Current trends in LET reserach
Learning diary
by
diego Guzman
day 1
PAper 1
The first study presented at the LET 2018 conference held at the University of Oulu, is given by Jonna Malmberg (Post Doctoral researcher, LET University of Oulu). The paper presented was titled: Are we together or not? Sequential interplay of monitoring and physiological synchrony during a collaborative exam (Malmberg, Järvelä, Iman, Ilkka & Seppänen, 2017)
This study is grounded in the context of Collaborative Learning (CL) and Regulated Learning (RL), more specifically in the role of monitoring processes as a prerequisite for a successful group task. Previous studies like Dillebourg (1999, in Malmberg, 2018) and Barron (2001 in Malmberg, 2018) in CL, Volet, Vauras, Salo and Khosa (2017 in Malmberg, 2018) approaches to explain group´s outcome in CL, Ku, Tseng and Akarasriwon (2013 in Malmberg, 2018) attitudinal factors in CL and Järvelä, Malmberg and Kovuniemi (2016 in Malmberg, 2018) methods to recognize Socially Shared Regulation of Learning (SSRL) from chats in Computer supported Collaborative Learning (CSCL) establish the framework for this study. However, also physiological data is used for understanding a possible source of group’s mental attunement. In this line, Kreibig´s studies of the autonomous nervous system activity in emotion are taken into account. The physiological values are read from individuals´ electro-dermal activity (EDA) for a further investigation of the role of group´s synchrony in monitoring processes (Kreibig, 2010 in Malmberg, 2018).
In this line, Malmberg (2018) highlighted the relevance of monitoring in collaboration for establishing mental attunement, to foster joint attention and for promoting interaction among group members. However, despite Malmberg (2018) stated that monitoring is an individual activity, evidences of this process cannot be seen unless is put in any form of communication, thus, it remains invisible for research. Having mentioned this, the aim of the study is to determine whether and how individuals are in synchrony in a collaborative task if it is not observable through their verbal communication and if this is reflected as physiological synchrony.
Regarding the methodology for the study, Malmberg (2018) exposed that two methods were used for the data collection. Firstly, the student’s EDA was obtained from the wristbands equipped with sensors and secondly, video recording of the session is used for further triangulation of data. Subsequently, for the data analysis, the video observations are used to determine the utterances in monitoring and those are divided in three main phases: task interpretation, experiment and reporting. Lately, the peaks in the EDA of each student are identified and displayed accordingly to the phase where it happened, and finally the group’s physiological synchrony index of each session (SSI) is calculated.
The findings from the study suggest that monitoring is reflected in the form of peaks in the students’ EDA. However, it cannot be concluded that higher levels of physiological synchrony do constitute groups’ attunement on the same topic, but that the group members are actively thinking. For this reason, it is assumed that physiological synchrony is not sufficient for establishing shared regulation.
With no doubt this conclusion is a big determinant for future research in the field of understanding “sharedness” in social regulation of learning. This study can help future researchers to identify the angle from which to tackle EDA. In this case, this value was not sufficient for determining the students´ attunement but the members’ active cognition, thus future research should focus on the identification of the target of those moments of active cognition. The values of mental activity can be an interesting indicator for educators for identifying reasons for the pitfalls in students’ learning. Even if at this moment the current methodology has not shown the qualitative aspects of that mental activity, teachers can use it at this stage as a prompt or call of attention for the identification of low cognitive activity.
PAper 2
The second presentation is given by Muhterem Dindar (Post Doctoral researcher, LET University of Oulu) under the title: Interplay of temporal changes in self-regulation, academic success and physiological synchrony (Dindar,Malmberg, Jarvela & Kirschner, 2017).
The study is framed in the theory of Self-Regulation of Learning (SRL) and in Collaborative Learning (CL) but also is grounded in the field of physiology (Dindar, 2018). For instance, this paper refers to previous studies regarding the use of physiological data from Henriques, Paiva, and Antunes (2013 in Dindar, 2018): accessing emotion patterns from affective interactions using electrodermal activity and the use of different methods to give meaning to the data, from Azevedo, Taub, Mudrick, Farnsworth and Martin (2016 in Dindar, 2018): interdisciplinary research methods used to investigate emotions with advanced learning technologies.
The main aim of the study is to analyze how physiological synchronicity (PS) of individuals when working in a collaborative task can show temporal changes in students’ self-regulatory skills concerning cognition, motivation, behavior and emotion. Previous studies and researchers have utilized self-reports to measure students´ regulatory skills before and after a specific task. These questionnaires are valuable for research purposes and easy to administer, but there are several disadvantages. The presenter identified the following issues as determinant for the use of other methods apart from questionnaires in this study. Self-reports can be rarely introduced during the collaborative activity and preferable they should be short, in order to avoid students´ random answers. Moreover, when questionnaires are used, there are variables that cannot be controlled, such as the respondent´s interpretation of what is desirable or acceptable to be answered, and other factors affecting their state of mind and mood, causing a source of bias for the study. Finally, self-reports cannot be considered as a reliable source for identifying what is happening on-line and thus, for this particular study, Dindar (2018) highlighted that for understanding what causes the need of regulation, physiological data can be a more reliable measure for filling the gap of questionnaires. Nevertheless, the necessity of contextualizing the events requires the use or other methods like self-reports (Dindar, 2018).
The presenter affirmed that the scope of the study is to understand what events cause a need of self-regulation in collaboration. For this purpose, the following questions were examined in the study. Firstly, it is researched if changes in areas of regulated learning have an impact on academic achievement (RQ1). Secondly, the possible relation between PS and self-reports in terms of the facets of regulated learning (RQ2) and thirdly, if the PS has any direct relationship with academic achievement (RQ3) (Dindar, 2018). The participants are high-school students and the context is an optional physics course. The topic is the reflection of light through water. There are 31 participants (M=23 & F=8). Regarding the methods, this study has collected data from three main sources. Firstly, from students’ questionnaires or self-reports (N=19). Secondly, from student’s EDA (N=12) and thirdly from academic achievement scores, as the outcome of their learning.
Additionally, for the data analysis only quantitative methods are used to explain the data obtained. The SSI is calculated from the EDA of each group and correlation matrixes are created: one comparing the dimensions of SRL from self-reports and the student’s academic achievement scores for RQ1. The second comparing PS, SRL dimensions and academic achievement for RQ2 and the third contrasting PS and academic achievement scores for RQ3. In this line, for obtaining the SSI, Dindar (2018) explained that the index is calculated utilizing a five-seconds window and a one-second span for continuing the analysis. The reason for using a one-second time-frame is for calculating the increase and decrease slopes in the EDA values.
The data obtained from the self-reports and the academic achievement scores is correlated and the main findings reflect that the motivational change is the only facet directly related to the final scores. Moreover, motivational change is also related to emotional change, which is the only connected facet of regulated learning related to the group task scores. However, the results also show that the facet which is most related to other aspects of SRL is behavioral change, which is connected to cognitive and motivational changes. For the RQ2, the results show that only cognitive change is related to the SSI. Dindar (2018) affirmed that this is due to the nature of the study and also to the topic of the course. The tasks are cognitively demanding and in physics, normally the exercises have one answer, which is not open to debate. Finally, in terms of the RQ3, no relation between the PS and the academic success has been found.
Limitations of this study have been found in the number of participants in general, which had an impact on the data obtained from correlational analysis. Moreover, other limitations have been the number of participants who gave the consent for measuring the EDA, the topic of the course and the nature of the experiment. For these reasons, this study cannot be extrapolated to other collaborative learning situations.
For further analysis of EDA peaks and physiological data Dindar (2018) remarked that multilevel modeling is necessary for a holistic and objective perspective of the phenomena. Moreover, other possible improvements of this study should be focused on the tasks given to the students. As the nature of the activities were cognitively demanding, Dindar (2018) claimed that more and diverse tasks, in terms of the topic, should be given to clarify if different peaks in other facets of regulated learning are obtained or not.
PAper 3
The final presentation in the first day of the conference is given by Hanna Järvenoja (Post Doctoral researcher, LET University of Oulu) under the title: Measuring motivation and emotion regulation on-line (Järvenoja, Järvelä, Malmberg, Näykki, Kurki, Mykkänen, Törmänen & Isohätälä, 2017).
The aim of this presentation is to depict state of art claims and developments in process-oriented methods and online measures in learning, in reference to emotion and motivation regulation. For this reason, there is not an explanation of an empirical study, but examples of methods used in empirical studies are used to support and contextualise the information.
Jävenoja (2018) affirmed that motivation and emotion regulation has been widely studied and examined in theory but there is lack of study in practice. More stress has to be put in the study of these two facets of regulated learning in varying situations. The reasons to examine motivation and emotion regulation yield on the understanding of the factors that trigger such processes and the implications those have on learning. Moreover, these two areas play a role in every of the steps of regulated learning, whether is individually or socially regulated, thus the importance to be also studied in collaboration. Nevertheless, as technology has become a main support for both learning and studying how people learn, Järvenoja (2018) mentioned that technology-enhanced learning (TEL) also constituted a part of the framework for the emotion and motivation regulation research in varying situations.
