The sensitive nature of feedback as an interactional phenomenon
Social actions such as advice and feedback are sensitive actions, often triggering emotional and defensive reactions. Moreover, just how to understand and describe such actions – for instance, as “constructive advice”, “complaints”, or even “insults” – is a potential point of contention and dispute for participants in interaction. Based on a unique corpus of video data from higher education (500h, from 12 different settings), the project aims to develop a conceptual and empirically grounded understanding of the social organization of feedback. We focus on how participants orient to the sensitive nature of feedback, and deal with situations where emotional reactions, resistance and defensiveness become focal. The diversity of the corpus also allows distinguishing generic from setting-specific phenomena and specify the relevance of contextual and situational characteristics, strengthening the potential for theoretical development. The project is a timely contribution not only empirically and theoretically, but also in its ambition to conduct analyses of a large corpus of existing video data. Encouraging secondary use of data is currently priority nationally and internationally. An important methodological contribution of the project is to develop workflows for secondary use of video data, potentially serving as a model for similar future initiatives.
Final report
The purpose of the project has been to develop a theoretically and empirically grounded understanding of the sensitive nature of feedback. The project has primarily focused on feedback in higher education and is based on a video corpus (500h) that includes critique in architectural education, supervision in essay writing, debriefing in simulation-based healthcare and maritime training, skill training in crafts and surgery, as well as feedback in restaurant kitchens and race car driver training. The diversity of the material has allowed for distinguishing between general and more specific phenomena and for specifying the relevance of contextual features of activities.
Key Results of the Project
1. Ambiguity of Critique and Feedback
In everyday contexts, the term "critique" is often used for actions that point out faults and shortcomings. Relationship columns in newspapers can be used as an illustrative example. One person writes that their partner interprets everything as criticism, while another feels oppressed by their partner's constant reminders of what has not been done or what should have been done. Such letters illustrate how the same action can be perceived differently by the two conversational partners. Advice on how something could be done differently suggests that the original was not good enough, while a reminder implies that something could be forgotten. Context is crucial for how the utterances are to be understood. Critique and remarks that are uncontroversial during a driving lesson become provocative when they come from a fellow passenger. During the driving lesson, the purpose and roles are clear – the driving instructor is there to educate – and comments, even those about mistakes, can be understood in terms of learning and teaching. It is not as clear what the fellow passenger's comments are aimed at.
The term "critique" also has a different meaning. The activities carried out at universities are largely based on a critical, scientific approach, where feedback typically is formulated as "constructive criticism." Nevertheless, there are many contexts where the boundary between the two meanings of critique is not sharp. From a teacher's perspective, feedback might be understood in terms of teaching, scientific approach, and critical thinking. However, it is not uncommon for students to respond to such feedback by apologizing or becoming defensive. In this way, students display an understanding of the teacher's contributions as complaints rather than constructive advice. The focus shifts from what could be done or what lessons could be drawn to what has been done, why, and, in some cases, who is to blame for the mistake. This tension has been examined from the perspectives of conversation analysis and ethnomethodology. Instead of explaining the ambiguity in critical feedback with social-psychological models and theories, the focus has been on describing how participants orient themselves towards statements as ambiguous and how they negotiate how critical comments should be produced and understood.
2. Sequential Organization of Feedback
A starting point for conversation analysis is the sequential organization of human interaction. In talk-in-interaction, sequences are typically organized as adjacency pairs, which is a unit consisting of two actions performed by different speakers, where the second action responds to the first: for instance, question/answer and greeting/greeting. A characteristic feature of these sequences is that the second turn displays an understanding of the first – by producing an answer, the second speaker shows that the prior utterance was heard as a question. Feedback does not consist of a specific type of action: it can be based on assessments, suggestions, questions, or various forms of directives. The project has specifically examined sequences where students respond to the feedback they receive by apologizing or defending themselves. In these cases, students show that they primarily understand the feedback as a summative assessment, or even complaint, rather than as instruction and formative assessment. The project has also examined how teachers and instructors produce critical feedback in various ways so that it is understood as constructive.
Another topic has been the differences between feedback given after an activity has been completed (e.g., a "debriefing") and feedback on an ongoing action (e.g., instruction during surgical training). Comparatively, criticism in the latter case is rarely perceived as a complaint regardless of how sharply the instructions are formulated. This can partly be explained by the sequences having a different organization. Instead of being based on two spoken statements (e.g., advice/acceptance), such interaction has a three-part organization, where an instruction is followed by an attempt to follow the instruction, which in turn is followed by an assessment of what has been done in relation to what has been instructed. Correction and assessment are thus part of the organization itself, and there is not the same space left for ambiguity as in the former case.
