Complex Systems and Education: From Theory to Research to Practice

Satellite of the Conference on Complex Systems 2021 (Lyon, France & Virtually)

Purpose of the Satellite


Over the past two decades, there has been a gradual but significant increase in the research and theoretical development concerning education as a complex process. This complexity entails the hierarchical relationship between systems and their constituent components, such as for example students within classrooms within school buildings; complex transformations such as hysteresis and second order change in the qualitative transitions in the child’s intellectual development, and emergent learning processes, fractal patterns, self-transcending constructions and uncomputability. The inherent complexity of the educational process has been long recognized, but the development of a research tradition that specifically utilizes methodologies that are grounded in complexity theory has been slow to materialize. Currently, however, interesting, and methodologically sophisticated research is being conducted, that not only enriches the field of education, but also enhances the complexity field with new applications and examples.


This satellite seeks to capitalize on these developments with a diverse set of studies that share a complexity lens in their attempts to understand the educational process but are distinct in the methodologies that are used to capture the complex aspects of the process, and include social network analysis, catastrophe theory, time series analysis, and orbital decomposition. The presentations included in this satellite will showcase some of these approaches, as well as summarize how our understanding of the educational process is enriched using them. Practitioners tend to take a holistic view of the field in which they work, with processes and variables being interconnected and outcomes not necessarily being predictable. There is an affinity between that perspective and complexity theory, which focuses as well on the adaptive and systemic features of the educational process in a dynamical holistic manner as well. An open discussion toward the end of the session will seek to bring out these similarities between theory and practice in education, based on the work presented during the day. This satellite follows the success of its past editions and seeks to reinforce these developments through the presentation of new research and new practical applications of complexity theory in education. We seek to enrich participants’ conference experience by engaging them in discussions of the application of complex dynamical systems in educational research and the improvement of educational practice. The participants will discuss how future research should embrace and operationalize complexity in education and further develop the conceptual and empirical frameworks that could guide productive research in this area in the future, thus adding to the interdisciplinary character of the CCS conference. The satellite will be a day-long session with a varied program and great network opportunities. 

 

The proceedings of this satellite session will be published in the International Journal of Complexity in Education

 

Facilitators

Matthijs Koopmans (Mercy College, USA; mkoopmans@mercy.edu)

Hiroki Sayama (Binghamton University, USA; sayama@binghamton.edu)

Dimitrios Stamovlasis (Aristotle University of Thessaloniki, Greece; stadi@edlit.auth.gr)

Schedule of Presentations

 

Thursday October 28, 2021

 

Note: All session times are Central European Time (CET). See table below for information on time differences.

14:30 – 16:15: Session I

14:30 - 14:45 Welcome and Opening Remarks

14:45 – 15:00: Using Quantum Semantic Networks in Constructing University Students’ Lexicons of Abstract Concepts from Written Reports: A Case of Photon Concept in Context of Double-Slit Experiment

Ismo T. Koponen, Karoliina Vuola, & Maija Nousiainen – University of Helsinki, Finland

15:00 – 15:15 Can Cognitive Data Science Highlight Mindset Change in Student Populations?

Massimo Stella – University of Exeter, UK

15:15 – 15:30: Mixed-Methods Social Network Analysis Sheds New Light on Second Language Acquisition

Michał B. Paradowski – University of Warsaw, Poland

15:30 – 15:45 On the Use of Agent-Based Models as Teaching Tools and Their Impact on the Characterization of Student-System Interaction

Luis E. Cortés-Berrueco – Instituto Politécnico Nacional, México

Gustavo Carreón-Vázquez – Universidad Nacional Autónoma de México, México

15:45 – 16:00: Complexity and Education: A Network Approach to Curriculum Design

Ashuwin Vaidya – Montclair State University, USA

16:00 – 16:15: Practice Guidance for Mixed Methods Complexity-Informed Research

 

Emma P. Bullock – Sam Houston State University, USA

Cheryl Poth – University of Alberta, Canada

 

16:15 – 16:45: Break

 

 

16:45 – 18:00: Session II

16:45 – 17:00: The Salience of Emotion in Role-based Reflections by STEM Camp Instructors

Joanna K. Garner – Old Dominion University, USA

Melissa Kuhn – Old Dominion University, USA

Erica Matheny – National Inventors’ Hall of Fame; USA

Alaina Rutledge – National Inventors’ Hall of Fame; USA

17:00 – 17:15: Dynamical Systems Measures of Group Functioning

Bernard P. Ricca – University of Colorado, USA

17:15 – 17:30: Challenges of the Leadership, Management, and Government in High School in Emergency (Covid19): Case Study

