View in EDS HTML Full Text PDF Full Text

Explanation strategies in humans versus current explainable artificial intelligence: Insights from image classification.

Saved in:
Bibliographic Details
Title: Explanation strategies in humans versus current explainable artificial intelligence: Insights from image classification.
Authors: Qi, Ruoxi (AUTHOR), Zheng, Yueyuan (AUTHOR), Yang, Yi (AUTHOR), Cao, Caleb Chen (AUTHOR), Hsiao, Janet H. (AUTHOR)
Source: British Journal of Psychology. May2026, Vol. 117 Issue 2, p479-502. 24p.
Subjects: Task performance, Research funding, Corn, Winter sports, Portable computers, T-test (Statistics), Artificial intelligence, Tennis, Multiple regression analysis, Eye movement measurements, Lemon, Cell phones, Analysis of covariance, Attention, Large-scale brain networks, Analysis of variance, Concepts, Visual perception, Insects, Mollusks, Horses, Mushrooms, Transducers
Geographic Terms: China
Abstract: Explainable AI (XAI) methods provide explanations of AI models, but our understanding of how they compare with human explanations remains limited. Here, we examined human participants' attention strategies when classifying images and when explaining how they classified the images through eye‐tracking and compared their attention strategies with saliency‐based explanations from current XAI methods. We found that humans adopted more explorative attention strategies for the explanation task than the classification task itself. Two representative explanation strategies were identified through clustering: One involved focused visual scanning on foreground objects with more conceptual explanations, which contained more specific information for inferring class labels, whereas the other involved explorative scanning with more visual explanations, which were rated higher in effectiveness for early category learning. Interestingly, XAI saliency map explanations had the highest similarity to the explorative attention strategy in humans, and explanations highlighting discriminative features from invoking observable causality through perturbation had higher similarity to human strategies than those highlighting internal features associated with higher class score. Thus, humans use both visual and conceptual information during explanation, which serve different purposes, and XAI methods that highlight features informing observable causality match better with human explanations, potentially more accessible to users. [ABSTRACT FROM AUTHOR]
Copyright of British Journal of Psychology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Psychology and Behavioral Sciences Collection
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: pbh
DbLabel: Psychology and Behavioral Sciences Collection
An: 192785876
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Explanation strategies in humans versus current explainable artificial intelligence: Insights from image classification.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Qi%2C+Ruoxi%22">Qi, Ruoxi</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zheng%2C+Yueyuan%22">Zheng, Yueyuan</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Yang%2C+Yi%22">Yang, Yi</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Cao%2C+Caleb+Chen%22">Cao, Caleb Chen</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Hsiao%2C+Janet+H%2E%22">Hsiao, Janet H.</searchLink> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22British+Journal+of+Psychology%22">British Journal of Psychology</searchLink>. May2026, Vol. 117 Issue 2, p479-502. 24p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Task+performance%22">Task performance</searchLink><br /><searchLink fieldCode="DE" term="%22Research+funding%22">Research funding</searchLink><br /><searchLink fieldCode="DE" term="%22Corn%22">Corn</searchLink><br /><searchLink fieldCode="DE" term="%22Winter+sports%22">Winter sports</searchLink><br /><searchLink fieldCode="DE" term="%22Portable+computers%22">Portable computers</searchLink><br /><searchLink fieldCode="DE" term="%22T-test+%28Statistics%29%22">T-test (Statistics)</searchLink><br /><searchLink fieldCode="DE" term="%22Artificial+intelligence%22">Artificial intelligence</searchLink><br /><searchLink fieldCode="DE" term="%22Tennis%22">Tennis</searchLink><br /><searchLink fieldCode="DE" term="%22Multiple+regression+analysis%22">Multiple regression analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Eye+movement+measurements%22">Eye movement measurements</searchLink><br /><searchLink fieldCode="DE" term="%22Lemon%22">Lemon</searchLink><br /><searchLink fieldCode="DE" term="%22Cell+phones%22">Cell phones</searchLink><br /><searchLink fieldCode="DE" term="%22Analysis+of+covariance%22">Analysis of covariance</searchLink><br /><searchLink fieldCode="DE" term="%22Attention%22">Attention</searchLink><br /><searchLink fieldCode="DE" term="%22Large-scale+brain+networks%22">Large-scale brain networks</searchLink><br /><searchLink fieldCode="DE" term="%22Analysis+of+variance%22">Analysis of variance</searchLink><br /><searchLink fieldCode="DE" term="%22Concepts%22">Concepts</searchLink><br /><searchLink fieldCode="DE" term="%22Visual+perception%22">Visual perception</searchLink><br /><searchLink fieldCode="DE" term="%22Insects%22">Insects</searchLink><br /><searchLink fieldCode="DE" term="%22Mollusks%22">Mollusks</searchLink><br /><searchLink fieldCode="DE" term="%22Horses%22">Horses</searchLink><br /><searchLink fieldCode="DE" term="%22Mushrooms%22">Mushrooms</searchLink><br /><searchLink fieldCode="DE" term="%22Transducers%22">Transducers</searchLink>
