Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Randomized Controlled Trial
. 2025 Jan 31:27:e59946.
doi: 10.2196/59946.

Effect of Artificial Intelligence Helpfulness and Uncertainty on Cognitive Interactions with Pharmacists: Randomized Controlled Trial

Chuan-Ching Tsai  1 , Jin Yong Kim  2 , Qiyuan Chen  2 , Brigid Rowell  1 , X Jessie Yang  2 , Raed Kontar  2 , Megan Whitaker  1 , Corey Lester  1
Affiliations
Randomized Controlled Trial

Effect of Artificial Intelligence Helpfulness and Uncertainty on Cognitive Interactions with Pharmacists: Randomized Controlled Trial

Chuan-Ching Tsai et al. J Med Internet Res. .

Abstract

Background: Clinical decision support systems leveraging artificial intelligence (AI) are increasingly integrated into health care practices, including pharmacy medication verification. Communicating uncertainty in an AI prediction is viewed as an important mechanism for boosting human collaboration and trust. Yet, little is known about the effects on human cognition as a result of interacting with such types of AI advice.

Objective: This study aimed to evaluate the cognitive interaction patterns of pharmacists during medication product verification when using an AI prototype. Moreover, we examine the impact of AI's assistance, both helpful and unhelpful, and the communication of uncertainty of AI-generated results on pharmacists' cognitive interaction with the prototype.

Methods: In a randomized controlled trial, 30 pharmacists from professional networks each performed 200 medication verification tasks while their eye movements were recorded using an online eye tracker. Participants completed 100 verifications without AI assistance and 100 with AI assistance (either with black box help without uncertainty information or uncertainty-aware help, which displays AI uncertainty). Fixation patterns (first and last areas fixated, number of fixations, fixation duration, and dwell times) were analyzed in relation to AI help type and helpfulness.

Results: Pharmacists shifted 19%-26% of their total fixations to AI-generated regions when these were available, suggesting the integration of AI advice in decision-making. AI assistance did not reduce the number of fixations on fill images, which remained the primary focus area. Unhelpful AI advice led to longer dwell times on reference and fill images, indicating increased cognitive processing. Displaying AI uncertainty led to longer cognitive processing times as measured by dwell times in original images.

Conclusions: Unhelpful AI increases cognitive processing time in the original images. Transparency in AI is needed in "black box" systems, but showing more information can add a cognitive burden. Therefore, the communication of uncertainty should be optimized and integrated into clinical workflows using user-centered design to avoid increasing cognitive load or impeding clinicians' original workflow.

Trial registration: ClinicalTrials.gov NCT06795477; https://clinicaltrials.gov/study/NCT06795477.

Keywords: AI helpfulness and accuracy; CDSS; artificial intelligence; clinical decision support system; cognition; cognitive interactions; cognitive processing; decision-making; eye-tracking; interaction; medication; medication verification; pharmacists; uncertainty visualization.

PubMed Disclaimer

Conflict of interest statement

Conflicts of Interest: None declared.

Figures

Figure 1
Figure 1
The medication verification system interface with AI Help. In trials without AI assistance, pharmacists only have access to the reference image (refImage) and fill image. With black box AI help, pharmacists saw an AI match plot, which shows the AI’s stance on the match status for 4 key medication characteristics (imprint, color, shape, and score). In uncertainty-aware AI help, in addition to the match plot, pharmacists also saw the AI histogram, which displays the probability distribution from 50 predictions, indicating the uncertainty of the prediction. Examples of high certainty (left-hand image) and low certainty (right-hand image) are shown. AI: artificial intelligence.
Figure 2
Figure 2
Percentage of trials by region of interest and AI help type. Left: first fixation; right: last fixation. P values were calculated for the 2-tailed z tests to determine the significance of the observed differences and are displayed above respective brackets. AI histogram was only present in the uncertainty-aware condition. No comparative statistics exist in this case. AI: artificial intelligence.
Figure 3
Figure 3
Boxplot representation of the number of fixations in each area per trial, categorized by help type. Bonferroni adjusted P values were calculated alongside Mann-Whitney U tests, and these statistics are displayed above respective brackets. AI histogram was only present in the uncertainty-aware condition. No comparative statistics exist in this case. AI: artificial intelligence.
Figure 4
Figure 4
Boxplot representation of fixation duration in each area, categorized by help type. Bonferroni adjusted P values were calculated alongside Mann-Whitney U tests, with a Bonferroni correction, and these statistics are displayed above respective brackets. AI histogram was only present in the uncertainty-aware condition. No comparative statistics exist in this case. AI: artificial intelligence.
Figure 5
Figure 5
Boxplot representation of dwell time in each area, categorized by help type. Bonferroni adjusted P values were calculated alongside Mann-Whitney U tests, with a Bonferroni correction, and these statistics are displayed above respective brackets. AI histogram was only present in the uncertainty-aware condition. No comparative statistics exist in this case. AI: artificial intelligence.
Figure 6
Figure 6
Dwell time in each area by case type. Compared with helpful advice trials, those with unhelpful advice resulted in significantly longer dwell times in fill and reference images. Bonferroni adjusted P values were calculated alongside Mann-Whitney U tests and these statistics are displayed above respective brackets. AI: artificial intelligence.
See this image and copyright information in PMC

References

    1. Moon J, Chladek JS, Wilson P, Chui MA. Clinical decision support systems in community pharmacies: a scoping review. J Am Med Inform Assoc. 2023;31(1):231–239. doi: 10.1093/jamia/ocad208.7329266 - DOI - PMC - PubMed
    1. Sutton RT, Pincock D, Baumgart DC, Sadowski DC, Fedorak RN, Kroeker KI. An overview of clinical decision support systems: benefits, risks, and strategies for success. NPJ Digit Med. 2020;3:17. doi: 10.1038/s41746-020-0221-y. https://doi.org/10.1038/s41746-020-0221-y 221 - DOI - DOI - PMC - PubMed
    1. Berner ES. Clinical decision support systems: theory and practice. New York, NY: Heidelberg; 2007.
    1. Smithburger PL, Buckley MS, Bejian S, Burenheide K, Kane-Gill SL. A critical evaluation of clinical decision support for the detection of drug-drug interactions. Expert Opin Drug Saf. 2011;10(6):871–882. doi: 10.1517/14740338.2011.583916. - DOI - PubMed
    1. Madaras-Kelly KJ, Hannah EL, Bateman K, Samore MH. Experience with a clinical decision support system in community pharmacies to recommend narrow-spectrum antimicrobials, nonantimicrobial prescriptions, and OTC products to decrease broad-spectrum antimicrobial use. J Manag Care Pharm. 2006;12(5):390–397. doi: 10.18553/jmcp.2006.12.5.390. https://europepmc.org/abstract/MED/16792446 2006(12)5: 390-397 - DOI - PMC - PubMed

Publication types

Associated data

LinkOut - more resources