Neuroanatomy of Control

1. Neuropsychology: Neglect and Extinction

There are undoubtedly several cortical and subcortical areas that participate in the control of spatial attention. The pulvinar nucleus of the thalamus, for example, is involved in filtering or suppressing irrelevant stimuli in a cluttered display. However, the cortical region that plays the most significant role in the control of spatial attention is the posterior parietal region. Damage to the parietal region (especially the right parietal region) in humans results in a profound attentional impairment referred to as neglect. Neu-ropsychological patients with neglect fail to pay attention to stimuli falling on the side of space opposite the lesion (the contralesional side). For example, a patient with damage to the right parietal lobe may fail to acknowledge a person sitting on the left, may fail to eat food on the left half of a plate, and may fail to read words on the left half of a page. Neglect occurs soon after damage to the parietal region. As a patient recovers and the neglect becomes less severe, patients can process a single stimulus presented in the con-tralesional visual field. These recovering patients show another disorder, however, referred to as extinction: When two stimuli are presented simultaneously in opposite visual fields, patients will extinguish, or fail to notice, the stimulus in the contralesional field. In other words, extinction patients exhibit neglect of contrale-sional stimuli only in the presence of ipsilateral stimuli. Both neglect and extinction appear to be disruptions of the ability to control spatial attention and deploy it to the contralesional field.

Neglect can be observed in spatial precuing tasks. In these tasks, extinction patients can detect and identify targets presented in the contralesional visual field; furthermore, these patients can use exogenous precues to allocate attention to the contralesional field prior to the appearance of the target. However, when a precue appears in the ipsilesional field and a target appears in the contralesional field, these patients are much slower to detect the target than when the contralesional field is cued and the target appears in the ipsilesional field. That is, they are impaired primarily when they are cued to the good field and then the target appears in the bad field. A straightforward interpretation of these results is that the contralesional and ipsilesional sides of space compete for attention. A bottom-up factor, such as a spatial precue, can bias the competition in favor of the cued field. The effect of parietal damage is to weaken the ability of the contralesional field to compete for attention. When both the cue and the target are in the same field, there is no competition and the target can be detected quickly. However, when the good field is cued and the target appears in the bad field, the good field wins the competition for attention even though the target is in the bad field, leading to abnormally slow responses.

What aspect of attentional control, bottom-up or top-down, is disrupted in these patients? Although parietal-damaged patients appear to have intact perceptual processing, the disorder of attention appears to involve bottom-up control parameters: Attention is not captured effectively by contralesional inputs. Furthermore, some forms of top-down atten-tional control appear to be intact in parietal-damaged patients. These patients can make use of top-down expectancies or task-relevant goals. For example, a contralesional stimulus may not be extinguished if the ipsilesional stimulus is irrelevant to a task and the patient is instructed to ignore this ipsilesional stimulus. Presumably, the top-down control of attention is intact in these patients and biases attention to select the relevant item in the bad field.

2. Neuroimaging Studies

Neuroimaging studies also indicate a central role for the posterior parietal lobe in spatial selection. Because the whole brain can be examined using some of these techniques, other brain regions that influence spatial attention can be observed. Observing multiple neural sites simultaneously may be useful for isolating the sources of bottom-up and top-down control.

Separate neuroanatomical sources for two forms of control is suggested by positron emission tomography (PET) studies of performance in the spatial precuing task. Observers were presented with sequences of visual targets that appeared in a predictable sequence that would engage endogenous attentional allocation. The predictable sequences were leftward or rightward appearances of the target; the target first appeared near fixation to the left or right and then continued to move in the same direction in a majority of the trials. For example, if the first target first appeared slightly to the right of fixation, the second target would likely occur to the right of the first target's position; similarly, the third target would likely occur to the right of the second target, and so forth. A predictable sequence allows observers to anticipate the next target location and endogenously allocate attention to that expected location. Thus, the peripheral targets in this task involve both exogenous and endogenous components—the appearance of the target is an exogenous luminance change and the predictable sequence allows observers to anticipate the next target and allocate attention endogenously. Two neural regions of interest exhibited increased blood flow during this task: the superior parietal lobe near the postcentral sulcus (near Brodmann's area 7) and the superior frontal cortex (near Brodmann's area 6).

To distinguish the superior parietal and superior frontal areas by their sensitivity to the exogenous and endogenous components, observers performed a control task. This control task presented the same peripheral targets in a predictive sequence, but instead ofdetecting these peripheral targets observers detected targets presented at fixation. Exogenous orienting would occur to the peripheral targets, even though these targets were irrelevant to the observers' task. However, endogenous attention would be directed to the central targets because these were the task-relevant stimuli. When endogenous shifts of attention were eliminated in this manner, the superior frontal areas were no longer active, but the superior parietal areas continued to be active. These results suggest that superior parietal regions are involved in the exogenous, bottom-up control of spatial attention and that superior frontal regions are involved in the endogenous, top-down control of spatial attention.

The bottom-up control of spatial attention coordinated by the superior parietal lobe appears to involve spatial selection only. If other visual attributes are selected, such as a color or a shape, the superior parietal lobe does not appear to show increased blood flow as measured by PET, although other extrastriate visual areas are activated in response to different visual attributes. For example, in searching for a target defined by color, blood flow increased in the left dorsolateral occipital lobe and in the left collateral sulcus. Searching for a target defined by movement or by shape resulted in increased blood flow in other extrastriate areas. Thus, there appears to be a large network of extrastriate visual areas that each mediate bottom-up aspects of attention to different stimulus attributes. Furthermore, PET results suggest that each of these areas could receive feedback from frontal lobe areas—areas that may represent task-relevant goals useful for allowing one stimulus attribute (e.g., color) to be selected from an image containing many attributes (e.g., color, shape, and movement).

There may exist some extrastriate visual regions that participate in attentional control across different visual attributes. Recent functional magnetic resonance imaging (MRI) results have found activation in two parietal lobe areas across three very different attention tasks: (i) a spatial shifting task similar to that described previously, (ii) an object matching task in which observers reported whether two attended objects were the same or different, and (iii) a nonspatial conjunction task in which observers searched for a target letter in a sequential stream of colored letters. These two parietal subareas seem to be involved in a wide range of visual selection, contrary to the PET results discussed previously that demonstrated no parietal involvement in visual search for targets defined by color, shape, or motion.

However, there is a resolution to the apparent discrepancy between a general attentional involvement of parietal areas and a spatial-specific role for parietal areas: Parietal lobe attention areas may control the suppression of visual distractors. Functional MRI studies that exhibit parietal activation across attention tasks required irrelevant stimuli to be ignored; PET visual search studies that did not exhibit parietal activation across tasks involved displays containing only task-relevant stimuli. In the biased competition account, parietal lobe areas may receive feedback from frontal lobe areas that allow parietal regions to suppress distractors and act as a "gate" for other extrastriate visual areas.



Is there a cause or cure for autism? The Complete Guide To Finally Understanding Autism. Do you have an autistic child or know someone who has autism? Do you understand the special needs of an autistic person?

Get My Free Ebook

Post a comment