Although extensive research has been done in motivation and emotion regulation in learning, Järvenoja (2018) affirmed that the main objective of studying them in varying situations is to obtain the changes “on the fly”, on the actual moment. For this matter, on-line measures can provide the picture of the situational variations and the variations in social contexts. On-line process-oriented methods, theoretically support the cyclicality of regulated learning. They differ from other retrospective methods that rely on subjective interpretations, such as self-reports, questionnaires and interviews. However, caution must be applied when researching, and emotion and motivation regulation has to be understood as a multi-layered, context-situated, individual and social phenomena. As a consequence, multiple methods have to be taken into consideration for obtaining an objective and comprehensive perspective (Järvenoja, 2018).
Turning now to the discussion of this presentation, Järvenoja (2018) defined four claims that can be considered as principles for the research: 1) Motivation is situation and context specific. There are variations along time. 2) Motivation in Learning is both individual and social. It does not mean “sharedness” but built in interaction. 3) Effects of motivation and emotions in learning is multi-layered. There are several facets and levels influencing, for instance cognitive and personal as well as collective situations and finally, 4) motivation and emotion regulation matters in successful learning (Järvenoja, 2018).
In conclusion, each of these principles highlighted by Järvenoja (2018) require different methodological approaches and the results will show different challenges. For instance, being motivation contextualised and situated, ecological valid methods must be applied not to interfere in the learning situation and also to capture the on-line variations. However, the generalisability of these findings may be limited and the complexity of learning situations can be difficult to interpret with only one method. Other challenges, such as merging individual and group level data, including other sources of information, also subjective interpretations, and combining information to obtain big and complex data have been identified by Järvenoja (2018).
What I found especially interesting? Why?
Personally, the three papers presented have demonstrated that related issues can be tackled from different perspectives. For instance, physiological synchrony has been used for two of the experiments but for different purposes. Consequently, it can be said that the methodology for a study not only depends of the nature of the experiment but also on the aim of the same.
Regarding the paper presented by Malmberg (2018), the idea of visualizing “sharedness” caught my attention. Undoubtedly, metacognitive monitoring in collaboration needs to be visible in order to be understood by others. Socially Shared Regulation and Collaborative Learning are two merging theories with different approaches but in this context, “sharedness” plays the same role in both. For this reason, previous studies have defined terms like grounding (Baker, Hansen, Joiner, & Traum, 1999) and shared conceptual space (Roschelle & Teasley, 1995) and had raised the importance of communication and participation for achieving a mutual understanding in the group. However, as Malmberg (2018) affirmed, when interaction is not verbalized, when the monitoring is not “visible” how is it “sharedness” understood different from “collective minds thinking”? The current methods of research search for ecological validity and non-intrusive ways of obtaining data. However, based on the results from the study carried out by Malmberg (2018) the methods have identified changes in the EDA values but not in the qualitative characteristics of the same. In this same line, as Järvenoja (2018) affirmed in her presentation, it is important to remark the validity of the methods specially in the data collection, not causing a source of bias or influencing the students’ performance.
However, in any case the three presentations have one idea in common even if the scope of each of the papers was different. Process-oriented methods put the focus on the in-situ happening, on the contextual and situated learning (Dindar, 2018; Järvenoja, 2018; Malmberg, 2018), thus the need of understanding the multiple layers of the self and the socially shared regulation, and the connections in between, are required for future research (Järvenoja, Jävelä & Malmberg, 2015). Commonly, analytical approaches have been used to identify the individual and social processes and the relations, however cross-validity challenges emerge due to the multiple layers and facets involved in such processes. Differences in operationalization of data may also arise and suppose another challenge when studying self and socially shared regulation or learning (Järvenoja et al., 2005). As a consequence, a posteriori triangulation of data is required to establish a common ground for a comprehensive analysis. For example, Malmberg (2018) made use of PS and video observation for their study, Dindar (2018) utilized self-reports and PS, and later discover that multilevel modelling would be required in future research and Järvenoja (2018) presented eye-tracking, EdX log data from questionnaires and EDA and heart rate tracking with wristband sensors.
Another interesting aspect was mentioned by Dindar (2018) when the future implications of his research were being discussed. Dindar claimed the need of multilevel modeling for a more comprehensive perspective of the phenomena being studied. In this case, he is referring to the different facets in self-regulated learning and their impact on academic achievement in the individual level. Multilevel modelling aims to approach the complexity of the group work phenomenon. It is a statistic tool that can be used both to account for the complex structure of the data and to incorporate variables at both the client and group levels (Selig, Trott & Lemberger, 2017 p.1). Two defining features of multilevel modelling are nested data and varying regression coefficients. Nested data signifies that the scores obtained from individuals are not independent one from each other, but are organised in a set of groups or clusters (Selig,et al., 2017). For example, in the research, the physics course tasks, the facets of regulated learning and the academic achievement are contained. On the other hand, varying regression coefficients permit the use of not lineally distributed data, which can show a more realistic picture of the phenomenon. Moreover, multilevel models can be used to analyse repeated measures of data (Buxton, 2008).
Regarding the topic of interest, I find appealing how interest is related to cognition and motivation. During Järvenoja´s presentation I personally asked how interest can be developed and she mentioned that it evolves and that one can never have internal interest in everything, but that through regulation one can deal with it. Literature in interest has shown the agreement among researchers in the two main classes of interest, situational and individual (see. Alexander, Kulikowich, & Schulze, 1994; Hidi, 1990, 2000; Krapp, 2000; Krapp, Hidi, & Renninger, 1992; Renninger, 1990, 2000; Schraw & Lehman, 2001, in Hidi & Renninger, 2006) and in the affective and cognitive facets of it interacting as merged systems (Hidi & Renninger, 2006). Situational refers to the context specific interest that raises as an affective response to an environmental stimulus, and individual interest when it this is prolonged and the person is likely to reengage due to a physiological state of mind that has been previously activated (Krapp & Fink, 1992; Renninger, 2000; Renninger & Wozniak, 1985, in Hidi & Renninger, 2006). Based on this, one cannot have interest in what is not known, neither if there is not stimulus, but how can teachers then help their students? Should interest be triggered? Should teachers focus in maintaining the interest of their students? What I find now really “interesting” is how that physiobiological reaction is different in each of the human beings, but somehow is in our system all procedures are the same and there are based on our cognition. Affective processes are caused by an action reaction effect, being the emotional response cognitive in nature. Thus, every interest has to be cognitive related. I personally believe that one can only be interested in what one knows, but the question is, is it needed to ask about what you could be interested in?
Another interesting point was raised by a student colleague. She asked the presenter how to measure motivation or interest in an online setting. She claimed that in face-to-face settings, it might be easier for researchers to measure, but not in online settings. As a response to this question, Järvenoja (2018) explained the need of multiple measures to make visible the metacognitive monitoring of emotions and motivation. In addition, the researcher claimed that in online settings the role of the verbalization is even more important because other clues are missing in the environment.
References
Presentations
Dindar, M. (2018, March). Interplay of temporal changes in self-regulation, academic success and physiological synchrony. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/muhterem_dindar_earli_presentation.pdf
Järvenoja, H. (2018, March). Measuring motivation and emotion regulation on-line. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/hanna_jc3a4rvenoja_-measuring-motivation-and-emotion-regulation-on-line.pdf
Malmberg, J. (2018, March). Are we together or not? Sequential interplay of monitoring and physiological synchrony during a collaborative exam. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/jonna_malmberg.pdf
Reflection part
Baker, M.J., Hansen, T., Joiner, R. & Traum, D. (1999). The role of grounding in collaborative learning tasks. In P. Dillenbourg (Ed.), Collaborative Learning : Cognitive and Computational Approaches, pp. 31-63. Amsterdam : Pergamon / Elsevier Science
Buxton, R. (2008). Statistics : Multilevel modelling. Retrieved from http://www.lboro.ac.uk/media/wwwlboroacuk/content/mlsc/downloads/Multilevel modelling.pdf
Hidi, S., & Ann Renninger, K. (2006). The four-phase model of interest development. Educational Psychologist, 41(2), 111–127. https://doi.org/10.1207/s15326985ep4102_4
Järvenoja, H., Järvelä, S. & Malmberg, J. (2015). Understanding the process of motivational, emotional and cognitive regulation in learning situations. Educational Psychologist, 50(3), 204-219.
Roschelle, J., & Teasley, S. D. (1995). The construction of shared knowledge in collaborative problem solving. In Computer supported collaborative learning (pp. 69-97). Springer, Berlin, Heidelberg.
