The increasing availability of experimental data and computational power have resulted in increasingly detailed and sophisticated models of brain structures. Biophysically realistic models allow detailed investigations of the mechanisms that operate within those structures. In this work, published mouse experimental data were synthesized to develop an extensible, open-source platform for modeling the mouse main olfactory bulb and other brain regions. A “virtual slice” model of a main olfactory bulb glomerular column that includes detailed models of tufted, mitral, and granule cells was created to investigate the underlying mechanisms of a gamma frequency oscillation pattern (“gamma fingerprint”) often observed in rodent bulbar local field potential recordings. The gamma fingerprint was reproduced by the model and a mechanistic hypothesis to explain aspects of the fingerprint was developed. A series of computational experiments tested the hypothesis. The results demonstrate the importance of interactions between electrical synapses, principal cell synaptic input strength differences, and granule cell inhibition in the formation of the gamma fingerprint. The model, data, results, and reproduction materials are accessible at https://github.com/justasb/olfactorybulb. The discussion includes a detailed description of mechanisms underlying the gamma fingerprint and how the model predictions can be tested experimentally. In summary, the modeling platform can be extended to include other types of cells, mechanisms and brain regions and can be used to investigate a wide range of experimentally testable hypotheses.

Body size plays a pervasive role in determining physiological and behavioral performance across animals. It is generally thought that smaller animals are limited in performance measures compared to larger animals; yet, the vast majority of animals on earth are small and evolutionary trends like miniaturization occur in every animal clade. Therefore, there must be some evolutionary advantages to being small and/or compensatory mechanisms that allow small animals to compete with larger species. In this dissertation I specifically explore the scaling of flight performance (flight metabolic rate, wing beat frequency, load-carrying capacity) and learning behaviors (visual differentiation visual Y-maze learning) across stingless bee species that vary by three orders of magnitude in body size. I also test whether eye morphology and calculated visual acuity match visual differentiation and learning abilities using honeybees and stingless bees. In order to determine what morphological and physiological factors contribute to scaling of these performance parameters I measure the scaling of head, thorax, and abdomen mass, wing size, brain size, and eye size. I find that small stingless bee species are not limited in visual learning compared to larger species, and even have some energetic advantages in flight. These insights are essential to understanding how small size evolved repeatedly in all animal clades and why it persists. Finally, I test flight performance across stingless bee species while varying temperature in accordance with thermal changes that are predicted with climate change. I find that thermal performance curves varied greatly among species, that smaller species conform closely to air temperature, and that larger bees may be better equipped to cope with rising temperatures due to more frequent exposure to high temperatures. This information may help us predict whether small or large species might fare better in future thermal climate conditions, and which body-size related traits might be expected to evolve.

The ability to detect and appropriately respond to chemical stimuli is important for many organisms, ranging from bacteria to multicellular animals. Responses to these stimuli can be plastic over multiple time scales. In the short-term, the synaptic strengths of neurons embedded in neural circuits can be modified and result in various forms of learning. In the long-term, the overall developmental trajectory of the olfactory network can be altered and synaptic strengths can be modified on a broad scale as a direct result of long-term (chronic) stimulus experience. Over evolutionary time the olfactory system can impose selection pressures that affect the odorants used in communication networks. On short time scales, I measured the effects of repeated alarm pheromone exposure on the colony-level defense behaviors in a social bee. I found that the responses to the alarm pheromone were plastic. This suggests that there may be mechanisms that affect individual plasticity to pheromones and regulate how these individuals act in groups to coordinate nest defense. On longer time scales, I measured the behavioral and neural affects of bees given a single chronic odor experience versus bees that had a natural, more diverse olfactory experience. The central brains of bees with a deprived odor experience responded more similarly to odorants in imaging studies, and did not develop a fully mature olfactory network. Additionally, these immature networks showed behavioral deficits when recalling odor mixture components. Over evolutionary time, signals need to engage the attention of and be easily recognized by bees. I measured responses of bees to a floral mixture and its constituent monomolecular components. I found that natural floral mixtures engage the orientation of bees’ antennae more strongly than single-component odorants and also provide more consistent central brain responses between stimulations. Together, these studies highlight the importance of olfactory experience on different scales and how the nervous system might impose pressures to select the stimuli used as signals in communication networks.

