This paper proposes a theoretical framework for understanding the dynamics of the tutoring relationship in a vocational education and training (VET) context. The model posits that the tutoring relationship is a matter of coordination involving the agency of both the mentee and the tutor (Filliettaz et al., 2024). The tutoring relationship is often characterised as asymmetrical in terms of knowledge and status (Wisker et al. 2013). In this perspective, an expert professional in a working context establishes a formal learning path or offers opportunistic learning situations for a novice, and organises guidance according to this choice. The extant literature on this relationship has focused on environmental conditions and methods offered by the tutor to the mentee, with the aim of determining whether these support or hinder the tutoree engagement, agency and learning in the workplace (see, for example, Minokken et al., 2017). The majority of studies to date have focused more on the tutor's activity, than on the mentee’s. The present study aims to address the aforementioned question by focusing on the activities of both parties in the tutoring relationship, and according to a social interaction theoretical perspective.

From a cognitive perspective, agency is defined as one of the most focal and pervasive beliefs that enable individuals to act in pursuit of desired outcomes, to influence the course of events, and to shape their own lives. Conversely, from a social interaction perspective, the concept of agency is understood to be more dynamic and context-dependent. Rather than considering agency as a stable attribute assigned to individuals, this perspective asserts that it is heavily shaped by intersubjective mechanisms and becomes visible through interactions individuals have with others in the circumstances of their social encounters (Goffman, 1956). AlZidjaly (2009) defines agentivity as an embedded and negotiated moment after moment through interactions. This concept is defined as "co-constructed, mediated, and continually negotiated in the collectivity," implying that all participants engaged in a joint action make adjustments or "attunement" (Filliettaz et al., 2024) Agency can also be regarded as an indexical and reciprocal process, whereby individuals recognise actors and their identities through the interpretation of local salient action (Parish & Hall, 2020).

According to this perspective, agency can be exercised in social interactions in different ways, and this paper will explore them. The following research questions will guide our analysis:

(I) Which actions do people accomplish through interactions in order to make themselves understandable? How are these actions understood by other participants, and what kind of response do they provide? (Drew 2013, Levinson 2013)

(II) How participants select semiotic resources (speech, prosody, gestures, body postures, material objects, etc.) available in the environment to accomplish a meaning-making process through interactions? How these resources are designed, used and combined reveal specific choices made by co-participants? (Kress et al., 2001)

(III) Who is knowledgeable about a specific topic and how are the other participants positioned in relation to this topic? Which epistemic stance arises and which epistemic status is endorsed and negotiated? (Heritage, 2012)

(IV) Considering the socially organised nature of the production of visible forms of the world through the direction of the gaze and constant communication, how is professional vision transmitted from an architect to a novice? (Goodwin, 2000)

To explore those questions, we have chosen to focus on an architect's office with nine employees and one trainee. In order to explore the aforementioned questions, the focus is directed towards a specific VET context, namely an architect's office comprising nine employees and one trainee. The trainee has previously taken a semester off in order to undertake an internship for a period of six months. She has also gained familiarity with computer programs designed for the purpose of creating three-dimensional images. Two weeks of her work were recorded during this period, and she was accompanied by three different architects.The data analysed here concerns one activity she accomplished with the guidance of an architect. The objective of this activity was to identify the points of view of a building on a PC screen using 3D software. Through this activity, participants were also able to identify and resolve a series of practical issues they had encountered. The activity alternated between individual work and collaboration with the architect. Ten moments of interaction were selected and analysed. The focus here is on three moments of interaction in which the architect and the mentee must adjust some curb lines on the picture. The proposal of an interactional analysis (Filliettaz et al. 2022) is intended to provide a response to the research questions posed.

The following discussion will emphasise the agentic and distributed nature of the tutorial relationship in this particular context. More specifically, the discussion will address the following points: - the complexity of the tutorial relationship as a collective achievement around a collective activity requiring coordination; - the mentee's active role in supporting the architect's activity and obtaining the information needed to carry out her own activity; and - the active role of the tutor in keeping his "global vision" of the activity and reformulating to make himself understandable.

Al Zidjaly, Najma (2009) Agency as an interactive achievement. Language in Society 38 (2): 177–200. Drew, Paul (2013) Conversation analysis as social action. Journal of Foreign Languages 37 (3): 2–20. Filliettaz, L., Zogmal, M., Garcia, S., Bimonte, A., Billett, S., & Flückiger, B. (2024). Negotiating parental agency in daily encounters with educators. Journal of Applied Linguistics and Professional Practice, 18(1), 66-103. Goffman, Erving (1956) The Presentation of Self in Everyday Life. Edinburgh: University of Edinburgh. Goodwin, C. (2000). Practices of seeing: Visual analysis: An ethnomethodological approach. Handbook of visual analysis, 157-182. Kress, Gunther, Carey Jewitt, Jon Ogborn and Charalampos Tsatsarelis (2001) Multimodal Teaching and Learning: The Rhetorics of the Science Classroom. London: Continuum. Heritage, John (2012) Epistemics in action: Action formation and territories of knowledge. Research on Language and Social Interaction 45: 1–29. Levinson, Stephen (2013) Action formation and ascription. In Jack Sidnell and Tanya Stivers (eds) Handbook of Conversation Analysis, 103–130. New York: Wiley. Mikkonen, S., Pylväs, L., Rintala, H., Nokelainen, P., & Postareff, L. (2017). Guiding workplace learning in vocational education and training: A literature review. Empirical Research in Vocational Education and Training, 9, 1-22. Parish, Ayden and Kira Hall (2020) Agency. In James Stanlaw (ed.) The International Encyclopedia of Linguistic Anthropology, vol. 1. Hoboken, NJ: Wiley Blackwell. Online. Wisker G, Exley K, Antoniou M, Ridley P (2013) Working one-to-one with students: supervising, coaching, mentoring, and personal tutoring. Routledge, New York