Selig, J. P., Trott, A., & Lemberger, M. E. (2017). Multilevel Modeling for Research in Group Work. Journal for Specialists in Group Work, 42(2). https://doi.org/10.1080/01933922.2017.1282571
day 2
Poster session 1
During the second day of the LET 2018 conference held at the University of Oulu on the 29th of March, the first two presentations are given by Marta Sobocinski (Doctoral student, LET University of Oulu). In the first presentation Sobocinski (2018) introduced the SLAM project and in the second one, the paper presented was titled: Exploring small-scale adaptation in socially shared regulation of learning (Sobocinski, 2017). This last presentation is the main field of study of Sobocinski’s doctoral studies.
Firstly, SLAM is a project founded by the Academy of Finland where research is placed on Regulated Learning. Sobocinski (2018) mentioned SLAM’s main objectives as (1) making situational characteristics of learning visible. (2) Find evidences of students’ strategic regulation of learning. (3) Strengthen areas of regulated learning for self- and socially shared by using learning analytics and available technologies and (4) design interfaces that support and reinforce regulated learning for individual students, groups and teachers.
Secondly, the study presented is framed in the field of Regulated Learning, more specifically on the Self-Regulation of Learning (SRL) but also in the field of physiology, as previous presenters Malmberg, Järvenoja and Dindar, with the difference of using heart rate indexes and not EDA values. Despite this fact, the same idea of groups’ attunement or “togetherness” (Noy, Levit-Binun, & Golland, 2015 in Sobocinski, 2018) is involved by using physiological data and synchronicity.
For a clear explanation of the background of the study, more in detail Sobocinski (2018) mentioned the relevance of the cyclicality of SRL as the key element for the learner’s adaptation, basing her arguments in previous studies from experts like Hadwin, Järvelä and Miller (2017). Notwithstanding, the presenter claimed that regulation is only activated when challenges are confronted, and successfully achieved when they have been metacognitively monitored. In addition, due to the cyclicality of the process and to the agency of individuals, an adaptation happens as a result. However, Sobocinski presented the idea of adaptation in line to what other researchers have based their investigations, on “the process on the fly” (Järvenoja 2018; Dindar, 2018). For this reason, the presenter claimed that, contrary to understanding the learning process ‘cyclicality’ as fixed and in the long term, the current study deals with small-scale adaptations in the learning process. This term, has been scarcely studied empirically because of the difficulty to conceptualize it, according to Pieschl, Stahl, Murray and Bromme, (2012 in Sobocinski, 2018) and the presenter affirmed that her research aims to stretch that gap and that the focus was in understanding the relationship between monitoring and small-scale adaptations of individuals in a collaborative learning task. For this purpose, there are three research questions: What types of monitoring do students use? (RQ1). How does small-scale adaptation occur during collaborative learning? (RQ2) and what is the relationship between monitoring, adaptation and physiological synchrony? (RQ3).
Regarding the data collection, two main sources of information: the heartbeat measurements from empatica S4 sensors and video data was collected. The participants (N=12) were high school students and the context was an advanced physics course. The students worked in groups of three. Subsequently, for the data analysis, the different facets of regulated learning were coded in regard to monitoring and also the phases where it occurred were included from the video data. An inductive approach to the data analysis was applied using and adaptation to the codes used by Hadwin, Järvelä & Miller, 2017 and Bakhtiar, Webster & Hadwin, 2016. For the analysis of the small-scale adaptations, the researcher observed the phase before and after to establish a causal relationship between monitoring and the phase where students are. Furhermore, the physiological data was taken into account when there was a match in the heartbeat rate (high synchrony heartrate episodes). Therefore, the physiological indexes were calculated. Finally, a sequential lag analysis was used to establish the relationship between the physiological synchrony, monitoring and adaptation. To do so, the analysis was conducted at lag 1 and lag 2, looking for a statistical answer to what is most common to be found after a monitoring event.
The results of this study show that, for RQ1 the most frequent events of monitoring happened in the task enactment and task definition phases, being the first phase the most emphasized. In both cases, cognition and behavior facets the most highlighted and where generally followed by a reaction (83.09%) and (81.75%) respectively. On the contrary, motivation and emotion monitoring was mostly found by an absence of reaction. As for RQ2, the most common small-scale adaptation found was the switch from task enactment phase to task definition (63.52%) or to goals and planning (17.14%). These results depict the nature of the activity, because students were constantly evaluating during the activity whether they are performing the experiment correctly and if they were accomplishing their objectives. Finally, for RQ3, the findings reveal that after a reaction event, it can be followed by monitoring cognition and behavior, and also adaptation, with the particularity of having physiological synchrony, whereas in the case of no reaction, physiological synchrony was not found in the sequential analysis.
In brief, the main findings of this study according to Sobocinski (2018) are: (1) monitoring behavior and cognition during task enactment are the most common monitoring events. (2) Small scale adaptation happens usually between task enactment and task definition phase. (3) Adaptation occurs after a monitoring event and is followed by a reaction and physiological synchrony and (4) physiological synchrony is associated with successful monitoring.
Personally, this study has shown how the task can condition target of the students’ interaction. In this case, cognition and behavior have been the main concerns for students to regulate because of the nature of the activity. In this fashion, if there was not a reaction after an attempt to monitor motivation and/or emotions it could mean that the proactivity of group members and the task itself can determine the physiological arousal. In this case the groups’ physiological index was used but it would be interesting to analyze if all the group members did not feel any increase in their arousal or if any did but did not want to express it. Thus, I think the results from Sobocinski (2018) can be used to analyze if the group dynamics can have an impact on the individuals’ beliefs of the groups’ needs.
Poster session 2
The following presentation is given by Eetu Haataja (Research asssistant, LET University of Oulu). The tittle was: Monitoring in collaborative learning and physiological synchrony – How they co-occur? (Haataja, 2017).
The background of this study relies on the theories of Self-Regulated Learning (SRL) and of Collaborative Learning (CL), but also, as a continuation of aforementioned studies (Malmberg, 2018; Dindar, 2018; Järvenoja, 2018 & Sobocinski, 2018), this study uses physiological data as part of the research. Haataja (2018) presented several methods used for measuring; self-regulatory skills, such as questionnaires, think aloud protocols and interviews, and in the context of collaborative learning, through observation. However, he posited that new approaches or methods need to be used to capture the temporal variations and reveal changes in regulation. He proposed as an example, process-oriented methods.
The main focus of Haataja’s research is also in the role of monitoring. He presented monitoring as being part of metacognition, which indeed, is in nature a cognitive activity. However, Haataja (2018) presented that in the self-regulated theory, cognition about emotions, motivation and behaviour is also a part of metacognition, and thus can be monitored. Especially meaningful that cognition, is for identifying and solving challenges that may arise in collaboration, for fostering knowledge construction and create awareness in the students about their own learning process. For this particular last reason, Haataja (2018) specifically highlighted the question how does monitoring temporarily unfold in groups? He also claimed that if monitoring reflects groups’ joint awareness, is it related to PS? This is the main question in the research.
This study was planned as a collaborative activity for high school students. Their task was to design the breakfast for a marathon athlete. The students made use of WeSPOT as their learning environment and three of the groups were chosen for further analysis. In this line, as part of the data collection, the researcher used physiological data in the form of EDA. These values are read from empathic sensors attached to wristbands and reflects sympathetic activity of users. Also, video data was used for data collection and later for coding monitoring events and their target. On the other hand, the physiological methods for the data analysis are similar to the ones previously mentioned by Dindar (2018) and Järvenoja (2018). First it was calculated the EDA values of each student with quantitative statistics. Then, the values were standardized, and physiological concordance of pairs was calculated from the slopes in the EDA values in a 5 second window. Finally, the SSI was obtained as an indicator of the whole session. As the aim is to capture the temporal variations of physiological data and monitoring, a 120 second moving window index was used to reveal those changes. In addition, this index was later used for the cross-correlation of monitoring and PS.
The results show a weak relation in between monitoring and PS when the facets of regulated learning are considered separately. However, if the overall monitoring activity is taken into account, the values are higher meaning that the relation is more significant. The facets most monitored are cognition and behaviour, which could explain the nature of this experiment. Personally, I believe that these results can be utilised for conducting a future research. It might be interesting to combine the dataset and qualitative methods from Sobocinski like the video data and the quantitative analysis of Haataja using the EDA values for a better understanding of the quality of monitoring and the source of that monitoring. Moreover, for the validity of the results of research perhaps it is also convenient to include another topic that is less cognitive-focused as it is physics. Consequently, other forms or monitoring may arise and be visible.
Poster session 3
The last presentation in the second day of the LET Conference 2018 is given by Hector Pijeira-Díaz (Doctoral student, LET University of Oulu). The tittle was: Investigating collaborative learning success with physiological coupling indices based on electrodermal activity (Pijeira-Díaz, Drachsler, Järvela & Kirschner, 2016).
