Information Science · Human–AI Decision Making
NIC research program / Kerry Thornhill
What makes a decision open to challenge?
Research on how people use evidence to evaluate AI-supported decisions—and what information makes a specific concern possible to investigate.
The Necessary Informational Conditions (NIC) program connects conceptual analysis, research on institutional answerability, and proposed empirical studies of human judgment. A related philosophical inquiry examines how institutions shape what counts as a reason.
MS in Interdisciplinary Studies · In progress · University of North Texas
Anuradha and Vikas Sinha College of Artificial Intelligence and Advanced Analytics
01 / The research problem
An explanation is a beginning.
A person may know an outcome and its stated reason yet lack the grounds to investigate a mistake. The NIC program asks what makes a material objection investigable, and how it can receive a consequential institutional response.
Understanding contestability02 / Papers and work
Four inquiries.
Distinct burdens of proof.
From Notice to Contestable Grounds
Conceptual article · Preprint, October 2026What informational tasks must remain possible where meaningful contestability is required or promised?
B / 02Private Power without Epistemic Preemption
Normative theory · Preprint planned January 2027When does a private institution owe informed challenge and consequential reconsideration?
E / 03Testing Informational Conditions for Meaningful AI Contestability
Empirical program · ProposedA proposed 36-month program: evidence integration, informational functions, and a contestability toolkit.
H / 04What Counts as a Reason: Institutional Formation and the Limits of Rational Reconstruction
Hermeneutics · Preprint planned December 2026How do institutions shape what counts as a reason, and how can those standards be examined?
Paper A preprint available · Two further papers in preparation · Empirical studies proposed
View all work →03 / Research program
One argument map.
Three phases, 36 months.
The argument map sets out how an institutionally operative determination, a plausible material objection, and the epistemic task needed to investigate it connect—and the six candidate informational functions that task may require. The research plan tests those functions in three phases and ends with a working contestability toolkit.
- 1Institutionally operative determination
- 2Plausible material objection
- 3Epistemic task needed to investigate it
- PDFArgument map v. 5 · Research plan v. 6 · 12 October 2026
04 / Empirical inquiry · Proposed studies
Studying how evidence changes judgment.
The proposed empirical work examines how people weigh independent evidence alongside AI or human advice. It then uses synthetic consumer-credit cases to test whether information matched to a specific concern helps people recognize genuine errors while preserving warranted reliance when no error is present.
The studies are designed to distinguish useful information from information that merely increases confidence or suspicion. Their findings could support, narrow, or revise the working framework.
Read the proposed study plan
Conceptual illustration · No study results
05 / A worked example
The record says one thing.
The person says another.
A credit determination refers to payment arrears. The applicant believes the account was already settled. What would let that concern become a useful question—and what would let the answer matter?
Synthetic case / Four stages / No study results
06 / Author & inquiry
A program open to criticism.
Kerry Thornhill is an MS student at the University of North Texas, developing a sustained academic research career in Information Science. The project welcomes substantive feedback on its arguments, methods, and scope.
About the researcher