Browsing by Subject "learning"

Now showing 1 - 16 of 16

Applications of GIS Methods for Environmental Learning in South Texas
(2007) Young, Lauren P.; Smith, Elizabeth
With the application of GIS being utilized in a variety of disciplines, it is appropriate that these new techniques should be incorporated into the classroom. In developing strategies for teaching with GIS in geography, earth science, and environmental studies in South Texas, teachers are offered another opportunity for innovative techniques to reach their students. By using an expanded approach from the ESRI book, Mapping Our World-GIS Lessons for Educators, selected topics will be presented to illustrate the design, development, and application of the approach. The following topics are targeted for development within the project: introductory GIS skills, climate, hydrology, wetlands, migratory birds, fisheries, and vegetation using regional and local datasets. This specialized learning experience will allow rural communities to increase their regional geographic knowledge, while also integrating state knowledge skills required at each grade level.
Daily Online Testing in Large Classes: Boosting College Performance while Reducing Achievement Gaps
(PLOS One, 2013-11-20) Pennebaker, James W.; Gosling, Samuel D.; Ferrell, Jason D.
An in-class computer-based system, that included daily online testing, was introduced to two large university classes. We examined subsequent improvements in academic performance and reductions in the achievement gaps between lower- and upper-middle class students in academic performance. Students (N = 901) brought laptop computers to classes and took daily quizzes that provided immediate and personalized feedback. Student performance was compared with the same data for traditional classes taught previously by the same instructors (N = 935). Exam performance was approximately half a letter grade above previous semesters, based on comparisons of identical questions asked from earlier years. Students in the experimental classes performed better in other classes, both in the semester they took the course and in subsequent semester classes. The new system resulted in a 50% reduction in the achievement gap as measured by grades among students of different social classes. These findings suggest that frequent consequential quizzing should be used routinely in large lecture courses to improve performance in class and in other concurrent and subsequent courses.
Effects of Mood on Directed Forgetting
(2023) Zheng, Sabrina; Lewis-Peacock, Jarrod A.
Working memory is the ability to retain and manipulate a limited amount of information. It’s limited in the amount of information it can hold simultaneously, therefore it would be beneficial to effectively remove information intentionally. The item method directed forgetting task is one way to investigate how information is removed from memory. Studies have continually shown that memory for items that are associated with a remember cue are remembered better than items associated with a forget cue, known as the directed forgetting effect. Mood states may be one factor that impacts the directed forgetting effect, but these effects have not been extensively studied. Negative mood states tend to enhance detailed processing leading to better memory for item level details, whereas positive mood states promote global processing leading to more categorical memory. Prior literature has found mixed results regarding the impact of mood on an item-method directed forgetting task; some finding that a negative mood reduces forgetting while others finding that negative mood has no effect on forgetting when compared to a positive mood condition. These previous studies did not specify methods of forgetting which may explain the inconsistent results. This study investigated the effects of sad and happy moods on the directed forgetting operational cues: maintain, clear, and suppress. Results suggested that mood was successfully manipulated only in the positive mood group which could explain the insignificant differences in memory between the two mood groups. Interestingly, memory was worse for clear cued memory items over maintain items in the positive mood group, suggesting that clear was a more effective forget cue than suppress. Future studies will investigate alternative methods to manipulate mood more effectively to fully explore our research question. These effects could help us better understand and treat patients suffering mental health conditions involving intrusive thoughts.
Electrical Stimulation of Lateral Habenula during Learning: Frequency-Dependent Effects on Acquisition but Not Retrieval of a Two-Way Active Avoidance Response
(PLOS One, 2013-06-28) Ilango, Anton; Shumake, Jason; Wetsel, Wolfram; Scheich, Henning; Ohl, Frank W.
The lateral habenula (LHb) is an epithalamic structure involved in signaling reward omission and aversive stimuli, and it inhibits dopaminergic neurons during motivated behavior. Less is known about LHb involvement in the acquisition and retrieval of avoidance learning. Our previous studies indicated that brief electrical stimulation of the LHb, time-locked to the avoidance of aversive footshock (presumably during the positive affective “relief” state that occurs when an aversive outcome is averted), inhibited the acquisition of avoidance learning. In the present study, we used the same paradigm to investigate different frequencies of LHb stimulation. The effect of 20 Hz vs. 50 Hz vs. 100 Hz stimulation was investigated during two phases, either during acquisition or retrieval in Mongolian gerbils. The results indicated that 50 Hz, but not 20 Hz, was sufficient to produce a long-term impairment in avoidance learning, and was somewhat more effective than 100 Hz in this regard. None of the stimulation parameters led to any effects on retrieval of avoidance learning, nor did they affect general motor activity. This suggests that, at frequencies in excess of the observed tonic firing rates of LHb neurons (>1–20 Hz), LHb stimulation may serve to interrupt the consolidation of new avoidance memories. However, these stimulation parameters are not capable of modifying avoidance memories that have already undergone extensive consolidation.