Animals must learn to ignore stimuli that are irrelevant to survival, which is a process referred to as ‘latent inhibition’. This process has been shown to be genetically heritable (Latshaw JS, Mazade R, Sinakevitch I, Mustard JA, Gadau J, Smith BH (submitted)). The locus containing the AmTYR1 gene has been shown through quantitative trait loci mapping to be linked to strong latent inhibition in honey bees. The Smith lab has been able to show a correlation between learning and the AmTYR1 receptor gene through pharmacological inhibition of the receptor. In order to further confirm this finding, experiments were designed to test how honey bees learn with this receptor knocked out. Here this G-protein coupled receptor for the biogenic amine tyramine is implemented as an important factor underlying latent inhibition in honey bees. It is shown that double-stranded RNA (dsRNA) and Dicer-substrate small interfering RNA (dsiRNA) that are targeted to disrupt the tyramine receptors specifically affects latent inhibition but not excitatory associative conditioning. The results therefore identify a distinct reinforcement pathway for latent inhibition in insects.

Pregnancy and childbirth are both natural occurring events, but still little is known about the signaling mechanisms that induce contractions. Throughout the world, premature labor occurs in 12% of all pregnancies with 36% of infant deaths resulting from preterm related causes. Even though the cause of preterm labor can vary, understanding alternative signaling pathways, which affect muscle contraction, could provide additional treatment options in stopping premature labor. The uterus is composed of smooth muscle, which is innervated, with a plexus of nerves that cover the muscle fibers. Smooth muscle can be stimulated or modulated by many sources such as neurotransmitters [i.e. dopamine], hormones [i.e. estrogen], peptides [i.e. oxytocin] and amines. This study focuses on the biogenic monoamine tyramine, which is produced in the tyrosine catecholamine biosynthesis pathway. Tyramine is known to be associated with peripheral vasoconstriction, increased cardiac output, increased respiration, elevated blood glucose and the release of norepinephrine. This research has found tyramine, and its specific receptor TAAR1, to be localized within mouse uterus and that this monoamine can induce uterine contractions at levels similar to oxytocin.

Neurotoxicology has historically focused on substances that directly damage nervous tissue. Behavioral assays that test sensory, cognitive, or motor function are used to identify neurotoxins. But, the outcomes of behavioral assays may also be influenced by the physiological status of non-neural organs. Therefore, toxin induced damage to non- neural organs may contribute to behavioral modifications. Heavy metals and metalloids are persistent environmental pollutants and induce neurological deficits in multiple organisms. However, in the honey bee, an important insect pollinator, little is known about the sublethal effects of heavy metal and metalloid toxicity though they are exposed to these toxins chronically in some environments. In this thesis I investigate the sublethal effects of copper, cadmium, lead, and selenium on honey bee behavior and identify potential mechanisms mediating the behavioral modifications. I explore the honey bees’ ability to detect these toxins, their sensory perception of sucrose following toxin exposure, and the effects of toxin ingestion on performance during learning and memory tasks. The effects depend on the specific metal. Honey bees detect and reject copper containing solutions, but readily consume those contaminated with cadmium and lead. And, exposure to lead may alter the sensory perception of sucrose. I also demonstrate that acute selenium exposure impairs learning and long-term memory formation or recall. Localizing selenium accumulation following chronic exposure reveals that damage to non-neural organs and peripheral sensory structures is more likely than direct neurotoxicity. Probable mechanisms include gut microbiome alterations, gut lining

damage, immune system activation, impaired protein function, or aberrant DNA methylation. In the case of DNA methylation, I demonstrate that inhibiting DNA methylation dynamics can impair long-term memory formation, while the nurse-to- forager transition is not altered. These experiments could serve as the bases for and reference groups of studies testing the effects of metal or metalloid toxicity on DNA methylation. Each potential mechanism provides an avenue for investigating how neural function is influenced by the physiological status of non-neural organs. And from an ecological perspective, my results highlight the need for environmental policy to consider sublethal effects in determining safe environmental toxin loads for honey bees and other insect pollinators.

In the honey bee antennal lobe, uniglomerular projection neurons (uPNs) transiently spike to odor sensory stimuli with odor-specific response latencies, i.e., delays to first spike after odor

stimulation onset. Recent calcium imaging studies show that the spatio-temporal response profile of the activated uPNs are dynamic and changes as a result

of associative conditioning, facilitating odor-detection of learned odors.

Moreover, odor-representation in the antennal lobe undergo reward-mediated plasticity processes that increase response delay variations

in the activated ensemble of uniglomerular projection neurons. Octopamine is necessarily involved in these plasticity processes. Yet, the cellular mechanisms are not

well understood. I hypothesize that octopamine modulates cholinergic transmission to uPNs by triggering translation

and upregulation of nicotinic receptors, which are more permeable to calcium. Consequently, this increased calcium-influx signals transcription factors that upregulate potassium

channels in the dendritic cortex of glomeruli, similar to synaptic plasticity mechanisms recently

shown in various insect species. A biophysical model of the antennal lobe circuit is developed in order to test the hypothesis that increased potassium channel expression in uPNs mediate response delays to first

spike, dynamically tuning odor-representations to facilitate odor-detection of learned odors.

