Page 45 - Artificial Intelligence in the Age of Neural Networks and Brain Computing

32 CHAPTER 2 Mind, Brain, Autonomous Agents, and Mental Disorders

1. TOWARDS A UNIFIED THEORY OF MIND AND BRAIN
A major scientific and technological revolution in understanding autonomous
adaptive intelligence is currently underway. How the brain works provides a critical
example of such intelligence. This revolution has been supported, in part, by
publications over the past 50 years of design principles, mechanisms, circuits,
and architectures that are part of an emerging unified theory of biological
intelligence. This emerging theory explains and predicts how brain mechanisms
give rise to mental functions as emergent properties.
This theory has clarified how advanced brains are designed to enable individuals
to autonomously adapt in real time in response to complex changing environments
that are filled with unexpected events. Its results hereby provide a blueprint
for designing increasingly autonomous adaptive agents for future applications in
engineering and technology. Many large-scale applications in engineering and
technology have already been developed; for example, http://techlab.bu.edu/
resources/articles/C5.
As part of the development of the biological theory, the data from thousands of
psychological and neurobiological experiments have been explained and predicted
in a unified way, including data about perception, cognition, cognitive-emotional
dynamics, and action. These results include an emerging unified theory of what
happens in an individual brain when it consciously sees, hears, feels, or knows
something; how seeing, hearing, feeling, and knowing can be integrated into unified
moments of conscious experience; and how unconscious processes can influence a
brain’s decision-making [1].
As sufficiently mature models of typical, or normal, behaviors became
understood, it also became possible to increasingly explain brain mechanisms and
behavioral symptoms of mental disorders. Applications to autism, schizophrenia,
and medial temporal amnesia were among the first to be made; for example,
Refs. [2e4]. Additional applications have been recently made towards explaining
how the dynamics of learning, memory, and cognition may break down during
Alzheimer disease, why slow wave sleep disorders are often correlated with
Alzheimer disease and other mental disorders, how symptoms of Fragile X syndrome
and autistic repetitive behaviors may arise [5,6], and how these insights may help to
guide new clinical therapies.
How did a theory that was developed to explain data about the learning and
performance of typical, or normal, behaviors lead to explanations of data about
mental disorders? This happened when it began to be noticed that, when various
model brain mechanisms become imbalanced in prescribed ways, then formal
analogs of behavioral symptoms of different mental disorders emerged. In autism,
these imbalances include underaroused emotional depression in the drive represen-
tations of regions like the amygdala, hypervigilant learning and narrowing of
attention in the recognition learning circuits of brain regions like the temporal
and prefrontal cortices, and a failure of adaptively timed learning in brain regions
like the hippocampus, basal ganglia, and cerebellum [4]. In this way, one could