oURspace (Univ. of Regina)
Not a member yet
15455 research outputs found
Sort by
Categorization catastrophe: applying hypodescent and contact theory to biracial photo lineups
No full textA Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of Bachelor of Arts (Honours) in Psychology, University of Regina. 19 p.Past research shows that biracial individuals tend to be categorized into the most marginalized group (i.e., hypodescent; Ho et al., 2011; Chao et al., 2013). Per contact theory, the quantity and quality of contact with Biracial people (Civile & McLaren, 2022) may predict outgroup categorization and biases (Dickter et al., 2015). The current study sought to examine the role of hypodescent and contact theory in understanding participants’ ability to categorize target faces. Participants were asked to categorize Black-White biracial targets on both a discrete and continuous scale, as well as complete the General Intergroup Contact Quantity and Contact Quality Scale. Results of the study generally followed the expected trajectory, indicating that estimates of ancestry roughly paralleled actual morph level. Neither Biracial nor Black contact was a significant predictor of categorization. Together, these findings may be used to craft photo lineup studies for a wider range of targets. We argue that Psycholegal researchers should expand their range of target groups to include biracial populations. Future researchers should attempt to replicate these findings with other mixedrace groups and control for observer race.Studentn
Child Maltreatment History, Deployment-Related Traumatic Events, and Past 12-Month Cannabis Use Among Veterans in Canada
No full textObjective
Cannabis use among veterans in Canada is an understudied public health priority. The current study examined cannabis use prevalence and the relationships between child maltreatment histories and deployment-related traumatic events (DRTEs) with past 12-month cannabis use including sex differences among Canadian veterans.
Method
Data were drawn from the 2018 Canadian Armed Forces Members and Veterans Mental Health Follow-up Survey (response rate 68.7%; veterans only n = 1,992). Five child maltreatment types and 9 types of DRTEs were assessed in relation to the past 12-month cannabis use.
Results
The prevalence of lifetime and past 12-month cannabis use was 49.4% and 16.7%, respectively. Females were less likely than males to report lifetime cannabis use (41.9% vs. 50.4%; odds ratio [OR] 0.71; 95% CI, – 0.59 to 0.86). No sex differences were noted for past 12-month cannabis use (14.1% vs. 17.0%; OR 0.80; 95% CI, 0.60 to 1.07). Physical abuse, sexual abuse, neglect, any child maltreatment, most individual DRTEs, and any DRTE were associated with increased odds of past 12-month cannabis use after adjusting for sociodemographic and military variables. Some models were attenuated and/or nonsignificant after further adjustments for mental disorders and chronic pain conditions. Sex did not statistically significantly moderate these relationships. Cumulative effects of having experienced both child maltreatment and DRTEs compared to DRTEs alone increased the odds of past 12-month cannabis use. Statistically significant interaction effects between child maltreatment history and DRTE on cannabis use were not found.
Conclusions
Child maltreatment histories and DRTEs increased the likelihood of past 12-month cannabis use among Canadian veterans. A history of child maltreatment, compared to DRTEs, indicated a more robust relationship. Understanding the links between child maltreatment, DRTEs, and cannabis use along with mental disorders and chronic pain conditions is important for developing interventions and improving health outcomes among veterans.The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Data collection of the Canadian Armed Forces Members and Veterans Mental Health Follow-up Survey was supported by the Canadian Institutes of Health Research Foundation grant (No. 333252; Principal Investigator: Jitender Sareen), Department of National Defense. True Patriot Love and the Canadian Institutes of Military Veterans Health Research funded the knowledge translation activities. Preparation of this article was supported by a Tier 1 Canada Research Chair in Childhood Adversity and Resilience at the University of Manitoba (Afifi), and a CIHR Catalyst Grant: Cannabis and Mental Health Grant (Afifi)
Dimensions of competition in urban cannabis markets
No full textAbstractWe report the results of a survey of legal cannabis retail operators in the two largest urban centres of a Canadian province that uses a fully private model to determine the dimensions of competition between retail outlets. We find that, in addition to expected price and location competition, customer loyalty is central in the opinion of retailers. Firms also did not see the illicit market as their primary competition, instead identifying nearby legal retailers as their main competitors. Restrictions on advertising loomed large in the minds of retailers as a policy‐based impediment to their success. These results are further supported by visual inspection of the exterior of all stores in one of the urban centres.Faculty of Graduate Studies and
Research, University of Regina, FGSR
Undergraduate Research Award
Investigations in plasmid biology at the levels of the gene, community, and ecosystem
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Science in Biology, University of Regina. xvii, 143 p.Bacteria are diverse microorganisms that interact with and impact human life in myriad
ways. Their genomes are typically comprised of a single chromosome on which core genes are encoded, and additional genetic structures that carry additional, accessory loci. Plasmids are one type of accessory genetic element that have a capacity to jump horizontally between bacteria,
and in doing so, provide genes and associated functions to new hosts.