– Name: SubjectGeographic
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22China%22">China</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Explainable AI (XAI) methods provide explanations of AI models, but our understanding of how they compare with human explanations remains limited. Here, we examined human participants' attention strategies when classifying images and when explaining how they classified the images through eye‐tracking and compared their attention strategies with saliency‐based explanations from current XAI methods. We found that humans adopted more explorative attention strategies for the explanation task than the classification task itself. Two representative explanation strategies were identified through clustering: One involved focused visual scanning on foreground objects with more conceptual explanations, which contained more specific information for inferring class labels, whereas the other involved explorative scanning with more visual explanations, which were rated higher in effectiveness for early category learning. Interestingly, XAI saliency map explanations had the highest similarity to the explorative attention strategy in humans, and explanations highlighting discriminative features from invoking observable causality through perturbation had higher similarity to human strategies than those highlighting internal features associated with higher class score. Thus, humans use both visual and conceptual information during explanation, which serve different purposes, and XAI methods that highlight features informing observable causality match better with human explanations, potentially more accessible to users. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of British Journal of Psychology is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=pbh&AN=192785876
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1111/bjop.12714
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 24
        StartPage: 479
    Subjects:
      – SubjectFull: Task performance
        Type: general
      – SubjectFull: Research funding
        Type: general
      – SubjectFull: Corn
        Type: general
      – SubjectFull: Winter sports
        Type: general
      – SubjectFull: Portable computers
        Type: general
      – SubjectFull: T-test (Statistics)
        Type: general
      – SubjectFull: Artificial intelligence
        Type: general
      – SubjectFull: Tennis
        Type: general
      – SubjectFull: Multiple regression analysis
        Type: general
      – SubjectFull: Eye movement measurements
        Type: general
      – SubjectFull: Lemon
        Type: general
      – SubjectFull: Cell phones
        Type: general
      – SubjectFull: Analysis of covariance
        Type: general
      – SubjectFull: Attention
        Type: general
      – SubjectFull: Large-scale brain networks
        Type: general
      – SubjectFull: Analysis of variance
        Type: general
      – SubjectFull: Concepts
        Type: general
      – SubjectFull: Visual perception
        Type: general
      – SubjectFull: Insects
        Type: general
      – SubjectFull: Mollusks
        Type: general
      – SubjectFull: Horses
        Type: general
      – SubjectFull: Mushrooms
        Type: general
      – SubjectFull: Transducers
        Type: general
      – SubjectFull: China
        Type: general
    Titles:
      – TitleFull: Explanation strategies in humans versus current explainable artificial intelligence: Insights from image classification.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Qi, Ruoxi
      – PersonEntity:
          Name:
            NameFull: Zheng, Yueyuan
      – PersonEntity:
          Name:
            NameFull: Yang, Yi
      – PersonEntity:
          Name:
            NameFull: Cao, Caleb Chen
      – PersonEntity:
          Name:
            NameFull: Hsiao, Janet H.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 00071269
          Numbering:
            – Type: volume
              Value: 117
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: British Journal of Psychology
              Type: main
ResultId 1