References and tools

Bonelli, M. (2026). ILS-Bench: Investor Language-to-Suitability Benchmark. Mendeley Data, V1. Licence CC BY 4.0. doi.org/10.17632/w48mh2dtg5.1

Training data for both advisors, the shared label vocabulary, the Human review outcome and the case browser.

Pratama, R., and co-authors (2026). Systematic literature review of trust and algorithm aversion in the choice between human and AI financial advisors. Presented at SSRAAI 2026.

The review that identified miscalibrated trust as the problem. AdviceIT is its design-side follow-up.

Nahidi, N., and Zarifis, A. (Eds.) (2026). AI, FinTech, and the Future of Robo-Advisory: Risk Management and Ethical Considerations. Contributions to Finance and Accounting. Springer Nature Switzerland. doi.org/10.1007/978-3-032-18109-1

Domain background on AI-driven robo-advisory: the framing of suitability, escalation to a human adviser, and the ethics note and debrief in the study design.

Lee, J. D., and See, K. A. (2004). Trust in automation: Designing for appropriate reliance. Human Factors, 46(1), 50 to 80.

The notion of appropriate reliance and calibrated trust that the dependent variables operationalise.

Dietvorst, B. J., Simmons, J. P., and Massey, C. (2015). Algorithm aversion: People erroneously avoid algorithms after seeing them err. Journal of Experimental Psychology: General, 144(1), 114 to 126.

Under-reliance, and the reason the flawed trials are debriefed.

Bansal, G., Wu, T., Zhou, J., Fok, R., Nushi, B., Kamar, E., Ribeiro, M. T., and Weld, D. S. (2021). Does the whole exceed its parts? The effect of AI explanations on complementary team performance. CHI 2021.

Explanations can increase reliance on wrong advice, the reason sound and flawed trials are both needed.

Szymanski, M., Keyaerts, S., Conati, C., De Croon, R., Vanden Abeele, V., Verbert, K. (2025). Designing and Personalising Hybrid Health Explanations for Lay Users. ACM Transactions on Interactive Intelligent Systems.

The visual, textual and hybrid modality factor, the need for cognition and ease-of-satisfaction scales, and the three-phase study structure.

Szymanski, M., Millecamp, M., Verbert, K. (2021). Visual, Textual or Hybrid: The Effect of User Expertise on Different Explanations. IUI '21.

The original modality comparison.

Samimi, R., Bhattacharya, A., Gosak, L., Stiglic, G., Verbert, K. (2025). Visual-Conversational Interface for Evidence-Based Explanation of Diabetes Risk Prediction. CUI '25. doi.org/10.1145/3719160.3736616

The two-path conversational explainer: analytical questions answered from computation, the rest by a language model.

Millecamp, M., Htun, N. N., Conati, C., Verbert, K. (2019). To Explain or Not to Explain: The Effects of Personal Characteristics When Explaining Music Recommendations. IUI '19.

Personal characteristics as moderators of whether an explanation helps.

Millecamp, M., Htun, N. N., Conati, C., Verbert, K. (2020). What's in a User? Towards Personalising Transparency for Music Recommender Interfaces. UMAP '20.

Interactive control over explanations.

Bhattacharya, A., Ooge, J., Stiglic, G., Verbert, K. (2023). Directive Explanations for Monitoring the Risk of Diabetes Onset. IUI '23.

Explanations that say what to do, the basis for the actionable steps.

Liao, Q. V., Gruen, D., Miller, S. (2020). Questioning the AI: Informing Design Practices for Explainable AI User Experiences. CHI '20.

The question bank the conversational explainer's supported questions come from.

Buçinca, Z., Malaya, M. B., Gajos, K. Z. (2021). To Trust or to Think: Cognitive Forcing Functions Can Reduce Overreliance on AI in AI-assisted Decision-making. Proceedings of the ACM on Human-Computer Interaction 5 (CSCW1). doi.org/10.1145/3449287

Reading the case before the advice is shown, and the analysis pause.

Zhang, Y., Liao, Q. V., Bellamy, R. K. E. (2020). Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Making. FAT* '20. doi.org/10.1145/3351095.3372852

The confidence condition, and why its effect is not assumed to be positive.

Schemmer, M., Kühl, N., Benz, C., Bartos, A., Satzger, G. (2023). Appropriate Reliance on AI Advice: Conceptualization and the Effect of Explanations. IUI '23. doi.org/10.1145/3581641.3584066

How appropriate reliance is defined and measured.

Springer, A., Whittaker, S. (2020). Progressive Disclosure: When, Why, and How Do Users Want Algorithmic Transparency Information? ACM Transactions on Interactive Intelligent Systems 10 (4). doi.org/10.1145/3374218

Detail on demand across the site.

Lins de Holanda Coelho, G., Hanel, P. H. P., Wolf, L. J. (2020). The Very Efficient Assessment of Need for Cognition: Developing a Six-Item Version. Assessment 27 (8). doi.org/10.1177/1073191118793208

The six need for cognition items asked before the cases.

Buell, R. W., Norton, M. I. (2011). The Labor Illusion: How Operational Transparency Increases Perceived Value. Management Science 57 (9). doi.org/10.1287/mnsc.1110.1376

The analysis screen, and why it is held constant across conditions.

Rudin, C. (2019). Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, 1, 206 to 215.

The interpretable-by-design position behind the interpretable rule-based advisor and the fidelity factor.

Lundberg, S. M., and Lee, S.-I. (2017). A unified approach to interpreting model predictions. NeurIPS 2017.

SHAP: Shapley values as feature attributions relative to a baseline, computed exactly for the neural network.

Wachter, S., Mittelstadt, B., and Russell, C. (2017). Counterfactual explanations without opening the black box. Harvard Journal of Law and Technology, 31(2).

The counterfactual content and the contrastive why-not panel: smallest change that alters the outcome, found by search.

Guo, C., Pleiss, G., Sun, Y., and Weinberger, K. Q. (2017). On calibration of modern neural networks. ICML 2017.

Temperature scaling and expected calibration error for the probabilities shown in the confidence content.

De Croon, R., and colleagues, Augment research group, KU Leuven (2025). Designing and personalising hybrid health explanations for lay users. ACM Transactions on Interactive Intelligent Systems.

The hybrid content and the adaptive delivery, and personalisation to a measured user characteristic.

Lusardi, A., and Mitchell, O. S. (2011). Financial literacy around the world: An overview. Journal of Pension Economics and Finance, 10(4), 497 to 508.

The Big Three financial literacy questions used as the moderator and by the adaptive delivery.