This study addresses the measurement of high school students' epistemic knowledge associated with scientific experimentation (EKSE) which concerns how scientific experimentation generates knowledge and why that knowledge is justified. Based on philosophical, educational standards, and literature analyses, an EKSE construct is characterized as (1) underlying students' decisions and reasoning elicited during experiment design, data collection/measurement, and data analysis/interpretation and (2) progressing in five levels of sophistication: no information, nascent, dogmatic, contextualized, and reflective. This study psychometrically examined the structure of the EKSE construct using the construct modeling approach. An instrument was designed to ask students decisions and reasoning encountered while carrying out the activities embedded in experimental inquiry. High school students took the instrument as a pretest and a posttest to a physics curriculum where they learned to carry out independent scientific experiments. Results indicate that (1) all decisions and reasoning taking place during experiment design, data collection/measurement, and data analysis/interpretation contribute to the EKSE construct, (2) epistemic knowledge develops in the hypothesized order, (3) scientifically aligned decisions are likely to occur when students' epistemic knowledge is at or higher than the dogmatic level, and (4) students' EKSE can improve after engaging in a curriculum that encourages independent scientific experimentation with materials. Implications of these results and further research are discussed.

Abstract Designing an effective feedback system that functions in real-world classrooms is both complex and painstaking. Using a conjecture map, we delineate the design research process related to an artificial intelligence (AI)-enabled automated feedback system called HASbot. In particular, we focus on whether and how automated feedback on scientific argument artifacts can improve students’ simulation-mediated scientific inquiry. We describe the design of HASbot in terms of domain model, student model, progress support model, and user interface. We investigated mediating processes using students’ simulation interactions from log data, changes in scientific arguments when students used HASbot, and students’ epistemic discourse. We discuss challenges in developing AI-afforded automated feedback systems to support science practices: (1) improving AI-based diagnostic models, (2) improving automated feedback design and strategies for science practices, (3) soliciting larger impacts through teacher buy-ins, and (4) addressing equity concerns in AI.

A design study was conducted to test a machine learning (ML)-enabled automated feedback system developed to support students’ revision of scientific arguments using data from published sources and simulations. This paper focuses on three simulation-based scientific argumentation tasks called Trap, Aquifer, and Supply. These tasks were part of an online science curriculum module addressing groundwater systems for secondary school students. ML was used to develop automated scoring models for students’ argumentation texts as well as to explore emerging patterns between students’ simulation interactions and argumentation scores. The study occurred as we were developing the first version of simulation feedback to augment the existing argument feedback. We studied two cohorts of students who used argument only (AO) feedback (n = 164) versus argument and simulation (AS) feedback (n = 179). We investigated how AO and AS students interacted with simulations and wrote and revised their scientific arguments before and after receiving their respective feedback. Overall, the same percentages of students (49% each) revised their arguments after feedback, and their revised arguments received significantly higher scores for both feedback conditions, p < 0.001. Significantly greater numbers of AS students (36% across three tasks) reran the simulations after feedback as compared with the AO students (5%), p < 0.001. For AS students who reran the simulation, their simulation scores increased for the Trap task, p < .001, and for the Aquifer task, p < 0.01. AO students who did not receive simulation feedback but reran the simulations increased simulation scores only for the Trap task, p < .05. For the Trap and Aquifer tasks, students who increased simulation scores were more likely to increase argument scores in their revisions than those who did not increase simulation scores or did not revisit simulations at all after simulation feedback was provided. This pattern was not found for the Supply task. Based on these findings, we discuss strengths and weaknesses of the current automated feedback design, in particular the use of ML.

In this paper, we propose a metric based on the sample entropy concept for measuring the systematicity of students’ experimentation patterns in an open-ended simulation environment where a number of parameters are at students’ disposal to explore. Unlike other indicators of systematicity proposed in the literature, the sample entropy metric provides a continuous scale and draws upon the up-to-date computational algorithm applied to dynamic processes involved in physical and biological systems. This sample entropy-based metric correlates significantly with student learning outcomes related to (1) how well students described the nature of relationship explored during their experimentation and (2) whether students coordinated between claim and data collected from their experimentation. Our analysis indicates that (1) the sample entropy metric captures the aspect of students’ experimentations that is not captured by several conventional measured and (2) it has potential for general application to a variety of simulation-based activities when assessing students.

Describing the normal state single particle spectral function line shapes of high temperature superconductors remains an important goal in condensed matter physics. Recently, we have proposed a phenomenological extremely correlated Fermi liquid (pECFL) model that promises to accomplish this goal and that is uniquely distinguished from other models. Here, we present an even more simplified phenomenological model, which we refer to as the aECFL model, that performs practically at the same level as the pECFL model. Noting the similarities of the aECFL model and the pECFL model, as well as the differences between the two models, we emphasize the universal significance of the $ω$-dependence of the so-called caparison factor in the ECFL model.

