SYMPOSIA...
|
| Mechanisms of Visual Attention |
| Jason B. Mattingley
- The University of Queensland, Australia:
Role of human parietal cortex in selective attention and awareness
The human parietal lobe receives input from each of the major sensory modalities, and is involved in the earliest stages of motor planning. Parietal neurons are an important source of attentional control signals in the brain, allowing information to be prioritized in the service of awareness and goal-directed action. I will discuss three recent studies from our laboratory which reveal some key properties of the human parietal cortex in selective attention. First, I show that acquired lesions of the parietal lobe result in a spatial gradient of perceptual impairment which is maximal at extreme contralesional locations. Crucially, the steepness of this pathological gradient increases with attentional load in a central task. Second, I show that stimulation of subregions of the inferior parietal lobule can selectively disrupt the ability to reorient attention in space, and that this effect follows a timecourse that implies distinct ‘early’ and ‘late’ parietal influences on selection. Third, I consider the potential effect of parietal control mechanisms on the representation of sensory information at the earliest stages of cortical processing. During a task requiring selection of rapidly sequential visual targets, we find that the widely reported ‘attentional blink’ is associated with attenuation of neural activity in the primary visual cortex. Taken together, our findings suggest a variety of roles for the human parietal lobe in modulating both spatial and non-spatial aspects of perception. |
|
Bruno van Swinderen - The University of Quensland, Australia
Neural correlates of visual attention in Drosophila melanogaster
The genetic model organism Drosophila melanogaster has been used to study mechanisms of learning and memory for thirty years. In contrast, selective attention has only very recently been addressed in this model, in part because most memory paradigms were olfactory while attention is best studied with visual paradigms. Nevertheless, flies do seem to display attention-like properties. Neural correlates of selective attention in flies were identified by recording brain activity during responses to competing visual stimuli. Brain responses displayed characteristic properties of attention such as salience effects, expectancy, and alternation dynamics. These attention-like processes appeared to be disrupted in short-term memory mutants, or when synaptic output from key brain structures or systems was transiently silenced. By exploiting Drosophila molecular-genetic tools within brain recording paradigms, we hope to investigate the basic architecture supporting visual perception and attention-like processes in a small model brain. |
|
Roger Remington - The University of Queensland, Australia:
Interacting systems in attentional and oculomotor capture
An external event can sometimes trigger an involuntary shift attention (attentional capture), or an involuntary saccade (oculomotor capture), toward the source of stimulation despite the intention to avoid being distracted. Accounts of this involuntary orienting differ in whether top-down modulation of such involuntary orienting can occur (e.g., Folk, Remington, & Johnston, 1992) or whether the response is completely stimulus-driven (e.g., Theeuwes, 1992). Evidence is emerging suggesting that multiple systems underlie involuntary orienting and that an understanding of their characteristics could clarify the roles of top-down and bottom-up mechanisms. I discuss experiments that allow comparison of top-down modulation of oculomotor capture and attentional capture under similar stimulus conditions. These experiments find a dissociation between attention and saccades: stimulus conditions that produce capture of attention do not produce capture of saccades, and vice versa. This dissociation suggests two interacting brain systems, one associated with the bottom-up orienting via pathways subserving saccades, the other with top-down-mediated orienting, via pathways that link orienting to task demands. I discuss the empirical evidence for distinguishing these systems, and how this distinction can help reconcile theoretical disputes. |
|
Philip Smith - The University of
Melbourne, Australia
The Computational Dynamics of Visual Attention
The simplest attentional task, detecting a cued stimulus in an otherwise empty visual field, produces unexpectedly complex data. Cues interact strongly with spatial uncertainty and visual masks, and there is a dissociation in the effects of these variables on sensitivity and response time (RT). I describe a theory that accounts for these effects at the level of the RT distributions. The theory links visual encoding, masking, spatial attention, visual short term memory (VSTM), and perceptual decision making in an integrated dynamic framework. Stimuli processed by early visual filters are encoded in VSTM under the control of spatial attention. The VSTM trace drives a diffusion process decision mechanism that determines the speed and accuracy of responses. VSTM trace formation is described by a two-channel shunting model, in which selection into VSTM depends on the total energy in the stimulus, but the decision in based on form information carried by an independent parallel pathway. The computational principles of the VSTM model are strikingly similar to the idea, currently influential in neuroscience, that visual experience is the confluence of information carried by parallel "where" and "what" pathways. |
|
Alan
Kingstone - University of British
Columbia, Canada
Rethinking human attention
Studies of human attention routinely seek to isolate "volitional" (top-down) processes from "reflexive" (bottom-up) processes. The Posner Cuing Paradigm represents the "state of the art" model task for this research enterprise. It is fair to say that this paradigm has been at the very centre of modern investigations of human attention, being applied in all the major disciplines of cognitive and social neuroscience to investigate the behavioural and the neural mechanisms of human attention-including, but not limited to, animal studies, developmental studies, patient studies, and functional neuroimaging investigations. Indeed, the results have formed the very building blocks of how we think about human attention. Unfortunately, this model task and its findings are fundamentally flawed. In a series of investigations I will show that this paradigm has failed entirely to isolate volitional attention. I will also argue that it has failed to provide an adequate way to compare volitional and reflexive attention. Finally, I will show how this model task can be fixed so that it can provide a solid theoretical and empirical basis for future studies of volitional and reflexive attention.
| | |
