Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
101 results
Search Results
Item MODELING ADAPTABILITY MECHANISMS OF SPEECH PERCEPTION Nika Jurov(2024) Jurov, Nika; Feldman, Naomi H.; Idsardi, William; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Speech is a complex, redundant and variable signal happening in a noisy and ever changing world. How do listeners navigate these complex auditory scenes and continuously and effortlessly understand most of the speakers around them? Studies show that listeners can quickly adapt to new situations, accents and even to distorted speech. Although prior research has established that listeners rely more on some speech cues (or also called features or dimensions) than others, it is yet not understood how listeners weight them flexibly on a moment-to-moment basis when the input might deviate from the standard speech. This thesis computationally explores flexible cue re-weighting as an adaptation mechanism using real speech corpora. The computational framework it relies on is rate distortion theory. This framework models a channel that is optimized on a trade off between distortion and rate: on the one hand, the input signal should be reconstructed with minimal error after it goes through the channel. On the other hand, the channel needs to extract parsimonious information from the incoming data. This channel can be implemented as a neural network with a beta variational auto-encoder. We use this model to show that two mechanistic components are needed for adaptation: focus and switch. We firstly show that focus on a cue mimics humans better than cue weights that simply depend on long term statistics as has been largely assumed in the prior research. And second, we show a new model that can quickly adapt and switch weighting the features depending on the input of a particular moment. This model's flexibility comes from implementing a cognitive mechanism that has been called ``selective attention" with multiple encoders. Each encoder serves as a focus on a different part of the signal. We can then choose how much to rely on each focus depending on the moment. Finally, we ask whether cue weighting is informed by being able to separate the noise from speech. To this end we adapt a feature disentanglement adversarial training from vision to disentangle speech (noise) features from noise (speech) labels. We show that although this does not give us human-like cue weighting behavior, there is an effect of disentanglement of weighting spectral information slightly more than temporal information compared to the baselines. Overall, this thesis explores adaptation computationally and offers a possible mechanistic explanation for ``selective attention'' with focus and switch mechanisms, based on rate distortion theory. It also argues that cue weighting cannot be determined solely on speech carefully articulated in laboratories or in quiet. Lastly, it explores a way to inform speech models from a cognitive angle to make the models more flexible and robust, like human speech perception is.Item HOW BILINGUALS' COMPREHENSION OF CODE-SWITCHES INFLUENCES ATTENTION AND MEMORY(2024) Salig, Lauren; Novick, Jared; Slevc, L. Robert; Neuroscience and Cognitive Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Bilinguals sometimes code-switch between their shared languages. While psycholinguistics research has focused on the challenges of comprehending code-switches compared to single-language utterances, bilinguals seem unhindered by code-switching in communication, suggesting benefits that offset the costs. I hypothesize that bilinguals orient their attention to speech content after hearing a code-switch because they draw a pragmatic inference about its meaning. This hypothesis is based on the pragmatic meaningfulness of code-switches, which speakers may use to emphasize information, signal their identity, or ease production difficulties, inter alia. By considering how code-switches may benefit listeners, this research attempts to better align our psycholinguistic understanding of code-switch processing with actual bilingual language use, while also inspiring future work to investigate how diverse language contexts may facilitate learning in educational settings. In this dissertation, I share the results of three pre-registered experiments with Spanish-English bilinguals that evaluate how hearing a code-switch affects attention and memory. Experiment 1a shows that code-switches increase bilinguals’ self-reported attention to speech content and improve memory for that information, compared to single-language equivalents. Experiment 1b demonstrates that this effect requires bilingual experience, as English-speaking monolinguals did not demonstrate increased attention upon hearing a code-switch. Experiment 2 attempts to replicate these results and establish the time course of the attentional effect using an EEG measure previously associated with attentional engagement (alpha power). However, I conclude that alpha power was not a valid measure of attention to speech content in this experiment. In Experiment 3, bilinguals again showed better memory for information heard in a code-switched context, with a larger benefit for those with more code-switching experience and when listeners believed the code-switches were natural (as opposed to inserted randomly, removing the element of speaker choice). This suggests that the memory benefit comes from drawing a pragmatic inference, which likely requires prior code-switching experience and a belief in code-switches’ communicative purpose. These experiments establish that bilingual listeners derive attentional and memory benefits from ecologically valid code-switches—challenging a simplistic interpretation of the traditional finding of “costs.” Further, these findings motivate future applied work assessing if/how code-switches might benefit learning in educational contexts.Item MULTIMODAL ANALYSIS OF NEURAL SIGNALS RELATED TO SOURCE MEMORY ENCODING IN YOUNG CHILDREN(2024) Lei, Yuqing; Riggins, Tracy; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The emergence of source memory is an important milestone during memory development. Decades of research has explored neural correlates of source