Pijeira-Díaz presented the topic of his paper referencing the previous speakers from that same day. He emphasized Collaborative Learning (CL) as the main focus of this research but also mentioned the use of physiological data. As a starting idea and for framing the research, Pijeira-Díaz claimed that the term physiological synchrony is not a state of the art technology, but it has been used and studied before under other terminology. For instance, among many others, synonims of PS are: physiological linkage (Levenson & Gottman, 1983 in Pijeira-Díaz, 2018), physiological compliance (Smith & Smith, 1987 in Pijeira-Díaz, 2018), physiological synchronization (Henning & Sauter, 1996 in Pijeira-Díaz, 2018) and physiological makers of togertherness (Noy, et al., 2015 in Pijeira-Díaz, 2018).
To begin with the presentation, Pijeira-Díaz (2018) started by claiming the benefits and the challenges of collaborative learning. As for this last issue, according to Baker, Järvelä & Andriessen (2013 in Pijeira-Díaz, 2018) three main strands can be differentiated: cognitive, socio-emotional and the operationalization of CL, which refers to “how to measure what was going on in that group” (Pijeira-Díaz, 2018).
The presenter is specially interested in this last challenge and mentioned the most widely used methods for that measurement. Traditionally the operationalization of CL has been evaluated with self-reports, but their disadvantages are well known. Also, by transcribing group interactions, but in here the researchers’ subjectivity might affect the reliability of the results. For these reasons, Pijeira-Díaz is particularly interested in biosensor data. He argued that study is focused in using physiological data to obtain objective information, in the form of numbers or values, that can be interpreted by researchers, but that do not come directly from their analysis or interpretation. In this sense, as presented by Sobocinski (2018), Haataja (2018), Malmberg (2018) and Dindar (2018) biodata has been used and the focus has been put specially in five indices found in literature (Elkins et al., 2009 in Pijeira-Díaz, 2018): SM, signal matching, which measures the amplitude of the signal. However, to use this measure it is essential to standardise each individual’s signal when being relaxed. IDM, instantaneous derivative matching, which measures the direction of the signal. DA, directional agreement, which shows the percentage of agreement of individual signals moving in the same direction. This index does not need to me standardized and also allows to compare more than two individuals without computation. PCC, pearson’s correlation coefficient, which is an indication of the linear relationship of the data between two individuals and finally, FTZ, fisher’s z-transform, which is a transformation of PCC for obtaining a normally distributed index.
Pijeira-Díaz’s aim with this study is to identify which PCI (physiological coupling index) better reflects the three main collaborative measures: Collaborative Will (CW), Collaborative Learning Product (CLP) and Dual Learning Gain (DLG). Regarding the methods, this study used only quantitative approaches. For the data collection, Pijeira-Díaz (2018) used MSQL questionnaires for measuring CW, group’s reports for CLP and empathica sensors. As for the data analysis, regressions analyses were conducted. The participants for this study were high school students (N=48, M=21, F=27) and their task was to design a healthy breakfast for an athlete. Finally, the results from this study show that the better predictor for CW is the IDM index, having a value of almost the double compared to the other indexes. Moreover, this same index seems to be a good indicator as well for CLP. Finally, the best indicator of DLG appears to be DA.
Personally, I think Pijeira-Díaz’s presentation is interesting, especially for the different terminology presented regarding physiological synchrony. However, I do not know exactly what to extract from this study, since in the presentation, Pijeira-Díaz only had time to explain the background and the importance of physiological synchrony. In this fashion, there was no time to discuss about the research itself, and thus, my understanding of the results is quite limited.
What I found especially interesting? Why?
From this day presentations there are two ideas that caught my attention. However, I must say that I could not find any particular issue that would lead me to make research from my interest as in the case of the other presentations. Nevertheless, this are my beliefs, concerns and insights from the presentations given by Haataja (2018), Sobocinski (2018) and Pijeira-Díaz (2018).
Starting with Haataja’s research, I find interesting the moving window approach for the data analysis. From all the methods presented, this is the one I consider the most interesting, because it captures temporal variations, but the values and valences can be different. For instance, the aim can be to measure the distance between to opposite values at a time (x) and at time (x +/- y). Also, I think that Pijeira-Díaz (2018) presentation of the indexes have helped me understanding Haataja’s moving window. When I was listening to Haatja’s presentation, apparently, I did not get the full idea of the method used for the data analysis. The reason why I could not understand it was because he mentioned he had need to standardize the vales. Not before Pijeira-Diaz had mentioned the different indexes used in their research, that I could understand what the “slopes” of the EDA values meant and thus, also what the comparison would signify. I think that the moving window approach mentioned by Haataja (2018) can be highly interesting for contextualizing the EDA group values. In addition, I would consider applying this approach in my master thesis if my research questions allow, since my interest is in groups’ interactions and the moving window might help for capturing the temporal variations.
Secondly, from Sobocinki’s presentation I found interesting the term adaptation as part of the cyclicality of SRL (Hadwin, Järvelä, & Miller 2017). As it is known that adaptation happens as part of regulation, I believe in the addition of the term “small adaptation” posited by Sobocinski (2018). I think that the results from that study highlight the context of the activity as cognitive demanding but lacks emotional and behaviour regulation. What I find as possible for future research is to investigate if there would be any type of temporal variation in between the areas of regulation and the adaptation process. Also, it might be interesting to triangulate sources from physiological measures.
Finally, I would like to remark the last idea mentioned by Sobocinski when explaining the SLAM project. The aim of developing a data visualisation tool that combine the multiple data channels and permit a better comprehension of the phenomena, is to me, simply amazing. I truly understand the collaboration here between the data signal technicians and the experts from learning sciences. This issue is mentioned by Järvelä (2018) as essential for the future development of the research, and in here, it takes form of a real software capable of using multimodal data.
References
Presentations
Haataja, E. (2018, April). Monitoring in collaborative learning and physiological synchrony – How they co-occur?. Poster session presented at the at the LET2018 Conference, Oulu.
Pijeira-Díaz, H. (2018, April). Investigating collaborative learning success with physiological coupling indices based on electrodermal activity. Poster session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/20180329_hc3a9ctor-j-pijeira-dc3adaz.pptx
Sobocinski, M. (2018, April). Exploring small-scale adaptation in socially shared regulation of learning. Poster session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/marta-presentation.pptx
Reflection part
Miller, M., Järvelä, S., & Hadwin, A. (2017). Self-regulation, co-regulation, and shared regulation in collaborative learning environments. In Handbook of self-regulation of learning and performance (pp. 99-122). Routledge.
day 3
Paper 4
In the third day of the LET Conference 2018 the first paper is presented by Tiina Törmänen (Doctoral student, LET University of Oulu). The tittle was: Exploring collaborative groups’ emotional states with video and physiological data (Törmänen, Järvenoja, Kurki, Devai, and Sanna Järvelä, 2018).
This presentation is part of Törmänen’s PhD study, which is framed under the name of Tracking the Role of Socially Shared Emotion Regulation for Collaborative Learning Progress. Thus, it can be said that this study is grounded on the theories of regulated learning: Socially Shared Regulation of Learning (SSRL) and Collaborative Learning (CL). Moreover, Törmänen (2018) affirmed that this study is part of a project called EmReg, which main aim is to research about emotion regulation in secondary school pupils. Törmänen’s doctoral study aims to address the following issues: (1) the relationship between the individual’s affective state and the emotion regulation within groups and (2) the role of emotion regulation for collaborative learning progress (Törmänen, 2018). The main question that is guiding her study is: how to capture the situation specificity and interconnection between emotional states and emotion regulation. Törmänen affirmed that emotions are situation specific and therefore the need to interpret them in authentic contexts.
However, she claimed the difficulty of measuring emotions because of being a multifaceted phenomenon, having different layers involving cognition, affection, motivation and physiological process, among others. Nevertheless, Törmänen presented a model that might be helpful for evaluating emotions. It uses two main dimensions: valence and activation. This model is an adaptation from the model of academic emotions by Pekrun, Goetz, Titz & Perry (2002 in Törmänen, 2018). In brief this model presents the two dimensions in two different axis, having an activation/de-activation axis and negative/positive axis.
For this study in particular, Törmänen made use of process-oriented methods to capture the level of activation of the situated emotions and also to contextualize them. In this fashion, similarly to the other presenters of this conference, Törmänen (2018) had used Skin conductance Response (SCR) to find peaks in the level of physiological arousal of students. The measurement she used is EDA. Moreover, to contextualize the emotions Törmänen used video data. The context of the research is a science assignment for 6th grade students (N=41) who are set in 12 groups. The activity was divided into four phases: individual work, brainstorming, planning and building.
Moving to the research questions, Törmänen aimed to find: (1) how the collaborative groups’ emotional state is varying in terms of observed valence and activation during a collaborative learning session. (2) Is there an association between groups’ observed emotional activation level and physiological activation level measured with EDA? (3) In what kind of situations is group members’ physiological activation level in synchrony? (3.1) Is there a need for group level emotion regulation in those situations?