3. Feedback as Education
In instructional settings that focus on highlighting and discussing students' mistakes, there is an inherent tension. Laughter and other framing techniques thereby become a resource for balancing pedagogical seriousness with a tolerance for individual mistakes. The project has also shown how students' and teachers' ways of providing feedback differ in such a way that students are more indirect and teachers more direct. The analyses demonstrate how participants' ways of handling the sensitivity of feedback through various types of explanations and excuses become expressions of their understanding of professional practice: for instance, criticism regarding deficiencies in communication in a simulated emergency care situation is toned down by referring to a reluctance to "disturb," which reveals an understanding of social actions in the medical context based on everyday logic. The example illustrates how participants' ways of handling the sensitive nature of feedback are closely intertwined with how they deal with the intended learning content – such as different forms of professional communication.
The project has also addressed learning and interaction in academic writing, highlighting the importance of direct interaction around texts in developing students' writing and their understanding of academic literacy. In direct interaction, a shared understanding of feedback can be developed in a different way than through written comments. At the same time, the supervision situation can be emotionally charged for students. This theme is further developed in a study of how students and teachers use and perceive generative AI to receive feedback on text, compared to feedback from a supervisor. The study highlights how the human encounter in the feedback situation brings with it unique opportunities to convey central knowledge content but is also characterized by a different sensitivity than computer-generated feedback.
New Research Questions Generated by the Project
The project has been highly generative in terms of new research questions. The focus of the project has been on the sensitive nature of feedback, but the analyses have also contributed to an increased understanding of the interactive organization of instructions more generally. There is much research on how conversations are organized, including conversations in classroom contexts. However, it turns out that instruction in bodily or craft skills has an organization that differs from that described in the research literature. Consequently, it becomes relevant to ask how such interaction is organized and how it differs in fundamental ways from ordinary conversations. The project's focus on emotions and feelings has also raised questions about interpersonal interaction in a time where we interact with generative AI in various ways. Students interviewed in the project report that they do not experience feedback from such technology as emotionally charged. A central question in relation to this is whether the very meaning of feedback is thereby changed.
International Anchoring of the Project
The project has a clear international anchoring through the participation of Professor Lorenza Mondada (Basel University). The project started during the pandemic, which initially made international collaboration difficult. The collaboration has partly taken place remotely and also been postponed to the later parts of the project. During the fall of 2022, extensive collaboration took place in Tokyo and Gothenburg. Results from the project have been presented at international workshops, invited presentations, and conferences. The project has also organized workshops in Tokyo and Gothenburg. This has strengthened and developed an international network consisting of leading researchers in ethnomethodology, conversation analysis, and multimodal interaction analysis. Some of the outcomes of this international collaboration have been prepared for and published in edited volumes.
Publications and Publication Strategy
The project has worked on several studies simultaneously. One reason for this is the comparative approach where we have worked with material from different contexts. In most cases, the individual studies have been presented at international workshops and conferences before being submitted to international journals or edited volumes. Due to delays in the publication process, this setup has meant that the studies have either been published recently or – in cases where only presentations are listed below – are being prepared for publication during the year (under review, revision, etc.). We expect at least eight book or article publications from the project before the end of 2024. All texts are published open access, either through library agreements or, in the case of chapters published in edited volumes, by paying open access fee.
Key Results of the Project
1. Ambiguity of Critique and Feedback
In everyday contexts, the term "critique" is often used for actions that point out faults and shortcomings. Relationship columns in newspapers can be used as an illustrative example. One person writes that their partner interprets everything as criticism, while another feels oppressed by their partner's constant reminders of what has not been done or what should have been done. Such letters illustrate how the same action can be perceived differently by the two conversational partners. Advice on how something could be done differently suggests that the original was not good enough, while a reminder implies that something could be forgotten. Context is crucial for how the utterances are to be understood. Critique and remarks that are uncontroversial during a driving lesson become provocative when they come from a fellow passenger. During the driving lesson, the purpose and roles are clear – the driving instructor is there to educate – and comments, even those about mistakes, can be understood in terms of learning and teaching. It is not as clear what the fellow passenger's comments are aimed at.
The term "critique" also has a different meaning. The activities carried out at universities are largely based on a critical, scientific approach, where feedback typically is formulated as "constructive criticism." Nevertheless, there are many contexts where the boundary between the two meanings of critique is not sharp. From a teacher's perspective, feedback might be understood in terms of teaching, scientific approach, and critical thinking. However, it is not uncommon for students to respond to such feedback by apologizing or becoming defensive. In this way, students display an understanding of the teacher's contributions as complaints rather than constructive advice. The focus shifts from what could be done or what lessons could be drawn to what has been done, why, and, in some cases, who is to blame for the mistake. This tension has been examined from the perspectives of conversation analysis and ethnomethodology. Instead of explaining the ambiguity in critical feedback with social-psychological models and theories, the focus has been on describing how participants orient themselves towards statements as ambiguous and how they negotiate how critical comments should be produced and understood.