memory using electroencephalography (EEG) and functional magnetic resonance imaging (fMRI). However, connections between findings from the two approaches, particularly within children, remain unclear. This dissertation identified fMRI-informed cortical sources of two EEG signals during memory encoding, the P2 and the late slow wave (LSW), that predicted subsequent source memory performance in a sample of children aged 4 to 8 years. Both P2 and LSW were source localized to cortical areas of the medial temporal lobe (MTL), reflecting MTL’s crucial role in both early-stage information processing and late-stage integration of memory, which also validated LSW’s suspected role in memory updating. The P2 effect was localized to all six tested subregions of cortical MTL in both left and right hemispheres, whereas the LSW effect was only present in the parahippocampal cortex and entorhinal cortex. P2 was additionally localized to multiple areas in the frontoparietal network, a cortical network known as the “attention network”, highlighting interactions between memory encoding and other cognitive functions. These results reflect the importance of considering both spatial and temporal aspects of neural activity to decode memory mechanism, and demonstrated the potential of combining multimodal measures in children, paving the way for future developmental research.Item INHIBITION IS KEY: A COGNITIVE APPROACH TO SUCCESSFUL WORD PROBLEM SOLVING(2024) Jaffe, Joshua Benjamin; Bolger, Donald J; Human Development; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Numerical competency and reading comprehension skills are necessary, but insufficient for word problem success. Depending on the word problem structure, successful problem solving may require inhibiting the seemingly obvious and correct answer. Inhibitory control plays a significant role in processing and solving word problems. Through classroom practices and textbook problems, I argue that individuals form associations between relational terminology and specific mathematical operations (“more” for addition and “less” for subtraction), and the notion that all numerical values in a problem must be used to produce an answer. In this study, I proposed an inhibitory performance-based model that posits two approaches to problem solving: (a) a successful approach where solvers inhibit mathematical associations and form appropriate set schemas to conceptualize semantic relations, and (b) an association approach where solvers do not inhibit associations and therefore may have an inaccurate understanding of the semantic relations. To test the model, data were analyzed from 105 undergraduate students at the University of Maryland. The study consisted of four sections: cognitive skills, word problems, domain-specific inhibitory control tasks, and a semi-structured interview. The word problem section included problems that were both consistent and inconsistent with an individual’s operational and numerical associations. Overall, the quantitative results identified that participants performed significantly worse on inconsistent problems. Further, the data suggest that failure to correctly answer inconsistent problems may be due to inhibitory control rather than other cognitive skills. The qualitative data indicated that a vast majority of participants believed in both mathematical associations and remembered classroom experiences that may have contributed to these beliefs. While inhibitory control has been suggested to play a significant role in word problem performance, this is one of the first studies to explicitly examine the relationship through domain-specific inhibitory control tasks and an interview. These results guide a path for future research to examine how individuals develop mathematical associations and for interventions to dissuade their usage.Item Determining the Mechanisms of Spoken Language Processing Delay for Children with Cochlear Implants(2023) Blomquist, Christina Marie; Edwards, Jan R; Newman, Rochelle S; Hearing and Speech Sciences; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The long-term objective of this project was to better understand how shorter auditory experience and spectral degradation of the cochlear implant (CI) signal impact spoken language processing in deaf children with CIs. The specific objective of this research was to utilize psycholinguistic methods to investigate the mechanisms underlying observed delays in spoken word recognition and the access of networks of semantically related words in the lexicon, which are both vital components for efficient spoken language comprehension. The first experiment used eye-tracking to investigate the contributions of early auditory deprivation and the degraded CI signal to spoken word recognition delays in children with CIs. Performance of children with CIs was compared to various typical hearing (TH) control groups matched for either chronological age or hearing age, and who heard either clear or vocoded speech. The second experiment investigated semantic processing in the face of a spectrally degraded signal (TH adult listeners presented with vocoded speech) by recording event-related potentials, specifically the N400. Results children with CIs show slower lexical access and less immediate lexical competition, and while early hearing experience supports more efficient recognition, much of these observed delays can be attributed to listening to a degraded signal in the moment, as children with TH demonstrate similar patterns of processing when presented with vocoded speech. However, some group differences remain, specifically children with CIs show slower speed of lexical access and longer-lasting competition, suggesting potential effects of learning from a degraded speech signal. With regards to higher-level semantic processing, TH adult listeners demonstrate more limited access of semantic networks when presented with a degraded speech signal. This finding suggests that uncertainty due the degraded speech signal may lead to less immediate cascading processing at both the word-level and higher-level semantic processing. Clinically, these results highlight the importance of early cochlear implantation and maximizing access to spectral detail in the speech signal for children with CIs. Additionally, it is possible that some of the delays in spoken language processing are the result of an alternative listening strategy that may be engaged to reduce the chance of incorrect predictions, thus preventing costly revision processes.Item Information Uncertainty Influences Learning Strategy from Sequentially Delayed Rewards(2023) Maulhardt, Sean Richard; Charpentier, Caroline; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)The problem of temporal credit assignment has long been posed as a nontrivial obstacle to identifying signal from data. However, human solutions in complex environments, involving repeated and intervening decisions, as well as uncertainty in reward timing, remain elusive. To this end, our task manipulated uncertainty via the amount of information given in their feedback stage. Using computational modeling, two learning strategies were developed that differentiated participants’ updates of sequentially delayed rewards: eligibility trace whereby previously selected actions are updated as a function of the temporal sequence - and tabular update - whereby additional feedback information is used to only update systematically-related rather than randomly related past actions. In both models, values were discounted over time with an exponential decay. We hypothesized that higher uncertainty would be associated with (i) a switch from tabular to eligibility strategy and (ii) higher rates of discounting. Participants’ data (N = 142) confirmed our first hypothesis, additionally revealing an effect of the starting condition. However, our discounting hypothesis had only weak evidence of an effect and remains an open question for future studies. We explore potential explanations for these effects and possibilities of future directions, models, and designs.Item ROLE OF PROJECTIONS FROM ORBITOFRONTAL CORTEX TO DORSAL STRIATUM IN INCUBATION OF OXYCDONE CRAVING(2023) Lin, Hongyu; Li, Xuan; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Oxycodone seeking progressively increases during abstinence and maintains for an extended period, a phenomenon termed incubation of oxycodone craving. We previously found that the orbitofrontal cortex (OFC) plays a causal role in this incubation. Here, we aimed to identify critical downstream regions of OFC in incubation of oxycodone craving by focusing on the central to medial portion of the dorsal striatum (DS), based on previous anatomical evidence. We first measured projection-specific activation on abstinence day 15 seeking test by using cholera toxin b (retrograde tracer, injected into DS) +Fos (activity marker) double-labeling in the OFC. Next, we determined the effect of pharmacological reversible inactivation of DS on incubated oxycodone seeking on abstinence day 15. We then used an anatomical asymmetrical disconnection procedure to determine whether OFC to DS projections contribute to incubated oxycodone seeking.Item EFFECTS OF DIVERSE INITIALIZATION ON BAYESIAN OPTIMIZERS(2023) Kamrah, Eesh; Fuge, Mark D; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Design researchers have struggled to produce quantitative predictions for exactly why andwhen diversity might help or hinder design search efforts. This thesis addresses that problem by studying one ubiquitously used search strategy—Bayesian Optimization (BO)—on different ND test problems with modifiable convexity and difficulty. Specifically, we test how providing diverse versus non-diverse initial samples to BO affects its performance during search and introduce a fast ranked-DPP method for computing diverse sets, which we need to detect sets of highly diverse or non-diverse initial samples. We initially found, to our surprise, that diversity did not appear to affect BO, neither helping nor hurting the optimizer’s convergence. However, follow-on experiments illuminated a key trade-off. Non-diverse initial samples hastened posterior convergence for the underlying model hyper-parameters—a Model Building advantage. In contrast, diverse initial samples accelerated exploring the function itself—a Space Exploration advantage. Both advantages help BO, but in different ways, and the initial sample diversity directly modulates how BO trades those advantages. Indeed, we show that fixing the BO hyper-parameters removes the Model Building advantage, causing diverse initial samples to always outperform models trained with non-diverse samples. These findings shed light on why, at least for BO-type optimizers, the use of diversity has mixed effects and cautions against the ubiquitous use of space-filling initializations in BO. To the extent that humans use explore-exploit search strategies similar to BO, our results provide a testable conjecture for why and when diversity may affect human-subject or design team experiments. The thesis is organized as follows: Chapter 2 provides an overview of existing studies that explore the impact of different initial stimuli. In Chapter 3, we explain the methodology used in the subsequent experiments. Chapter 4 presents the results of our initial study on the diverse initialization of BO (Bayesian Optimization) applied to the wildcat wells function. In this chapter we also investigate the conditions under which less diverse initial examples perform better and expand on these findings in Chapter 5 by considering additional ND continuous functions. The final chapter discusses the limitations of our findings and proposes potential areas for future research.Item GENERATING AND MEASURING PREDICTIONS IN LANGUAGE PROCESSING(2023) Nakamura, Masato; Philips, Colin; Linguistics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Humans can comprehend utterances quickly, efficiently, and often robustly against noise in the inputs. Researchers have argued that such a remarkable ability is supported by prediction of upcoming inputs. If people use the context to infer what they would hear/see and prepare for likely inputs, they should be able to efficiently process the predicted inputs.This thesis investigates how contexts can predictively activate lexical representations (lexical pre-activation). I address two different aspects of prediction: (i) how pre-activation is generated