Can “less” create “more” in analogical reasoning?
Successful analogical reasoning requires an analogue in a source domain to have high degrees of structural and surface similarity with a learning task in a target domain. It also requires learners to have sufficient source- and target-domain knowledge. We review the literature and speculate that “less” might create “more”; in some situations, analogies that have fewer degrees of similarity may be more effective for learning. In this exploratory study, we engaged eight school leaders in dyads to develop a bottom-up perspective on innovation diffusion through analogical reasoning. The qualitative data in the study appears to echo our speculation. The dyads that have less prior target-domain knowledge face challenges with regard to innovation diffusion when they learn with analogues that have more degrees of similarity – both structural and surface. They, however, are able to learn with analogues that have fewer degrees of similarity. Learning was shown to take place when the dyads reflected on an analogue first, before they compared the analogue and innovation diffusion to make any analogical inferences. Although constrained by the exploratory nature of the study, the findings provide preliminary evidence that “less” is possible to create “more” in analogical reasoning under certain conditions, implying an interesting direction for experimental examination in future.
Integrating artificial intelligence into science lessons: Teachers’ experiences and views
Background
In the midst of digital transformation, schools are transforming their classrooms as they prepare students for a world increasingly automated by new technologies, including artificial intelligence (AI). During curricular implementation, it has not made sense to teachers to teach AI as a stand-alone subject as it is not a traditional discipline in schools. As such, subject matter teachers may need to take on the responsibility of integrating AI content into discipline-based lessons to help students make connections and see its relevance rather than present AI as separate content. This paper reports on a study that piloted a new lesson package in science classrooms to introduce students to the idea of AI. Specifically, the AI-integrated science lesson package, designed by the research team, provided an extended activity that used the same context as an existing lesson activity. Three science teachers from different schools piloted the lesson package with small groups of students and provided feedback on the materials and implementation.
Findings
The findings revealed the teachers’ perceptions of integrating AI into science lessons in terms of the connection between AI and science, challenges when implementing the AI lesson package and recommendations on improvements. First, the teachers perceived that AI and science have similarities in developing accurate models with quality data and using simplified reasoning, while they thought that AI and science play complementary roles when solving scientific problems. Second, the teachers thought that the biggest challenge in implementing the lesson package was a lack of confidence in content mastery, while the package would be challenging to get buy-in from teachers regarding curriculum adaptation and targeting the appropriate audience. Considering these challenges, they recommended that comprehensive AI resources be provided to teachers, while this package can be employed for science enrichment programs after-school.
Conclusions
The study has implications for curriculum writers who design lesson packages that introduce AI in science classrooms and for science teachers who wish to contribute to the development of AI literacy for teachers and the extension of the range of school science and STEM to students.
Building the science of research management: What can research management learn from education research?
Research management is an emerging field of study and its development is significant to the advancement of research enterprise. Developing the science of research management requires investigating social mechanisms involved in research management. Yet, studies on social mechanisms of research management is lacking in the literature. To address this gap, this paper proposes importing methodologies and theories from other social science disciplines to study the social mechanisms of research management and to build the science of research management. The paper first articulates what constitutes the science of research management, then proposes to appropriate Design-Based Research (DBR), a methodology in education research, for building the science of research management while at the same time strengthening the theory-practice nexus. A study of education research is then presented to illustrate how DBR is used to enact the theory of homophily which is imported from sociology. It reveals an opportunity to use social designs to develop social relationships among teachers from different schools for networked learning. Such a research endeavour also has potential to advance theories of relationship-building in sociology. Inferring from the example as an analogue to what is suggested for research management, the paper advocates a way to reciprocally connect research management as an emerging research field with more established social science disciplines at large and to advance both the theory and practice of research management.
Investigating analogical problem posing as the generative task in the productive failure design
Research on Productive Failure and preparatory mechanisms has consistently demonstrated a positive learning effect when students generate problem solutions before receiving formal instruction. However, it has been less examined whether the effect still holds when the generative task does not involve problem solving. Using a 2x2 experimental design, this study investigated the effects of generative tasks that involve analogical problem posing (without solving) on learning and transfer. Pedagogical sequence (i.e., generation-first or instruction-first) and type of analogical reasoning task (i.e., generating one’s own analogical problems or generating analogical mappings between given analogical problems) were the two factors manipulated. Preliminary analysis revealed no multivariate effects of the factors. Thus, we discuss the learning mechanisms enacted by analogical reasoning, reliability of the instruments, and the participants’ prior condition as possible reasons and to inform future studies.
Leadership in times of pandemics: Reflections from Singapore
The COVID-19 pandemic is compressing the timeline for Singapore’s digital transformation in education. Reflecting on the implementation of Home-Based Learning (HBL) during the pandemic, we examine three barriers that inhibit digital transformation and technological implementation in education with leadership considerations: the first order barrier is infrastructural and can be mitigated by leadership foresight; the second order barrier concerns design capabilities of teachers which can be mitigated by tight-but-loose calibration; and the third order barrier deals with sustainability which can be mitigated by ecological leadership. The tight-but-loose calibration optimises the ‘tight’ system-led innovations such as Student Learning Space (SLS) for efficient deployment and for equitable access of high quality online resources for students; and ‘loose’ opportunities for teacher-led innovations on learning designs within and beyond system-led innovations to nurture teacher agency and professionalism. We posit that ecological leadership is key to sustaining deep change together with the ‘tight-but-loose’ system calibration.
