哺乳动物大脑在漫长的演化过程中维持着基本对称的左右双半球形态。大脑半球镜像区域为同伦脑区，具有功能相似性，称为功能同伦。胼胝体、前连合等白质纤维束为功能同伦提供了跨物种保守的解剖和结构连接基础，为理解人类心理行为的脑机制提供了演化视角。同伦功能连接方法基于同伦区域神经活动时间序列的同步性来量化功能同伦水平，以往研究已据此揭示了功能同伦水平在表征人脑功能组织及其发育与健康等方面的有效性。然而，该方法缺乏同伦脑连接理论的指导，未能对功能同伦的空间特性和解剖结构等进行考虑，从而限制了从拓扑、空间和时间三个不同维度对功能同伦进行系统研究。本文基于跨物种脑连接组双因子生成理论，提出一种全新的功能同伦图谱方法，暨同伦功能亲和度图谱（研究一），进一步考察了同伦功能亲和度图谱的多模态脑图谱空间关联（研究二），解析了同伦功能亲和度图谱的发育和演化模式（研究三和研究四）。

研究一借鉴社会网络和生态网络研究中“物以类聚、人以群分”的基本连接原则，基于跨物种保守的脑结构连接生成理论，以脑区的全脑空间连接模式表征其功能，进而以同伦脑区空间连接模式的同质性量化其功能的相似性。与社会网络类比，将这一相似性定义为同伦功能亲和度，表征同伦脑区之间的功能亲近程度。本研究首先研发了同伦功能亲和度的快速皮层计算方法，利用“美国人脑连接组计划”和“中国人脑连接组计划”的大规模静息态功能磁共振成像数据，绘制了人脑同伦功能亲和度图谱，揭示出同伦功能亲和度沿初级皮层到联合皮层递减的皮层分布模式。利用个体的重测数据，证实了该图谱方法在个体差异测量中的可靠性。通过选择演化等级最高的颞顶联合区作为感兴趣区域，展示了同伦脑区之间低功能亲和度水平是其空间连接模式差异性的体现，验证了该脑区功能亲和度的空间分布层次能有效体现认知功能的半球偏侧化，个体同伦功能亲和度能有效地预测其语言和社会认知功能表现。上述结果验证了同伦功能亲和度图谱方法对研究半球间信息处理的有效性，为后续研究奠定了方法学基础。

研究二基于不同空间尺度的脑图谱数据库，通过与同伦功能亲和度图谱之间的空间关联分析，解析其潜在的神经生物学基础。本研究在磁共振影像的毫米级尺度上，发现：皮层厚度低的区域，髓鞘化度高，便于信息的快速传递，体现出较高的同伦功能亲和度，反之则体现出较低的同伦功能亲和度，促使其沿初级—联合皮层主梯度方向降低，反映出所进行的半球间信息处理复杂度的提升。进一步在细胞构筑水平的关联分析发现：信息主要输入层（第IV层）中越厚的区域同伦功能亲和度水平越高，而信息输出层（第I/II/III/V/VI层）则呈现相反的趋势，这一趋势在深层皮质（第V/VI层）中更为明显，这为进一步理解各层神经元在半球信息处理中的贡献提供了初步基础。在更微观的神经元信息传递环路中，高转运体密度区域表现出较高的同伦功能亲和度，而高受体密度区域表现出较低的同伦功能亲和度。最后，在基因表达的尺度上发现：遗传调控更强的后侧脑区，表现出更高的同伦功能亲和度，而腹侧及前侧脑区来自遗传的调控更弱，可能使其更易受环境影响，表现出较低的同伦功能亲和度。上述结果为人脑同伦功能亲和度视角下的半球间信息处理过程提供了多层次生物学关联解释。

研究三利用来自“中国人彩巢计划”的加速纵向追踪队列样本及“美国人脑连接组发育计划”的横断面样本，分别计算了不同年龄组人群的同伦功能亲和度图谱，建模了皮层及其大尺度功能网络的同伦功能亲和度发育轨线，发现：随着年龄增加，皮层同伦功能亲和度整体水平逐渐降低，反映了同伦脑区在全脑功能连接模式上的相似性逐渐降低，半球间信息整合需求降低，从而体现出更强的功能偏侧化，这一发育趋势在联合皮层网络（默认网络和额顶网络）最为明显，并且跨样本可重复。在脑区层面，颞顶联合区较低的同伦功能亲和度水平所反映的半球功能特异性在学龄早期就已经存在，并在发育过程中持续增强。最后，智力关联分析揭示出超高智商儿童的同伦功能亲和度具备更快的发育速度，从而实现发育早期高水平半球间信息融合和后期高水平半球间信息分离，体现出高效而复杂的认知功能发展。综合研究二，以上结果为从微观“基因—神经元—环路—结构”到宏观“心理行为”全链条理解学龄儿童青少年脑智发育提供了活体且无创的功能影像学证据，具有多层次的发育生物学内涵。

