Connectionist modeling is one approach to understanding human intelligence using simulated networks of neuron-like processing units. In this article, we report on recent progress in connectionist models that simulate empirical data of higher-level cognitive processes, these being memory, learning, language, thinking, cognitive development, and social cognition. We also review and summarize the advantages and disadvantages of these connectionist models. The computational framework of connectionist modeling has the potential to integrate specialized psychological findings of different areas using the same architectures and local functions of units and connections, inspired from neuroscience. In particular, the problems of dealing with structured information in distributed form, and doing tasks that require variable binding in connectionist networks are discussed from several different perspectives. As one possible solution to treat systematic mental representations properly, the symbolic connectionist model, which is a hybrid approach using symbolic representations and connectionist architectures, is explained. We argue that connectionist computer simulation offers significant benefits for today's psychological researches, and that connectionist modeling is likely to have an important influence on future studies.