this was the case, rather than when a single target was presented in his sighted field. This suggests that although he was not consciously aware of the presence of the object in his blind field, its combined input with an object in his sighted field facilitated the bottom up processing of the object. This implicit visual processing exemplifies the phenomenon of blindsight.

Motion Detection in the Cortically Blind
Motion detection is mediated by two dissociable pathways:


 * a route from the retina to area V5 via subcortical nuclei that bypasses area V1 and is specialized for processing fast motion


 * indirect cortical route via area V1 for processing slow motion

There is the notion of dynamic parallelism that allows us to being to comprehend how cortically blind patients are able to detect motion. There’s a double dissociation in which patients with lesion affecting V1 are able to detect and discriminate reliably the direction of the object at greater speeds, but not at lower speeds. And the same is opposite for patients with bilateral lesions of area V5 in which they can detect and discriminate the direction of movement at lower speeds, but not at greater speeds.

The dorsal pathway or 'where’ pathway projects from early visual areas to the posterior parietal cortex, and is traditionally thought to use vision for spatial perception. Others, however, believe that it is not restricted to spatial processing in and of itself,but more so in the guidance and control of actions using visual information, or “how” to interact with an object. The second visual pathway, the ventral, 'what' stream projects from early visual areas to the inferotemporal (IT) courtex. This stream computes vision for object perception.

A study by Fiehler et al. (2009), tested the hypothesis that the dorsal pathway can evolve in the absence of visual information. The evidence that dorsal stream activity is associated with working memory maintenance of kinesthetic information, and also that this pathway supports control of action by integrating all movement relevant signals, not solely vision, was the basis for their research. This study applied fMRI to study the dorsal pathway contributions to action control of kinaesthetic movement in congenitally blind humans and sighted controls. A large activation overlap was found for the found large activation overlap in congenitally blind and sighted participants in areas of the dorsal stream during performance and working memory maintenance of kinesthetically guided hand movements. These results provide support that the development of the dorsal action system is not restricted to visual experience during the development of an individual. In this experiment, participants layed in the scanner, and used their right arm to reach a pen apparatus in order to trace different line patterns. There was an overlap in dorsal activation patten in sighted controls, specifically the working memory maintenance of kinaesthetic information activated the left anterior intraparietal sulcus. There are mutual interactions between primary sensory cortex and superior parietal cortex. The superior parietal cortex in dorsal pathway is involved in online action control, and the superior parietal cortex active during hand and finger movements requiring kinaesthetic control. In the anterior intraparietal sulcus (also in dorsal pathway region) which is an area that transforms sensory input into motor commands, allowing for control of action, and continuously updates the ongoing movement with respect to the aspired movement goal. The activation varied with memory load as seen by an increase in the hemodynamic response with larger mnemonic demands.


 * Key: substantial overlap of activation of dorsal pathway structure (anterior intraparietal sulcus & superior parietal cortex) in CB and SC participants. All in all this supports the assumption that dorsal pathway functions related to contrl of immediate and delayed actions do not depend on early visual input, absence of vision..

Some patients with brain damage affecting primary visual cortex, though clinically blind in their field defects, can still discriminate visual stimuli when forced choice procedures are used. In a study by Azzopardi and Cowey (2001) it was found that patients were sensitive to moving stimuli in their scotomata. In more detail, the patients could detect movement in any kind of stimulus, and could discriminate the direction of single bars, for example, but cannot discriminate in more complex stimuli. Motion discrimination impairment caused by brain damage affecting V1 is inconsistent with the proposed existence of a subcortical pathway to extrastriate cortical motion areas, such as MT, which bypasses the VA and is specialized for analysing fast motion.

In the Azzopardi and Cowey (2001), they focused on three subjects:
 * G.Y. A 42 year old man with unilateral lesion in left medial occipital cortex due to a car accident
 * G.M. A 69 year old man who suffered a mild stroke, destroying his left ventral striate cortex, causing contralateral right superior quadrantanopia
 * S.P. A 43 year old woman who suffered a stroke at the age of 23, resulting in a left upper quadrantanopia

All the patients were presented with stimuli such as first-order bars, random dot kinematograms, first-order plaids and gratings...etc, at different speeds. All three subjects could detect every type of motion and discriminate direction easily in normal part of visual field, but GM and GY could detect isolated first-order moving bars in ‘blind’ area, and were best at intermediate speeds. SP was relatively impaired at detecting isolated moving bars compared with the other two subjects for reasons unknown. They could all discriminate first-order, 100% coherent random dot kinematograms from static ones. GY and GM could discriminate moving stimulus from static one, SP was not tested. GM and GY could reliable discriminate the direction of isolated first-order moving bars, but not in random dot kinematograms, which was the easiest condition. The fact that they can’t discriminate direction in the random dot kinematograms is consistent with finding that the MT of monkeys with V1 lesions, which are sensitive to the direction of moving bars presented in the scotoma, are nevertheless insensitive to the direction of motion of random dot kinematograms, gratings and plaids. Low-level spatiotemporal mechanisms are completely disabled as result of damaged V1, and motion perception is therefore completely reliant on feature-tracking mechanisms.