Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics
With the increasing adaptability and complexity of advisory artificial intelligence (AI)-based agents, the topics of explainable AI and human-centered AI are moving close together. Variations in the explanation itself have been widely studied, with some contradictory results. These could be due to u...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Multimodal Technologies and Interaction |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2414-4088/6/12/106 |
| _version_ | 1797456039933640704 |
|---|---|
| author | Carolin Wienrich Astrid Carolus David Roth-Isigkeit Andreas Hotho |
| author_facet | Carolin Wienrich Astrid Carolus David Roth-Isigkeit Andreas Hotho |
| author_sort | Carolin Wienrich |
| collection | DOAJ |
| description | With the increasing adaptability and complexity of advisory artificial intelligence (AI)-based agents, the topics of explainable AI and human-centered AI are moving close together. Variations in the explanation itself have been widely studied, with some contradictory results. These could be due to users’ individual differences, which have rarely been systematically studied regarding their inhibiting or enabling effect on the fulfillment of explanation objectives (such as trust, understanding, or workload). This paper aims to shed light on the significance of human dimensions (gender, age, trust disposition, need for cognition, affinity for technology, self-efficacy, attitudes, and mind attribution) as well as their interplay with different explanation modes (no, simple, or complex explanation). Participants played the game <i>Deal or No Deal</i> while interacting with an AI-based agent. The agent gave advice to the participants on whether they should accept or reject the deals offered to them. As expected, giving an explanation had a positive influence on the explanation objectives. However, the users’ individual characteristics particularly reinforced the fulfillment of the objectives. The strongest predictor of objective fulfillment was the degree of attribution of human characteristics. The more human characteristics were attributed, the more trust was placed in the agent, advice was more likely to be accepted and understood, and important needs were satisfied during the interaction. Thus, the current work contributes to a better understanding of the design of explanations of an AI-based agent system that takes into account individual characteristics and meets the demand for both explainable and human-centered agent systems. |
| first_indexed | 2024-03-09T16:01:44Z |
| format | Article |
| id | doaj.art-daf83433b2584712868fd83fa456631f |
| institution | Directory Open Access Journal |
| issn | 2414-4088 |
| language | English |
| last_indexed | 2024-03-09T16:01:44Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Multimodal Technologies and Interaction |
| spelling | doaj.art-daf83433b2584712868fd83fa456631f2023-11-24T17:02:38ZengMDPI AGMultimodal Technologies and Interaction2414-40882022-11-0161210610.3390/mti6120106Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual CharacteristicsCarolin Wienrich0Astrid Carolus1David Roth-Isigkeit2Andreas Hotho3Psychology of Intelligent Interactive Systems, University of Würzburg, 97070 Würzburg, GermanyMedia Psychology, University of Würzburg, 97070 Würzburg, GermanyCenter for Social Implications of Artificial Intelligence, University of Würzburg, 97070 Würzburg, GermanyData Science, University of Würzburg, 97070 Würzburg, GermanyWith the increasing adaptability and complexity of advisory artificial intelligence (AI)-based agents, the topics of explainable AI and human-centered AI are moving close together. Variations in the explanation itself have been widely studied, with some contradictory results. These could be due to users’ individual differences, which have rarely been systematically studied regarding their inhibiting or enabling effect on the fulfillment of explanation objectives (such as trust, understanding, or workload). This paper aims to shed light on the significance of human dimensions (gender, age, trust disposition, need for cognition, affinity for technology, self-efficacy, attitudes, and mind attribution) as well as their interplay with different explanation modes (no, simple, or complex explanation). Participants played the game <i>Deal or No Deal</i> while interacting with an AI-based agent. The agent gave advice to the participants on whether they should accept or reject the deals offered to them. As expected, giving an explanation had a positive influence on the explanation objectives. However, the users’ individual characteristics particularly reinforced the fulfillment of the objectives. The strongest predictor of objective fulfillment was the degree of attribution of human characteristics. The more human characteristics were attributed, the more trust was placed in the agent, advice was more likely to be accepted and understood, and important needs were satisfied during the interaction. Thus, the current work contributes to a better understanding of the design of explanations of an AI-based agent system that takes into account individual characteristics and meets the demand for both explainable and human-centered agent systems.https://www.mdpi.com/2414-4088/6/12/106explainable AIhuman-centered AIrecommender agentexplanation complexityindividual differences |
| spellingShingle | Carolin Wienrich Astrid Carolus David Roth-Isigkeit Andreas Hotho Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics Multimodal Technologies and Interaction explainable AI human-centered AI recommender agent explanation complexity individual differences |
| title | Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics |
| title_full | Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics |
| title_fullStr | Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics |
| title_full_unstemmed | Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics |
| title_short | Inhibitors and Enablers to Explainable AI Success: A Systematic Examination of Explanation Complexity and Individual Characteristics |
| title_sort | inhibitors and enablers to explainable ai success a systematic examination of explanation complexity and individual characteristics |
| topic | explainable AI human-centered AI recommender agent explanation complexity individual differences |
| url | https://www.mdpi.com/2414-4088/6/12/106 |
| work_keys_str_mv | AT carolinwienrich inhibitorsandenablerstoexplainableaisuccessasystematicexaminationofexplanationcomplexityandindividualcharacteristics AT astridcarolus inhibitorsandenablerstoexplainableaisuccessasystematicexaminationofexplanationcomplexityandindividualcharacteristics AT davidrothisigkeit inhibitorsandenablerstoexplainableaisuccessasystematicexaminationofexplanationcomplexityandindividualcharacteristics AT andreashotho inhibitorsandenablerstoexplainableaisuccessasystematicexaminationofexplanationcomplexityandindividualcharacteristics |