Its known that the age-related decline in cognitive facilities is not due to the loss of neurons but more subtle changes in specific areas of the brain. Structural and morphological changes in cellular alignment in the minicolumns correlate with increased prevalence of neurological diseases and in aging. In the rhesus monkey, cognitive decline is similar to what humans experience in aging. In the monkey prefrontal cortex, Brodmann area 46, an important region for executive functioning, cognitive decline correlates with changes in cellular alignment or “columnar strength” as studies by Cruz et al., (2009). Using the density maps method in Area 46, the ventral bank was identified to be the most susceptible to structural changes. Minicolumns, are defined by the cellular alignment of neurons in the cortex and some believe that the dendritic bundles of neurons in the cortex is also considered an integral part of the columns. The functional role of apical dendrites, is not well understood, however, given the that repeated organized bundles transverse through the laminae could be further support for their inclusion in minicolumns with possible functional importance. If structural changes such as loss of columnar strength (neuronal displacement) that correlates with cognitive aging, it is possible that the dendritic organization may also be affected in this area. In this thesis, it is hypothesized that the dendritic bundles in this area could also be related to the cognitive deficits associated with normal aging. Using double- fluorescence labeling for dendrites (MAP-2) and neurons (Neu-N), 3D confocal reconstructions of the dorsal and ventral banks of Area 46 were used to investigate structural/morphological changes in the dendritic bundles in young and old rhesus monkeys. While cortical thickness and apical dendritic length between both banks did not change, we found a significant increase in inter-bundle spacing at layer 6A in the older monkeys in the ventral bank. Inter-bundle spacing for bundles in layer 5 was measured and showed that the young consistently have smaller inter-bundle spacing. Future studies with larger sample size will also investigate whether changes in dendritic bundles and their organization also correlate with age-related cognitive deficits.

Traditionally, working memory and its neural underpinnings have been studied in the visual domain. A rich and diverse amount of research has investigated the lateral prefrontal cortex (lPFC) as a primary area for visual working memory, while another line of research has found the neurotransmitter acetylcholine (ACh) to be involved. This dissertation used auditory cues and found similar patterns of activity for processing auditory working memory information within a task compared to visual working memory processes. The first two experimental chapters demonstrated that the cholinergic system is involved in auditory working memory in a comparable fashion to its role in visual working memory. In chapter 2, blocking ACh impaired performance on an auditory working memory task in a dose dependent manner. Chapter 3 investigated the specificity of the effect of blocking ACh by administering an ACh agonist (physostigmine) at the same time as an ACh antagonist (scopolamine). When both drugs were administered together performance on the delayed matching-to-sample task (DMTS) task improved compared to performance on scopolamine alone. These results support the hypothesis that ACh is involved in auditory working memory. Chapter 4 investigated the neural correlates of auditory working memory in area 46 and found that this region of the lPFC contains neurons that are responsive to auditory working memory components in a very similar way to how it this region encodes information during visual working memory tasks. Neurons in the lPFC are responsive to visual or auditory cues, the delay portion of tasks, the wait time (i.e. decision making period), response, and reward times. This type of coding provides support for the theories that position the lPFC as a key player in recognition and working memory regardless of modality.