视觉客体识别是人类一项重要的生存技能，我们通过不断累积客观知识才能适应更丰富的生存和生活需要。腹侧枕颞叶视觉皮层被认为是参与客体识别的重要脑区。视觉颜色、形状等信息在腹侧枕颞叶视觉通路的初级视觉皮层及外纹状皮层进行处理，更复杂的语义信息在如颞叶前部、额叶等高级区域进行处理。随着研究的深入，人们发现在视觉客体识别过程中也存在着“自上而下”的“指导”作用，其中一种表现为过去的经验和知识会对视觉客体的识别与表征产生影响。然而以往关于客体知识经验对视觉客体表征的研究多为行为实验，在脑机制方面的研究仍有待更丰富、更全面和直接系统的探索。本研究希望更系统地对客体先验知识对知觉加工区视觉客体神经表征的调控进行考察。为此，本研究以腹侧枕颞叶视觉皮层中的颜色加工区为主要研究对象，通过探讨客体知识对该区域视觉客体神经表征的影响，揭示客体知识经验与客体知觉表征之间的关系。研究设计了常见蔬菜客体自然颜色组，及非自然颜色组两种实验条件，在脑激活水平和激活模式两种维度去考察自然颜色组和非自然颜色组引起的差异，并认为对这两种实验条件有差异表现反映了客体颜色经验对认知判断的影响，这些有差异的区域即为受到客体颜色知识调控的脑区。同时从功能连接的角度探究对初级加工区域的调控作用是否来自更高级的区域。针对脑激活水平的差异比较（研究一），我们对比自然颜色组和非自然颜色组实验条件进行了组水平的单变量分析，分别在全脑和感兴趣区（颜色加工区）进行差异比较，差异显著的区域即为受到客体知识经验调控的脑区。针对激活模式的差异比较（研究二），我们采用支持向量算法，训练分类器区分自然组和非自然组的客体多变量激活模式，在全脑上和感兴趣区中去检测分类的正确率，显著高于随机水平的脑区被认为受到客体知识经验的影响。针对初级皮层受调控区域和高级区域的交互关系（研究三），我们以研究二中找到的多变量模式分类成绩显著的区域作为种子点，用心理-生理交互（Psycho-physiological Interaction, PPI）的方法计算受调控的初级区域和其他区域在任务中的交互关系。 研究一的单变量分析结果一定程度上揭示了这种不同，距状皮层在两种条件上的激活差异趋势正是受到客体经验知识的影响。 研究二从客体多变量表征模式的角度，考察了相对初级的视觉区域受到的经验调控状况，发现位于大脑后部包括距状皮层、舌回、梭状回在内的腹侧枕颞叶皮层，以及位于大脑前部包括双侧腹内侧前额叶、左侧额下回、左侧颞叶前部在内的额颞皮层，视觉客体的神经表征模式受到了客体知识的影响。此外，研究较为全面的总结了这些区域各自对知觉特征与客体知识经验的依赖程度，并发现从后部枕颞叶到前部额颞叶呈现了客体知识经验作用逐渐增加、知觉信息编码逐渐减少的功能组织模式，进一步帮助我们理解了客体识别的过程。 研究三试图从功能连接角度去观察初级知觉皮层受到的调控是否来自高级区域的信息反馈，但遗憾的是我们尚未找到呈现自然颜色和非自然颜色客体时与初级知觉皮层有功能连接差异的脑区，可能受限于此方法对精细差异的探测还相对困难。未来的研究或许可以进一步精细地探讨客体经验对初级皮层的影响反馈具体借由何种途径来实现。 综上，本研究使用单变量分析、多变量模式分析和PPI分析的方法在腹侧枕颞叶皮层发现初级视觉加工区的神经表征受到客体与颜色交互经验的影响，并且各区域间对经验调控存在一定的过渡规律，呈现出由后侧枕颞叶向前部额颞叶知觉特征作用由强到弱、客体知识经验作用由弱到强的功能组织模式，对我们进一步理解客体识别有很好的帮助。

Visual object recognition is the ability to perceive an object’s physical properties and apply sematic attributes to it, which is an important survival skill for human acclimatization by acquiring knowledge about the world. Based on brain injury studies in patients and primates, researchers found the ventral occipital-temporal cortex(VOTC) a critical region for object recognition. And visual recognition processing has been typically viewed as a bottom-up hierarchy in which information is processed sequentially with increasing complexities(Marr, 1976). For instance, color, shape and other object information are processed in the primary visual cortex and external striate cortex of the VOTC visual pathway, while more complex semantic information is processed in advanced regions such as the anterior part of temporal lobe and frontal lobe. Meanwhile, there are also top-down regulations from high-level regions, shown as that past experience will have an impact on the cognitive process(Humphreys, Riddoch, & Price, 1997). However, these studies mainly focused on behavior effects, and the exploration of the role of object knowledge in the perception process in the human brain is not yet comprehensive and direct. Thus, we expect to examine the neural mechanisms of perceptual changes caused by prior knowledge of objects more comprehensively. Besides, this study tried to explore the extent to which the various areas of object recognition are empirically modulated, in order to better understand the process of object identification. Furthermore, from the point of view of functional connectivity, it is worth exploring whether the modulation from object experience to perception is achieved by means of enhanced interaction between advanced regions and the low-level visual areas. Thus this study aims to identify the influence of object knowledge on neural representation through perception within primary visual regions of VOTC, and to focus on the neural changes of color perception. Specifically, we designed natural color group and unnatural color group to test the differences between the two experimental conditions in both brain activation (Study 1) and neural pattern (Study 2). It is believed that the differences in the performance of these two conditions reflects the influence of object*color experience on cognitive judgements. And we further explored whether the regulation towards color perception comes from advanced areas using PPI (Psycho-physiological Interaction) algorithm (Study 3). Study 1 investigated how object*color experience modulates the brain activation. Univariate result to some extend revealed the difference in brain activation between two conditions in calcarine cortex, which reflected the influence by the object’s experience. Study 2 examined the empirical regulation on neural representation of primary visual regions within VOTC based on object multivariate patterns. We found that in the posterior part of the brain, including the calcarine cortex, lingual gyrus, and fusiform gyrus of the ventral occipital-temporal cortex, as well as in the anterior part of the brain, including the bilateral ventral medial prefrontal lobe, the left inferior frontal gyrus, and the left anterior temporal lobe of the frontal-temporal cortex, the neural representation of visual objects is influenced by object knowledge. Thus, this study summarized regulated areas, in addition, their dependence on object experience, and from the occipital-temporal areas to frontal-temporal areas, an increased dependence of object experience was found. Study 3 intended to observe whether the modulation on neural representation of the primary sensory cortex came from feedback of high-level areas. However, we have not yet found any brain areas showing differentiate functional connectivity with defined ROIs when presenting natural and unnatural color objects. It may be difficult to detect such high fine differences using PPI method, and further research could keep exploring how object experience feedback to primary visual areas can be achieved through more elaborate ways. In summary, this study used univariate analysis, multivariate pattern analysis, and PPI analysis to identify the how the experience of interactions between object and color affects the neurological representation of primary visual processing areas within ventral occipital-temporal cortex. And our results suggest a certain transitional law, showing a trend of increased dependence on the experience, and at the same time, a decreased dependence trend on perceptual attributes, from occipital-temporal lobe in the posterior brain to frontal-temporal lobe in the anterior brain, which are helpful for us to better understand visual object recognition.