Computational neuroscience is an active field of study, spanning the traditional fields of biological neuroscience, electrical engineering, and computer science (CS). Although it has yet to achieve much by way of concrete applications that bring benefit to society or the individual, it is nonetheless much prized for its perceived potential.

This book is an artifact that has been developed as a result of a cross-disciplinary effort by the authors, who come from different backgrounds. It attempts to provide a good foundation for the idea that correlation, in an existential as well as procedural sense, is a central theme of learning in the human brain, as well as in artificial machines.

The book includes eight chapters: “The Correlative Brain,” “Correlation in Signal Processing,” “Correlation-Based Neural Learning and Machine Learning,” “Correlation-Based Kernel Learning,” “Correlative Learning in a Complex-Valued Domain,” “ALOPEX: A Correlation-Based Learning Paradigm,” “Case Studies,” and “Discussion.” There are also four appendices on some of the mathematical background needed to understand the material.

Perhaps because none of the authors is from a CS background, algorithms are not presented in the pseudocode fashion that is now popular, or even the older flowchart style still common in some disciplines and application domains. A reader therefore needs to devote some time to extra thought and analysis in order to discover the likely algorithmic structure of the learning paradigms present in the form of mathematical equations supported by discussion in plain English. Perhaps for the same reason, the authors do not deal with computational complexity issues that would doubtless assume importance in any practical setting.

The book also skips over many controversial issues, such as the disputed relationship between the computational neural network model common in artificial intelligence (AI) studies and the actual biological and cognitive architecture of consciousness. While the work presented is valuable within a certain context, many of the claims made to provide motivation, as well as some of the underlying assumptions, would tend to attract the ire of thinkers such as Martin Gardner [1] and Roger Penrose [2], while even others, such as Daniel Dennett [3], might not agree wholeheartedly with the importance this book purports to give to correlation as a cognitive paradigm. There is also no effort to compare correlation with other machine learning paradigms, such as those arising from genetic algorithms.