Thèse la Génétique de l'Interface Vers une Méthodologie pour l'Innovation dans la Conception des Interactions H/F - Doctorat.Gouv.Fr

Établissement : Ecole Nationale de l'Aviation Civile École doctorale : SYSTEMES Laboratoire de recherche : ENAC-LAB - Laboratoire de Recherche ENAC Direction de la thèse : Catherine LETONDAL ORCID 0000000349355008 Début de la thèse : 2027-10-01 Date limite de candidature : 2026-11-23T23:59:59 Dans cette thèse, vous analyserez comment l'innovation se produit dans la conception des interfaces. L'objectif est de proposer une nouvelle méthodologie fondée sur la compréhension des schèmes de fonctionnement des techniques, vecteurs de l'innovation et conduisant à des avancées telles que les interactions tactiles efficaces. Alors que les approches actuelles ont tendance à adopter une vision statique, fondée sur l'état de l'art ou les besoins des utilisateurs, cette thèse adoptera une approche génétique qui retraçera les processus internes d'évolution au sein même des techniques d'interaction. Cette approche s'inspire des travaux du philosophe Gilbert Simondon, qui montre que les objets techniques sont caractérisés par des modes de fonctionnement et des contraintes matérielles qui induisent le processus d'invention. Ce type d'analyse génétique offre un meilleur moyen de comprendre le large éventail de facteurs qui stimulent l'innovation technique, et peut aider les concepteurs à identifier des pistes inexplorées de manière plus systématique et plus efficace. Much of the research into human-computer interaction focuses on the development of new interaction techniques, as evidenced by papers presented at conferences such as UIST or CHI, or by reference works such as (Myers 2024). An interaction technique is a method which, by combining input modalities (gesture, speech, etc.) and output modalities (2D or 3D graphics, sound, etc.), enables basic interaction tasks to be performed, such as selection, data entry, activation, scrolling, or the specification of arguments and properties. For example, scrolling through a video can be achieved through direct manipulation-not via a scroll bar, but by grabbing and moving an object that is moving within the video (Dragicevic et al 2008). The simultaneous specification of arguments in a drawing tool can be achieved with a single gesture spanning several palettes (CrossY, Apitz et al 2004). Selecting a target by pointing at it within a set of shapes of different sizes can be achieved more efficiently by dynamically resizing the selection cursor (Bubble Cursor, Grossman et al 2005). Selection within an area can be carried out using both hands by sizing and selecting within the area (WritLarge, Xia et al 2017). A final, well-known example is the input of geometric shapes, which can be achieved through a combination of voice commands and pointing gestures (Bolt 1980).
Reflection on the design of interaction techniques is driven both by a desire to explore (new paradigms, new systems, etc.) and by the need to resolve usability issues or meet new requirements. However, the solutions explored also involve a dimension that depends on the internal possibilities of interaction techniques: it is therefore partly technical schemata-which serve an internal, rather than an external, functional purpose-that will drive the exploration of one design approach or another, such as being based on a click-through or on a look-through mechanism in the see-through tools lineages (Bier et al 1994, Letondal et al 2026). According to Gilbert Simondon (1958), understanding the genesis of a technical object requires an analysis of internal schemata, as opposed to an analysis of functions, with these schemata belonging to technical lineages or trends. This has also been studied by the anthropologist André Leroi-Gourhan (1943, 1945). From this perspective, for example, the internal combustion engine and the spring-driven engine do not belong to the same lineage (even though they share the same function); rather, it is the spring-driven engine and the bow that share the same schema.
These developments generally result in what Simondon describes as concretisation', or functional convergence, involving the integration or exaptation of functions made possible by technical schemata. Examples of such developments can be found in nature (such as the bird's feather, which was initially adapted for protection against the cold and, through exaptation, proved highly effective for flight) or in the artificial world (such as the printing press, which was exapted from wine presses, or the World Wide Web, which arose from a convergence between hypertext and the Internet). Such a genetic analysis of digital objects began in the field of programming and software, with, for example, object-oriented programming (Kurtov 2016), or website programming (Ferrarato 2019), or other examples of analyses (Hui 2016, Duhem 2013), but to date no analysis of interaction techniques as technical objects operating according to schemata has been proposed.
Another particularly relevant aspect of G. Simondon's philosophy is the consideration of two poles of the technical schema: an objective' pole, with its technical dimension, and a subjective' pole, linked to the cognitive dimension of technical manipulation and invention. Indeed, the technical schema is perceived cognitively and can inform both technical understanding and exploration, which are capable of incorporating and manipulating operational technical schemas. This analysis can usefully be compared with the theory of technical reasoning (Osiurak 2014, Renom 2022, 2023), which describes, in particular, the cognitive schemata underpinning the acquisition of tools.
Interaction techniques form the technical core of the field of HCI: conferences, books and course modules are devoted to them. As illustrated by the Museum of Interaction (https://hci-museum.lisn.upsaclay.fr), they are the defining feature of this field of innovation and invention. How, then, should we approach interaction techniques as genuine technical objects rather than merely as sequences of reciprocal actions occurring between a human and a computer? The paradigm of instrumental interaction proposed by Michel Beaudouin-Lafon (2000) offers a relevant model for such an understanding, by analysing interaction techniques as instruments, or tools, which mediate the relationship between the user and the computer system.