Regarding the data analysis, Törmänen used a deductive approach. She divided the video data in segments of 30 seconds and analysed each segment allocating socioemotional disclosures. For example, expressing positive or negative emotions verbally or by charged interaction. Subsequently Törmänen coded the groups interaction in terms of valence and activation for RQ1. Four categories composed the valence; positive, negative, mixed and unclear, and two the valence: de-activating and activating. As for RQ2 and RQ3 Törmänen will use the EDA values to explore the coded interaction and physiological data.
The results from this study are in a preliminary stage. However, it can be concluded that different groups experienced different emotions and thus socioemotional challenges. Also, that the emotions varied throughout the session. In this line, for instance one group score showed a 94% of socioemotional challenge whereas another just 22%. In addition, it is remarkable that for some groups more negative activation than positive was found, even for this last group, 44% of the activity was denoted with negative activating emotions. In brief, more activating than de-activating emotions were found in this study, and most of them were negatively charged. However, there were also group differences in terms of the valence of the emotions and the socioemotional interactions.
Personally, this presentation is a great example of what is a starting point for a PhD study. I firmly believe that Törmänen is using similar methods as the other presenters in this conference but for understanding a broad topic as is “successful collaboration”. I am convinced that after answering to RQ2 and RQ3, many new questions will come to Törmänen’s mind, which perhaps will reconduct her PhD study. In this sense, what I would highlight from this study at this point is that the group’s atmosphere is undoubtedly a result of the individuals’ cognitive and emotional state, thus it can vary along time. The point I want to make here is that even the results obtained from this kind of research, have to be explicitly detailed and contextualised in order to provide an objective picture of the phenomena. Teachers and students can see this as a learning in process that is eventually situated in a specific time and location, with different resources and attributions. Therefore, making people metacognitively aware of how the learning and personal environment are affecting the “learning” itself is a genuine but interesting point that this presentation has raised on me.
Paper 5
The next presentation on the third day of the LET Conference 2018 is given by Pirkko Siklander (Ph.D Docent, LET University of Oulu). The tittle was: Playfulness as a teachers’ and educator’s competence. (Siklander, 2018)
This presentation was slightly different from the others due to the topic which was playfulness. Siklander begun the presentation with the claim that playful learning is a need in every learning context, therefore it is also a pedagogical tool that every teacher should have in their tool box. The benefits of play are diverse and unexpected, but it is known that during childhood, playing is part of the development of problem-solving skills and creative thinking (Banaji & Burn, 2007; 2010 in Siklander, 2018a). Moreover, research has been done in the field and has also shed light on the issue of its influence on developmental patterns. It has been found that can enhance collaboration and regulation skills and support a state of well-being (Bateson & Martin, 2013 in Siklander, 2018a). However, Siklander (2018a) affirmed that play is nowadays decreasing in children’s lives but new the new Finnish curriculum stresses the need of implementing play in learning, not only for students but also for teachers. Notwithstanding the difficulty of applying it is considerable.
Siklander presented the poster and study: I like to make people laugh: adult playfulness among educators. The aim of this research is to evaluate how people see their own playful skills. To do so, the research questions were: (1) How do adult educators evaluate their own playfulness? (2) What are the most general qualities of playfulness among educators? (3) Are there differences in experienced playfulness among educators of different ages and gender? The participants were teachers and students in the educational field (N=123).
For the data collection several methods were utilized. Firstly, a Finnish adaptation of Staempli’s (2005;2007) Adult Playfulness Scale (APS). In second place, GuilFord’s test of divergent thinking- Alternative uses task (Guildford & Hoepfner, 1971) and finally, The Remote Associates test (RAT) (Mednick, 1960). The aim of Guilford’s test is to derive multiple answers from a single issue. It has four dimensions: originality, fluency, flexibility and elaboration. On the other hand, Mednick’s test deals with the ability to associate seemingly unconnected words and with the establishment of relations between those. Regarding the data analysis, multivariate tests were run to examine the influence of gender and age.
The results from this study show that participants rate quite highly themselves in terms of playfulness, including also the joy of being with others and their creativity. Moreover, highly playful members are stimulated by original ideas and thoughts and are also challenged by ambiguous problems. Finally, in terms of age differences, this study did not come with any significant differences, but it did in terms of gender. Male participants are slightly less playful than their female colleagues.
I believe that this presentation has stirred old fashion ideas from our heads such as if you are having fun maybe it is not challenging enough. I am in total agreement with Siklander (2018a) when it comes to make learning playful. I think that not only cognition matters in this case, but also the human relation between the teacher and the pupil. I think that this study can raise awareness on the need of making things more human and thus, also playful. Teachers have to get to know better students not from mere observation but from experimentation in their natural environment. However, I also feel that I might need to reflect upon one particular mater. Playful learning does not advocate for clownish learning. Maybe during a period of time but not for a large extension. The point I want to make here is that young students can see a blurry line between being playful and being an entertainer if the teachers get extremely into being playful all the time.
What I found especially interesting? Why?
These two presentations have triggered one question in my mind. When researching about learning, we set the environment in a classroom or laboratory, but we rarely analyse play as a collaborative learning activity. It might be interesting to research groups’ free play in terms of emotion and cognitive regulation. When students play collaborative games, they are commonly assuming the role of their character and that requires some special cognitive and emotion capacity. In addition, it can be interesting to consider also process oriented methods to understand the “on the fly” (see. Järvenoja, 2018; Dindar, 2018) variations along time.
Another idea that caught my attention was the relationship between playfulness and divergent thinking. Lieberman (1965) established relationships between creativity, divergent thinking and playfulness from a behavioural perspective. It is assumed that as a result of exploratory activities in a rich and stimulant environment children develop creative thinking abilities (Lieberman 1965; Chapman, 1978). Therefore there is also an individual cognitive component in play, thus the environment can be considered as the static material but also can encompass human resources, like teachers. In this sense, in addition to what Siklander (2018) framed as “teachers’ playfulness”, I believe that it is essential to place play within a constructivistic and holistic pedagogy, so that teachers are not considered as entertainers but as co-players and supporters. A good example can be pedagogies like Emilia Reggio and Montessori, which support the children’s active approach of learning. In addition, this pedagogies also aim to foster students independence and inner drive, which are shaped in real and safe contexts to play and explore (Samuelsson & Carlsson, 2008). However, Montessori’s pedagogy has been relatively criticised because of its restriction of play to specific materials and of the absence of pretend play (Lillard, 2013). Obviously, freedom and imagination are essential characteristics of children’s play, specially because at the very early stages of their development, when infants act, they do not separate between play and learn (Samuelsson & Carlsson, 2008). Therefore, there is an imperative need of an enriching environment, in terms of material and human resources that support children’s free play and learning.
In my opinion, the connection between divergent thinking and playful learning yields on children access to rich environments where they have the opportunity to explore, and through iterative, non-restrictive and spontaneous play, children create their own version of the world. Therefore, from my perspective play should be included in schools as a mean, not as an end.
References
Presentations
Törmänen, T. (2018, April). Exploring collaborative groups’ emotional states with video and physiological data. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/tt_currenttrendsspring2018.pptx
Siklander, P. (2018a, April). Playfulness as a teachers’ and educator’s competence. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/playfulness-current-trend-2018.pdf
Reflection part
Chapman, J. A. (1978). PLAYFULNESS AND THE DEVELOPMENT OF DIVERGENT THINKING ABILITIES. Child: Care, Health and Development, 4(6), 371–383. https://doi.org/10.1111/j.1365-2214.1978.tb00096.x
Lieberman, J. N. (1965). Playfulness and divergent thinking: An investigation of their relationship at the kindergarten level. Journal of Genetic Psychology, 107(2), 219–224. https://doi.org/10.1080/00221325.1965.10533661
Lillard, A. S. (2013). Playful learning and Montessori education. American Journal of Play, 5(2), 157–186. Retrieved from http://www.journalofplay.org/sites/www.journalofplay.org/files/pdf-articles/5-2-article-play-learning-and-montessori-education_0.pdf
Samuelsson, I. P., & Carlsson, M. A. (2008). The playing learning child: Towards a pedagogy of early childhood. Scandinavian Journal of Educational Research, 52(6), 623–641. https://doi.org/10.1080/00313830802497265
day 4
Paper 6
In the fourth day of the LET Conference 2018 the first paper is presented by Arttu Mykkänen (PhD, LET University of Oulu). The tittle was: Students’ interpretations of a group awareness tool in a collaborative learning setting. (Mykkänen, Koivuniemi, Järvenoja & Järvelä, 2017).
