The Brain Mechanisms Of Information Binding

Integrating different and discrete information into coherent object and event is anessential function of the human brain. My thesis work is investigating the bindingproblem in perception and working memory with behavioral, EEG and fMRIapproaches. The first experiment investigates whether neural synchronization isengaged in binding of verbal and spatial information in working memory. Weanalyzed oscillatory power and phase synchronization of electroencephalography(EEG) recorded from subjects performing a working memory task. Subjects wererequired to maintain both verbal (letters) and spatial (locations) information of visualstimuli while the verbal and spatial information were either bound or separate. Wefound that frontal theta power, and large scale theta phase synchronization betweenbilateral frontal regions and between the left frontal and right temporal-parietalregions were greater for maintaining bound relative to separate information. However,the same effects were not observed in the gamma band. These results suggest thatworking memory binding involves large scale neural synchronization at the theta band.The second experiment investigates whether illusory contour (IC) perception involvesearly induced gamma activity. The temporal binding hypothesis proposes that visualfeature binding is achieved by neuronal synchronization. Nevertheless, the existinghuman neurophysiological evidence for the neuronal synchronization in visual featurebinding - the oscillatory induced gamma activity (IGA) is under suspicion, becausethe previously observed IGA occurs at a later stage after 200 ms thus leading to theobjection that IGA is related to some later top-down processes rather than the earlyperceptual processing. However, the present EEG study identified an IGA as early as90 ms after stimulus onset stronger for a Kanizsa-type illusory contour than for acontrol stimulus, suggesting that IGA is involved in IC perception at a relatively earlystage. The IC is a classical example of feature binding. Therefore, the finding of theearly IGA withstands the doubts about the role of IGA in visual feature binding andprovides the new early human evidence for the temporal binding hypothesis. The thirdexperiment investigates whether the binding processing and the object representation are dissociable. This was investigated in the present study by recoding EEG and fMRI when subjects performed a feature binding task. Kanizsa-type illusory counters (IC) and corresponding control stimuli were presented with the support ratio being 0.45 or 0.75. The binding related activity should decrease with the support ratio while the object related activity should increase with the support ratio. This dissociation was found in EEG/ERP results at occipital sites. We found that the induced gamma activity occurring at about 100～150 ms after stimulus onset decreased with the support ratio, and the ERP component including P1 and N170 increased with the support ratio. The fMRI results identified the object related activity at the lateral occipital cortex. These results suggest the dissociable binding processing and the object representation, which are respectively represented by the induced and evoked activities at occipital sites. The LOC object related BOLD activation as well as the absence of binding related brain region in the fMRI results may indicate that the binding processing and the object representation involve the same brain regions while different neuronal populations are responsible for them respectively.