using contextual information and stored knowledge, and (ii) how pre-activation is reflected in different measures. I first assess the linking hypothesis of the speeded cloze task, a measure of pre-activation, through computational simulations. I demonstrate that an earlier model accounts for qualitative patterns of human data but fails to predict quantitative patterns. I argue that a model with an additional but reasonable assumption of lateral inhibition successfully explains these patterns. Building on the first study, I demonstrate that pre-activation measures fail to align with each other in cases called argument role reversals, even if the time courses and stimuli are carefully matched. The speeded cloze task shows that “role-appropriate” serve in ... which customer the waitress had served is more strongly pre-activated compared to the “role- inappropriate” serve in ... which waitress the customer had served. On the other hand, the N400 amplitude, which is another pre-activation measure, does not show contrasts be- tween the role-appropriate and inappropriate serve. Accounting for such a mismatch between measures in argument role reversals provides insights into whether and how argument roles constrain pre-activation as well as how different measures reflect pre-activation. Subsequent studies addressed whether pre-activation is sensitive to argument roles or not. Analyses of context-wise variability of role-inappropriate candidates suggest that there are some role-inappropriate pre-activations even in the speeded cloze task. The next study at- tempts to directly contrast pre-activations of role-appropriate and inappropriate candidates, eliminating the effect of later confounding processes by distributional analyses of reaction times. While one task suggests that role-appropriate candidates are more strongly pre- activated compared to the role-inappropriate candidates, the other task suggests that they have matched pre-activation. Finally, I examine the influence of role-appropriate competitors on role-inappropriate competitors. The analyses of speeded cloze data suggest that N400 amplitudes can be sensitive to argument roles when there are strong role-appropriate competitors. This finding can be explained by general role-insensitivity and partial role-sensitivity in pre-activation processes. Combined together, these studies suggest that pre-activation processes are generally insensitive to argument roles, but some role-sensitive mechanisms can cause role-sensitivity in pre-activation measures under some circumstances.Item Relations Between Latent Episodic Memory, Nap Habituality, and the Cortex During Childhood(2023) Allard, Tamara Lynn; Riggins, Tracy; Psychology; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)During childhood, episodic memory demonstrates marked improvements that are supported by the protracted development of the hippocampus and a larger network of cortical regions. To date, most research has focused on associations with the hippocampus in this age group. Few studies have explored the contribution of cortical regions and no studies have explored this longitudinally. Thus, the first aim of this dissertation was to examine the longitudinal co-development of cortical thickness and surface area in memory-related cortical regions with a latent episodic memory variable in 4- to 8-year-old children (N = 177). Findings, uncorrected for multiple comparisons, demonstrated that a thinner cortex in multiple episodic memory network regions (i.e., inferior frontal gyrus, inferior parietal sulcus, lingual gyrus, middle temporal gyrus, precuneus, lateral occipital cortex, superior frontal gyrus, superior parietal lobule, superior temporal gyrus, and temporal pole) at age 4 predicted more rapid improvements in memory performance from age 4 to 6 years. Similarly, greater surface area in the precuneus and less surface area in the medial orbitofrontal gyrus at age 4 also predicted more rapid improvements in memory performance from age 4 to 6 years. Additionally, results revealed that several regions demonstrate parallel co-development with latent episodic memory performance from age 4 to 8 years. Specifically, greater changes in cortical thickness and surface area of the entorhinal cortex were associated with greater changes in memory from age 4 to 6 years. Furthermore, cortical thickness of entorhinal cortex and surface area of anterior cingulate cortex, entorhinal cortex, inferior parietal sulcus, lingual gyrus, and superior temporal gyrus showed co-development with latent episodic memory from age 6 to 8 years. Together, these findings suggest that cortical thickness and surface area of the episodic memory network support improvements in memory performance during childhood. However, these findings did not survive correction for multiple comparisons. Although age-related differences were one focus of this investigation, individual differences were another. Specifically, during childhood children transition away from afternoon napping. This transition has previously been associated with differences in memory consolidation abilities and hippocampal maturation. These associations suggest that habitual nappers require more regular sleep to consolidate memories due to an immature episodic memory network. However, limited work has examined these associations outside the hippocampus. Therefore, the second aim of this dissertation was to examine whether regions that support longitudinal memory development differ as a function of nap habituality (N = 44). Findings revealed significant differences in cortical thickness of right inferior frontal gyrus and surface area of lateral occipital cortex, such that non-nappers demonstrated a thinner cortex and greater surface area in these regions compared to nappers, though these findings did not survive correction for multiple comparisons. Thus, although there is some evidence that memory-related cortical regions may differ based on nap habituality, additional work is needed to support this claim. Together this dissertation provides new data on the co-development of memory with brain structure in the episodic memory network and identifies individual differences that may be associated with these brain structures.