Mykkänen (2018) begun the presentation giving the theoretical framework of the study. This study is built upon Collaborative Learning (CL) and Regulated Learning (RL), similarly to Dindar’s (2018), Malmberg’s (2018), Sobocinki’s (2018), Haataja’s (2018) and Törmänen's (2018). After the background, the presenter continued with the affirmation that success in collaboration require specific conditions and that by merely putting students together, collaboration does not always have a positive result. However, Mykkänen (2018) affirmed that one way to support students in collaboration is making them aware of the state of their group, and to do so, there are several tools. Some of these tools, under the name of group awareness tools, are meant to prompt the progress of collaboration and SSRL, to inform the group about members’ participation, to support the feedback process and to support the cognitive aspects of the groups’ performance. However, despite the number of tools available, Mykkänen (2018) affirmed that there is no research in the students’ perceptions about the use of those tools.
This study aims to understand what students perceive from the use of the tools. The research questions were: (1) What types of advantages relating to the group awareness tool use do students describe after a series of collaborative tasks? and (2) What types of disadvantages relating to the group awareness tool use do students describe after a series of collaborative tasks? The participants were second-year teacher education students (N=44, F=36, M=8) in a math didactics course for a period of seven weeks. The participants were paired in groups of five members and the activities took place in two different settings. Firstly, the students solved problems related to a lecture and secondly, they created a mid-term plan for a mathematical topic.
Regarding the data collection, each member filled in the S-REG every working session. This tool was created with the purpose of measuring the individuals’ perceptions about the cognitive, emotional and motivational aspects of the task. After doing this, the software would show, in the form of a traffic light, the cognitive, emotional and motivational states of the group and also would prompt the aspects where the group is having mismatches. In addition, another method used for the data collection was semi-structured interviews. 43 interviews were carried out by two different researchers investigating about the groups’ collaboration. However, there were also two questions regarding the students’ perceptions about the awareness tool: (1) on a scale 1 to 10, how beneficial was the awareness tool to your group? and (2) how did the awareness tool help your group’s performance? (Mykkänen, 2018).
Continuing with the data analysis, deductive approach or data driven approach was utilised along with content analysis. The aim was to identify how students interpreted the use of the awareness tool. As a result, the two independent coders obtained an inter-rater reliability of k= 72 and three main categories: positive interpretations, negative interpretations and needs of improvement. In general, students reported more positive than negative interpretations, but some of the negative aspects seem important to clarify. Despite these awareness tools can help to understand others’ state of mind, prompt discussions and set a proper start for the activity, they can also be an extra source of frustration and might even feel intrusive for some students. From the interviews, in 28 situations students affirmed that the use of the S-REG was not beneficial, and this would lead them to ignore it (f=24).
In summary, this study is in line with previous research in CL and CSCL. Mykkänen’s findings have indicated that increasing students’ awareness can support and prompt discussion and help in recognizing others’ state of mind, which leads to better collaboration (Eligio et al., 2012; Järvelä et al., 2015 & Phielix, et al., in Mykkänen, 2018). Moreover, it is important to establish the borderline and respect co-workers’ privacy, especially in terms of negative and personal issues. Highlighting these aspects or force people to talk can be seen as over controlling and restrictive for a successful collaboration (Dillenbourg, 2002 & Rummel et al., 2012 in Mykkänen, 2018). Finally, Mykkänen (2018) has found that students mostly devoted more time in the task related activities rather than in the relational aspects of collaboration. As a conclusion, Mykkänen (2018) portrayed Janssen and Bodemer 2013 model of spaces in collaboration. The content space relates the purely cognitive aspects of the task, whereas the relational space contains the socioemotional aspects. Mykkänen (2018) affirmed that students have raised aspects from both the content and relational spaces as a result of the use of these tools. Finally, the presenter suggested the future work of this study as a need to improve the awareness tools as an embedded part of the collaboration and not as a “must do”.
Personally, the results that can be extracted from this paper must be taken with precaution. I believe that making the group aware of their own needs is essential for a successful collaboration, or at least necessary for benefiting from the opportunities that this kind of learning offers. For collaboration to be efficient, not effective or successful, I am totally convinced that the social part can be skipped, notwithstanding that the consequences might be considerable and detrimental. However, as Mykkänen has mentioned, it is the right of the students to choose whether they want to establish deeper relationships or not. Some students do not want to establish any kind of social relationship apart from the mere work. I consider this idea as respectable, but I truly advice collaborators to undergo managing the socioemotional aspects of the collaboration for a fully embedded learning experience. When doing collaborative work, not only content is learnt, and social skills are developed, but also learning about how to manage oneself in the environment with others takes place, which is essential for the future in the workplace.
On the other hand, I agree with Mykkänen’s future implications in his research. Group awareness tools are great scaffolds for a more personalised and meaningful collaboration. Despite the fact that it can prompt controversial situations it is always at the students’ hand to whether and how to tackle that issue. Therefore, it can also foster self-regulatory skills. Teachers have the duty of supporting the students when doing collaborative work, but the groups’ socioemotional and cognitive state is complicated to visualize unless is spoken and shown in a form that everyone understands. These tools can be used by teachers to help and scaffold students to become better and more regulated learners.
Paper 7
The next presentation of the LET Conference 2018 is given by Kristiina Kurki (Ph.D, LET University of Oulu). The tittle was: Exploring regulatory interactions among young children and their teachers in day-care context – focus on teachers’ monitoring (Kurki, Järvenoja & Järvelä, 2017).
This presentation was part of the EmReg project and it is grounded in the field of Regulated Learning (RL), particularly in early childhood teachers’ regulatory skills. Kurki (2018) begun the presentation explaining what emotion and behavior regulation mean, and also highlighted the fact that in early stages of children’s development, this regulation is highly external or co-regulated by parents, teachers, caregivers… Therefore, the importance these figures have on the children’s development is unquestionable because kids internalize regulatory skills from their surroundings. Regulatory skills develop during childhood and interactions with others and equals is a cornerstone for this to successfully progress.
The aim of this study is to explore the types of emotion and behavior regulation strategies students use independently or with the teacher’s support. Moreover, the focus is also put on the relationship between teachers’ level of monitoring and the use of strategies by children. The context of this research is the early childhood school or kindergarten. The participants are 30 children between 2 and 5 years of age in a day-care facility. Kurki (2018) presented the three variables in her research and the previous studies and literature on this topic. For instance, mentioned Gross (2014) when referring to children’s emotion and behavior regulation strategies. Also, cited Hadwin (2013) and Bryce & Whitebread 2012 when referring to the adaptation of the regulatory skills and finally, quoted Thompson and Meyer (2007) when referring to teachers’ level of monitoring.
The methods for collecting data are video and audio recordings from authentic situations. There was no preparation and no activities were pre-defined. As for the data analysis, a theory-based approach was followed. Four phases encompassed the video coding. Firstly, the socioemotional challenges where identified. Secondly, the strategies where coded according to Kurki et al. (2017) and Gross (2014) coding scheme. Five main categories where operationalised: SM, situation modification. SS, situation selection. PI, providing information. RA, redirecting one’s own activity/attention and RM, response modulation. Thirdly, the teachers monitoring was coded for that particular moment, being active or weak monitoring. Finally, connections between teachers’ support and monitoring and children strategies use were conducted using statistical tests.
The findings from this study show that generally, teachers’ involvement, monitoring and support make a difference in the use of strategies, but not in the quality of those. These results are in line with previous studies mentioned by Kurki (2018) (see. Cole et al., 2009; Fox & Calkins, 2003; Spinrad, Stifter, Donelan-McCall & Turner, 2004; Kopystynska et al., 2016, Lengua et al., 2013) In addition, adaptation strategies seem to be more common in students who have been supported by active monitoring from teachers. In conclusion, the amount and quality of the strategies used goes in line with the challenges students face. Kurki (2018) concluded that these challenges are distinctive and thus the active monitoring of teachers can help students to gain access to more complex strategies and to rehearse the ones already known. Finally, Kurki (2018) finalised the presentation exposing the future implications of this study. With no doubt, this research highlights the importance of monitoring skills for teachers to; get insights about strategy use, to get a more personalised and closer trait to children and also to follow a healthy and right development in small children. However, the presenter affirmed that more emphasis has to be placed on finding the instant and long-term effects of teachers’ active monitoring and students’ strategy use. Also, according to Kurki (2018) it is important to identify the connections between teachers’ skills and children adaptation and strategy use.
In my opinion, this study is very interesting from the teachers’ perspective. Having the right balance between giving enough freedom and establish a safe socioemotional atmosphere in the classroom can be a complicated issue. I find appealing the use of “socioemotional challenges” as the main target of the research, but it could be also interesting to see how small children deal with positive emotions alone and in groups or pairs. I remember that in one question during the conference, the presenter was asked if it would be more authentic if also the parents could be seen as mediators of those “challenges”, in order to contrast as well to what extent parents, influence the adaptation and use of strategies. In summary, it is my belief that the results from this study support the idea that teachers have a great impact on the students’ regulatory skills and thus the need of establishing a good grounding for this in the very early stages of development. In brief, this study can be useful for future educators and for anyone dealing with children in general in any kind of situation.
