69238 research outputs found
Sort by
Identifying Emotion Categories in te reo Māori Speech through a Community-Oriented Approach
No full textWe, as humans, use emotions to convey our thoughts and interact with the world. How emotions are categorised is essential in various domains such as emotion-based technology development, psychological studies and emotional well-being. In the psychology literature, emotions are often classified into six basic categories: anger, disgust, fear, happiness, sadness, and surprise. Many emotion-based studies in various languages adopt these categories or their direct translations, assuming that emotions are universal. However, it has been argued that studies attempting to verify the universality of emotions have not been conducted in a culturally appropriate manner. Therefore, it remains questionable whether emotions are categorised similarly or differently across cultures. Furthermore, these emotion categories have been verified using facial expressions based on images but have not yet been validated for speech.
Aotearoa New Zealand’s only Indigenous language, te reo Māori (the Māori language), spoken by the Māori people, coexists with English. Generally, Māori speakers are bilingual in both English and te reo Māori, and they may describe emotions using a mix of both languages. These days, with emotion analysis informing technology development, it is essential to evaluate the universality of presumptive universal emotions while understanding common social emotions across different cultures. No study has yet confirmed that these presumptive universal emotions hold for Māori. Research indicates that Māori and Pākehā (New Zealand Europeans) experience different patterns of emotional reactions to social situations. This finding suggests that the above emotion categories might not fully capture Māori emotions. Te reo Māori includes a wide range of emotion terms without a direct English translation, and scholars have suggested that the nuances of Māori emotions can be lost in translation. Thus, some emotions may not be fully captured even when emotion categories are directly translated from English to te reo Māori.
To address these gaps, this study takes a community-oriented approach to identify emotion categories in te reo Māori speech. Initially, a questionnaire was conducted with te reo Māori speakers, where participants were presented with te reo Māori speech recordings and asked to write down emotions they could identify. The questionnaire resulted in the identification of 218 emotion terms in both English and te reo Māori. Then, a focus group with experts in te reo Māori was conducted to group the identified emotions into several broader categories. The findings from the focus group will be presented at the conference. All the steps in this study were designed in close collaboration with te reo Māori researchers to maintain cultural appropriateness. The identified emotion categories in this study will be used to develop emotion-based speech technology in future
The Validity of New Zealand Corporate Transactions Undertaken Contrary to the Interests of the Company
No full textNew Zealand law relating to the impact on corporate transactions of a breach of the directors’ duty to act in the best interests of the company is complex and not well understood. This makes it difficult for parties to commercial transactions to know where they stand.
This thesis sets out and analyses the current law, including uncertainties in the law. It also suggests how the law might appropriately be reformed through amendments to the Companies Act 1993.
The thesis suggests that the security of commercial transactions would be enhanced by clarifying that company directors will not be considered to lack actual authority to enter into contracts as a matter of agency law just because they had a subjective motivation to act contrary to the company’s interests.
Such subjective mismotivation will, however, amount to a breach of fiduciary duty, giving rise to the equitable remedy of rescission (avoidance of transactions). That remedy provides the company with the right to avoid the contract except where the contracting third party is unaware of the breach of duty. The thesis recommends that the availability of the remedy of rescission, and the circumstances in which a company loses the right of rescission, should be spelled out in the Act.
In addition, the thesis recommends that the Act also clarifies the circumstances in which a company can effectively ratify (affirm) a contract that is voidable due to a breach of directors’ duty. This legislative clarification would include specifying that shareholders associated with the directors in breach cannot vote on a shareholder resolution affirming a voidable transaction.