Finally, research in HCI focusing on innovation in interaction techniques has long included what are known as design spaces' (e.g. Card and Mackinlay 1990) or characterisation spaces' (Baglioni et al 2009). The aim of these design spaces' is to characterise, from a technical and functional perspective, the properties of interaction techniques and the dimensions of the problems involved. However, none of these approaches offers a study of the genesis of interaction techniques; the taxonomic approach by Fleck et al (2018) on tangible interfaces is the one that comes closest to this. Furthermore, these studies do not establish a link between the dimensions of a design space and the methods of invention and creativity employed. Yet, to be comprehensive, adopting and adapting a genetic analysis approach to interaction techniques must include the dimension of invention (Inie & Daalsgard 2017), as it is through invention that a technical object and a technical lineage take concrete form (Simondon 1958).
This thesis sets out to analyse the patterns that drive technical invention within the field of human-computer interaction (HCI). This will guide the design of a new methodology based on a better understanding of HCI's technical objects-namely, interaction techniques-and of the concrete processes of exploration. The proposed approach consists of describing their technical genesis: studying their evolution through the stages that lead to improved functioning, both internal - in terms of structure and reusability - and external - in terms of efficiency and adaptation to needs. This will be approached with an emphasis on technical objects not as factual 'givens' - finished and inevitable facts of the world - but as the results of processes in which we can intervene: as a part of an ongoing cycle of human design. Such a genetic analysis also enables, through a better understanding of the factors driving evolution, the discovery of unexplored avenues. Thus, whilst HCI generally focuses primarily on usage, this new angle of analysis allows for a better understanding of interaction techniques as objects that are also subject to internal constraints, thereby fostering the development of more appropriate design and creative methods.
The scientific issues addressed in this thesis are of a theoretical nature, focusing primarily on the relationship, in the genesis of technical objects, between the constraints imposed by technical elements and the mechanisms of invention. Clarifying this issue will then enable the development of methodological principles for creativity and design, that is to say, for the exploration and development of new techniques.
The candidate will be required to explore the methodologies of classification, design spaces and cladistics (Fleck et al 2018) as applied to technical objects, and potentially propose adaptations to account for the specific characteristics of interactive technical objects. The Museum of Interaction (Mackay 2015) is a catalogue that offers an initial classification of these technologies, as does the cladistic classification of tangible interactive objects proposed by Fleck et al (2018). Furthermore, descriptions of the patterns involved in the evolution of interaction techniques are, to some extent, present in the scientific literature on HCI, particularly in the state of the art' sections, which serve to situate the published interaction technique within evolutionary lineages. An initial study of these lineages could therefore draw inspiration from an analysis of these sections. Technical frameworks are also present in course materials and in the diagrams produced by engineers and researchers.
At the same time, an analysis of interaction techniques and existing models will need to be carried out. The model of instrumental interaction (Beaudouin-Lafon 2000), which makes it possible to specify an instrument's degrees of indirectness, integration and compatibility, provides a formalisation of what concretisation' would entail for an interactor. This is a static measure, but one could envisage using it to measure the degree of concretisation of an object evolving within a technical lineage. This model is an example of a schema, but there are likely to be others whose genesis would demonstrate concretisation, such as liaison' (Magnaudet et al 2018), a schema that unifies several forms of interaction. The IDID (Investigation / Diagrammatisation / Inverse invention / Data Learning) method for managing inventive knowledge, developed at the CEA by Larsim, offers several tools for evaluating performance throughout the evolution of technical lineages, notably the performance matrix' (Bontems 2016), which can be used to formalise the progress of a schema across multiple dimensions.
Software design patterns, and particularly their implementations in libraries (Ferrarato 2019), would also be a good starting point for such an analysis, notably because they describe established structures-refined over time-of software components, including their relationships, dependencies, synergies and multifunctionality (Gamma et al 1995, Borchers 2000a). This is also the case with Zimmermann's framing constructs', which are built from recurring patterns and function as specific perspectives that designers can adopt when applying a theory (such as that of product attachment) (Zimmermann 2009). For programmers, this realisation corresponds to what is sometimes described as elegance (Feenberg 2016).
Furthermore, in order to understand the scope of creativity and invention, the study will need to incorporate field-based analyses, including interviews with researchers in the field of interaction techniques, where appropriate, conducted in accordance with the IDID method (Bontems 2016) to enable the diagrammatisation of technical lineages or phases of innovative design, supplemented by the organisation of mini-hackathon-style workshops to characterise the questions raised by participating designers or programmers.
Finally, the exploration of the technical patterns employed in interaction techniques will involve a significant amount of programming, both to analyse the impact of implementing these patterns on the invention process and to implement the technical patterns through coding.

Le candidat idéal sera titulaire d'un master en IHM ou en informatique et manifestera un vif intérêt pour la conception d'interfaces et l'interaction homme-machine.
Il est important de s'intéresser aux questions théoriques, par exemple à la philosophie de la technologie ou aux théories de l'invention.
Des compétences en programmation sont requises pour pouvoir explorer et mettre au point des mécanismes techniques dans le domaine des techniques d'interaction.

Application link : https://edd-projets.utoulouse.fr/