Henrietta Lyons

Henrietta Lyons

Auto-generated profile, based on publicly available information

Henrietta Lyons

School of Computing and Information Systems, University of Melbourne, Melbourne, Australia

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 December 2023 in ACM Transactions on Interactive Intelligent Systems
Verified
Journal Article
Open access
RSRonal Singh
Ronal Singh
TMTim Miller
Tim Miller
HLHenrietta Lyons
Henrietta Lyons
LSLiz Sonenberg
Liz Sonenberg
EVEduardo Velloso
Eduardo Velloso
PDPaul Dourish
Paul Dourish
In this article, we show that explanations of decisions made by machine learning systems can be improved by not only explainingwhya decision was made but also explaininghowan individual could obtain their desired outcome. We formally define the concept ofdirective explanations(those that offer specific actions an individual could take to achieve their desired outcome), introduce two forms of directive explanations (directive-specific and directive-generic), and describe how these can be generated computationally. We investigate people’s preference for and perception toward directive explanations through two online studies, one quantitative and the other qualitative, each covering two domains (the credit scoring domain and the employee satisfaction domain). We find a significant preference for both forms of directive explanations compared to non-directive counterfactual explanations. However, we also find that preferences are affected by many aspects, including individual preferences and social factors. We conclude that deciding what type of explanation to provide requires information about the recipients and other contextual information. This reinforces the need for a human-centered and context-specific approach to explainable AI.

12 June 2023
Verified
Conference Paper
Open access
HLHenrietta Lyons
Henrietta Lyons
TMTim Miller
Tim Miller
EVEduardo Velloso
Eduardo Velloso
In this paper, we explore why decision subjects generally express a preference for human reviewers of algorithmic decisions over algorithmic reviewers. We theorise that decision subjects desire control over the decision-making process in order to increase their chance of receiving a favourable outcome. To this end, human reviewers will be seen as easier to influence than algorithmic reviewers, thus providing more control. Using an online study we find that: (1) people who have a greater Desire for Control over their lives exhibit a stronger preference for human review; (2) interaction with a reviewer is important because it enables influence and ensures understanding; and (3) the higher the impact of a decision, the greater the incentive to influence the outcome, and the greater the preference for human review. Our qualitative results confirm that outcome favourability is a driver for reviewer preference, but so is the desire to be treated with dignity.

11 November 2024
Verified
Conference Paper
Open access
ABAgathe Balayn
Agathe Balayn
YPYulu Pi
Yulu Pi
DWDavid Gray Widder
David Gray Widder
KAKars Alfrink
Kars Alfrink
MYMireia Yurrita
Mireia Yurrita
UGUjwal Gadiraju
Ujwal Gadiraju
This workshop will grow and consolidate a community of interdisciplinary CSCW researchers focusing on the topic of contestable AI. As an outcome of the workshop, we will synthesize the most pressing opportunities and challenges for contestability along AI value chains in the form of a research roadmap. This roadmap will help shape and inspire imminent work in this field. Considering the length and depth of AI value chains, it will especially spur discussions around the contestability of AI systems along various sites of such chains. The workshop will serve as a platform for dialogue and demonstrations of concrete, successful, and unsuccessful examples of AI systems that (could or should) have been contested, to identify requirements, obstacles, and opportunities for designing and deploying contestable AI in various contexts. This will be held primarily as an in-person workshop, with some hybrid accommodation. The day will consist of individual presentations and group activities to stimulate ideation and inspire broad reflections on the field of contestable AI. Our aim is to facilitate interdisciplinary dialogue by bringing together researchers, practitioners, and stakeholders to foster the design and deployment of contestable AI.

16 October 2025 in Proceedings of the ACM on Human-Computer Interaction
Verified
Journal Article
Open access
HLHenrietta Lyons
Henrietta Lyons
TMTim Miller
Tim Miller
EVEduardo Velloso
Eduardo Velloso
High-stakes algorithmic decisions should be contestable, yet there is limited guidance on how to design for contestability, resulting in inadequate contestation processes. Using an algorithmic university admissions decision scenario, we investigate what type of review process people prefer when contesting algorithmic decisions. Through a choice-based conjoint experiment, we explore review process preferences, asking participants to choose their preferred review process from a choice of two that differ across five design attributes: who the reviewer is, how independent the review is, the ability to communicate with the reviewer, the style of the review, and how much the review costs. Expanding on existing studies that focus on the perspective of the decision subject, we introduce eight different scenario perspectives to explore the extent to which particular design attributes matter to different stakeholders. People consistently prefer human reviewers who can make fresh decisions and communicate directly, while also prioritising affordable review processes. These findings highlight the importance of human involvement and cost-effectiveness in review processes. While the impact of perspective was relatively small, our qualitative analysis reveals important underlying tensions between efficiency, fairness, and individual needs. These subtle variations underscore the complexity of designing universally acceptable review processes. We propose design considerations that can help decision-makers to design review processes that are tailored to the specific decision-making context.