Skip to Main content Skip to Navigation
Theses

Recognized Cyber Picture et Confiance interpersonnelle : optimiser les activités collaboratives cyber militaires

Abstract : Since the early 2000s, civilian and military networks are faced with an increasing number of cyber-attacks. In order to facilitate decision making and to better coordinate actions in case of attacks coming from the cyberspace, it has become necessary for the French Air Force (armée de l’air, AA) to put in place a system or tool that would allow a shared representation of the cyber state of its infrastructures. Although this system (or tool), named Recognized Cyber Picture (RCP), is not yet entirely defined, it will probably be deployed in the command and control centres (C2) of the AA. The role of the RCP will be to identify and highlight potential problems in the information received by the C2 (modified, added or lost information). Therefore, operators of C2 equipped with RCP may be confronted with a dissonance in terms of information processing and transmission. To facilitate decision making, C2 operators would need to put in place mechanisms to “compensate” for this dissonance. Among these mechanisms, interpersonal trust has been identified in the literature as a decision making facilitator in highly uncertain environments. This thesis proposes to study the role of trust on the behaviors of operators present in C2 integrating this cyber dimension. In order to answer this (research) question, we developed and then tested an experimental protocol that allows to induce several levels of trust and to investigate their impact on behaviours. Based on this protocol, 4 experiments were carried out in the laboratory with potential operators of the RCP. The first experiment (conducted on 200 participants from the AA, including 160 operational and 40 students from the AA Academy) focused on the role of psychosocial characteristics of individuals, and their behaviours when working with unknown and more or less trustworthy collaborators. The second experiment (conducted with 40 participants from the AA Academy), investigated the differences in behaviour when activity relies on trust between two human operators or between a human operator and an automated system (an emulated artificial intelligence). The third experiment (conducted with 35 engineering students) explored the influence of self-confidence of an individual on his behaviour towards more or less trustworthy co-workers. These first three experiments highlight, in particular, behavioural changes induced by the level of trust within the relationship, which have been little studied to date. This includes supervision, quality of supervision, time management, but also the acceptance of the information transmitted by the co-worker. The last experiment (conducted with 40 engineering students) explored the influence of both the level of confidence in a relationship and the uncertainty in the data on operator decision making. This experiment shows that the level of coherence of available information and trust levels play an important role in the operator’s decisional processes. The results of these experiments have been confronted to models from the literature, and have allowed to bring into light new elements to adjust and adapt these models to the case of a C2 with a cyber component. This thesis makes a theoretical contribution to the understanding of trust mechanisms, and a practical contribution to support the implementation of the RCP in Command and Control centers.
Document type :
Theses
Complete list of metadatas

https://tel.archives-ouvertes.fr/tel-03097633
Contributor : Abes Star :  Contact
Submitted on : Tuesday, January 5, 2021 - 2:09:29 PM
Last modification on : Wednesday, January 6, 2021 - 3:28:04 AM

File

BOLLON_FLORENT_2020.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03097633, version 1

Citation

Florent Bollon. Recognized Cyber Picture et Confiance interpersonnelle : optimiser les activités collaboratives cyber militaires. Automatique. Université de Bordeaux, 2020. Français. ⟨NNT : 2020BORD0113⟩. ⟨tel-03097633⟩

Share

Metrics

Record views

51

Files downloads

48