Search for: All records

The use of writing-based exercises in a circuit analysis course has shown promise in aiding students likely to struggle in the course by enhancing their conceptual understanding of topics related to DC circuit analysis [1]. As grading of writing samples and providing personalized feedback can be time-intensive, automating the evaluation and feedback processes through use of emerging techniques in natural language processing (NLP) could open the door for more widespread use of such writing exercises across STEM courses, thus benefiting students in most need of assistance. In this paper, the development and initial testing of two web-based writing activities that leverage a basic NLP technique to probe student writing related to DC circuits are described. The first writing exercise has students describe what happens to the power of various elements in a resistive circuit as the value of one of the resistors decreases. The second exercise has students consider situations in which the ideal independent voltage and current source models might fail. Both writing exercises are built from a template that includes several metacognitive prompts to spur self-reflection on the part of the user. A rule-based approach was taken to detect evidence of common misconceptions [2] and errors in student responses, as well as to identify sentences that revealed the student was correctly addressing the problems. Based on identified misconceptions or correct concepts in a student’s writing, the web-based application selects appropriate directed line of reasoning (DLR) feedback paths to attempt to lead the writer to an accurate understanding of the behavior of the circuits in question. Key features of the web-based application template as well as details regarding misconception detection and personalized feedback are described. Student impressions of the value of the DLR feedback is assessed using comments provided by the student within the applications. Planned modifications of the web-based writing exercise template based on this formative assessment will be given and address a broader goal of this work – to develop a web-based template that instructors across STEM disciplines, even those without a background in coding, could use to implement their own conceptual writing exercises. 

Contribution: This article describes the implementation, assessment, and evaluation of conceptual-based writing exercises in an introductory course on electric circuit analysis. Background: Students' struggles in gateway courses such as circuit analysis are often traced to inadequate metacognitive skills on the part of the student as well their misconceptions regarding fundamental phenomena related to the course. Writing is known to be a powerful tool for insight into a student's thought process and to foster metacognitive activity. Research Questions: What effect does the use of short writing exercises have on students' understanding of fundamental concepts related to the behavior of electric circuits operating at dc? What effect does the use of the conceptually based writing exercises have on students' ability to justify their responses when answering conceptual questions related to basic electric circuit concepts? Methodology: In the first semester of the study, a single writing exercise was given and in the second semester, a total of five such exercises were administered. In each semester, students were separated into ``at-risk'' and ``not at-risk'' groups based on their responses to the first writing exercise. A 2 x 2 x (2) mixed analysis of variance (ANOVA) was conducted, with at-risk/not at-risk and semester/semester between-subjects factors and pre-test/post-test on a multiple-choice conceptual-based exam a within-subjects factor. Findings: Results suggest that only the at-risk group may have benefited in terms of deepened conceptual understanding and the ability to justify their responses from the use of multiple conceptual-based writing exercises. 

Background: Biomarkers for Alzheimer’s disease (AD) are crucial for early diagnosis and treatment monitoring once disease modifying therapies become available. Objective: This study aims to quantify the forward magnetization transfer rate (kfor) map from brain tissue water to macromolecular protons and use it to identify the brain regions with abnormal kfor in AD and AD progression. Methods: From the Cardiovascular Health Study (CHS) cognition study, magnetization transfer imaging (MTI) was acquired at baseline from 63 participants, including 20 normal controls (NC), 18 with mild cognitive impairment (MCI), and 25 AD subjects. Of those, 53 participants completed a follow-up MRI scan and were divided into four groups: 15 stable NC, 12 NC-to-MCI, 12 stable MCI, and 14 MCI/AD-to-AD subjects. kfor maps were compared across NC, MCI, and AD groups at baseline for the cross-sectional study and across four longitudinal groups for the longitudinal study. Results: We found a lower kfor in the frontal gray matter (GM), parietal GM, frontal corona radiata (CR) white matter (WM) tracts, frontal and parietal superior longitudinal fasciculus (SLF) WM tracts in AD relative to both NC and MCI. Further, we observed progressive decreases of kfor in the frontal GM, parietal GM, frontal and parietal CR WM tracts, and parietal SLF WM tracts in stable MCI. In the parietal GM, parietal CR WM tracts, and parietal SLF WM tracts, we found trend differences between MCI/AD-to-AD and stable NC. Conclusion: Forward magnetization transfer rate is a promising biomarker for AD diagnosis and progression. 

Summary The net reclassification improvement (NRI) and the integrated discrimination improvement (IDI) were originally proposed to characterize accuracy improvement in predicting a binary outcome, when new biomarkers are added to regression models. These two indices have been extended from binary outcomes to multi-categorical and survival outcomes. Working on an AIDS study where the onset of cognitive impairment is competing risk censored by death, we extend the NRI and the IDI to competing risk outcomes, by using cumulative incidence functions to quantify cumulative risks of competing events, and adopting the definitions of the two indices for multi-category outcomes. The “missing” category due to independent censoring is handled through inverse probability weighting. Various competing risk models are considered, such as the Fine and Gray, multistate, and multinomial logistic models. Estimation methods for the NRI and the IDI from competing risk data are presented. The inference for the NRI is constructed based on asymptotic normality of its estimator, and the bias-corrected and accelerated bootstrap procedure is used for the IDI. Simulations demonstrate that the proposed inferential procedures perform very well. The Multicenter AIDS Cohort Study is used to illustrate the practical utility of the extended NRI and IDI for competing risk outcomes. 

The multivariate normative comparison (MNC) method has been used for identifying cognitive impairment. When participants' cognitive brain domains are evaluated regularly, the longitudinal MNC (LMNC) has been introduced to correct for the intercorrelation among repeated assessments of multiple cognitive domains in the same participant. However, it may not be practical to wait until the end of study for diagnosis. For example, in participants of the Multicenter AIDS Cohort Study (MACS), cognitive functioning has been evaluated repeatedly for more than 35 years. Therefore, it is optimal to identify cognitive impairment at each assessment, while the family‐wise error rate (FWER) is controlled with unknown number of assessments in future. In this work, we propose to use the difference of consecutive LMNC test statistics to construct independent tests. Frequency modeling can help predict how many assessments each participant will have, so Bonferroni‐type correction can be easily adapted. A chi‐squared test is used under the assumption of multivariate normality, and permutation test is proposed where this assumption is violated. We showed through simulation and the MACS data that our method controlled FWER below a predetermined level.