Shifting attention within or between feature dimensions. Neural temporal dynamics in human early visual cortex

点击翻译按钮获取中文摘要

Shifting attention within or between feature dimensions is an elementary requirement for many visual processes, including visual search. Currently, the neural temporal dynamics of these shifts are unknown. While the Dimensional Weighting Account claims that shifts between feature dimensions last longer compared to within-dimensional shifts, the finding of a reversed top-down backward progression has led some researchers to hypothesize that shifts towards a feature that is processed upwards in the traditional visual hierarchy (i.e. color) occur earlier than shifts towards features that are processed further down (i.e. orientation). The feature similarity gain model has no clear predictions with regard to shifting times, but one may derive that there are no distinct latency differences. In a number of our studies, we showed that frequency-tagged stimuli eliciting the Steady state visual evoked potential (SSVEP) serve as a powerful tool to uncover temporal dynamics of top-down modulated attentional shifting in early visual cortex. To uncover the temporal dynamics we will present frequency-tagged stimuli that contain two feature dimensions, such as color and orientation. Cues will prompt participants to shift attention either within (i.e. color to color) or between feature dimensions (i.e. color to orientation). Time course analysis of SSVEP amplitudes provides a near-real-time measurement of shifting dynamics in early visual cortex as a function of time after a shifting cue. Knowing the anatomical activation patterns related to frequency-tagged stimuli is crucial for testing the controversial hypotheses regarding attention shifting. To test whether attending to a certain feature of a frequency-tagged stimulus will result in a selective activation of areas with high affinity to that feature, we will conduct (i) an attention task combined with retinotopic mapping in a 7 Tesla scanner, and (ii) combined fMRI-EEG recordings (at 3 Tesla). The first will provide information about area-specificity relative to the nature of the attended/ignored features at a fine anatomical scale, within the extend visual cortex. At present, it is unclear whether the increase in SSVEP amplitude with attention is based on a general facilitation effect, i.e. an increase of activation in all visual areas that are activated by the flickering stimuli, or if attending to a specific feature results in a feature-specific activation involving a subset of visual areas. Knowing these processes however is crucial for a clear interpretation of temporal shifting time differences between vs. across feature dimensions on the basis of SSVEP amplitudes. We are convinced that at the end of the proposed project, we can resolve the controversy mentioned above with regard to the mechanism mediating feature-based attention shifts. We will thus make a critical and novel contribution to our understanding of the neural mechanisms underlying feature-based visual selective attention.