Electrophysiological characterisation of spatial memory cells in the anterior thalamic nuclei
(2012) Tran, Thao (Emma)
The Papez circuit appears to play a crucial role in the formation of episodic memory. In particular, the interconnections between the hippocampus and the anterior thalamus seem to be vital for an animal’s ability to orientate itself to and navigate through its environment. This is further supported by electrophysiological studies in freely moving rats, which have discovered neurons that encode for location (place cells) in the hippocampus and directional heading (head direction cells) in the anterodorsal and anteroventral nuclei of the thalamus. Due to the relative inaccessibility of the anteromedial nucleus of the thalamus, its electrophysiological properties are not well-characterised, and its role in spatial processing still remains elusive. In this study, we were able to successfully record from a population of head direction (20%) and theta rhythmic neurons (15%) in the anteromedial thalamic nucleus, suggesting that this structure plays an important role in the temporal–diencephalic pathways that regulate spatial processing.
Examining the role of Dif B in the learning and memory of Drosophila melanogaster using a simplified aversive phototaxis suppression assay
(2023) Gedamu, Hanna; Atkinson, Nigel
Drosophila melanogaster, the common fruit fly, is a great model for investigating learning and memory. As such, they are used to model human disorders involving neurological deficits, including Alzheimer’s disease and Fragile-X syndrome (1). This is made possible by the existence of various analog proteins and processes between Drosophila and humans. For example, the Toll-like receptor (TLR) pathway is a signaling pathway involved in human innate immunity and was initially discovered in D. melanogaster as the Toll pathway, hence the nomenclature. One of the actors downstream of the Toll signaling pathway of the fly is a nuclear factor kappa B (NF-κB) called Dorsal-like-immunity factor (Dif). Dif mRNA undergoes differential splicing to produce two protein isoforms, DifA and DifB. Between these two splice isoforms, the absence of the DifB variant has been seen to dramatically alter various fly behaviors and is thus the focus of this paper. Neurological deficits like those presumably caused by DifB mutations can be assayed using an existing experiment that allows researchers to leverage an organism’s natural affinity for light. It involves aversive phototaxic suppression and has been found to be effective with D. melanogaster. The version of the assay used in this paper has been modified from previously published methods to be a more efficient and accessible way to test learning. In general, the assay utilizes operant conditioning to train flies before they are tested for memory. By using an aversive chemical stimulus, flies can be trained to exhibit a photophobic response and avoid light, which is contrary to their natural behavior of being attracted to light. Flies that are mutant for different genes can be scored on their ability to learn and recall the punishment to help identify genes or proteins that are involved in learning processes. To explore this and test the efficacy of the newly developed methods, flies mutant for DifB expression were tested for learning and memory deficits using the modified procedure. It was ultimately found that the data trended as hypothesized and will likely be ideal with the collection of more data. With recent studies suggesting that NF-κBs play a role in adult mammalian memory and brain plasticity (2), the expectancy is that the revealed importance of NF-κBs in the learning and memory of arguably the most foundational model organism will support human learning and memory disorder research, especially upon discovery of a Dif or DifB mammalian analog.
Fast Food, Slow Learning
(The Texas Scientist, 2016) The Texas Scientist
High resolution fMRI reveals distinct forms of associative novelty in the medial temporal lobe
(2012) Manthuruthil, Christine; Preston, Alison
Both Alzheimer’s Disease and Parkinson’s Disease involve alterations to the structure of the medial temporal lobe (MTL). Varying patterns of neuronal connectivity, however, suggest that not only does the MTL support learning and memory, but that its subregions play distinct roles in these processes as well. The exact nature of these contributions remains an area of active investigation. Examinations of associative novelty may offer an important tool for characterizing the processes carried out by different subregions. Associative novelty can be further broken down into associative novelty per se, which are simply novel stimulus configurations, and associative mismatch novelty, which are novel stimulus configurations that violate existing expectations. In this study, we used high resolution fMRI to characterize different associative novelty signals across the MTL; specifically, we were interested in whether there was a dissociation of associative novelty signal types between MTL subregions, or instead, a functional specialization for associative novelty signal types distributed across these subregions. Establishing subregional function could help elucidate the spectrum of cognitive deficits manifest in both Parkinson’s and Alzheimer’s patients.