Learning innovation diffusion as a complex adaptive system : case studies on developing knowledge about and knowledge in doing for education leaders through cognitive conflicts
Conceiving innovation diffusion as a Complex Adaptive System (CAS) broadens the approaches to foster scalable and sustainable diffusion of innovation in schools. This study adopted an Interpretivist Paradigm through multiple case studies to understand education leaders’ learning with regard to innovation diffusion as a CAS. There were four cases involving eight education leaders learning in four dyads. The purpose was to seek a descriptive and interpretative understanding of their learning processes, which included the knowledge development trajectories and the learning patterns through which the learning activities induced the knowledge development trajectories. Cognitive conflict was engaged as the learning strategy and learning activities included building an agent-based model, simulating the researcher’s model and playing a first-person role-play game. Analogical reasoning was engaged in reflections.
The grounded theory coding and analysis and episodic uptake analysis revealed nonlinearity of the knowledge development trajectories. The findings suggest that, prior to the learning, the dyads maintained ten simple Newtonian knowledge elements. After learning, the dyads made progression in developing ten comprehensive CAS knowledge elements. Analyzing the knowledge development trajectories indicated that the dyads experienced difficulties in constructing the CAS knowledge elements when they maintained prior Newtonian knowledge elements. The development trajectory of each knowledge element was gradual, progressing from conceptualization to comprehension and towards knowledge application in the real world. The development trajectories of the different knowledge elements were intertwined.
Further data analysis on how cognitive conflict enabled the learning processes implied nonlinear learning patterns. Cognitive conflict induced by analogical reasoning helped the dyads conceptualized the first CAS knowledge element, whereas cognitive conflict induced by model building and simulation did not. For the dyads who had conceptualized the first CAS knowledge element through analogical reasoning, the discrepancies generated from model building and simulation, from knowledge incompatibility and from game play induced cognitive conflicts and sustained the intertwined knowledge development trajectories.
There are sources of evidence in this study that could potentially extend the conventional conditions for analogical reasoning and cognitive conflict. The literature of analogical reasoning suggests that effective analogical reasoning requires learners’ adequate knowledge on the analogy and the target domain, and analogy’s sufficient degrees of similarities with the target. The learning patterns in this study seemed to suggest that, when the dyads, who lacked CAS knowledge, were conceptualizing the first CAS knowledge element, they did not learn through analogies that had more degrees of similarities; however, they learnt through analogies that had fewer degrees of similarities. Similarly, the findings may also suggest alternative views about the premises for cognitive conflict. The literature suggests that meaningful cognitive conflict requires discrepancies to be credible and relevant, and learners to have more prior knowledge. In this study, the agent-based models provided by the researcher were direct modeling of innovation diffusion, whereas the analogies used in reflections were from domains different from innovation diffusion and were only plausible to be mapped for the dyads to learn innovation diffusion. While the analogies induced valid discrepancy and meaningful cognitive conflict in this study, the model building and simulation did not. The findings in this study also alluded that salient knowledge that mediates learning Complex Systems is perhaps learner dependent and domain specific. The study thus advocates that research on analogical reasoning and cognitive conflict should focus on understanding learners’ meaning-making processes, rather than conditions for analogical reasoning and cognitive conflict.
Based on the findings, this study also espouses that learning Complex Systems perhaps involves meta-complexity: the learning of individual knowledge elements is complex, and the intertwined knowledge development trajectories also exhibit characteristics similar to scale-free network topologies. The implications of meta-complexity for learning design are also discussed to complement the design principles for learning Complex Systems.
Exploring interactions between learners and ChatGPT from a learner agency perspective: A multiple case study on historical inquiry
A noticeable surge in students’ widespread adoption of ChatGPT in the past year brought attention to the need for a deeper understanding of their interactions with this new technology. While attempts at theorising learner-ChatGPT interactions have been made, few studies offer empirical accounts of the interactions between learners and ChatGPT. This study aims to address this gap by utilising Emirbayer and Mische’s Choral Triad of Agency as an analytical framework to investigate secondary school students’ self-initiated interactions with ChatGPT in the context of historical inquiry. Through an in-depth examination of three cases, we unpacked three distinct types of learner-ChatGPT interactions—ChatGPT-as-historical source, ChatGPT-as-feedback, and principled non-use. Although students presented unique interaction patterns with ChatGPT, each case was found to have limited routined interactions with ChatGPT. Our analysis revealed that the students held static agentic orientations in their use of ChatGPT due to their limited experiences with ChatGPT and inadequate ideation for alternative ways of utilising it. Implications of this study propose the need for deliberate interventions to encourage students to have more diverse and meaningful interactions with ChatGPT.
Development of a tool for decision making on subject placement in secondary schools
Cultivating laterality in learning communities in Singapore education system: Scaling of innovation through networked learning community
Cultivating teachers to be active and agentic learners is crucial for contemporary teacher education (Lipponen & Kumpulainen, 2011). Those teachers’ qualities are essential in preparing students’ future readiness in an increasingly complex world (P21 Framework Definitions, 2015). In fact, both learning principles and evidence from practice inform us that purposeful collaboration in networked learning communities (NLCs) encourage teacher agency to learn (Lieberman & Wood, 2003; Muijs, West & Ainscow, 2010). As a complement to the literature, we are interested in the development of social relationships among teachers, which enables and facilitates their learning. We propose “laterality” – the relations and networks among peers (e.g., teachers) as an important concept to characterize NLCs.
Studies on laterality, which have shown to support teacher learning, are usually found in the decentralized systems where individuals are the best entities to form these networks to support each other’s growth (Hargreaves & Goodman, 2006; Muijs et al., 2010). Thus, developing laterality from the bottom-up becomes natural in the decentralized contexts (Granovetter, 1973). Despite considerable theoretical promise of laterality and its increasing prevalence in practice, we wonder whether teacher laterality matters in the centralized education systems, and if it does, how it grows.