Navigation through natural environments requires continuous sensory guidance. In addition to coordinated muscle contractions of the limbs that are controlled by spinal cord, equilibrium, body weight bearing and transfer, and avoidance of obstacles all have to happen while locomotion is in progress and these are controlled by the supraspinal centers.

For successful locomotion, animals require visual and somatosensory information. Even though a number of supraspinal centers receive both in varying degrees, processing this information at different levels of the central nervous system, especially their contribution to visuo-motor and sensory-motor integration during locomotion is poorly understood.

This dissertation investigates the patterns of neuronal activity in three areas of the forebrain in the cat performing different locomotor tasks to elucidate involvement of these areas in processing of visual and somatosensory information related to locomotion. In three studies, animals performed two contrasting locomotor tasks in each and the neuronal activities were analyzed.

In the first study, cats walked in either complete darkness or in an illuminated room while the neuronal activity of the motor cortex was recorded. This study revealed that the neuronal discharge patterns in the motor cortex were significantly different between the two illumination conditions. The mean discharge rates, modulation, and other variables were significantly different in 49% of the neurons. This suggests a contextual correlation between the motor cortical activity and being able to see.

In two other studies, the activities of neurons of either the somatosensory cortex (SI) or ventrolateral thalamus (VL) were recorded while cats walked on a flat surface (simple locomotion) or along a horizontal ladder where continuous visual and somatosensory feedback was required (complex locomotion).

We found that the activity of all but one SI cells with receptive fields on the sole peaked before the foot touched the ground: predictably. Other cells showed various patterns of modulation, which differed between simple and complex locomotion. We discuss the predictive and reflective functionality of the SI in cyclical sensory-motor events such as locomotion.

We found that neuronal discharges in the VL were modulated to the stride cycle resembling patterns observed in the cortex that receives direct inputs from the VL. The modulation was stronger during walking on the ladder revealing VL’s contribution to locomotion-related activity of the cortex during precision stepping.

Many behaviors are organized into bouts – brief periods of responding punctuated by pauses. This dissertation examines the operant bouts of the lever pressing rat. Chapter 1 provides a brief history of operant response bout analyses. Chapters 2, 3, 5, and 6 develop new probabilistic models to identify changes in response bout parameters. The parameters of those models are demonstrated to be uniquely sensitive to different experimental manipulations, such as food deprivation (Chapters 2 and 4), response requirements (Chapters 2, 4, and 5), and reinforcer availability (Chapters 2 and 3). Chapter 6 reveals the response bout parameters that underlie the operant hyperactivity of a common rodent model of attention deficit hyperactivity disorder (ADHD), the spontaneously hypertensive rat (SHR). Chapter 6 then ameliorates the SHR’s operant hyperactivity using training procedures developed from findings in Chapters 2 and 4. Collectively, this dissertation provides new tools for the assessment of response bouts and demonstrates their utility for discerning differences between experimental preparations and animal strains that may be otherwise indistinguishable with more primitive methods.

Warning coloration deters predators from attacking prey that are defended, usually by being distasteful, toxic, or otherwise costly for predators to pursue and consume. Predators may have an innate response to warning colors or learn to associate them with a defense through trial and error. In general, predators should select for warning signals that are easy to learn and recognize. Previous research demonstrates long-wavelength colors (e.g. red and yellow) are effective because they are readily detected and learned. However, a number of defended animals display short-wavelength coloration (e.g. blue and violet), such as the pipevine swallowtail butterfly (Battus philenor). The role of blue coloration in warning signals had not previously been explicitly tested. My research showed in laboratory experiments that curve-billed thrashers (Toxostoma curvirostre) and Gambel's quail (Callipepla gambelii) can learn and recognize the iridescent blue of B. philenor as a warning signal and that it is innately avoided. I tested the attack rates of these colors in the field and blue was not as effective as orange. I concluded that blue colors may function as warning signals, but the effectiveness is likely dependent on the context and predator.

Blue colors are often iridescent in nature and the effect of iridescence on warning signal function was unknown. I reared B. philenor larvae under varied food deprivation treatments. Iridescent colors did not have more variation than pigment-based colors under these conditions; variation which could affect predator learning. Learning could also be affected by changes in appearance, as iridescent colors change in both hue and brightness as the angle of illuminating light and viewer change in relation to the color surface. Iridescent colors can also be much brighter than pigment-based colors and iridescent animals can statically display different hues. I tested these potential effects on warning signal learning by domestic chickens (Gallus gallus domesticus) and found that variation due to the directionality of iridescence and a brighter warning signal did not influence learning. However, blue-violet was learned more readily than blue-green. These experiments revealed that the directionality of iridescent coloration does not likely negatively affect its potential effectiveness as a warning signal.