Higher Education in the New Digital Ecosystem
(2016-05-10) Bass, Randy
Membership query removal implies non-trivial cryptographic primitives
(2010-05) Wright, John; Adam Klivans
Understanding learning and discovering to what extent it can be automated is one of the major problems in computer science. This is of obvious relevance when trying to create intelligent robots, for example, and fields as diverse as biology and economics use learning algorithms to help manage and understand their extremely large data sets. There has even been work devoted to analyzing computational learning systems in order to gain intuition about natural learning systems: authors of a '60s textbook on learning machines called perceptrons give voice to their hope that among their readership are "psychologists and biologists who would like to know how the brain computes thoughts" [MP69]. More recently, we have seen the development of a mathematical model for analyzing evolution as a learning process [Val09]. One major approach taken to analyzing learning has been to look at classification problems. As an example of such a problem, consider the scenario of teaching a computer to tell when an image has a bird in it. Say you go about doing this by showing the computer a ton of pictures and telling it which ones have birds in them. After giving it some time to compute, you might hope that it can now figure out whether or not images that it hasn't seen before have birds in them. But how can you be sure that this is a fruitful way of teaching image recognition? Computational learning theory attempts to answer this question in a mathematically rigorous way by formalizing these classification scenarios in models that de fine exactly the interaction between the learner and what it's trying to learn and give a metric for evaluating the learner's success. When learning is so formalized in models, we can begin to answer natural questions about learning, such as to what extent does asking questions help. Two of these models that we will focus on here are Leslie Valiant's Probably Approximately Correct (PAC) model of learning and Dana Angluin's exact learning model. Researchers have carefully studied the issue of whether giving the learner the ability to ask questions (or to pose membership queries, in learning theory terminology) improves its ability to learn, and the answer is dependent on the learning model under consideration. In the agnostic learning model, for instance, a learner which works with membership queries can be converted to one which works without membership queries, although in the distribution-specifi c agnostic setting membership queries do increase the power of the learner [Fel09]. In both the PAC model and the exact learning model, on the other hand, it has been shown that there are some concepts that can be learned with membership queries but cannot be learned without. A well known example of such a class is the class of deterministic nite automatas (DFAs), which were shown to be learnable with membership queries in [Ang87] but not learnable without membership queries in [KV94] given suitable assumptions. Another major development of computer science has been that of modern cryptography. Starting from certain basic and currently unavoidable assumptions, powerful cryptographic protocols have been developed which multiple consenting parties can use to engage in secure communication, which has a wide variety of uses, including electronic commerce. A major part of the cryptographic research e ffort has been to further understand the assumptions that form the foundation of these cryptographic protocols, with one eventual aim being perhaps to construct cryptographic protocols that do not require any assumptions. An example of a protocol that will be used in this paper is the signature scheme, which allows a user to sign any message they send so that whoever receives it knows who sent it. The link between cryptography and learning theory was recognized very early on. The very first paper on PAC learning rules out learning polynomial-size circuits given cryptographic assumptions [Val84]. At a high level, a cryptographic protocol proven to be secure has a guarantee attached to it which prevents a malicious user from breaking the security, no matter how well that user may learn. So provably-secure cryptography bounds the ability to learn. One such link was shown by Angluin and Kharitonov in [AK95]. They showed that giving learners membership queries in the PAC model does not help them for a large number of concepts being learned if signature schemes exist. In this paper, we investigate whether the converse of this statement can be proven, i.e. whether it can be shown that if making membership queries does not help learners, then signature schemes exist. We look at severely weakened versions of the converse and prove a couple of results. On the whole, however, our attempt at converting their result is largely unsuccessful.
Museums---Their Use and Place in Learning and in the Transmission of Culture
(University of Texas at Austin, 1921-06-10) University of Texas at Austin
On Waves, Clusters, and Diffusion: A Conceptual Framework
(2005) Elkins, Zachary; Simmons, Beth
This article makes a conceptual and theoretical contri- bution to the study of diffusion. The authors suggest that the concept of diffusion be reserved for processes (not outcomes) characterized by a certain uncoordinated interdependence. Theoretically, the authors identify the principal sources of clustered policy reforms. They then clarify the characteristics specific to diffusion mecha- nisms and introduce a categorization of such processes. In particular, they make a distinction between two types of diffusion: adaptation and learning. They argue that this categorization adds conceptual clarity and distinguishes mechanisms with distinct substantive consequences.
Opportunity Creation in Innovation Networks: Interactive Revealing Practices
(2014) Jarvenpaa, Sirkka L.; Valikangas, Liisa; Jarvenpaa, Sirkka L.
Innovating in networks with partners that have diverse knowledge is challenging. The challenges stem from the fact that the commonly used knowledge protection mechanisms often are neither available nor suitable in early stage exploratory collaborations. This article focuses on how company participants in heterogeneous industry networks share private knowledge while protecting firm-specific appropriation. We go beyond the prevailing strategic choice perspectives to discuss interactive revealing practices that sustain joint opportunity creation in the fragile phase of early network formation.