研究四通过对人、猕猴、大鼠的皮层同伦功能亲和度图谱及其发育模式进行分析，进一步解析功能同伦发育的演化规律。首先绘制了人类、猕猴、大鼠的成年皮层同伦功能亲和度图谱，发现：灵长类皮层同伦功能亲和度均沿着其功能处理的层级而由高到低分布，暨在信息处理功能层级低的皮层中更高，在信息处理功能层级高的皮层则更低，反映出其内在半球间信息处理机制具有演化保守的特性；与之形成对比的是：随着演化等级越高，皮层同伦功能亲和度的空间变化的谱系就越宽，反映了半球间信息处理模式随着演化而趋于丰富多样。通过将发育过程分为儿童期、青春前期、青春后期以及成年早期四个阶段，进一步的跨物种比较功能同伦模式发现：这一演化特征随发育而加强，相比灵长类皮层，大鼠皮层不同区域同伦功能亲和度差异较小，各阶段间发育变化有限，负责复杂信息处理功能的部分区域也表现出了较高水平的功能同伦水平。这一表现可能与皮层演化起源以及不同物种的生态位和功能需求有关，也可能反映了动物活体影像技术的限制（如成像分辨率有限、麻醉的影响等）。

本文基于脑连接组生成理论研发了同伦功能亲和度图谱方法，验证了其作为评估人脑功能同伦影像学指标的有效性，展示了其用于刻画群体和个体间差异、揭示与发育有关的个体内变异以及与演化有关的物种间变异的潜力。本文初步研究表明，同伦功能亲和度图谱方法可作为半球间信息加工功能层级及其认知内涵的有效表征，为解析脑智发育和演化机制提供新的途径。本文以功能同伦为例，表明功能亲和度图谱方法有望为脑功能研究提供新的方法学视角，促进功能影像学在临床、教育等领域的转化应用。

Throughout the lengthy evolutionary process, the mammalian brain has maintained a fundamentally symmetrical form with left and right hemispheres. Homotopic brain areas, known as mirror areas in the two hemispheres, exhibit functional similarity, referred to as functional homotopy. White matter fiber tracts such as the corpus callosum and anterior commissure facilitate information processing between homotopic areas, underpinning the cross-species conservation of functional homotopy's structural basis. This foundation aids in understanding the brain mechanisms behind human psychological behavior from developmental and evolutionary perspectives. Homotopic Functional Connectivity (HFC) assesses the level of functional homotopy based on the temporal synchronicity of homotopic neural activities. Previous studies have demonstrated the effectiveness of HFC in characterizing the functional organization of the human brain and its development and health. However, HFC relies solely on experimental time series data and lacks the guidance of a theoretical framework for brain connectivity, neglecting the spatial and anatomical features of functional homotopy, thus limiting systematic research from topological, spatial, and temporal dimensions. This thesis introduces a novel method for evaluating functional homotopy, the Homotopic Functional Affinity (HFA) (Study 1), based on the dual-factor generative theory of cross-species brain connectivity. Then it examines the association between the HFA maps and multimodal brain maps involving information from different spatial scales (Study 2) and further analyzes the developmental and evolutionary patterns of HFA maps (Studies 3 and 4).

Study 1 employs the foundational principle of "birds of a feather flock together" from social and ecological network research and based on the cross-species conserved structural brain connectivity generative theory, characterizes brain function of homotopic areas by their global spatial connectivity patterns. It quantifies the functional similarity based on the homophily of spatial connectivity patterns of homotopic areas, defining this similarity as HFA, which represents the functional proximity between homotopic areas. This study developed a fast cortical computation method for HFA, utilized large-scale resting-state functional magnetic resonance imaging data from the American and Chinese Human Connectome Projects, and charted the human cortical HFA maps. These maps reveal a cortical HFA distribution pattern that decreases from primary to associative cortices. Then, leveraging the test-retest data, the study confirmed the reliability of the HFA method in measuring individual differences. By focusing on the temporo-parietal junction, an area of the highest evolutionary hierarchy, the study demonstrated that low levels of functional affinity between the homotopic areas reflect their spatial connectivity pattern diversity, validating that this area's functional affinity spatial distribution hierarchy effectively represents the lateralization of the associated cognitive functions. Individual HFA maps can predict functional performance in language and social cognition processes. These findings confirm the method's effectiveness in studying inter-hemispheric information processing, establishing a methodological foundation for follow-up studies.

