Can artificial intelligence separate the wheat from the chaff in systematic reviews of health economic articles?

Wolcherink, Martijn J. Oude; Pouwels, Xavier; Dijk, S. H. B. van; Doggen, Carine J.M.; Koffijberg, Hendrik

doi:10.1080/14737167.2023.2234639

Cited by 11 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With ASReview the references of all 9.975 papers were ranked and screened. ASReview is relatively new but has been used in several systematic reviews [ 122 , 123 ] and was considered accurate and efficient [ 124 , 125 ].…”

Section: Methodsmentioning

confidence: 99%

Tunneling, cognitive load and time orientation and their relations with dietary behavior of people experiencing financial scarcity – an AI-assisted scoping review elaborating on scarcity theory

van der Veer,

Madern,

van Lenthe

2024

Int J Behav Nutr Phys Act

View full text Add to dashboard Cite

Background The concept of a financial scarcity mindset has raised much attention as an explanation for poor decision-making and dysfunctional behavior. It has been suggested that financial scarcity could also impair dietary behavior, through a decline in self-control. Underlying cognitive mechanisms of tunneling (directing attention to financial issues and neglecting other demands), cognitive load (a tax on mental bandwidth interfering with executive functioning) and time orientation (a shift towards a present time horizon, versus a future time horizon) may explain the association between financial scarcity and self-control related dietary behavior. The current scoping review gathers recent evidence on how these mechanisms affect dietary behavior of people experiencing financial scarcity. It builds on a theoretical framework based on insights from behavioral economics and health psychology. Methods A literature search was executed in six online databases, which resulted in 9.975 papers. Search terms were tunneling, cognitive load and time orientation, financial scarcity, and dietary behavior. Screening was performed with ASReview, an AI-ranking tool. In total, 14 papers were included in the scoping review. We used PRISMA-ScR guidelines for reporting. Results Limited evidence indicates that a scarcity mindset could increase tunneling, through attentional narrowing on costs of food, which then directly impacts dietary behavior. A scarcity mindset involves experiencing financial stress, which can be understood as cognitive load. Cognitive load decreases attentional capacity, which could impair self-control in dietary choices. Financial scarcity is related to a present time orientation, which affects dietary choices by shifting priorities and decreasing motivation for healthy dietary behavior. Conclusions A scarcity mindset affects dietary behavior in different ways. Tunneling and a shift in time orientation are indicative of an attentional redirection, which can be seen as more adaptive to the situation. These may be processes indirectly affecting self-control capacity. Cognitive load could decrease self-control capacity needed for healthy dietary behavior because it consumes mental bandwidth. How a changing time orientation when experiencing financial scarcity relates to motivation for self-control in dietary behavior is a promising theme for further inquiry.

show abstract

Section: Methodsmentioning

confidence: 99%

Tunneling, cognitive load and time orientation and their relations with dietary behavior of people experiencing financial scarcity – an AI-assisted scoping review elaborating on scarcity theory

van der Veer,

Madern,

van Lenthe

2024

Int J Behav Nutr Phys Act

View full text Add to dashboard Cite

show abstract

“…This level of customization allows researchers to tailor the model to their specific research needs and objectives. Indeed, the simulation mode has been used to compare the workload savings achieved by ASReview with manual screening (Ferdinands et al, 2020;Oude Wolcherink, 2023). To work in the simulation mode, researchers need to first upload a fully labeled dataset, say 50 relevant records out of a dataset of 500 publications, and select a specific model configuration.…”

Section: A Quick Start Guidementioning

confidence: 99%

Study screening for meta-research: An overview of AI tools

Quan,

Hui

2023

Preprint

View full text Add to dashboard Cite

Second language acquisition (SLA) has seen a rapid increase in meta-research, such as systematic reviews and meta-analyses. This line of work has important implications, as it provides comprehensive overviews, establishes consensus, and identifies gaps in the literature on a given topic, both methodologically and substantively. However, the manual processes involved in meta-research can be resource intensive and, perhaps more importantly, prone to human errors. In this chapter, we provide an overview of artificial intelligence (AI) tools that have the potential to streamline meta-research processes, specifically in the context of literature screening. We discuss the strengths and weaknesses of these tools and share our experience using them. We conclude by recognizing that these AI tools are not a substitute for human expertise and judgements but offer exciting opportunities for researchers engaging in meta-research. Our overall goal is to help improve the efficiency, reproducibility, and transparency of meta-research in SLA.

show abstract

An evaluation of the performance of stopping rules in AI‐aided screening for psychological meta‐analytical research

König,

Zitzmann,

Fütterer

et al. 2024

Research Synthesis Methods

View full text Add to dashboard Cite

Several AI‐aided screening tools have emerged to tackle the ever‐expanding body of literature. These tools employ active learning, where algorithms sort abstracts based on human feedback. However, researchers using these tools face a crucial dilemma: When should they stop screening without knowing the proportion of relevant studies? Although numerous stopping rules have been proposed to guide users in this decision, they have yet to undergo comprehensive evaluation. In this study, we evaluated the performance of three stopping rules: the knee method, a data‐driven heuristic, and a prevalence estimation technique. We measured performance via sensitivity, specificity, and screening cost and explored the influence of the prevalence of relevant studies and the choice of the learning algorithm. We curated a dataset of abstract collections from meta‐analyses across five psychological research domains. Our findings revealed performance differences between stopping rules regarding all performance measures and variations in the performance of stopping rules across different prevalence ratios. Moreover, despite the relatively minor impact of the learning algorithm, we found that specific combinations of stopping rules and learning algorithms were most effective for certain prevalence ratios of relevant abstracts. Based on these results, we derived practical recommendations for users of AI‐aided screening tools. Furthermore, we discuss possible implications and offer suggestions for future research.

show abstract

Can artificial intelligence separate the wheat from the chaff in systematic reviews of health economic articles?

Cited by 11 publications

References 17 publications

Tunneling, cognitive load and time orientation and their relations with dietary behavior of people experiencing financial scarcity – an AI-assisted scoping review elaborating on scarcity theory

Tunneling, cognitive load and time orientation and their relations with dietary behavior of people experiencing financial scarcity – an AI-assisted scoping review elaborating on scarcity theory

Study screening for meta-research: An overview of AI tools

An evaluation of the performance of stopping rules in AI‐aided screening for psychological meta‐analytical research

Contact Info

Product

Resources

About