Neuroanatomy

Brain atlases have been created across species from flies to humans in order to obtain a better understanding of neuroanatomical morphology. Although these brain atlases allow for analysis of neuroanatomy they do not give insight about how the morphology adapt to fit challenges brought on by unique environments. Here I developed a brain atlas for Astyanax mexicanus, a species known to have populations that differ in various behaviors, to gain a better understanding about how populations of the same species, derived from different environments, evolve to be best suited for the challenges they face. By creating a brain atlas for adult surface fish and three populations of cavefish I was able to examine differences in neuroanatomical structures implicated in regulating behavior. My findings show significant differences in neuroanatomical regions known to regulate behavior. Along with these findings, the brain atlases created are a tool for researches to use and expand on in the future.

The thalamus has been traditionally viewed as a structural relay to specific cortical
areas behaviorally associated with sensory or motor functions, and thalamic nuclei that
function in this manner are referred to as 'relay nuclei·. However. the parts of the
thalamus interconnecting limbic association cortices (functionally involved in memory.
reward, emotion. and decision making) comprise the midline and intralaminar nuclei. The
midline thalamus has not been examined fully at the anatomical, physiological. or
behavioral level, and may serve as an important relay between cortical and subcortical
structures and the limbic system. The work incorporated into this dissertation included
five axonal tract tracing projects that were conducted in the rat. to explore and test the
hypothesis that the midline thalamus serves as an important interface between limbic
structures including the amygdala. nucleus accumbens. medial prefrontal cortex and
hippocampal formation.
An important finding was the demonstration of a closed anatomical loop between the
hippocampal formation, the ventral medial prefrontal cortex and the ventral midline
thalamus: CA 1/subiculum > PLIIL > RE > CA 1/subiculum. Another finding was that 1) the hippocampal formation innervates the entire medial prefrontal cortex; and 2) the
hippocampal formation projects more heavily to ventral as compared to dorsal cortices in
the mPFC. The paraventricular, parataenial, rhomboid and reuniens nuclei of the midline
thalamus were shown to distribute to limbic structures important for cognitive
processing: the amygdala, nucleus accumbens, hippocampal formation, parahippocampal
cortex, and the prefrontal cortex. Present results demonstrate that the ventral midline
nuclei (reuniens and rhomboid) extensively innervate limbic cortical structures (the
medial prefrontal cortex and hippocampal formation) whereas dorsal midline nuclei
(paraventricular and parataenial) distribute more heavily to subcortical limbic structures
(the amygdala and the nucleus accumbens). These midline nuclei may, therefore, relay
information between these limbic areas. This connectivity suggests that the midline
nuclei could further be subdivided from the intralaminar and relay groups. The midline
thalamic nuclei would, therefore, comprise the limbic thalamus.

This research aims at proposing a model for visual pattern recognition inspired by the neural circuitry in the brain. Our attempt is to propose few modifications in the Alopex algorithm and try to use it for the calculations of the receptive fields of neurons in the trained network. We have developed a small-scale, four-layered neural network model for simple character recognition as well as complex image patterns, which can recognize the patterns transformed by affine conversion. Here Alopex algorithm is presented as an iterative and stochastic processing method, which was proposed for optimization of a given cost function over hundreds or thousands of iterations. In this case the receptive fields of the neurons in the output layers are obtained using the Alopex algorithm.