Lic Carlos Cosentino – University of San Andrés, Argentina

17:30 – 17:45: Online Graduate Program in Systems Science at Binghamton University, State University of New York

Hiroki Sayama – Binghamton University, USA

17:45 -- 18:00: Concluding Remarks and Next Steps

Information About Time Differences by Presenters' Location

Session times indicated above are in Central European Time. The conversion table below is based on the location of the session presenters:

  • 1 Hour Later: Helsinki, Finland; Thessaloniki, Greece

  • No Time Difference: Lyon, France; Warshaw, Poland

  • 1 Hour Earlier: Exeter, UK

  • 5 Hours Earlier: Buenos Aires, Argentina

  • 6 Hours Earlier: Eastern United States

  • 7 Hours Earlier: Texas and Mexico City

  • 8 Hours Earlier: Edmonton, AB

Session Abstracts (Listed Alphabetically by First Authors’ Last Name)

 

Practice Guidance for Mixed Methods Complexity-Informed Research

Emma P. Bullock – Sam Houston State University, USA

Cheryl Poth – University of Alberta, Canada

 

Mixed methods research (MMR) requires the integration of both quantitative and qualitative data and assumes that their collective contribution mitigates inherent weaknesses in either type of data. The usefulness of MMR for solving complex societal issues is well established (e.g., Mertens, 2015; Poth, 2018) yet researchers continue to lack practical design guidance. In this presentation, we leverage our collective MMR experiences and complexity theory expertise for managing dynamic research conditions and the resulting ‘messiness’ in educational settings. As such, we advance four complexity-informed design practices with examples from two mixed methods studies. The first investigated the role of school leaders in students’ mathematics achievement (Bullock, 2017). The second examined the experiences learners attributed to their learning outcomes in a doctoral MMR class (Poth, 2021). The comparison of these studies describes and illustrates how: (a) the need for integration informs unit of analysis and initial design decisions, (b) attention to interactions informs data source and sampling decisions, (c) adaptations to emergent conditions inform integration and procedural decisions, (d) transparency in methodological integrity informs the findings interpretations and results decisions. This points to important implications for guiding researchers studying complex systems using MMR to: a) identify among the possible units of analysis to focus attention, based on the need for integration, b) select data sources and sampling strategies, capturing the outcomes of interactions which may lead to an understanding of emergent properties not otherwise understood, c) adapt initial design decisions to capture revelations of the nonlinear nature of the analysis and integration decisions, and d) document nonlinear design procedures to transparently reveal how the interpretation of results were informed by evidence of methodological integrity. We posit that all researchers can be better equipped for managing the messiness inherent in our work through the integration of complexity theory with MMR design practices.

References

Bullock, E. P. (2017). An explanatory sequential mixed methods study of the school leaders' role in students' mathematics achievement through the lens of complexity theory (Order No. 10288685). Doctoral thesis. Utah State University. Available at: https://digitalcommons.usu.edu/etd/6096/ (Accessed: 14 December 2020).

Mertens, D. M. (2015). Mixed methods and wicked problems. Journal of Mixed Methods, Research, 9, 1-6.

Poth, C. (2018). Innovation in mixed methods research: A practical guide to integrative thinking with complexity. Sage.

Poth, C. (2021). Adaptive practices for a complexity-sensitive approach to mixed integration. In J. Hitchcock, & A. Onwuegbuzie (Eds.) Handbook for

advancing integration in mixed methods research (pp. 305–329). Sage.

On the Use of Agent-Based Models as Teaching Tools and Their Impact on the Characterization of Student-System Interaction

 

Luis E. Cortés-Berrueco – Instituto Politécnico Nacional, México

Gustavo Carreón-Vázquez – Universidad Nacional Autónoma de México, México

Learning is a complex task [1], and to better understand this process, it is essential the collaboration between education and complexity researchers [2]. However, more robust bridges between these two sciences are needed, aiming to improve communication, the unification of concepts, and the development of methodologies [3]. In this work it is presented a case study of two different agent- based models used in high school education activities designed to teach fundamental concepts of the complexity science such as: system, system´s dynamics, parameters, emergence, interactions, and parameter space. The models were specially designed for: easily illustrate some of the fundamental dynamics of an epidemy and the segregation phenomena, let the students to experiment by changing parameter´s values and, to register interactions between student and graphic user interface elements. The student-GUI interactions information registered let us identify how the activities were fulfilled by the students and after processing the data it was also possible to identify emergent patterns from the student’s behavior. We will show the number of interactions for each of the GUI elements for one of the activities of the epidemic model, as well as mouse tracking for each of the experiments conducted by one of the students during one of the activities of the epidemic model.

