We studied the visual field distribution of speed and accuracy of manual responses to small brief light flashes, in patients with left hemineglect or extinction resulting from right hemisphere vascular lesions and in brain-damaged and healthy control subjects. All patients with right hemisphere lesions showed a greater impairment in both the speed of response and the detection rate in the contralesional than in the ipsilesional hemifield. This interfield difference increased with the eccentricity of stimulus presentation and was especially pronounced in neglect patients who showed a paradoxical increase in speed of response and detection rate at increasingly larger eccentricities in the ipsilesional hemifield. We hypothesize that both the contralesional slowing down and the ipsilesional speeding up of the response depends upon an exaggerated gradient of attention towards the ipsilesional hemifield. To assess whether these abnormalities concern automatic or controlled attentional processes, in a second experiment, we manipulated the predictability of the side of the stimulus presentation by using blocked rather than randomized stimulus presentations. This resulted in a speeding up of responses in both hemifields thus showing that the patients were able to focus attention to the side of stimulus presentation voluntarily. However, there was no modification of the contra-ipsilesional differences which, therefore, are likely to be related to abnormal automatic processes rather than controlled attention.

The aim of this study was to investigate the neural correlates of the functional distinction underlying attentional mechanisms of endogenous-sustained and exogenous-transient spatial selection. We recorded event related potentials (ERPs) and used functional magnetic resonance imaging (fMRI) in separate experiments while subjects performed a simple reaction time (RT) to the same visual stimulus displayed to one of several field locations. Endogenous-sustained or exogenous-transient focusing of attention onto target location were obtained by presenting the stimulus in blocks of same-point vs. randomised-point trials, respectively. Same-point stimuli yielded overall faster RT than randomised stimuli, indicating a facilitating effect of endogenous-sustained spatial attention on the perceptual processing of the impending stimulus. Moreover, same-point vs. randomised presentations revealed significant increases in the fMRI signal in the bilateral lingual and fusiform gyri as well as in the right calcarine sulcus, in conjunction with a larger amplitude of the posterior P1 component of ERPs, but no modulation of the amplitude of the N1 component. Rather, a larger amplitude of N1 was found in the reverse contrast, randomised minus same-point trials, which revealed increases in the fMRI signal along the posterior left superior frontal sulcus and bilaterally in the superior precuneus. These findings indicate that N1 indexes exogenous orienting of attention and is likely to represent the activity of frontal and parietal components of the attention network involved in eliciting attention changes. In contrast, the effects of those changes, resulting in a modulation of activation in visual occipital areas, are indexed by P1.

No abstract

The control of visuo-spatial attention entails the joint contribution of goal-directed (endogenous) and stimulus-driven (exogenous) factors. However, little is known about the neural bases of the interplay between these two mechanisms. To address this issue, we presented endogenous (spatially informative) and exogenous (noninformative) visual cues sequentially within the same trial (double-cue paradigm) during fMRI, crossing factorially the validity of the two cues. We found that both endogenous and exogenous cues affected behavioral performance, speeding-up or slowing-down target discrimination when valid and invalid, respectively. Despite the double-cue paradigm maximizes the interplay between endogenous and exogenous factors, the two types of cue affected responses in an independent manner without any significant effect of congruence. The imaging data revealed increased activation in separate cortical areas following invalid endogenous and invalid exogenous cues. A fronto-parietal system was activated during invalid endogenous trials, whereas a region at the temporo-occipital junction was activated during invalid exogenous trials. Within both circuits, activity was unaffected by the validity of the other cue. These results indicate the existence of separate, noninteracting neural circuits for endogenous and exogenous reorienting of visuo-spatial attention.

No abstract