Paper 8
The last presentation on the fourth day of the LET Conference 2018 is given by Pirkko Siklander (Ph.D, Docent, LET University of Oulu). The tittle was: Hiking in the nature to promote learner’s agency and competences (Kangas, Vuojärvi & Siklander, 2017).
This presentation aims to show the state of the research conducted by Kangas, Vuojärvi and Siklander in the field of education in nature. In a nutshell, it is grounded in Technology Enhanced Learning (TEL) by using gamification and other social media practices. In formal and informal education, indoors and outdoors activities and also it shares certain commonalities with the field of Regulated Learning due to the fact that learning in nature advocates to foster learners’ agency by multiple and meaningful experiences.
As an introduction to the topic, Siklander (2018b) explained that the new curriculum aims to emphasize more the students’ competences and skills, and thus the need to engage in situations that support through that learning. Siklander (2018b) also mentioned that informal learning, and more in particular outdoors learning, is a playful and meaningful student-centered approach. In addition, Siklander (2018b) mentioned the previous studies on outdoor learning as being focused on the impact this kind of learning has on the social and personal state of students. Results have shown that students become more active and participative through the affordances of nature (see. Beames, Atencio, & Ross, 2009; Dolan, 2016; Higgins, 1995). In this line, Siklander (2018b) claimed the importance of agency for learning. In traditional schooling students have little agency when everything is given to them, but in outdoor learning students become social agents and creators of their learning (Hyvönen et al, 2014; Hujala, Helenius & Hyvönen, 2010 in Siklander, 2018b). Moreover, Siklander (2018b) presented the term “efficacious agency” as an umbrella term for concepts such as self-efficacy, resilience and including the areas and repertoire of strategies of regulated learning.
This study aims to investigate how students’ agency and competences are promoted in outdoors learning during a hiking course. Other learning outcomes for the course are biology and geography related, content wise and promote personal growth, CL skills and self steem, competence wise. The research questions are: In which ways students’ agency emerges during a hiking course? RQ1 and how the competences needed in the activities are manifested? RQ2. The participants were eighth-grade students (N=21) and subject teachers (N=2).
Regarding the methods for data collection, the researcher used a participatory approach, combining field notes and recordings with the learning diaries or individual portfolios and later with interviews. For the data analysis, qualitative inductive content analysis was conducted placing emphasis on the activities that reflected agency. The result showed for main categories: responsibility, resiliency, collaboration and feeling of success. Secondly, another analysis was conducted with the focus on the competences needed for each activity.
The a priori results show for RQ1 that students undergone through experiences where they collaborated and cooperated, thus negotiation and contributions were evident during the course. In addition, students reported difficult moments, but their resiliency skills were fostered and as a result, students held positive attitudes and also feelings of success. Finally, as part of a group or community, learners were responsible for their individual and the group accountabilities, therefore this course also fostered students’ responsibility. As for RQ2, according to Siklander (2018b), the competences were manifested in the forms of solving problems, of social negotiation and of personal and social responsibility.
Finally, the presenter stated the future implication of this research. There is an urgent need of finding comprehensive ways to evaluate the competences gained by students because traditional assessment tools might not be suitable for out-of -school activities. In addition, it seems important to identify triggers that will help students to be engaged in this kind of activities, like for example technology of gamification (Siklander, 2018b).
Personally, this research reinforces the message that learning is an active process and thus the need to let students be proactive agents on it. In addition, I believe that Siklander’s ideas are relevant for the future of schooling, especially for those students who like nature and outdoors activities. However, this type of courses can also help to include students who normally would not go outdoors by themselves. Nowadays people tend to spend less time outdoors than in the past and also, even if using technology, we tend to engage less in real face-to-face activities. I also disagree that technology can be used to engage students in outdoors activities. However, I would make a distinction. If being in the nature is a mean in itself, then the use of technology is understood, but if the aim is to enjoy and be aware of the environment, then technology can be a double edge sword. Easily, I could see students not paying attention to the landscape, forestry, animals… and just playing with their gadgets, which in turn, could still develop social, emotional and cognitive skills.
What I found especially interesting? Why?
From these presentations, I consider that Kurki’s research will shed light on the need of monitoring children behavior. However, during the question round, I asked personally to Kurki if she would consider including parents in future research, but she mentioned that for the situation to be authentic parents will not be consider unless the research could also be conducted at homes. This issue made me think about the role that parents have in children’a development, and specially in children’s emotional development.
Research has been conducted in this matter and the evidences suggest that children’s emotional development is highly related to advances in cognitive skills. For instance, the sophistication of cognitive capacities create complex thinking structures which allows children to identify themselves and compare with others. This later shape children’s mentalism or theory of mind and self-understanding (Eisenberg, Fabes, & Spinrad, 2007). In a longitudinal study it has been found that parents relationship with children is asymmetrical along time (Salsich, 2001). This means that during the first years of children development, parents have more power to determine the interactions with children. As Kurki (2018) mentioned in her presentation, most of the children’s regulatory strategies are assimilated from an external source which, in most cases acts also as a co-regulator. In this sense, parents, teachers and other caregivers become children’s emotional coaches. Not only children can copy the strategies but also it is important to make them visible in order to support their autonomy. In this line Salsich (2001) affirms that parents should put verbal labels to children’s emotions and should support the disclosure of learning experiences, in order to contribute to children’s regulation. In this same line, Lengua, Honorado and Bush (2007) highlight the importance of reflecting parent’s responsiveness and scaffolding chidren’s autonomy. This means that parents have to engage in situations where children might need a corregulator and disengage when they can function independently. However, is important to clarify that parents views regarding emotions and emotion regulation, are highly influenced by their own culture and subculture and also by their own personal emotional state (Salsich, 2001).
Specially interesting is the fact that parents’ closeness, warmth and positive expressivity towards children promote better regulatory skills and better socioemotional state, but also foster children’s empathy which is translated to a better social functioning (Zhou et al., 2002). Why I consider this meaningful is because in the past and still in some cultures, the tendency is the opposite. For instance, in certain ethnic groups it is not permitted to display negative emotions, neither children are allowed to express anger or disagreement in public. Also worth mentioning is the old belief that children that affirmed that children should be taught to be “strong” in order to overcome possible bullies and harsh conditions (Salsich, 2001).
Also as an educator, I think that emotions have to be one of the most important aspects to deal during the first years of school. Supporting students to establish a correct self-image and esteem and to have a a sufficient repertoire of strategies for self-regulation and co-regulation is one of my most important and personal goals as a teacher.
This particular issue has raised more questions that I aim to further explain in a future. For instance, one of my interests is in researching about empathy in collaboration and in free play situations, among children and adults. I want to identify temporal patterns of empathic responses, qualities of empathic interactions and the connection between empathic interactions and learning outcomes in adults, and between empatic interactions and role play in children.
References
Presentations
Mykkänen, A. (2018, April). Exploring collaborative groups’ emotional states with video and physiological data. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from:https://letmaster.files.wordpress.com/2018/03/awareness_mykkanen_earli_2017_2_9.pptx
Kurki, K. (2018, April). Exploring regulatory interactions among young children and their teachers in day-care context – focus on teachers’ monitoring. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/earli2017-presentation-kristiina-kurki.pdf
Siklander, P. (2018b, April). Hiking in the nature to promote learner’s agency and competences. Paper session presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/hiking-course-current-trends.pdf
Reflection part
Eisenberg, N. , Fabes, R. A. and Spinrad, T. L. (2007). Prosocial Development. In Handbook of Child Psychology (eds W. Damon, R. M. Lerner and N. Eisenberg). doi:10.1002/9780470147658.chpsy0311
Lengua, L. J., Honorado, E., & Bush, N. R. (2007). Contextual risk and parenting as predictors of effortful control and social competence in preschool children. Journal of Applied Developmental Psychology, 28(1), 40–55. http://doi.org/10.1016/j.appdev.2006.10.001
von Salisch, M. (2001). Children’s emotional development: Challenges in their relationships to parents, peers, and friends. International Journal of Behavioral Development, 25(4), 310–319. https://doi.org/10.1080/01650250143000058
Zhou, Q. , Eisenberg, N. , Losoya, S. H., Fabes, R. A., Reiser, M. , Guthrie, I. K., Murphy, B. C., Cumberland, A. J. and Shepard, S. A. (2002), The Relations of Parental Warmth and Positive Expressiveness to Children's Empathy‐Related Responding and Social Functioning: A Longitudinal Study. Child Development, 73: 893-915. doi:10.1111/1467-8624.00446
day 5
Keynote presentation
Finally, on the last day of the LET Conference 2018, a keynote presentation is given by Prof. Sanna Järvelä (Head of LET department, University of Oulu). The tittle was: Multimodal data to understand students’ cognition, metacognition, motivation and emotions in a learning process (Järvelä, 2018).