 

These models were helpful communication tools between the education and complexity researchers involved. Prototype models were used by both teams of researchers to specify the goal concepts to be learned by the students. In the next phase, a feedback work frame was established between the education and the complexity teams, the education activities were designed according to the models’ capabilities, and in turn, the models’ capabilities were modified (when possible) according to the activities’ requirements. Finally, after the students fulfilled the activities, both teams were able to take measurements according to their corresponding methodologies, analyze their results and cross information with the other team.

References:

 

  1. Epstein, J.M. (2006). Generative Social Science. Studies in Agent-Based Computational Modeling. Princeton University Press.

  2. Dickes, A. C., & Sengupta, P. (2013). Learning natural selection in 4th grade with multi-agent based computational models. Research in Science Education, 43(3), 921 – 953.

  3. Resnick, M., &Wilensky, U. (1998). Diving into complexity: Developing probabilistic decentralized thinking through roleplaying activities. Journal of the Learning Sciences, 7, 153–172.

 

 

Challenges of the Leadership, Management, and Government in High School in Emergency (Covid19): Case Study

 

Lic Carlos Cosentino – University of San Andrés, Argentina

The epistemological, ontological, and methodological approach from the complexity theory allows to build a greater understanding of the school as a complex dynamic system (CDS) and the equilibrium dynamics in relation to the concepts of management, government and leadership embodied in the directors of three schools of the province of Buenos Aires during pandemic. This article presents the preliminary advances of an ongoing investigation in Argentina on high school management, governance and leadership as CDS in emergency. In this study we wonder if the disturbances that impact school and besides change in flows between the subsystems provoke such a destabilization that it generates transformation. Fundamentally, we evaluate about director´s interventions as catalysts for this CDS in emergency, being key to achieve greater inclusion and educational coverage navigates uncertainties and generates resilience in the face of shocks.

The main milestones that show the disturbances:

  • As a result of Covid19, the school discontinues class attendance in March 2020.

  • Directors face the problem of managing schools remotely, seeking to strengthen the pedagogical nucleus and redefining the basic conditions: time and space organization, grouping of students and conditions of attendance.

  • We detect a new way of inhabiting space-time by establishing new coordinates.

  • The interrelationships developed by its actors are disturbed, hindered, and recreated in other areas, adopting the capacity for self-organization.

  • The school adopts a remote way building ways of assembling communication and digital platforms.

  • Director, teachers and students develop new practices and patrons

  • Emerging and configurations mediated by techno-pedagogical assemblages that shape the school mode

 

The Salience of Emotion in Role-based Reflections by STEM Camp Instructors

Joanna K. Garner – Old Dominion University, USA

Melissa Kuhn – Old Dominion University, USA

Erica Matheny – National Inventors’ Hall of Fame; USA

Alaina Rutledge – National Inventors’ Hall of Fame; USA

 

Grounded in complex systems principles, the Dynamic Systems Model of Role Identity (DSMRI) posits that individuals draw on formal and informal social roles as both a frame for interpreting experiences and a lens through which future actions are considered. The model portrays action as an emergent phenomenon resulting from the interplay between four identity components including self-perceptions, purposes and goals, epistemological and ontological beliefs, and action possibilities. These components are interdependent and are situated within individual-dispositional and contextual-sociocultural factors that act as control parameters on the system. Whereas previous research has revealed the ways in which role identity components may change over time, and the degree to which contexts can raise and lower the salience of particular social roles, a fifth component of the DSMRI – emotion – has not been widely investigated. In this presentation, we highlight how emotions emerged as a strong theme in educators’ reflections on their activities during a STEM and invention camp for elementary and middle school students. We will consider how students’ emotions in response to the camp curriculum served as a salient source of information for educators, and how emotions were described in statements of purpose, beliefs, and self-perceptions. We will draw attention to the role of positive, learning-related emotions as an integral component of identity change through a model-informed analysis of the ways in which emotions appeared in educators’ responses to prompts about what they learned by facilitating the camp. Finally, we will consider the connections that were articulated by educators between students’ emotions in the camp setting and their perceived action possibilities for applying their experience to teaching STEM and invention in their typical, school based context.