Plasmids have established impacts on bacterial physiology and evolution, but many ques-
tions remain about how plasmids persist, spread, and change in nature. Furthermore, since plas-
mids evade standard sequencing and assembly approaches, scientists have uncovered only a frac-
tion of total plasmid diversity in nature.
In this thesis I describe my efforts to investigate plasmids at the levels of gene, commu-
nity, and ecosystem. I used computer-based methodology to analyze existing plasmid sequences,
took traditional microbiological approaches to test plasmid traits in the lab, and worked toward
sequencing plasmids from an ecologically important natural environment.
By analyzing the distribution of a conserved protein on public plasmid sequences, I dis-
covered a long-term association between the locus and a specific backbone. I used bioinformatic methodology to investigate and predict roles that the protein might fulfill and related these to the plasmid’s immense success across several bacterial genera. This investigation furthered our understanding of plasmid biology at the level of a single gene. Then, I used further computer-based
methodology to chart a population of unique and pervasive plasmid sequences. I described sev-
eral lineages within this population and related these to bacterial hosts in which they are found,
then identified one group that appears to have mobilized between several hosts in recent history.
This study centred upon a population of plasmids and how they might interact with bacterial
communities. Taking my investigations on plasmids to the lab, I used molecular techniques to
create an experimental system to track the mobility of a Salmonella plasmid. I found that the plasmid is more frequently transferred between bacteria when it encodes a toxin-antitoxin sys-
tem. This analysis further related to plasmids in bacterial communities as it identified a modula-
tor of plasmid transmission. Finally, I made efforts to sequence plasmids from freshwater agri-
cultural ponds in Saskatchewan, Canada. Although I did not characterize complete plasmid se-
quences, I took steps toward reliably sequencing environmental plasmids from an important eco-
system.
Together, these research projects contribute to an enhanced understanding of the evolu-
tionary success strategies and natural diversity of bacterial plasmids. My work highlights that re-
search in the field of plasmid biology is essential and exciting.Studentye
Improving semantic segmentation of difficult classes in autonomous driving using Attention-Convolution Feature Extractor
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Applied Science in Software Systems Engineering, University of Regina. viii, 68 p.Deep learning has indeed brought about a revolution in computer vision applications, showcasing
impressive outcomes across tasks like object recognition and autonomous driving systems. Nevertheless,
there is still a demand for further research and the development of models that can achieve human-level
performance while efficiently utilizing hardware and data resources. Specifically, in the realm of
semantic segmentation for autonomous driving systems, challenges arise when it comes to accurately
detecting and segmenting important object classes, such as Persons, that may be obstructed or only
partially visible within street scene images. The challenges of handling these classes, coupled with the
current approach of treating all object classes equally, necessitates focused research efforts. Addressing
these challenges and improving the detection and segmentation of crucial classes forms the central focus
of this thesis. By exploring innovative approaches and advancing the state of the art, this research aims
to contribute to the development of more accurate and robust semantic segmentation models for
autonomous driving systems.
In the field of semantic segmentation, two prominent architectures, UNet and Fully Convolutional
Networks (FCN), have served as the foundations for many subsequent models. Recent breakthroughs in
deep learning have highlighted the effectiveness of attention models in various applications, including
natural language processing (NLP). Building upon these advancements, this thesis introduces the concept
of Attention Convolutional Block (ACB) integrated into the encoder sections of both UNet and FCN.
The primary objective of this thesis is to enhance the performance of semantic segmentation models
when confronted with important object classes, such as a Person situated in the far background of an
input image, or a cyclist obstructed by a car or a bush. By incorporating ACB blocks, which leverage
attention mechanisms, into the encoder sections of UNet and FCN, the proposed approach aims to
improve the segmentation of these important object classes. The integration of ACB blocks allows the
models to focus their attention on crucial regions of the input image, capturing long-range dependencies
and effectively capturing the features necessary for accurate segmentation. By specifically addressing
the challenges associated with important object classes, this research aims to push the boundaries of
semantic segmentation models and improve their performance in complex scenarios.
