Skip to Main content Skip to Navigation
Theses

Joint agency in human-machine interactions : how to design more cooperative agents?

Abstract : System automation has steadily created a gap between the human operators and the loop of control (i.e., “out-of-the-loop” (OOTL) problem), disconnecting them from the machines’ actions and outcomes (Kaber, Onal, & Endsley, 2000). In this thesis, we aimed at investigating how to keep the human operators in the loop of control. We based our investigations on the theoretical framework of the science of Agency. Interestingly, it has been shown that during human-human interactions, individuals could exhibit a sense of agency for other-generated actions and outcomes (or sense of “we-agency”) while such ability was impaired for machine-generated actions and outcomes (Obhi & Hall, 2011b). The first stage of the thesis sought to finely examine the cognitive processes underlying individuals’ loss of agency during joint tasks with automated artificial systems, both at the behavioral (Experiment 1) and at the cerebral (Experiment 2) levels. The second stage of the thesis sought to investigate on which characteristics of the machine it was possible to act in order to regain the human operator’s sense of agency using a top-down approach (Experiment 3) and a bottom-up approach (Experiment 4). Implications of our findings are discussed in regard with the literature on the sense of agency and the operational OOTL issue.
Document type :
Theses
Complete list of metadatas

Cited literature [152 references]  Display  Hide  Download

https://tel.archives-ouvertes.fr/tel-02437237
Contributor : Abes Star :  Contact
Submitted on : Thursday, May 28, 2020 - 9:37:06 AM
Last modification on : Monday, November 16, 2020 - 2:43:29 PM

File

Sahai-2019-These.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-02437237, version 2

Collections

Citation

Aïsha Sahaï. Joint agency in human-machine interactions : how to design more cooperative agents?. Computer science. PSL Research University, 2019. English. ⟨NNT : 2019PSLEE025⟩. ⟨tel-02437237v2⟩

Share

Metrics

Record views

196

Files downloads

97