Dataset Information

Exploring Feature Dimensions to Learn a New Policy in an Uninformed Reinforcement Learning Task.

ABSTRACT: When making a choice with limited information, we explore new features through trial-and-error to learn how they are related. However, few studies have investigated exploratory behaviour when information is limited. In this study, we address, at both the behavioural and neural level, how, when, and why humans explore new feature dimensions to learn a new policy for choosing a state-space. We designed a novel multi-dimensional reinforcement learning task to encourage participants to explore and learn new features, then used a reinforcement learning algorithm to model policy exploration and learning behaviour. Our results provide the first evidence that, when humans explore new feature dimensions, their values are transferred from the previous policy to the new online (active) policy, as opposed to being learned from scratch. We further demonstrated that exploration may be regulated by the level of cognitive ambiguity, and that this process might be controlled by the frontopolar cortex. This opens up new possibilities of further understanding how humans explore new features in an open-space with limited information.

SUBMITTER: Choung OH

PROVIDER: S-EPMC5732284 | biostudies-literature | 2017 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Exploring Feature Dimensions to Learn a New Policy in an Uninformed Reinforcement Learning Task.

Choung Oh-Hyeon OH Lee Sang Wan SW Jeong Yong Y

Scientific reports 20171215 1

When making a choice with limited information, we explore new features through trial-and-error to learn how they are related. However, few studies have investigated exploratory behaviour when information is limited. In this study, we address, at both the behavioural and neural level, how, when, and why humans explore new feature dimensions to learn a new policy for choosing a state-space. We designed a novel multi-dimensional reinforcement learning task to encourage participants to explore and l ...[more]

PMID: 29247192

Similar Datasets

Project description:BackgroundIn recent years, point-of-care ultrasound (POCUS) has become an essential field of medical education. Bedside ultrasound has become a necessary skill for clinical physicians. Previous studies have already discussed the importance of advancements in ultrasound education. However, learning motivations for ultrasound education have seldom been analyzed in the literature. For medical students, learning ultrasound could have a relevance for their future career. The Existence, Relatedness and Growth (ERG) theory extended Maslow's hierarchy of needs through these three concepts. This theory has been widely used in the workplace to analyze employee job performance but has not yet been applied in medical education. In this study ERG theory was applied to analyze pre-clinical medical students' learning motivation toward ultrasound education.MethodThis mixed method study used online questionnaires consisting of open-ended questions as a data collection tool, and based on these results, both qualitative and quantitative analysis were conducted. Participants answered a series of neutral and open-ended questions regarding their motivations to learn ultrasonography. After data collection, a three-step analysis was conducted based on the grounded theory approach. Finally, the results of the thematic coding were used to complete additional quantitative analysis.ResultsThe study involved 140 pre-clinical medical students, and their responses fell into 13 specific categories. The analysis demonstrated that students' motivations toward ultrasound education were unbalanced across the three ERG domains (F = 41.257, p < .001). Pairwise comparisons showed that students mentioned existence motivation (MD = 39.3%; p < .001) and growth motivation (MD = 40.7%; p < .001) more frequently than relatedness motivation. However, there was no significant difference between existence motivation and growth motivation (MD = - 1.4%; p = .830).ConclusionThe results revealed that students placed a high value on existence and growth needs rather than relatedness based on the survey. In addition, the findings suggest that ERG theory can be a useful tool to conduct medical education motivation analysis.

Dataset Information

Exploring Feature Dimensions to Learn a New Policy in an Uninformed Reinforcement Learning Task.

Publications

Exploring Feature Dimensions to Learn a New Policy in an Uninformed Reinforcement Learning Task.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets