sh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • harvard-anglia-ruskin-university
  • apa-old-doi-prefix.csl
  • sodertorns-hogskola-harvard.csl
  • sodertorns-hogskola-oxford.csl
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Human-Centered Artificial Intelligence: A Field Experiment
Stockholm School of Economics, Sweden.ORCID iD: 0000-0001-5441-2463
Uppsala University, Sweden.ORCID iD: 0000-0003-0491-2122
University of Zurich, Switzerland.ORCID iD: 0000-0001-5877-5448
Södertörn University, School of Social Sciences, Business Studies.ORCID iD: 0000-0003-4483-7673
Show others and affiliations
2026 (English)In: Management science, ISSN 0025-1909, E-ISSN 1526-5501, Vol. 72, no 1, p. 57-72Article in journal (Refereed) Published
Abstract [en]

Humans and artificial intelligence (AI) algorithms increasingly interact on unstruc-tured managerial tasks. We propose that tailoring this human-AI interaction to align withindividuals’ cognitive preferences is essential for enhancing performance. This hypothesisis examined through a field experiment in a multinational pharmaceutical firm. In the exper-iment, we manipulated four contextual parameters of human-AI interaction—work proce-dures, decision-making authority, training, and incentives—to align with sales experts’cognitive styles, categorized as either adaptors or innovators. Our results show that tailoredinteraction significantly improves sales performance, whereas untailored interaction resultsin negative treatment effects compared with both the tailored and control conditions. Quali-tative evidence suggests that this negative outcome arises from role conflicts and ambigui-ties in untailored interaction. Exploring the mechanisms underlying these outcomes further,a mediation analysis of AI login data reveals that human-AI interaction tailoring leads salesexperts to adjust their AI utilization, which contributes to the observed performance out-comes. These findings support a human-centered approach to AI that prioritizes indivi-duals’ information-processing needs and tailors their interaction with AI accordingly.

Place, publisher, year, edition, pages
Institute for Operations Research and the Management Sciences (INFORMS), 2026. Vol. 72, no 1, p. 57-72
Keywords [en]
artificial intelligence, algorithms, behavioral theory, decision making, field experiment
National Category
Business Administration Human Computer Interaction
Identifiers
URN: urn:nbn:se:sh:diva-58033DOI: 10.1287/mnsc.2022.03849ISI: 001662263900004Scopus ID: 2-s2.0-105029738353OAI: oai:DiVA.org:sh-58033DiVA, id: diva2:1994302
Funder
Familjen Erling-Perssons StiftelseJan Wallander and Tom Hedelius Foundation and Tore Browaldh Foundation, W20-0036Marianne and Marcus Wallenberg Foundation, 2021.0074Marianne and Marcus Wallenberg Foundation, 2021.0133Available from: 2025-09-02 Created: 2025-09-02 Last updated: 2026-03-02Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Pashkevich, Natallia

Search in DiVA

By author/editor
Krakowski, SebastianHaftor, DarekLuger, JohannesPashkevich, NatalliaRaisch, Sebastian
By organisation
Business Studies
In the same journal
Management science
Business AdministrationHuman Computer Interaction

Search outside of DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 192 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • harvard-anglia-ruskin-university
  • apa-old-doi-prefix.csl
  • sodertorns-hogskola-harvard.csl
  • sodertorns-hogskola-oxford.csl
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf