The relationship between learners' cognitive and affective states has become a topic of increased interest, especially because it is an important component of self-regulated learning (SRL) processes. This paper studies sixth grade students' SRL processes as they work in Betty's Brain, an agent-based open-ended learning environment (OELE). In this environment, students learn science topics by building causal models. Our analyses combine observational data on student affect with log files of students' interactions within the OELE. Preliminary analyses show that two relatively infrequent affective states, boredom and delight, show especially marked differences among high and low performing students. Further analysis shows that many of these differences occur after receiving feedback from the virtual agents in the Betty's Brain environment. We discuss the implications of these differences and how they can be used to construct adaptive personalized scaffolds.

Representation and prediction of student navigational pathways, typically based on neural network (NN) methods, have seen their potential of improving instruction and learning under insufficient human knowledge about learner behavior. However, they are prominently studied in MOOCs and less probed within more institutionalized higher education scenarios. This work extends such research to the context of online college courses. Comparing student navigational sequences through course pages to documents in natural language processing, we apply a skip-gram model to learn vector embedding of course pages, and visualize the learnt vectors to understand the extent to which students' learning pathways align with pre-designed course structure. We find that students who get different letter grades in the end exhibit different levels of adherence to designed sequence. Next, we fit the embedded sequences into a long short-term memory architecture and test its ability to predict next page that a student visits given her prior sequence. The highest accuracy reaches 50.8% and largely outperforms the frequency-based baseline of 41.3%. These results show that neural network methods have the potential to help instructors understand students' learning behaviors and facilitate automated instructional support.

We model coordination and coregulation patterns in 33 triads engaged in collaboratively solving a challenging computer programming task for approximately 20 minutes. Our goal is to prospectively model speech rate (words/sec) - an important signal of turn taking and active participation - of one teammate (A or B or C) from time lagged nonverbal signals (speech rate and acoustic-prosodic features) of the other two (i.e., A + B -> C; A + C -> B; B + C -> A) and task-related context features. We trained feed -forward neural networks (FFNNs) and long sshort-term memory recurrent neural networks (LSTMs) using group level nested cross-validation. LSTMs outperformed FFNNs and a chance baseline and could predict speech rate up to 6s into the future. A multimodal combination of speech rate, acoustic prosodic, and task context features outperformed unimodal and bimodal signals. The extent to which the models could predict an individual's speech rate was positively related to that individual's scores on a subsequent posttest, suggesting a link between coordination/coregulation and collaborative learning outcomes. We discuss applications of the models for real-time systems that monitor the collaborative process and intervene to promote positive collaborative outcomes.

The field of psychological science has seen major advances in the development of formal models of perceptual classification learning; however, little work has tested such models in real-world natural-category domains. In our current project, we aim to fill that gap by testing the ability of a formal exemplar model of classification to predict learning of rock categories in the geologic sciences. As a prerequisite for testing the model in this domain, we have conducted extensive work to derive a high-dimensional feature-space representation for the rock stimuli. An eight-dimensional representation yields good accounts of naive participants' judgments of similarity among a large battery of rock-picture samples; furthermore, the eight dimensions have natural psychological interpretations. We then use the exemplar model in combination with the derived feature-space representation to successfully predict participants' learning and generalization of a variety of scientifically defined rock categories. We discuss further steps for making use of the model and its associated feature-space representation to search for effective techniques of teaching categories in the science classroom.

Background: Interactive learning environments often provide help strategies to facilitate learning. Hints, for example, help students recall relevant concepts, identify mistakes, and make inferences. However, several studies have shown cases of ineffective help use. Findings from an initial study on the availability of hints in a mathematics problem-solving activity showed that early access to on-demand hints were linked to lack of performance improvements and longer completion times in students answering problems for summer work. The same experimental methodology was used in the present work with a different student sample population collected during the academic year to check for generalizability. Results: Results from the academic year study showed that early access to on-demand-hints in an online mathematics assignment significantly improved student performance compared to students with later access to hints, which was not observed in the summer study. There were no differences in assignment completion time between conditions, which had been observed in the summer study and has been attributed to engagement in off-task activities. Although the summer and academic year studies were internally valid, there were significantly more students in the academic year study who did not complete their assignment. The sample populations differed significantly by student characteristics and external factors, possibly contributing to differences in the findings. Notable contextual factors that differed included prior knowledge, grade level, and assignment deadlines. Conclusions: Contextual differences influence hint effectiveness. This work found varying results when the same experimental methodologywas conducted on two separate sample populations engaged in different learning settings. Further work is needed, however, to better understand how on-demand hints generalize to other learning contexts. Despite its limitations, the study shows how randomized controlled trials can be used to better understand the effectiveness of instructional designs applied in online learning systems that cater to thousands of learners across diverse student populations. We hope to encourage additional research that will validate the effectiveness of instructional designs in different learning contexts, paving the way for the development of robust and generalizable designs.

BackgroundA major concern in science education involves the under-representation of many groups in science and technology fields, especially by gender (Brotman and Moore, J Res Sci Teach 45:971-1002, 2008; Clark Blickenstaff, Gend Educ 17:369-386, 2006), stemming from an intersection of systemic obstacles (Cantu, Equity Excell Educ 45:472-487, 2012; Rosa and Mensah, Phys Rev Phys Educ Res 12:020113, 2016). Research on persistence of minoritized populations within science trajectories has often highlighted identity as particularly important (Archer et al., Sci Educ 94:617-639, 2010; Barton and Calabrese, Am Educ Res J 50:37-75, 2007; Barton et al., Am Educ Res J 50:37-75, 2013; Merolla and Serpe, Soc Psychol Educ 16:575-597, 2013).ResultsThis study quantitatively investigated the nature of science identity in over 1300 seventh and ninth grade students from a range of urban US public schools using survey data on science identity, choice preferences, and optional science experiences. Factor analyses validated this conceptualization of science identity as integrating perceived internal and external identity components. Regression analyses revealed the importance of this conceptualization of science identity for driving students' choices at this crucial developmental period. Furthermore, science identity had a complex differential function in supporting students' optional science choices by gender.ConclusionsThe novel contribution to the science identity field highlights the specific multi-component ways in which students endorse science identity in middle school and early high school. There was an important finding that science identity has a complex differential function in supporting student's optional science choices by gender. Thus, at this age, developing a strong science identity is especially critical for girls.

Countries, research institutions, and scholars are interested in identifying and promoting high-impact and transformative scientific research. This paper presents a novel set of text-and citation-based metrics that can be used to identify high-impact and transformative works. The 11 metrics can be grouped into seven types: Radical-Generative, Radical-Destructive, Risky, Multidisciplinary, Wide Impact, Growing Impact, and Impact (overall). The metrics are exemplified, validated, and compared using a set of 10,778,696 MEDLINE articles matched to the Science Citation Index Expanded (TM). Articles are grouped into six 5-year periods (spanning 1983-2012) using publication year and into 6,159 fields constructed using comparable MeSH terms, with which each article is tagged. The analysis is conducted at the level of a field-period pair, of which 15,051 have articles and are used in this study. A factor analysis shows that transformativeness and impact are positively related (rho = .402), but represent distinct phenomena. Looking at the subcomponents of transformativeness, there is no evidence that transformative work is adopted slowly or that the generation of important new concepts coincides with the obsolescence of existing concepts. We also find that the generation of important new concepts and highly cited work is more risky. Finally, supporting the validity of our metrics, we show that work that draws on a wider range of research fields is used more widely.

With the nationwide emphasis on improving outcomes for STEM undergraduates, it is important that we not only focus on modifying classroom instruction, but also provide students with the tools to maximize their independent learning time. There has been considerable work in laboratory settings examining two beneficial practices for enhancing learning: spacing and self-testing. In the current study, we examine biology students' study practices, particularly in the context of these two behaviors. We specifically investigate whether a light-touch study skills intervention focused on encouraging spacing and self-testing practices impacted their utilization. Based on pre- and post-course surveys, we found that students report utilizing both beneficial and ineffective study practices and confirm that usage of spacing and self-testing correlates with a higher course grade. We also found that students in the section of the course which received the study skills intervention were more likely to report continued use or adoption of spacing and self-testing compared to students in control sections without the intervention. Surprisingly, we found that underrepresented minorities (URMs) under-utilize self-testing, and that our intervention helped to partially ameliorate this gap. Additionally, we found that URMs who reported self-testing earned similar course grades compared to non-URMs who also self-tested, but that there was a much larger drop in performance for URMs who did not self-test relative to non-URMs who also did not selftest. Overall, we would encourage instructors to dedicate class time towards discussing the merits of beneficial study practices, especially for students that have historically underperformed in STEM disciplines.

Massive Open Online Courses (MOOCs) have the potential to democratize education by providing learners with access to high-quality free online courses. However, evidence supporting this democratization across countries is limited. We explored the question of MOOC democratization by conducting cross-national comparisons of gender differences in the enrollment in and completion of science, technology, engineering, and mathematics (STEM) MOOCs. We found that while females were less likely than males to enroll in STEM MOOCs, they were equally likely to complete them. Further, a higher probability to enroll in STEM MOOCs and smaller gender gaps in STEM MOOC enrollment and completion were found in less gender-equal and less economically developed countries.