The Study On Neural Mechanism Of Inhibition Of Return In Visual Modality

Since Posner et al (1984) first observed the inhibition of return effect (IOR), there are a lot of studies focus on how does IOR affect information processing and what mechanisms underline it. Early theory of IOR suggested that IOR acts on the processing of target at cued location by biasing attention away from the previously attended location and consequently, impairs perceptual processing of the target at that location, resulting in slower responses. A complementary view is that apart from the impairment of perceptual processing at the cued location, the stimulus-response mapping or perception-response link is also temporarily interrupted or blocked at this location, resulting in slower responses (Inhibitory tagging). The current study is aim to test directly whether inhibitory tagging acts to disconnect the stimuli-response mapping temporarily.In experiment 1, we employed the typical cue-target paradigm and presented Stroop stimuli at the cued or uncued location. Such a design would enable us to examine the modulation of spatial cueing on the conflict processing-related N450 effect. Behaviorally, the Stroop interference effect tended to be smaller under the cued condition than under the uncued condition. Electro-physiologically, the conflict-related N450 effect emerged later and had a smaller size at the cued location than at the uncued location. The delay of the N450 suggests that the processing of target is temporally blocked at the previously attended location.In experiment 2, we used a two response-key version of the Stroop task, with two colors associated with each response key. Such a task allowed for breaking the typical Stroop effect down into the semantic and the response components. Behaviorally, the IOR is interacted with the semantic levels of conflict rather than the response levels of conflict. And consistent with this, IOR is interacted with the N450 related to the semantic conflict rather than related to the response conflict processing. The results, therefore, suggests that inhibitory tagging may work on the semantic representation rather than the stimuli-response mapping.In experiment 3, an fMRI study was carried out to isolate the neural underpin of the inhibitory tagging and IOR. While the semantic conflict related areas, including the dorsal ACC medial frontal gyrus and intraparietal lobe had smaller activity under the cued than uncued condition, the response conflict related areas, including the DLPFC showed a reversed pattern of activity. The results are consistent with the hypothesis that inhibitory tagging acts on processes before the response processing (e.g., the semantic processing).The experiment 3 also observed two dissociable mechanisms underline IOR. While the orbitofrontal cortex and the hippocampus gyrus had smaller activity under the cued than uncued condition, the frontal eye field and supplement eye field area expressed reversed patterns of activity. Give the former areas is usually involved in encoding new information and episodic memory and the later areas is believed to be correlated with the orenting of attention, the result consistent with a recently object file based theory that IOR result from a competition between integrating the target into the old cue opened object file in cued locations and building a new object file in uncued locations.In sum, the current study provides a direct evidence for the exit of temporary inhibitory tagging and further suggests that it works on the semantic processing rather than the stimuli-response mapping. Besides, the fMRI study demonstrates an involvement of episodic memory system in the production of IOR and thus provides the first neuroimaging evidence for the object file-based IOR theory.