 

Using Quantum Semantic Networks in Constructing University Students’ Lexicons of Abstract Concepts from Written Reports: A Case of Photon Concept in Context of Double-Slit Experiment

 

Ismo T. Koponen, Karoliina Vuola, & Maija Nousiainen – University of Helsinki, Finland

 

In many research settings within science education researchers analyze students’ written text, reports and protocols. In such a research, written texts are often rendered in a form of a semantic network, 1,2 where words and terms are connected and form a network like lexicon. Several semi-automated methods exist for constructing such lexicon 1. The existing methods, however, are based in a way or another on fixed counting of connections or frequencies of appearance of words (i.e., objectively countable probabilities) either in proximity in text strings 1 or in a more complicated way by taking into account text structure 2. Such methods miss the important complexity emerging from different subjective ways to weight the importance of words and terms in different contexts. Here, we use quantum semantics approach 3 to construct semantic networks from students’ written reports. The quantum semantics provide means to take into account subjective factors, through its capability to model entanglement of terms and words. Such a feature provides first steps to include variability of meaning. In what follows, we show how quantum semantics is applied in to analyze university students’ written reports related to concept of photon, in context of double-slit experiment. We explore the extent of subjectivity in 12 reports. Attention is paid how closely the semantic networks can be brought into agreement

(concurrence) by suitable tuning the subjective entanglement factors. The results suggest that quantum semantics may be a promising and highly viable way to model students’ lexicons in form of semantic networks, with inbuilt possibilities to take into account subjective factors in construction of meaning of abstract terms.

 

References

 

[1] Ifenthaler, D., Hanewald, R. (Eds). Digital Knowledge Maps in Education: Technology-Enhanced Support for Teachers and Learners. (Springer: New York, NY, USA, 2014).

[2] Koponen, I.T.; Nousiainen, M. Concept networks of students’ knowledge of relationships between physics concepts: Finding key concepts and their epistemic support, Appl. Netw. Sci. (2018) 3:14.

[3] Surov, I. A. et al. Quantum semantics of text perception, Sci. Rep. (2021) 11:4193.

Mixed-methods Social Network Analysis Sheds New Light on Second Language Acquisition

Michał B. Paradowski – University of Warsaw, Poland

 

Social networks play a vital role in SLA. Combining computational and anthropological Social Network Analysis (SNA), we investigate the influence of peer interaction dynamics and social graph topology on measurable outcomes in two intensive language courses: a 5-week course of German for Erasmus+ exchange students in Baden-Württemberg (n=40), and two editions of a 4-week summer course of the Polish language and culture in Warsaw (n=332). Unlike studies focusing on the micro-level of individual participants’ egocentric networks, presenting an emic view only, we demonstrate how and why peer learner networks can be examined in their entirety, complementing an etic perspective. In particular, we focus on the moderating role of the social network (mesoscopic explanatory variable)—in turn influenced by engagement with the TL culture (macroscopic explanatory variable)—on L2 progress (microscopic response variable).

 

The quantitative component of the project showed among others:

  1. that outgoing interactions in the TL are a stronger predictor of progress than incoming interactions,

  2. a clear detrimental effect of interactions with same-L1 speakers (routgoing=−.31[-0.63, 0.00], p=.048),

  3. the strongest influence of the network in the domains of pronunciation and lexis, where degree centrality in the TL positively correlates with progress (routdegree=.258, p=.001 for pronunciation; routdegree=.304, p=.0002 and rindegree=.263, p=.001 for vocabulary), while betweenness in total communication is significantly anticorrelated (r=−.242, p=.003 and r=−.204, p=.01, respectively).

  4. This mirrors the impact of closeness centrality (ease of access to other students).

  5. Combined with the deleterious influence on SLA of a high in-degree, this underscores the importance of the network’s structural properties.

 

In turn, structured interviews carried out with course participants and their instructors yielded valuable information on the formation and types of the networks the learners engaged in and the purposes these networks served. The presentation will thus illustrate the benefits of combining computational (quantitative) and anthropological (qualitative) social network analysis. Lastly, we shall also compare two face-to-face iterations of one of the courses with its online edition during the COVID-19 pandemic.

Dynamical Systems Measures of Group Functioning

Bernard P. Ricca – University of Colorado – USA

Small collaborative group work is part of many classrooms, but the objective evaluation of group functioning is not well established. Two dynamical systems measures, burstiness and power law fits of data distributions, have been used in other psychological contexts to explore system health and overall functioning and those measures are used here to explore small collaborative groups. Burstiness, short time intervals between events in a group of events followed by long periods without those events, has been shown to be a measure of the extent to which a system is self-organized. Burstiness is a particularly useful metric when studying complex systems because it is often challenging to discern self-organization patterns in such systems. Power law distributions have similarly been posited to be indicators of self-organization in systems; the extent to which a data distribution is well modeled by a power law has also been posited to be an indicator of the health of a system.