The suggested legislative amendments will assist in advancing the original objective of the Law Commission in making New Zealand company law more accessible. The amendments would also draw an appropriate balance between policy objectives of enhancing the certainty and security of commercial transactions, and encouraging integrity and honesty in commercial dealings
Dementia Identification: Machine Learning and Routinely Collected Health Data
No full textDementia is a growing global public health concern, affecting an estimated 55 million people worldwide, with approximately 10 million new cases diagnosed annually, according to the World Health Organization (WHO). In Aotearoa New Zealand, prevalence estimates based on national datasets suggest that dementia affects 3.8%–4.0% of individuals aged 60 and older. When accounting for undiagnosed cases using a capture-recapture method, this estimate increases to 9.2% (95% CI: 8.9%–9.6%), with higher prevalence rates observed among Māori and Pacific populations. However, overseas research indicates approximately 60% of dementia cases remain unidentified, leading to an underestimation of true prevalence. Dementia identification is crucial due to the significant impact on individuals, the stigma associated with the disease, the burden placed on caregivers, the financial strain on healthcare systems, and the potential to optimise brain health even after diagnosis.
This study aimed to develop and evaluate machine learning models for identifying dementia using routinely collected health data while considering public attitudes and preferences regarding health data usage in research.
Routinely collected health data were obtained from the Te Whatu Ora Counties Manukau population aged 65 and older at the time of data extraction. Sociodemographic and clinical data, including both longitudinal and cross-sectional information, were included. Dementia status was determined using pharmacy records (antidementia drug prescriptions), interRAI assessments (dementia-related evaluations), and hospitalization records (International Classification of Diseases, Tenth Revision – ICD-10 codes related to dementia). Patients with at least one dementia-related record in these datasets were classified as living with dementia. A nested one-to-one case-control design was implemented, retaining pre-diagnostic information across six time windows prior to dementia diagnosis (0 days, 6 months, 1 year, 3 years, 5 years, and 8 years). Deep learning models were trained and evaluated using a training/validation/testing framework.
The models achieved approximately 80% accuracy (80.47, 95% CI: 80.23–80.72) in identifying dementia when using data immediately preceding diagnosis. Model performance declined as the time window extended further from the diagnosis date. One model achieved an area under the ROC curve (AUC) ranging from 0.71 (95% CI: 0.68–0.74) to 0.85 (95% CI: 0.83–0.86) across time windows. The best performance was obtained when all available features were included.
The most influential features in dementia identification included Aged Residential Care (ARC)-related factors, comorbidities (ICD-10 codes), and ethnicity. Among pharmacy-related variables, analgesics, diuretics, antithrombotics, anti-epileptic drugs, beta-adrenoceptor blockers, long-acting beta-adrenoceptor agonists, and diabetes treatments were identified as important. Additionally, the number of ophthalmology and cardiology specialist appointments, as well as delirium-related features, contributed significantly to model performance.
These findings highlight the potential of machine learning models as case-finding tools for dementia in New Zealand, providing opportunities for earlier identification and intervention. The application of such models could improve diagnostic accuracy and enhance the quality of life for individuals currently living with undiagnosed dementia
Protein structure characters in the light of phylogenetic systematics
No full textProtein structure characters have great potential for improving phylogenetic inference, especially for deep nodes where amino acid sequences are highly diverged. The combination of AlphaFold structure predictions and Foldseek’s “3Di” structural alphabet makes it relatively easy to conduct model-based phylogenetic inference that includes a partition of slow-evolving 3Di characters. However, we show that even identical amino acid sequences can produce substantially different 3Di codes, depending on the source of structural model and whether inter-chain interactions are considered. We argue that such variability can be addressed with key concepts from traditional organism-based phylogenetic systematics: semaphoront, hypodigm, and character ascertainment method. To illustrate this, we develop an analogy between organismal development, taphonomy, and subsequent description and character coding by a systematist, and the process of protein synthesis, folding, and interaction and subsequent extraction, experimentation, and structural modeling by a biochemist. We conclude that differences in 3Di codes between semaphoronts and are not intrinsically a problem, but they do require that the researcher uses the same replicable method on all proteins in the phylogenetic analysis. The guiding principle should be to maximize the chance that character differences in the data matrix are the results of underlying evolutionary changes, rather than artefactual differences between proteins due to differences in the methods used for obtaining semaphoronts and coding characters