The Bayesian Knowledge Tracing (BKT) model is a popular model used for tracking student progress in learning systems such as an intelligent tutoring system. However, the model is not free of problems. Well-recognized problems include the identifiability problem and the empirical degeneracy problem. Unfortunately, these problems are still poorly understood and how they should be dealt with in practice is unclear. Here, we analyze the mathematical structure of the BKT model, identify a source of the difficulty, and construct a simple Monte Carlo BKT model to analyze the problem in real data. Using the student activity data obtained from the ramp task module at the Concord Consortium, we find that the Monte Carlo BKT analysis is capable of detecting the identifiability problem and the empirical degeneracy problem, and, more generally, gives an excellent summary of the student learning data. In particular, the student activity monitoring parameter M emerges as the central parameter.

An interactive learning task was designed in a game format to help high school students acquire knowledge about a simple mechanical system involving a car moving on a ramp. This ramp game consisted of five challenges that addressed individual knowledge components with increasing difficulty. In order to investigate patterns of knowledge emergence during the ramp game, we applied the Monte Carlo Bayesian Knowledge Tracing (BKT) algorithm to 447 game segments produced by 64 student groups in two physics teachers' classrooms. Results indicate that, in the ramp game context, (1) the initial knowledge and guessing parameters were significantly highly correlated, (2) the slip parameter was interpretable monotonically, (3) low guessing parameter values were associated with knowledge emergence while high guessing parameter values were associated with knowledge maintenance, and (4) the transition parameter showed the speed of knowledge emergence. By applying the k-means clustering to ramp game segments represented in the three dimensional space defined by guessing, slip, and transition parameters, we identified seven clusters of knowledge emergence. We characterize these clusters and discuss implications for future research as well as for instructional game design.

The so-called "strange metal phase" [1] of high temperature (high Tc) superconductors remains at the heart of the high Tc mystery. Better experimental data and insightful theoretical work would improve our understanding of this enigmatic phase. In particular, the recent advance in angle resolved photoelectron spectroscopy (ARPES) [2, 3], incorporating low photon energies (about 7 eV), has given a much more refined view of the many body interaction in these materials. Here, we report a new ARPES feature of Bi2Sr2CaCu2O8+d that we demonstrate to have the key ability to distinguish between different classes of theories of the normal state. This feature—the anomaly in the nodal many body density of states (nMBDOS)—is clearly observed in the low energy ARPES data, but also observed in more conventional high energy ARPES data, when a sufficient temperature range is covered. We show that key characteristics of this anomaly are explained by a strong electron correlation model; the electron-hole asymmetry and the momentum dependent self energy emerge as key required ingredients.

Providing a full theoretical description of the single-particle spectral function observed for high-temperature superconductors in the normal state is an important goal, yet unrealized. Here, we present a phenomenological model approaching towards this goal. The model results from implementing key phenomenological improvement in the so-called extremely correlated Fermi-liquid model. The model successfully describes the dichotomy of the spectral function as functions of momentum and energy and fits data for different materials (Bi2Sr2CaCu2O8+δ and La2-xSrxCuO4), with an identical set of intrinsic parameters. The current analysis goes well beyond the prevalent analysis of the spectral function as a function of momentum alone.

Fully describing the single particle spectral function observed for high temperature superconduc- tors in the normal state is an important goal, yet unachieved. Here, we present a phenomenological model that demonstrates the capability to meet such a goal. The model results from employing key phenomenological improvement of the so-called extremely correlated Fermi liquid (ECFL) model, and is shown to successfully describe the data as a function of momentum as well as energy, for different materials (Bi2212 and LSCO), with an identical set of intrinsic parameters. This work goes well beyond the prevalent analysis of momentum dependent curves.

The normal-state single particle spectral function of the high temperature superconducting cuprates, measured by the angle-resolved photoelectron spectroscopy (ARPES), has been considered both anomalous and crucial to understand. Here, we report an unprecedented success of the new extremely correlated Fermi liquid theory by one of us [B. S. Shastry, Phys. Rev. Lett. 107, 056403 (2011)] to describe both laser and conventional synchrotron ARPES data (nodal cut at optimal doping) on Bi(2)Sr(2)CaCu(2)O(8+δ) and synchrotron data on La(1.85)Sr(0.15)CuO(4). It fits all data sets with the same physical parameter values, satisfies the particle sum rule and successfully addresses two widely discussed kink anomalies in the dispersion.

We study the electronic structure of Ca2-xNaxCuO2Cl2 and Bi2Sr2CaCu2O8+d samples in a wide range of doping, using angle-resolved photoemission spectroscopy, with emphasis on on the Fermi surface (FS) in the near anti-nodal region. The "nesting wave vector", i.e., the wave vector that connects two nearly flat pieces of the Fermi surface in the anti-nodal region, reveals a universal monotonic decrease in magnitude as a function of doping. Comparing our results to the charge order recently observed by scanning tunneling spectroscopy (STS), we conclude that the FS nesting and the charge order pattern seen in STS do not have a direct relationship. Therefore,the charge order likely arises due to strong correlation physics rather than FS nesting physics.