An official website of the United States government
Sensemaking is conceptualized as a trajectory to develop better understanding and is advocated as one of the fundamental practices in science education. However, the field is lacking of a framework to view the prolonged process of sensemaking that starts from a raise of uncertainty of a target phenomenon to a grasping of a better understanding of a target phenomenon. The process requires teachers to recognize the role of scientific uncertainty in different phases of sensemaking and develop responsive instructional supports to help students navigate the uncertainties. With an attention on student scientific uncertainty as a potential driver of the trajectory of sensemaking, this study aims to identify different phases of sensemaking that can be developed with students’ scientific uncertainty. This study especially attends to two types of scientific uncertainty—conceptual and epistemic uncertainties. Conceptual uncertainty refers to student struggle of using conceptual understanding (e.g., mastery of content and everyday knowledge) to respond to an encountered phenomenon. Epistemic uncertainty emerges from struggles in using epistemic understanding to generate new ideas. Based on the multiple case study method, we examined sensemaking activities in two Korean science classrooms and one American science classroom and identified three phases of sensemaking: (a) focusing on a driving question related to a target phenomenon, (b) delving into multiple resources to develop plausible explanation(s), and (c) examining the successfulness of the new understanding and concretizing it. Based on the findings, we discuss two emerging themes. First, sensemaking progresses through three distinctive phases driven by students’ dynamically evolving scientific uncertainty. Second, attending to both epistemic and conceptual uncertainties can support developing sensemaking coherent with students’ view.
more »
« less
Student Epistemic Agency and Coherence-seeking through Laboratory Experiments
Overly simplistic school science laboratories constrain student agency. We share and discuss a case from 9th grade science classroom in which students all conducted highly varied independent investigations that were each highly coherent and scientifically well-motivated. We discuss the conditions that led to their experiments in terms of instability and uncertainty. Our findings suggest that it may be beneficial to support and recognize multiple forms of uncertainty simultaneously to encourage multiple forms of investigation to respond to those uncertainties. Finally, an “instability” caused by having multiple candidate models or explanations in play may be more generative than uncertainties based on gaps in knowledge.
more »
« less
- Award ID(s):
- 1640054
- PAR ID:
- 10309670
- Editor(s):
- de Vries, E.; Hod, Y.; Ahn, J.
- Date Published:
- Journal Name:
- Proceedings of the 15th International Conference of the Learning Sciences - ICLS 2021
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Science practice introduces inevitable uncertainties that are desirable for learning. Yet, navigating student scientific uncertainties can be a challenge for teachers. This qualitative study explored how teachers perceive and utilize uncertainty during science instruction. Analysis of interviews and classroom observations collected from 14 middle school teachers in the United States indicated limited awareness of uncertainty’s use as a resource in science. Teachers perceived uncertainty as a way to induce curiosity and persist through struggle; however, they were quick to reduce students’ scientific uncertainty throughout lessons. Findings suggest that teachers need support to understand how uncertainty navigation can benefit student learning.more » « less
-
null (Ed.)Uncertainty is a common feature in first-principles models that are widely used in various engineering problems. Uncertainty quantification (UQ) has become an essential procedure to improve the accuracy and reliability of model predictions. Polynomial chaos expansion (PCE) has been used as an efficient approach for UQ by approximating uncertainty with orthogonal polynomial basis functions of standard distributions (e.g., normal) chosen from the Askey scheme. However, uncertainty in practice may not be represented well by standard distributions. In this case, the convergence rate and accuracy of the PCE-based UQ cannot be guaranteed. Further, when models involve non-polynomial forms, the PCE-based UQ can be computationally impractical in the presence of many parametric uncertainties. To address these issues, the Gram–Schmidt (GS) orthogonalization and generalized dimension reduction method (gDRM) are integrated with the PCE in this work to deal with many parametric uncertainties that follow arbitrary distributions. The performance of the proposed method is demonstrated with three benchmark cases including two chemical engineering problems in terms of UQ accuracy and computational efficiency by comparison with available algorithms (e.g., non-intrusive PCE).more » « less
-
Abstract Conceptualizing, assessing, and managing disaster risks involve collecting and synthesizing pluralistic information—from natural, built, and human systems—to characterize disaster impacts and guide policy on effective resilience investments. Disaster research and practice, therefore, are highly complex and inherently interdisciplinary endeavors. Characterizing the uncertainties involved in interdisciplinary disaster research is imperative, since misrepresenting uncertainty can lead to myopic decisions and suboptimal societal outcomes. Efficacious disaster mitigation should, therefore, explicitly address the uncertainties associated with all stages of hazard modeling, preparation, and response. However, uncertainty assessment and communication in the context of interdisciplinary disaster research remain understudied. In this “Perspective” article, we argue that in harnessing interdisciplinary methods and diverse data types in disaster research, careful deliberations on assessingType IIIandType IVerrors are imperative. Additionally, we discuss the pathologies in frequentist approaches, calling for an increasing role for Bayesian methods in uncertainty estimations. Moreover, we discuss the potential tradeoffs associated with information and uncertainty, calling for deliberate consideration of the role of diversity of information prior to setting the scope in interdisciplinary modeling. Future research guided by further reflections on the ideas raised in this article could help push the frontiers of uncertainty estimation in interdisciplinary hazard research and practice.more » « less
-
Abstract Robust quantification of predictive uncertainty is a critical addition needed for machine learning applied to weather and climate problems to improve the understanding of what is driving prediction sensitivity. Ensembles of machine learning models provide predictive uncertainty estimates in a conceptually simple way but require multiple models for training and prediction, increasing computational cost and latency. Parametric deep learning can estimate uncertainty with one model by predicting the parameters of a probability distribution but does not account for epistemic uncertainty. Evidential deep learning, a technique that extends parametric deep learning to higher-order distributions, can account for both aleatoric and epistemic uncertainties with one model. This study compares the uncertainty derived from evidential neural networks to that obtained from ensembles. Through applications of the classification of winter precipitation type and regression of surface-layer fluxes, we show evidential deep learning models attaining predictive accuracy rivaling standard methods while robustly quantifying both sources of uncertainty. We evaluate the uncertainty in terms of how well the predictions are calibrated and how well the uncertainty correlates with prediction error. Analyses of uncertainty in the context of the inputs reveal sensitivities to underlying meteorological processes, facilitating interpretation of the models. The conceptual simplicity, interpretability, and computational efficiency of evidential neural networks make them highly extensible, offering a promising approach for reliable and practical uncertainty quantification in Earth system science modeling. To encourage broader adoption of evidential deep learning, we have developed a new Python package, Machine Integration and Learning for Earth Systems (MILES) group Generalized Uncertainty for Earth System Science (GUESS) (MILES-GUESS) (https://github.com/ai2es/miles-guess), that enables users to train and evaluate both evidential and ensemble deep learning. Significance StatementThis study demonstrates a new technique, evidential deep learning, for robust and computationally efficient uncertainty quantification in modeling the Earth system. The method integrates probabilistic principles into deep neural networks, enabling the estimation of both aleatoric uncertainty from noisy data and epistemic uncertainty from model limitations using a single model. Our analyses reveal how decomposing these uncertainties provides valuable insights into reliability, accuracy, and model shortcomings. We show that the approach can rival standard methods in classification and regression tasks within atmospheric science while offering practical advantages such as computational efficiency. With further advances, evidential networks have the potential to enhance risk assessment and decision-making across meteorology by improving uncertainty quantification, a longstanding challenge. This work establishes a strong foundation and motivation for the broader adoption of evidential learning, where properly quantifying uncertainties is critical yet lacking.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.22318/icls2021.689