In our experiments, we evaluate our proposed models using the Cityscape Dataset, which provides
annotated images of everyday street scenes with 20 different object classes. We compare the results of
our models with other existing models based on Intersection over Union (IoU) scores and visual
representations of the output images. Specifically, we investigate the impact of incorporating the ACB
and compare the performance of UNet and FCN with and without the ACB block. The evaluation metrics
include mean IoU scores, per-class IoU scores, and qualitative examination of the segmented output
images. Our findings demonstrate that the use of the ACB block leads to improvements in both mean
IoU scores and per-class IoU scores for challenging object classes such as pedestrians and bicycles. For
instance, the mean IoU score of FCN improves by 0.1 when the ACB block is used. In the case of the
Bicycle class, FCN-32 without the ACB block achieves an IoU score of 0.64, whereas AC-FCN achieves
an IoU score of 0.69. Furthermore, the qualitative examination of the segmented output images reveals
more accurate segmentation of these difficult object classes when the ACB block is incorporated. This
demonstrates the effectiveness of our proposed approach in enhancing the detection and segmentation
of objects in street scenesStudentye
Attentional blink: Restricted attention necessitates conscious perception
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Science in Experimental and Applied Psychology, University of Regina. vii, 77 p.In attention capture studies, a target stimulus is poorly identified when presented
200-500 ms after the presentation of a task-relevant distractor, a phenomenon referred to
as the attentional blink. Evidence regarding the necessity of consciously perceiving the
distractor in yielding an attentional blink is inconsistent in the literature. The goal of this
thesis was to clarify under which conditions conscious perception of a distractor is
necessary to trigger an attentional blink. It was hypothesized that the presence of a blink
would vary according to the predictions of the novel restricted attention hypothesis, such
that additional conditions must be met for a distractor to capture attention and cause a
blink. The results provide evidence for this hypothesis, such that attention becomes
restricted when attention is deployed broadly. Critically, it was found that conscious
perception was a necessary condition in triggering a blink only when attention was
deployed broadly. It was suggested that conscious perception becomes necessary in such
conditions because it is responsible for shifting attention to the location of interest. This
finding was interpreted through the novel diffused attention model. The present thesis
clarifies when and why conscious perception becomes necessary.
Keywords: attentional blink, attention capture, contingent capture, conscious perceptionStudentye
A narrative exploration of the right to health in the lives of Indigenous women
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Kinesiology and Health Studies , University of Regina. xvi, 306 p.This study explores, through critical narrative analysis, the understanding of Indigenous
women about conditions that they need to be healthy, and how their stories and reflections
provide a critique that can inform thinking around the right to health. Literature from varied
disciplinary perspectives describes the right to health, and a body of health human rights, as
conceptual tools that identify the conditions all people require to be healthy, encompassing not
just health care and access to material goods, but equality, culture, power and participation.
Literature also provides some critique of human rights, and indicates that their alignment with
dominant discourses and powers may be a barrier to their effectiveness for Indigenous people.
With reference to Habermas’ theories of communicative action, including the colonization thesis,
the lifeworlds of 14 Indigenous women were explored in relation to the institutional discourse of
health human rights. The study finds that in the stories that women shared there was some
validation of human rights instruments relating to health, which identify as rights violations
health harms such as violence, disruption of families, experiences of racism, and lack of support
for mental health. However, women’s interpretation of these events often differed from
institutional discourse in that they located responsibility for violations less in the people or
organizations that harmed them, and more in processes of colonization carried out by successive
Canadian governments, that effectively undermined their rights and their health. As they
reflected on their stories, women identified a right to knowledge about history and the impact of
colonization on Indigenous people as being important to their physical and mental health. Other
findings are that a rights-based assessment of women’s health that focuses on experiences of
violations and harms may be perceived as deficit-based. To be more meaningful to Indigenous
women, a discourse of human rights in health could speak to their strengths and resources, and
support broadly defined goals in physical, spiritual and mental health by removing barriers to
agency. This study joins a body of other research in finding that explicit rights-based
participation in service delivery and health policy development and evaluation may help to avoid
abuses in the future, but may require more autonomous forms of governance and service delivery
to address longstanding power imbalance and distrust. The study concludes that a discourse of
health human rights can better meet the needs of Indigenous women when colonialism is named
as a human rights abuse and the primary cause of health inequity that affects their families and
communities, reinforcing their life world knowledge with rights-based accountability, and
creating common understanding in the public sphere.Studentye
Development of high-resolution climate projections over Canada in the 21st century
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Environmental Systems Engineering, University of Regina. xii, 175 p.In this research, high-resolution climate projections over Canada have been developed through the WRF model. The spatial and temporal variations of Canada’s temperature, precipitation, and precipitation extremes in the 21st century have been comprehensively analyzed. Potential mechanisms in terms of temperature, precipitation, and precipitation extremes over Canada have been investigated, and effects of anthropogenic warming on these variables have been revealed. In addition, permafrost degradation under climate change is extended to a global scale, and a dynamic longterm classification scheme for discontinuous permafrost is proposed.
The annual mean temperature over Canada is projected to increase by [1.53, 1.98], [2.51, 3.86], and [2.94, 6.19]°C in the 2030s, the 2050s, and the 2080s under RCP4.5 and RCP8.5 respectively, with largest increase in winter. The annual total precipitation is projected to increase by [16.33, 68.96], [64.80, 121.62], and [123.62, 184.33] mm in three future periods. It is found that the westerlies have significant impacts on Canada’s precipitation variations, and extreme precipitation frequency are related to ENSO phases.
A dynamic SOM-based permafrost classification scheme (DSOMPCS) is developed to provide robust, objective, and detailed classifications for discontinuous permafrost. Linear relationship is found between global temperature increase and permafrost loss. When global temperature increases by 1°C, there will be a 2.5 million km2 of permafrost loss.
This research will help provide insights into effects of anthropogenic warming on multiple aspects of climatology. The results can provide valuable information for mitigation and adaptation of climatic changes in a Canadian and global context.Studentye
Modeling and simulation of methane reforming reactors under sulfur poisoning conditions
No full textA Thesis Submitted to the Faculty of Graduate Studies and Research In Partial Fulfillment of the Requirements for the Degree of Master of Applied Science in Process Systems Engineering, University of Regina. xx, 324 p.The production of hydrogen from residual gases (i.e., solution gas produced from heavy oil, or biogas produced from anaerobic digestion of biomass) presents an opportunity to utilize a low-value feedstock otherwise contributing to climate change. Such gas streams may contain hydrogen sulfide, which is well-known to poison catalysts and reduce reaction rate. This has driven development of sulfur-tolerant catalysts, yet the incorporation of sulfur poisoning in fixed-bed reactor modeling – a critical tool used in the engineering design process – has received very limited attention.
This thesis focuses on the development of a reactor modeling framework that incorporates sulfur poisoning. Specifically, two model variations are developed: a simplified heterogeneous model, and a conventional heterogeneous model. Both the simplified and conventional heterogeneous models consider the transient energy and species mass balances via partial differential equations, but differ in the treatment of catalyst particles (wherein the simplified model considers only thin film transport to the catalyst surface and uses an effectiveness factor to approximate the catalyst’s performance, while the conventional model explicitly accounts for the energy and mass balances within the particle). Both models are discretized using the finite volume method and programmed in MATLAB, and are then validated against numerical results from the literature. Interestingly, suspected assumptions concerning the modeling of heat transfer within the literature reference used in validation are identified using the proposed modeling frameworks.
Both the simplified and heterogeneous models are then validated against experimental data available in the literature describing the loss of methane conversion
seen as a function of hydrogen sulfide concentration in the feed gas. A Maxted-type deactivation expression is used to calculate catalyst activity based on sulfur coverage, itself calculated using a Temkin isotherm. The reaction rate of the literature reference is approximated using an adaptation of methane reforming kinetics over nickel-alumina catalyst, showing activation energy corrections of 24.4 kJ/mol and 27.0 kJ/mol (using the simplified and conventional heterogeneous model, respectively) for the chosen rate expression. This is then used to scrutinize the experimental data further to assess any potential transport limitations that may have existed, noting some limitations appear above 1173 K. Lastly, the models are then used to estimate the order of deactivation based on the referenced experimental reactor, both finding =1 with an average absolute error of 27.2% and 26.2% (for simplified and conventional modeling, respectively). This contrasts the more commonly assumed order of =3, though may be reasonable in light of variability observed in the order of deactivation for methanation, and in considering similar deactivation from other experimental studies.
Lastly, assessment of modeling performance under transient and parametric conditions is made. Simplified heterogeneous modeling is noted to undergo a convergent behaviour (when mesh size and time increments are too large) that – while not affecting steady-state results – appears to modify transient results in a manner not reflective of experimental studies in the literature. Conventional modeling did not show this behaviour. Further, conventional modeling was shown to predict higher catalyst effectiveness once poisoning had taken place, predicting higher methane conversion than simplified modeling. Generally, conventional modeling is preferred when considering sulfur poisoning in methane reforming reactors.Studentye