A computer scientist can never convince students in computer science that how a computer works by just looking into the interconnected gates of the CPU chip without sophisticated principles and theories. Although all connections in a CPU have already been known, nobody may be able to explain the massive connections of gates (see attached figures), if high-level logical (mathematical) models remain unknown. Nobody would understand a computer by just observing its active parts and locations for certain functions because the causality is not that intuitive or straight forward in any super complex system.
This is the same situation across brain science, cognitive science, neurology, physiology, cognitive psychology, and natural intelligence. A hierarchical model of the brain at all the levels, such as the Layered Reference Model of the Brain (LRMB) [ICIC, http://www.ucalgary.ca/icic/], is required in order to enable the reductive/synthesized studies from the top down as well as from the bottom up.
Brain imaging technologies such as EEG, fMRI, MEG, and PET have provided many promising results on explaining the cognitive functions of the brain. However, they are highly limited to simple functions as revealed in the Layered Reference Model of the Brain (LRMB) [Wang et al., 2006]. There is a lack of a systematic knowledge about what roles particular group of neurons may play in complex cognitive functions such as learning and memorization, because neuroimages cannot yet pinpoint to detailed relationships between structures and functions in the brain.
The limitations of current brain studies seem to be equivalent to the problem to examine the functions of a computer by looking at its layout and locations where they are active using imaging technologies. It is well recognized that without understanding the logical and functional models and mechanisms of the CPU, nobody can explain its functions by only using fine pictures of the intricate interconnections of millions of transistors (gates). Further, it would be more confusing because the control unit (CU) and arithmetic and logic unit (ALU) of the CPU and related buses are always active for almost any different kind of operations. So do, unfortunately, in brain science and neurobiology. Without a rational guide to the high-level life functions and cognitive processes, it is hard to pinpoint rational functional relationship between a brain image and a specific cognitive function such as an action of learning and its effect in memory, a recall of a particular knowledge retained in LTM, and a mapping of the same mental object from STM to LTM.
An article on “On Abstract Intelligence and Brain Informatics: Mapping Cognitive Functions of the Brain onto its Neural Structures,” has recently been published in International Journal of Cognitive Informatics and Natural Intelligence, 6(4), 54-80 [Wang, 2012]. It presented the following:
“A key notion in abstract intelligence and cognitive informatics is that the brain and natural intelligence may only be explained by a hierarchical and reductive theory that maps the brain through the embodied neurological, physiological, cognitive, and logical levels from bottom-up induction and top-down deduction. This paper presents an abstract intelligence framework for modeling the structures and functions of the brain across these four levels. A set of abstract intelligent model, cognitive functional model, and neurophysiological model of the brain is systematically developed. On the basis of the abstract intelligent models of the brain at different levels, the conventionally highly overlapped, redundant, and even contradicted empirical observations in brain studies and cognitive psychology may be rigorously clarified and neatly explained. The improved understanding about the brain has led to the development of a wide range of novel technologies and systems such as cognitive computers, cognitive robots, and other applied cognitive systems.
- Badges
- Science topic
- Similar topics
- Psychology
- Cognitive Science