Search

Transportation sustainability is one of the most widely discussed topics by transportation agencies across the U.S. as the transportation sector generates the largest share of greenhouse gas emissions, accounting for 27% of 2020 greenhouse gas emissions. Effective measures should be taken in the transportation industry to reduce carbon dioxide emissions the vulnerability of the system and the people served by these systems to climatic change. Past literature, however, rarely examined how to enhance the efficiency of behavior change tactics and improve equity while reducing costs by assessing people's perceptions, motivations, and irrationality in their decision-making processes. The findings of this dissertation could facilitate the development of a human-centered and cost-effective framework for behavior change that state DOTs can utilize to establish demand management plans. By utilizing cutting-edge behavior measurement tools, including online surveys and field experiments, this dissertation aims to investigate the relationship between individual and contextual factors and the behavior to switch modes of transportation. The target audience of the transportation demand management programs, their perceived barriers to active transportation (i.e., any self-propelled, human-powered mode of transportation, such as walking or bicycling) and shared transportation (i.e., transportation services and resources that are shared among users), as well as the dynamic impact of their cognitive biases on mode shift, were examined in-depth using behavior science methodologies. Advanced statistical and machine learning algorithms were utilized to interpret the findings. This dissertation's outcomes include a comprehensive framework for travel behavior investigation, demographic and geological characterization of the target audiences, and a deeper understanding of how mental and physical biases keep people from switching to more sustainable modes of transportation. This work contrib

Space and time are two cornerstones of memory and navigation. However, how spatial and temporal information contribute to spatial and mnemonic representations is still poorly understood in humans. One of the neural signatures in humans related to memory and navigation are neural oscillations in cortical and subcortical regions (such as the hippocampus). The presence of cortical and hippocampal theta oscillations predicts better subsequent memory and more efficient spatial navigation, but the exact relationship between neural oscillations and the coding of spatial distances and temporal durations are unknown. In this dissertation, I will provide empirical evidence to help fill this gap and to better the understanding of how cortical and hippocampal neural oscillations support the coding of spatiotemporal information. In Chapter 1, I provide a brief summary of scalp electroencephalogram (EEG), intracranial EEG, mobile EEG, hippocampal theta oscillations, frontal midline theta oscillations, and behavioral models for spatial and temporal cognition. In Chapter 2, I report the feasibility of recording navigation-related frontal-midline theta oscillations using noninvasive scalp EEG in healthy humans using a mobile EEG approach. In Chapter 3 and 4, I report two investigations of how neural oscillations code spatial distances and temporal durations in healthy humans using mobile scalp EEG (Chapter 3) and in patients with hippocampal implanted electrodes (Chapter 4). In Chapter 5, I summarize how the understanding of oscillatory codes for space and time can inspire the behavioral and clinical applications for future research.

Perceptual Decision Making involves making a simple decision about some feature of a perceptual stimulus. How do different contextual influences change the underlying dynamics of this task? Here I use the Clicks task, a simple perceptual decision-making task where participants must report which side they heard more clicks, in order to study how different contextual influences can shape the underlying dynamics of this decision-making process. This task is subject to a number of different suboptimalities which can drive errors, such as side bias, integration shape, choice kernel, reinforcement learning, and noise. Here I address how various contextual influences can modulate these suboptimalities in order to change decision-making behavior. In the first study, I show that when participants are under higher motivation to perform the task accurately, noise in the decision process is subject to rapid modulation in order to achieve higher performance. However, no other suboptimalities were modulated in order to achieve this higher performance. The second study aims to find other sources of contextual influence that can modulate the other suboptimalities, such as the integration shape. Results show that through changing the underlying statistics of the stimulus, participants can indeed modulate their integration shape, or the way they weight information over time, to match these stimulus statistics. These two studies provide evidence that noise in the decision process, and integration shape are two underlying dynamics of the decision process that are subject to modulation. Future research can address further contextual influences that may be able to modulate the remaining suboptimalities such as side bias, reinforcement learning, and choice kernel.