Resonating Minds—Emergent Collaboration Through Hierarchical Active Inference

Working together on complex collaborative tasks requires agents to coordinate their actions. Doing this explicitly or completely prior to the actual interaction is not always possible nor sufficient. Agents also need to continuously understand the current actions of others and quickly adapt their own behavior appropriately. Here we investigate how efficient, automatic coordination processes at the level of mental states (intentions, goals), which we call belief resonance, can lead to collaborative situated problem-solving. We present a model of hierarchical active inference for collaborative agents (HAICA). It combines efficient Bayesian Theory of Mind processes with a perception–action system based on predictive processing and active inference. Belief resonance is realized by letting the inferred mental states of one agent influence another agent’s predictive beliefs about its own goals and intentions. This way, the inferred mental states influence the agent’s own task behavior without explicit collaborative reasoning. We implement and evaluate this model in the Overcooked domain, in which two agents with varying degrees of belief resonance team up to fulfill meal orders. Our results demonstrate that agents based on HAICA achieve a team performance comparable to recent state-of-the-art approaches, while incurring much lower computational costs. We also show that belief resonance is especially beneficial in settings where the agents have asymmetric knowledge about the environment. The results indicate that belief resonance and active inference allow for quick and efficient agent coordination and thus can serve as a building block for collaborative cognitive agents.

Multimedia Annotations for Practical Collaborative Reasoning

University education requires students to be trained both at university and at external internship centres. Because of Covid-19, the availability of multimedia resources and examples of practical contexts has become vital. Multimedia annotation can help students reflect on the professional world, collaborating and interacting with colleagues online. This study aims to encourage collaborative practical thinking by using new video annotation technologies. 274 students participated in an experiment of task design focusing on the analysis of a technology-based, award-winning educational innovation project. With mixed research design, qualitative and quantitative data exported from the video annotation platform used was collected and analysed. The results show differences in the quality and quantity of the answers: in the tasks with broad Folksonomy they are more numerous but more dispersed in their analysis, and vice versa. The quality of the answers given with narrow Folksonomy is also higher in both texts and videos modes. Producing multimedia annotations is a practical way to encourage students to practise reflective reasoning about the professional reality.

Embodied Displays of “Doing Thinking.” Epistemic and Interactive Functions of Thinking Displays in Children's Argumentative Activities

This study investigates moments in which one participant in an interaction embodies that he is “doing thinking,” a display that is commonly referred to as “thinking face. ” From an interactional perspective, it is assumed that embodied displays of “doing thinking” are a recurring social practice and serve interactive functions. While previous studies have examined thinking faces primarily in word searches and storytelling, the present study focuses on argumentative activities, in which children engage in processes of joint decision-making. The paper has two interrelated aims. The first aim is to describe how multiple modalities—beyond the face—are temporally coordinated to create multimodal gestalts of “doing thinking.” It is shown that thinking displays not only involve dynamic imaginative gaze but also stylized bodily postures. The second aim is to generate knowledge about the functions of thinking displays in children's argumentative activities. The analysis describes how both speakers and recipients use thinking displays in different turn positions and align them with verbal talk or silence. The data for this study comprise video recordings of decision-making processes in groups of older children. Drawing on a multimodal approach to situated interaction, it will be proposed that embodied displays of “doing thinking” provide a resource to shape participation frameworks, mark epistemic stances and create epistemic ecologies for collaborative reasoning. By investigating thinking displays in a particular conversational activity, the study sheds light on the diversity and context-sensitive functionality of thinking displays. It also contributes to recent research on children's collaborative reasoning as an embodied discursive practice.

Collaborative reasoning in the context of group competition

A key skill in collaborative problem-solving is to communicate and evaluate reasons for proposals to arrive at the decision benefiting all group members. Although it is well-documented that collaborative contexts facilitate young children’s reasoning, less is known about whether competition with other groups contributes to children’s collaborative reasoning. We investigated whether between-group competition facilitates children’s within-group collaborative reasoning, regarding their production of reasons and their use of transacts, communicative acts that operate on one another’s proposals and reasoning. We presented 5- and 7-year-old peer dyads with two collaborative problem-solving tasks (decorating a zoo and a dollhouse). In one task, children competed against another group (the competitive condition); whereas in the other task, they did not (non-competitive condition). Our results suggest that children’s sensitivity to group competition as reflected in their reasoning changed depending on the task. When they decorated a house, they produced more transacts in the competitive condition than in the non-competitive condition; whereas when they decorated a zoo, this pattern was reversed. Thus, our results highlight that group competition did not influence children’s collaborative reasoning consistently across different contexts.

Kemampuan Bertanya Siswa dalam Kegiatan Diskusi Kelompok pada Materi Rasio Trigonometri

<p>Small group discussions (group work) are an important part of the learning process though students are rarely skilled at asking a question. The purpose of this study is to examine student interactions and questioning skills during group work. This research study uses a qualitative descriptive approach. The research subjects consist of eight students who were divided heterogeneously into two groups, with four members for each group. The research instrument consists of the researcher herself, a task in the form of a trigonometric ratio group worksheet, and three video cameras to observe discussion activities. One camera was focused on all class activities and two cameras were pointing at each group being observed. All conversation transcripts during the discussion are deciphered, coded, and then analyzed qualitatively. The results of this study show the interaction of conveying opinion/rebuttals, asking questions, and providing answers, with the percentage of giving opinions/ objections were more dominant than other interactions. The ability to ask questions was obtained by 50% of the students having the medium questioning ability and 50% having the low questioning ability, with the level of questions asked at the LOTS level, namely C1 and C3 levels. None of the students had high questioning skills. Of the two groups observed, group A was more active in interacting both in terms of exchanging opinions/rebuttals, asking questions, and providing answers. Suggestions for further research need to be carried out an in-depth analysis of discussion activities both in terms of asking questions or providing feedback to see the emergence of collaborative reasoning.</p>

Improving Analytic Reasoning via Crowdsourcing and Structured Analytic Techniques

How might analytic reasoning in intelligence reports be substantially improved? One conjecture is that this can be achieved through a combination of crowdsourcing and structured analytic techniques (SATs). To explore this conjecture, we developed a new crowdsourcing platform supporting groups in collaborative reasoning and intelligence report drafting using a novel SAT we call “Contending Analyses.” In this paper we present findings from a large study designed to assess whether groups of professional analysts working on the platform produce better-reasoned reports than those analysts produce when using methods and tools normally used in their organizations. Secondary questions were whether professional analysts working on the platform produce better reasoning than the general public working on the platform; and how usable the platform is. Our main finding is a large effect size (Cohen’s d = 1.37) in favor of working on platform. This provides early support for the general conjecture. We discuss limitations of our study, implications for intelligence organizations, and future directions for the work as a whole.