In short, this presentation did not reflect a particular empirical study carried out by the presenter, but it made use of two studies to highlight the main message and findings.
The topic of this presentation is grounded in the theory of Regulated Learning, more specifically in Self-Regulated Learning (SRL). As the main issues, Järvelä (2018) wanted to highlight three ideas. Learning is active, is a process and is adaptive. Every self-regulated learner is aware of that and make use of strategies to succeed in this matter. However, Järvelä (2018) posited that learning is not always individual and other social forms of learning, like collaborative learning also require to be regulated. Also, it is essential to pay attention to the fact that regulation is not always visible, and thus researchers face methodological challenges.
In line with this last idea, the main aim of the current research on Järvelä’s field of study is to find methodological solutions that give a comprehensive and valid view of the regulatory processes in authentic collaborative learning situations, for example in CSCL and SSRL. For this issue, Järvelä (2018) presented the use of multimodal data as a solution. Multimodal data aims to gather information from different channels in order to triangulate and give a holistic and more objective view of the phenomena researched. In addition to this, multimodal data can help to capture the temporal and cyclical process of regulated learning, but the amount of data obtained can result extremely large and decontextualised. For that reason, it is still needed that researchers interpret the objective data to contextualise the situation.
Regarding the data collection methods, Järvelä (2018) gave examples of the different ways to obtain data. For example, mobile eye tracking systems, which have the small disadvantage of being intrusive. Secondly, sensors that measure physiological reactions, like EDA, BPM and temperature. In third place, Järvelä (2018) mentioned Log Data and the platform (EdX) used by Järvenoja (2018). Fourthly, online evaluation forms and retrospective dashboards and finally video data, like 360º video. However, despite of the benefit of using multiple channels for collecting data, at the end the vast amount of it results difficult to proceed and to understand the phenomena. For this reason, Järvelä (2018) called for multidisciplinary collaboration with experts in learning sciences and analytics, data-mining, signal processing and technology to sort this challenge.
In this fashion, multimodal data in regard to Regulated Learning, can depict at this moment patterns and temporal progress in strategy use (Malmberg, Järvenoja & Järvelä, 2013 in Järvelä, 2018) and patterns of regulated learning in CL (SRL, CoRL and SSRL) (Malmberg, Järvenoja & Järvelä, 2017 in Järvelä, 2018). Also, multimodal data gave insights of the criticality of regulation for collaborative learning progress and outcomes (Isohätälä et al., 2016; Järvelä et al., 2016 in Järvelä, 2018).
Finally, Järvelä, 2018 concluded her presentation stating the future implications for research. The need of understanding deeply learning processes call for the use of more sophisticated methods, like artificial intelligence and machine learning. Järvelä, 2018 claimed that the “traditional boundaries of learning” have to be broken in order to advance increasing the competence of students and research. For instance, adaptive support of technological artifacts and environments and more efficient processing methods and machines are some of the expectations from the use of AI and ML. In addition, creativity and smart thinking are also areas to place future research, according to Järvelä (2018).
I personally believe that Järvelä’s presentation is a good explanation of the future research on the field of Regulated Learning. Multimodal data brings new approaches to tackle the phenomena but also to put in practice what the researchers are investigating about. Now, researchers have to collaborate and jointly co-construct the meaning and support each other with their own expertise in their field. Also, I believe that this is what the learning sciences are advocating for, expertise in learning but also expertise collaboration to advance learning. Also, I would highlight as well from this Keynote the concept of triangulation. I think that Järvelä (2018) explained very well how and why triangulation of data is one of the musts of using multimodal data.
What I found especially interesting? Why?
Järvelä’s presentation has triggered my biggest research interest, empathy. Taking into account the concept of SRL, it is my belief that metacognitive knowledge of oneself emotions imply, at least, the awareness of other’s emotional state, which is a general definition of social sensitivity and empathy (Salovey & Grewal, 2005; Lawrence et al., 2004).
In this sense Multimodal data would definitely fit into this topic because the current gap in researching empathy relies in the fact that most of the methods used for measuring it, rely on self-interpretations. Empirical studies have reported empathy does not correlate with prosocial behavior if only measured with verbal indexes (Eisenberg & Miller, 1987; Eisenberg & Fabes, 1990) and also highlighted the need of using other methods like videotaped interaction (Angera & Long, 2006). In addition, with the development of more intricate and sophisticated technologies, physiological measurements have started to gain more space in the field of empathy, but this approach remains as less common (Gerdes et al., 2010). For this reason, multimodal data can lead the way for conducting research in the SRL field considering empathy as a metacognitive construct, at hand of self-regulated learners and required for successful socially-shared and co-regulation.
In addition, from this presentation I find especially interesting the field of artificial intelligence. From the past I have heard about this concept when referring to earlier studies like Azevedo’s MetaTutor (Azevedo, Witherspoon, Chauncey, Burkett & Fike, 2009). From their research, Azevedo, Moos, Joshnson & Chauncey (2010) describe MetaTutor as a hypermedia environment where an artificial agent scaffolds students in the use of SRL techniques. An agent is a computer system situated in an environment capable of autonomous action to meet the design objectives (Woorldrige & Jennings, 1995 in Miranda and Aldea, 2004 p.332). However, MetaTutor’s only focus is on cognitive and metacognitive SRL process for acquiring declarative knowledge. Azevedo et al. (2010) claim that some of the future implications of MetaTutor concern the agent-learner dialogue, where MetaTutor can prompt scaffolds based on the students’ linguistic, visual or spoken input. However, MetaTutor limitations make complicated to understand how this AI can be seen as promoter of SRL concerning cognition, behavior, motivation and emotions (Panadero, 2017). Martínez-Miranda and Aldea (2004) describe that since the creation of AI, one of the main objectives has been to build software systems that act and react to humans giving intelligent responses, also including emotions. However, these authors question the need of AI systems that include emotions, and judge that the current emotional artificial models are only valuable for the system or environment they were created for (Martínez-Miranda, 2004). Nevertheless, what if the environment is not as important as the fact that the agent is being a co-regulator and at the same time providing useful information about the subject?
It is my belief, that more comprehensive environments are needed if the aim is to tackle all the facets of Regulated Learning. Cognitive utterances can be traced from students’ disclosures, but emotions remain unseen unless new methodological approaches and technologies are used in combination of multimodal data.
References
Presentations
Järvelä, S. (2018, April). Multimodal data to understand students’ cognition, metacognition, motivation and emotions in a learning process. Keynote presented at the at the LET2018 Conference, Oulu. Retrieved from: https://letmaster.files.wordpress.com/2018/03/jc3a4rvelc3a4-let-24042018.pdf
Reflection part
Angera, J. J., & Long, E. C. J. (2006). Qualitative and Quantitative Evaluations of an Empathy Training Program for Couples in Marriage and Romantic Relationships. Journal of Couple & Relationship Therapy, 5(1), 1–26. https://doi.org/10.1300/J398v05n01_01
Azevedo, R., Moos, D. C., Johnson, A. M. Y. M., & Chauncey, A. D. (2010). Measuring Cognitive and Metacognitive Regulatory Processes During Hypermedia Learning: Issues and Challenges. Educational Psychologist, 45(4), 210–223. https://doi.org/10.1080/00461520.2010.515934
Azevedo, R., Witherspoon, A. M., Chauncey, A., Burkett, C., & Fike, A. (2009, October). MetaTutor: A MetaCognitive Tool for Enhancing Self-Regulated Learning. In AAAI Fall Symposium: Cognitive and Metacognitive Educational Systems.
Eisenberg, N., & Fabes, R. A. (1990). Empathy: Conceptualization, measurement, and relation to prosocial behavior. Motivation and Emotion, 14(2), 131–149. https://doi.org/10.1007/BF00991640
Eisenberg, N., & Miller, P. (1987). The Relation of Empathy to Prosocial and Related Behaviors. Psychological bulletin (Vol. 101). https://doi.org/10.1037/0033-2909.101.1.91
Gerdes, K. E., Segal, E. A., & Lietz, C. A. (2010). Conceptualising and measuring empathy. British Journal of Social Work, 40(7), 2326-2343.
Lawrence, E. J., Shaw, P., Baker, D., Baron-Cohen, S., & David, A. S. (2004). Measuring empathy: reliability and validity of the Empathy Quotient. Psychological medicine, 34(5), 911-920.
Martı́nez-Miranda, J., & Aldea, A. (2005). Emotions in human and artificial intelligence. Computers in Human Behavior, 21(2), 323–341. https://doi.org/https://doi.org/10.1016/j.chb.2004.02.010
Panadero E. (2017) A Review of Self-regulated Learning: Six Models and Four Directions for Research. Frontiers in Psychology.8:422. doi:10.3389/fpsyg.2017.00422.
Salovey, P., & Grewal, D. (2005). The science of emotional intelligence. Current Directions in Psychological Science. https://doi.org/10.1111/j.0963-7214.2005.00381.x