Some principles in the brain analysis of important signals: mapping and stimulus recognition

This essay addresses two questions: What are the meanings of different kinds of brain maps? Can the recognition of important communicative signals and other stimuli known to the system be adequately explained by an ensemble code? Several kinds and grades of brain maps are distinguished. Before we can properly state the meaning, especially of the higher forms of multiple and multifactor maps, we need to learn where they project. At the higher levels this is a labor-intensive task, requiring also ingenuity and ethological thinking to devise stimuli. The neuronal basis of recognition in those cases where behavior or perception is triggered in an either-or manner has chiefly been attributed to one or the other of two classes of models: a large spatiotemporal configuration of many kinds of cells in which there is little convergence of input, and a small set of nearly equivalent cells after successive convergences. The latter is known to exist but it is not clear how far it goes or for what classes of stimuli. The former is a more popular view but carries several burdens in required assumptions and is essentially not demonstrable or disprovable. I believe both exist and operate in sequence: specific ensembles are inputs to small sets of equivalent recognition units and specific arrays of different recognition units at higher and higher levels of abstraction may constitute the specific configurations for the most sophisticated recognitions. Our information base is pitifully small, especially in comparative physiology of higher integrative functions of the brain and in comparative behaviour.