2. Sequential Organization of Feedback
A starting point for conversation analysis is the sequential organization of human interaction. In talk-in-interaction, sequences are typically organized as adjacency pairs, which is a unit consisting of two actions performed by different speakers, where the second action responds to the first: for instance, question/answer and greeting/greeting. A characteristic feature of these sequences is that the second turn displays an understanding of the first – by producing an answer, the second speaker shows that the prior utterance was heard as a question. Feedback does not consist of a specific type of action: it can be based on assessments, suggestions, questions, or various forms of directives. The project has specifically examined sequences where students respond to the feedback they receive by apologizing or defending themselves. In these cases, students show that they primarily understand the feedback as a summative assessment, or even complaint, rather than as instruction and formative assessment. The project has also examined how teachers and instructors produce critical feedback in various ways so that it is understood as constructive.
Another topic has been the differences between feedback given after an activity has been completed (e.g., a "debriefing") and feedback on an ongoing action (e.g., instruction during surgical training). Comparatively, criticism in the latter case is rarely perceived as a complaint regardless of how sharply the instructions are formulated. This can partly be explained by the sequences having a different organization. Instead of being based on two spoken statements (e.g., advice/acceptance), such interaction has a three-part organization, where an instruction is followed by an attempt to follow the instruction, which in turn is followed by an assessment of what has been done in relation to what has been instructed. Correction and assessment are thus part of the organization itself, and there is not the same space left for ambiguity as in the former case.
3. Feedback as Education
In instructional settings that focus on highlighting and discussing students' mistakes, there is an inherent tension. Laughter and other framing techniques thereby become a resource for balancing pedagogical seriousness with a tolerance for individual mistakes. The project has also shown how students' and teachers' ways of providing feedback differ in such a way that students are more indirect and teachers more direct. The analyses demonstrate how participants' ways of handling the sensitivity of feedback through various types of explanations and excuses become expressions of their understanding of professional practice: for instance, criticism regarding deficiencies in communication in a simulated emergency care situation is toned down by referring to a reluctance to "disturb," which reveals an understanding of social actions in the medical context based on everyday logic. The example illustrates how participants' ways of handling the sensitive nature of feedback are closely intertwined with how they deal with the intended learning content – such as different forms of professional communication.
The project has also addressed learning and interaction in academic writing, highlighting the importance of direct interaction around texts in developing students' writing and their understanding of academic literacy. In direct interaction, a shared understanding of feedback can be developed in a different way than through written comments. At the same time, the supervision situation can be emotionally charged for students. This theme is further developed in a study of how students and teachers use and perceive generative AI to receive feedback on text, compared to feedback from a supervisor. The study highlights how the human encounter in the feedback situation brings with it unique opportunities to convey central knowledge content but is also characterized by a different sensitivity than computer-generated feedback.
New Research Questions Generated by the Project
The project has been highly generative in terms of new research questions. The focus of the project has been on the sensitive nature of feedback, but the analyses have also contributed to an increased understanding of the interactive organization of instructions more generally. There is much research on how conversations are organized, including conversations in classroom contexts. However, it turns out that instruction in bodily or craft skills has an organization that differs from that described in the research literature. Consequently, it becomes relevant to ask how such interaction is organized and how it differs in fundamental ways from ordinary conversations. The project's focus on emotions and feelings has also raised questions about interpersonal interaction in a time where we interact with generative AI in various ways. Students interviewed in the project report that they do not experience feedback from such technology as emotionally charged. A central question in relation to this is whether the very meaning of feedback is thereby changed.
International Anchoring of the Project
The project has a clear international anchoring through the participation of Professor Lorenza Mondada (Basel University). The project started during the pandemic, which initially made international collaboration difficult. The collaboration has partly taken place remotely and also been postponed to the later parts of the project. During the fall of 2022, extensive collaboration took place in Tokyo and Gothenburg. Results from the project have been presented at international workshops, invited presentations, and conferences. The project has also organized workshops in Tokyo and Gothenburg. This has strengthened and developed an international network consisting of leading researchers in ethnomethodology, conversation analysis, and multimodal interaction analysis. Some of the outcomes of this international collaboration have been prepared for and published in edited volumes.
Publications and Publication Strategy
The project has worked on several studies simultaneously. One reason for this is the comparative approach where we have worked with material from different contexts. In most cases, the individual studies have been presented at international workshops and conferences before being submitted to international journals or edited volumes. Due to delays in the publication process, this setup has meant that the studies have either been published recently or – in cases where only presentations are listed below – are being prepared for publication during the year (under review, revision, etc.). We expect at least eight book or article publications from the project before the end of 2024. All texts are published open access, either through library agreements or, in the case of chapters published in edited volumes, by paying open access fee.