The Study On Neural Mechanism Of Crowding Effect

We are exposed to a large number of visual information every day. How we recognize and select useful objects from the cluttered information, it is an important question in visual research. However, in the peripheral visual field, objects can be identified in isolation but are more difficult to be recognized when surrounded by other objects, this effect is called crowding. Crowding is a general phenomenon in vision, and understanding crowding will help to understand the mechanisms of object recognition.The crowding has been researched for nearly one century. However, the mechanisms underlying crowding are still as poorly understood as they are today. A lot of research using behavior research techniques and functional magnetic resonance imaging (fMRI) to investigate the mechanisms of crowding, and proposed many theories to explain the crowding effect, such as faulty-integration model, centroid model, compulsory averaging model, Positional uncertainty proposal and attentional proposal and so on. However, each of these theories cannot explain all of the results of studies, and they all lack of direct evidence. The feature-based processing mechanism suggests that the features from targets and flankers are averaged compulsorily or erroneously combined into a jumbled percept and thus the crowded targets cannot be perceived correctly. While the Attentional theory and Positional uncertainty proposal suggests that the crowded target can be perceived as a correct stimulus. By summarizing the previous research, this study assumed that the crowded multi-feature target can be perceived as a correct object, but it was restrict to access to conscious awareness.In this study, the ERP technology was used to explore the mechanism of crowding effect. The study one investigated whether the crowded multi-feature stimulus can be integrated into a correct object, and it included 3 experiments. Experiment 1 investigated whether the crowded Chinese words can be detected accurately, and provides reasonable parameters for the following experiments. Experiment 2 employed the semantic priming paradigm to investigate whether the crowded prime Chinese word can produce semantic priming effect (indexed by N400 component). Experiment 3 further investigated whether the crowded target word can be integrated into a correct word and produce semantic priming effect. The study two investigated why the crowded stimulus cannot be reported, and it included 2 experiments. Experiment 4 used Across-modal delayed response oddball task, and investigated the processing of the crowded stimulus and explore whether it can be accessed to conscious perception. Experiment 5 employed Oddball paradigm to investigate whether the crowded stimulus can induce P3 component, and provide a further evidence for the accessing of crowded stimulus to conscious perception. The results indicated that:(1) Under the crowded condition, same as the single feature stimuli, the multi-feature stimuli can be integrated into correct objects, and the semantic information of these stimuli can be accessed and induced significant semantic priming effect.(2) Under the uncrowded condition, the P2 ERP component was induced when the uncrowded Chinese word was recognized and it mainly distributed in the anterior scalp, and recognizes the uncrowded word will activate the left occipital lobe, parietal lobe and frontal area. While the P2 ERP component in the crowded condition mainly distributed in the posterior scalp, and recognized the crowded word only activated the left occipital lobe, this suggests that the crowded information cannot be accessed to conscious perception.(3) The uncrowded low probability Chinese word can induce significant P3 component, while the crowded low probability Chinese word cannot induce P3 component, it indicates that the crowded word cannot be perceive consciously. Combining with previous research and the current results, we suggest that the crowded multi-feature stimuli can be integrated correctly, but the flankers will be served as background noise and degrade the visibility of the crowded stimuli, it may degrade the visibility of the whole of stimuli or degrade the visibility of part of features, and thus induce the crowded stimuli cannot be accessed consciously.This study firstly used the Event-related potentials (ERP) to investigate the mechanism of crowding effect. We found that the crowded multi-feature stimulus can be integrated into a correct object, but it cannot be accessed consciously. These results questioned the feature-based processing mechanism, and provide evidence for the attentional theory. However, the current study only reveals the tip of the iceberg of the crowding, and a lot of problems need to be discussed in the future study.