Using S cones to investigate the role of superior colliculus in automatic visual processes

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

Of the two main visual pathways from the retina, the geniculostraite to the cortex and the retinotectal to the superior colliculus, it is the latter that has become associated with automatic and non-conscious aspects of vision and visually guided behaviour. The retinotectal pathway is widely accepted to mediate reflexive saccades, exogenous attentional orienting and inhibition of return (IOR). More recently it has been suggested to subserve the automatic processing of fear-conditional stimuli and fearful faces. However, functional dissociation of the human retinotectal and geniculostraite pathways is very difficult to achieve. Here, we propose to effectively create a functional lesion of the human retinotectal pathways by employing stimuli visible only to short-wave sensitive (S) cones, to which the superior colliculus is blind. Using this novel test for retinotectal mediation, we have begun to show that the established view is incorrect. Our preliminary data show that S cone stimuli are quite capable of eliciting exogenous orienting of attention, as well as IOR. Thus both aspects of automatic attentional orienting seem not to require signals in the retinotectal pathway. By contrast, S cone stimuli did not cause reflexive saccade effects, consistent with retinotectal mediation of saccade reflexes. We propose to use S cone stimuli to investigate the role of the superior colliculus in automatic orienting of gaze and attention, as well as in automatic fear processing. We will first test behaviourally whether S cone stimuli can interfere with reflex eye movements at all, whether cortical pathways contribute to reflex eye movements as well as to automatic attention, by titrating stimulus onset times to allow for relative delays in the S cone pathway. We will then test proposals that the frontal and supplementary eye fields (FEF and SEF) are responsible for suppressing reflex eye movements. If S-cone stimuli do not evoke reflexive saccades, we would predict that in conditions requiring reflex suppression, such as the antisaccade task, S cone stimuli will produce less fMRI activation then luminance stimuli. Converging evidence will be provided by studying the eye movements of two patients with focal lesion of the SEF. Using a similar combination of s cone stimuli, high resolution eye tracking during fMRI and patients with focal cortical lesions, we will test whether IOR is mediated by the superior colliculus and whether it is generated by the suppression of eye movements, as is widely believed. Our preliminary evidence suggests that there are distinct cortical and collicular mechanisms of IOR, but exactly how distinct these components might be, and whether either is associated with saccade inhibition, remains to be established. Lastly, we will use similar techniques to test the recent and controversial theory that fearful faces can be processed via the superior colliculus, without the involvement of visual cortex. Masked fear-conditioned vs non fear-conditioned S cone and grey scale faces will be tested for GSR, differential papillary constriction, simple activation of the amygdala, and correlated activation of amygdala, pulvinar and superior colliculus. Investigation of patients with hemianopia may provide a potential double dissociation with S cone stimuli. In the process of these studies, we will further characterise the S cone technique to provide a powerful psychophysical tool for achieving functional dissociation between human visual pathways.