Peripheral Vision: Disentangling the Pineal Gland and Peripheral Circadian Influence of Neutrophil Immunity in Zebrafish
No full textEntrained by the light-dark cycle, circadian rhythms regulate the immune response to infection. In larval zebrafish, light exposure enhances bacterial clearance and survival by increasing bactericidal activity within neutrophils, the most abundant innate immune cell. However, whether neutrophils directly sense light, or light is detected indirectly through the light-sensing pineal gland (PG) is unknown. This thesis investigates whether zebrafish neutrophils are entrained indirectly via the PG or possess intrinsic light sensitivity consistent with the decentralised circadian model in zebrafish. To evaluate pineal involvement, two independent ablation strategies were employed: a transgenic metronidazole ablation approach and CRISPR/Cas9 knockout of the transcription factor bsx that is essential for PG development. PG loss was confirmed by absence of restricted aanat2 expression, and dampened behavioural rhythmicity. Despite PG ablation, larvae exhibited normal survival following infection with GFP-tagged Salmonella enterica serovar Typhimurium (Sal-GFP), indicating that the PG is dispensable for light-enhanced immune protection. To confirm that a cell-intrinsic circadian clock operates in neutrophils to kill bacteria, a dominant-negative version of a core clock gene called bmal1a was overexpressed specifically within neutrophils. Larvae with neutrophil-specific expression of the dominant-negative bmal1a exhibited significantly reduced survival, elevated bacterial burden, and their neutrophils couldn’t eliminate intracellular bacteria efficiently. These findings support a decentralised circadian model in which peripheral clocks within innate immune cells are directly entrained by environmental light. Together these results provide functional and molecular evidence that zebrafish neutrophils possess cell-intrinsic circadian clocks and may be directly photosensitive, enabling them to act as a light-gated effectors of innate immunity
Neural Architecture of Social Punishment: Insights from a Queue-jumping Scenario
No full textPunishment in social settings is crucial for maintaining collective interests, yet the underlying mechanisms remain unclear. To address this, we developed a paradigm, the queue-jumping task, where participants imagine experiencing a queue-jumping event through vivid pictorial scenarios. Behavioral findings revealed that individuals prioritized collective interests over personal ones when punishing, highlighting the altruistic nature of social punishment. Neuroimaging results demonstrated that social punishment activated multiple neural circuits associated with social norms (e.g., fusiform gyrus and posterior cingulate cortex), self-related processing (e.g., ventromedial prefrontal cortex and middle cingulate cortex), and punishment implementation (e.g., anterior dorsolateral prefrontal cortex and middle temporal gyrus). Brain network analyses uncovered a social punishment network whose efficacy in information transmission forecasts individuals’ tendency to punish. This study provides valuable insights into the cognitive and neural mechanisms involved in social punishment. The current paradigm closely reflects real-life queue-jumping situations and daily punitive behaviors, demonstrating its generalizability and validity
Synthesis and Derivatisation of Cyclic Lipodepsipeptides for the Treatment of Infectious Diseases
No full textThe rapid increase in antibiotic consumption, coupled with misuse and over-reliance, has fuelled the rise of antimicrobial resistance (AMR), diminishing the effectiveness of existing antibiotics in an already declining antibiotic arsenal. To combat the spread of AMR, naturally occurring antimicrobial peptides (AMPs) have emerged as promising agents with novel mechanisms of action. Among these, cyclic lipodepsipeptides—a subclass of AMPs—exhibit broad-spectrum activity, offering diverse structural frameworks and mechanisms that show particular efficacy against resistant pathogens. This Thesis reports the chemical synthesis and derivatization of three cyclic lipodepsipeptide families.
The synthesis of cyclic lipodepsipeptides often presents significant challenges, especially in the incorporation of the ester (depsi) bond. The first chapter explores key factors influencing on-resin depsipeptide bond formation to establish a systematic framework for troubleshooting esterification in total synthesis of depsipeptides. Solvent choice emerged as the most critical factor, followed by the positioning of the reactive alcohol within the peptide sequence, and the presence of protecting groups.