In this study, data were collected from three groups of fifth graders undertaking an open-ended engineering design challenge. The groups were recorded, the audio was transcribed, and then talk turns were metacognitively coded. The burstiness and distribution of metacognitive codes are compared to the phases of the engineering design cycle to determine how burstiness and distribution might be used as measures of group functioning. Because there is little prior theory regarding applying these measures to classroom situations, the findings indicate potential fruitful areas for theorizing.

 

Online Graduate Program in Systems Science at Binghamton University, State University of New York

 

Hiroki Sayama – Binghamton University, USA

Concepts, methods, and tools developed in Complexity Science are highly applicable to a wide variety of application domains, including education research and practice. However, there are still only a small number of educational opportunities available for learners to gain advanced knowledge and skills on complexity in a systematic manner, not to mention formal graduate-level degree programs. To meet this educational need, Binghamton University's Systems Science program has been offering both Master's and PhD level graduate degree programs through a continuing-education distance learning platform over the last several decades, historically for working professionals in Engineering domains. This program has recently received a formal approval from the New York State Department of Education to be an officially designated online-only degree program. In this presentation, we will present the curricular contents, instructional methodologies, graduation requirements and other pedagogical/logistic details of our online Systems Science program. The impact the program has been producing and its future directions will also be discussed, especially in the context of the global COVID-19 pandemic that has changed the landscape of people's interest, need and mode of learning significantly.

Can Cognitive Data Science Highlight Mindset Change in Student Populations?

Massimo Stella – University of Exeter, UK

 A key aspect of learning is changing how we connect ideas and concepts in our minds. Forma mentis, a mindset, can capture how individuals perceive topics, trends, and experiences over time. To detect changes in the mindsets of students we use cognitive data science in the form of forma mentis networks (FMNs), which enable a data-driven, direct access to individuals’ ways of thinking. FMNs build cognitive representations of stances through two psycholinguistic tools: (1) conceptual associations from semantic memory (free associations, i.e., one concept eliciting another) and (2) affective norms (valence, i.e., how attractive a concept is). We test FMNs by investigating how Norwegian nursing and engineering students perceived innovation and health before and after a 2-month research project in e-health. We built and analysed FMNs by six individuals, including 1,000 associations between 730 concepts in total. When investigating changes before and after the training, individual FMNs highlighted drastic improvements in all students’ stances towards “teamwork”, “collaboration”, “engineering” and “future”, indicating the acquisition and strengthening of a positive belief about innovation. Nursing students improved their perception of ‘robots” and “technology” and related them to the future of nursing. A group-level analysis related these changes to the emergence, during the project, of conceptual associations about openness towards multidisciplinary collaboration, and a positive, leadership-oriented group dynamics. The whole group identified “mathematics” and “coding” as highly relevant concepts after the project. We also investigated persistent associations, i.e. those conceptual links that did not change before and after the training. Using network distance entropy and closeness we identified as pivotal in the students’ mindsets those concepts related to “personal well-being”, “professional growth” and “teamwork”. This result aligns with and extends previous studies reporting the relevance of teamwork and personal well-being for Norwegian healthcare professionals, also within the novel e-health sector. Our analysis indicates that cognitive data and forma mentis networks are promising proxies for detecting individual- and group-level mindset changes due to learning. 

Reference 

Stella M, & Zaytseva A. 2020. Forma mentis networks map how nursing and engineering students enhance their mindsets about innovation and health during professional growth. PeerJ Computer Science 6:e255 https://doi.org/10.7717/peerj-cs.255

 

Complexity and Education: A Network Approach to Curriculum Design

 

Ashuwin Vaidya – Montclair State University, USA

 

In this presentation we hypothesize that education, especially at the scale of curriculum, should be treated as a complex system composed of different ideas and concepts which are inherently connected. The task of a good teacher lies in elucidating these connections and helping students make their own connections. Such a pedagogy allows students to personalize learning and strive to be ‘creative’ and make meaning out of old ideas. To investigate our hypothesis, we take the example of a precalculus course curriculum. Precalculus is an important ‘bridge course’ taken by students in high schools and college alike linking school mathematics curriculum to that of a college course. However, the skills acquired in this course are useful to students in all STEM disciplines. Therefore, we feel that an examination of the curricular complexity of such a course can be widely revealing. We treat precalculus textbooks as exemplars of a specific pedagogy and map several texts into networks of isolated (nodes) and interconnected concepts (edges) thereby permitting computations of metrics which have much relevance to the education theorists, teachers and all others involved in the field of education. We contend that these network metrics provide valuable insights to teachers and students about the kind of pedagogy which encourages good teaching and learning.