Study 2 based on brain map databases at different spatial scales, analyzed the spatial correlations between different brain maps and the HFA map, and explored the underlying neurobiological basis. At the millimeter scale of magnetic resonance imaging, regions with lower cortical thickness exhibit higher myelination, facilitating rapid information transmission and exhibited higher HFA, while the opposite regions exhibited lower HFA, indicating increased inter-hemispheric information processing complexity along the "primary-associative" cortical principal gradient. Further laminar architecture correlation analysis showed that areas with thickness in the primary input layer (layer IV) had higher HFA levels, whereas the opposite trend was observed in the output layers (layers I/II/III/V/VI), especially in the deeper cortical layers (layers V/VI). This provides a preliminary basis for understanding the contribution of neurons at different layers to inter-hemispheric information processing. In the micro-scale neuronal information transmission circuits, the analysis found that regions with higher transporter density exhibit higher HFA, whereas areas with high receptor density exhibited lower HFA. Finally, at the genetic expression level, posterior brain regions with stronger genetic regulation showed higher HFA, whereas ventral and anterior brain regions, which are less genetically regulated and more susceptible to environmental influences, displayed lower HFA. These results offer a multi-level biological explanation for the inter-hemispheric information processing process through the lens of the human brain HFA.

Study 3 used accelerated longitudinal samples from the "Chinese Color Nest Project" and cross-sectional samples from the "Human Connectome Project - Development" to calculate the HFA maps for different age groups, modeling the HFA developmental trajectories of cortical and large-scale functional networks. It was found that as age increases, the overall cortical HFA gradually decreases, reflecting a reduced similarity in the global functional connectivity pattern of homotopic brain areas and a diminished need for inter-hemispheric information integration, thus demonstrating stronger functional lateralization. This developmental trend was most pronounced in the associative cortical networks (default mode and frontoparietal control networks) and was consistent across samples. At the regional level, the strong hemispheric functional specialization in the temporo-parietal junction was present in early school age and continued to strengthen with development. Lastly, the analysis of the intellectual association with HFA development showed that children with superior intelligence had a faster pace of HFA development, achieving early high-level inter-hemispheric integration and later high-level inter-hemispheric specialization, indicating efficient and complex cognitive function development. Together with Study 2, these findings provide in vivo, non-invasive functional imaging evidence for understanding brain-mind development in school-aged children and adolescents, offering insights from the micro "gene-neuron-circuit-structure" to the macro "psychological behavior" levels of developmental biology.

Study 4 analyzed the HFA maps and their developmental patterns in humans, macaques, and rats, further elucidating the evolutionary trends of functional homotopy. Initially, the study mapped the cortical HFA of adults in these species, finding that in primates, the HFA distribution aligns with the hierarchy of functional processing, being higher in areas with primary functional processing and lower in areas with higher-level processing, indicating an evolutionary conservation of the inter-hemispheric information processing mechanism. In contrast, as the evolutionary hierarchy increases, the spatial variation of cortical HFA broadens, reflecting a diversification of inter-hemispheric information processing with evolution. Subsequent cross-species comparisons of functional homotopy patterns during four developmental stages—childhood, early adolescence, late adolescence, and early adulthood—showed an intensifying evolutionary trend with development. Compared to primate cortices, rat cortical regions exhibited smaller differences in HFA and limited developmental changes, with some complex information processing areas showing relatively high levels of functional homotopy. This characteristic may relate to the evolutionary origins of the cortex, the ecological niches and the functional needs of rodents compared to primates, or it may reflect the limitations of in vivo imaging technology in animals (resolution, anesthesia).

In conclusion, this thesis developed the HFA method based on the brain connectivity generative theory, validating it as an effective neuroimaging marker for assessing human brain functional homotopy. We demonstrated the potential of HFA in characterizing differences between populations and individuals, revealing intra-individual variations related to development and cross-species variations related to evolution. The preliminary studies in this thesis indicate that the HFA map can effectively represent the functional hierarchy and cognitive content of inter-hemispheric information processing, providing new avenues for analyzing mechanisms of brain-mind development and evolution. Using functional homotopy as an example, this thesis suggests that the functional affinity methods hold promise for providing a new methodological perspective in brain function research, facilitating the translational applications of functional imaging in clinical and educational settings.