The brevicidine family of cyclic lipodepsipeptides displays potent activity against clinically relevant Gram-negative bacteria. In the second chapter, using the Cysteine Lipidation on a Peptide or Amino acid (CLipPA) methodology, developed by Brimble et al., a small library of S-lipidated brevicidine analogues was synthesised. Additionally, the synthesis, structural analysis, bioactivity and nephrotoxicity evaluation, and structure activity relationship (SAR) study of the recently discovered brevicidine B was reported.
In the third chapter, a chemoenzymatic approach to daptomycin derivatization, which possesses a calcium-dependent antimicrobial activity against Gram-positive bacteria, is explored. Enzymatic deacylation of the native lipid tail by Actinoplanes utahensis, followed by S-lipidation using CLipPA yielded a small library of daptomycin analogues.
Finally, the total synthesis of verlamelins A and B—cyclic lipodepsipeptides with antifungal and antiviral properties—was successfully completed using a combination of solid-phase peptide synthesis (SPPS) and solution-phase macrolactamisation
School Experiences: Overcoming Challenges
No full textThere is no universal experience of school, and no one is likely to have an easy time at school all of the time. Young people face many challenges in and out of school, and there is significant variation in what those challenges are and who experiences them.
One thing many young people agree on however is that having friends and ‘fitting in’ are key factors in the experience of school – anyone who is seen as not fitting in or is in any way ‘different’ becomes a target for bullying. Friendships are protective against bullying – having friends means that you are less likely to stand out from the crowd, and friends can stand up to bullies.
Our young people have good ideas about how to make school a more inclusive place to support those students who are having a tough time. It is important that initiatives to increase school engagement focus not only on learning and achievement but also students’ experiences in order to make school a positive place to be
A systematic review and future agenda on continuance intentions in mobile apps
No full textTechnology changes at ever increasing speeds. Therefore, it is crucial for practitioners and academics to understand why users’ intend to continue or discontinue their usage. This paper presents a current and comprehensive systematic literature review on continuance intentions for mobile applications. The review analyzes 119 studies from the Scopus database (January 2019–December 2023) using the PRISMA, SPAR, and TCCM frameworks. It identifies key theoretical models, determinants of mobile app continuance intention, research methods, existing gaps, and future research directions. Findings reveal that several well-recognised theoretical models are frequently applied in the literature on continuance intention. Consequently, the variables derived from these models are among the most commonly measured by researchers. Additionally, the majority of studies in this area employ quantitative methods, with structural equation modelling being most widely used. This review categorises the literature based on mobile application classifications and six distinct sets of factors influencing continuance intention: psychological, technical, social, behavioural, contextual, and barriers. Furthermore, it explores the outcomes associated with continuance intention. The paper identifies two primary areas for future research: the development of a conceptual framework and research design. It also highlights research opportunities related to emerging technologies and the gap between intentions and actual behaviours
Firewalls or frontlines: Geopolitical tensions and multinationals' digital technology upgrading
No full textHow do geopolitical tensions affect digital technology upgrading? Digital technology upgrading is widely recognized as a crucial driver of the Industry 4.0 era. As multinational enterprises (MNEs) increasingly seek to absorb or transfer cutting-edge digital technologies from overseas, the recent escalation of geopolitical tensions has
introduced substantial uncertainty into this process. Drawing from the techno-geopolitical uncertainty literature in general and the geographic relational view in particular, this study explores how Chinese MNEs are impacted
by the rise of geopolitical tensions in their efforts to transform digitally. Our findings reveal that geopolitical
tensions between home and host countries adversely impact the digital technology depth and breadth of the focal
firm. Also, these relationships are reinforced by the firm’s contextual feature – state ownership. Yet the foreign
presence of these MNEs positively moderates the relationship between geopolitical tensions and digital technology depth, but the effect is not observed for digital technology breadth. Our exploration of the interrelationships between geopolitical tensions, digital technologies, and firm governance advances the literature
and provides practical implications for policymakers and managers in the new era of geopolitics