Reinforcement learning strategies support generalization of learned hierarchical knowledge
(2021) McKee, Connor; Preston, Alison
In our everyday lives, we must learn and utilize context-specific information to inform our decision making. How do we learn what choices to make based on our memories? Prior rodent work has demonstrated that after learning, knowledge becomes organized hierarchically in a context-dependent manner. Here, we quantify the emergence of context-dependent hierarchical knowledge during learning and examine the flexible use of that knowledge to generalize across different scenarios. Participants learned about objects with context-dependent reward values in an X-shaped virtual environment consisting of an elongated, contextually-varying hallway with decision points on either end. First, participants learned the context-dependent object-reward pairings for one set of three objects. Next, they learned the context-dependent object-reward pairings for a new set of three objects. We hypothesized that prior knowledge of the hierarchical structure would generalize to the second set of objects as evidenced by a facilitation in learning rates. Participants gradually learned the context-dependent object-reward pairings during learning. When introduced to the new object set, learning rates did not significantly differ, indicating generalization of the hierarchical reward structure to the new object set. To further quantify how decision making unfolded, we applied three types of reinforcement learning (RL) models to our behavioral data: model-free, model-based (MB), and combination model-based model-free (MBMF). The MB model performed the best at using participants’ past selections to successfully predict future decisions and reward value expectations, indicating that current decisions were guided by prior selections. The MBMF model was best able to represent changes in participant learning across runs, possibly due to the model’s ability to assess different learning strategies. Overall, our results demonstrate that participants learned to flexibly decide which actions were the most adaptive, promoting correct decision-making in a given context. Furthermore, the structure of prior knowledge may support the generalization of learned experience.
The Efficacy of Film as a Tool for Social-Emotional Learning
(2021-05) Alley, Ellen
The purpose of this study is to test the efficacy of using film and television clips as a tool for teaching emotion in conjunction with standard written materials. Developing and assessing the effectiveness of an educational tool for social-emotional learning is critical to address the gap in the literature regarding ways to improve emotional recognition, understanding, and labeling skills in adults. The specific goals of this study are threefold: 1. Develop an enhanced tool for social-emotional learning (as outlined in the ability model of emotional intelligence) that utilizes film and television clips to enhance learning. 2. Assess the efficacy of the enhanced learning approach by comparing measures on emotional granularity and recognition to the standard group. 3. Assess the retention of the acquired skills over time by additional comparisons between the enhanced learning and standard groups. For the purpose of testing the efficacy of this tool, four emotions have been selected to be tested: shame, guilt, humiliation, and embarrassment. These emotions were selected because existing research demonstrates a positive relationship between emotional recognition of these affects and adaptive mental health outcomes. Additionally, the literature shows increased emotional regulation benefits for distinguishing negative emotions compared to positive emotions. Last, these emotions are commonly used, conflated, and confused words. The study, which includes a pretest, standard and enhanced learning, posttest, and two-week follow-up posttest, will collect quantitative data from a sample of 118 participants. This thesis will first examine the relevant literature regarding emotional intelligence, storytelling and emotion, and teaching tools. Second, it will describe the creation of the current tool, including the selection of emotions and film clips. Next, it will describe how emotional granularity, understanding, and recognition will be tested. Finally, it will examine the results of the study and the larger implications for the findings. This study attempts to explore address the gap in the literature regarding strategies for teaching emotion recognition, labeling, and understanding.
Training in action video games can increase the speed of perceptual processing. However, it is unknown whether video-game training can lead to broad-based changes in higher-level competencies such as cognitive flexibility, a core and neurally distributed component of cognition. To determine whether video gaming can enhance cognitive flexibility and, if so, why these changes occur, the current study compares two versions of a real-time strategy (RTS) game. Using a meta-analytic Bayes factor approach, we found that the gaming condition that emphasized maintenance and rapid switching between multiple information and action sources led to a large increase in cognitive flexibility as measured by a wide array of non-video gaming tasks. Theoretically, the results suggest that the distributed brain networks supporting cognitive flexibility can be tuned by engrossing video game experience that stresses maintenance and rapid manipulation of multiple information sources. Practically, these results suggest avenues for increasing cognitive function.
(PLOS One, 2013-08-07) Glass, Brian D.; Maddoz, Todd W.; Love, Bradley C.
Administrative deposit of works to UT Digital Repository: This works author(s) is or was a University faculty member, student or staff member; this article is already available through open access at http://www.plosone.org. The public license is specified as CC-BY: http://creativecommons.org/licenses/by/2.5/. The library makes the deposit as a matter of fair use (for scholarly, educational, and research purposes), and to preserve the work and further secure public access to the works of the University