View in EDS HTML Full Text PDF Full Text

Assessing readability of explanations and reliability of answers by GPT-3.5 and GPT-4 in non-traumatic spinal cord injury education.

Saved in:
Bibliographic Details
Title: Assessing readability of explanations and reliability of answers by GPT-3.5 and GPT-4 in non-traumatic spinal cord injury education.
Authors: García-Rudolph, Alejandro1,2,3 alejandropablogarcia@gmail.com, Sanchez-Pinsach, David1,2,3, Wright, Mark Andrew1,2,3, Opisso, Eloy1,2,3, Vidal, Joan1,2,3
Source: Medical Teacher. Aug2025, Vol. 47 Issue 8, p1336-1343. 8p.
Subject Terms: *Educational tests & measurements, *Evaluation, *Generative artificial intelligence, *Medical education, *Readability (Literary style), *Textbooks, *Certification, *Educational technology, Success, Research evaluation, Spinal cord injuries, Descriptive statistics, Reliability (Personality trait), Chatbots
Abstract: Purpose: Our study aimed to: i) Assess the readability of textbook explanations using established indexes; ii) Compare these with GPT-4's default explanations, ensuring similar word counts for direct comparisons; iii) Evaluate GPT-4's adaptability by simplifying high-complexity explanations; iv) Determine the reliability of GPT-3.5 and GPT-4 in providing accurate answers. Material and methods: We utilized a textbook designed for ABPMR certification. Our analysis covered 50 multiple-choice questions, each with a detailed explanation, focusing on non-traumatic spinal cord injury (NTSCI). Results: Our analysis revealed statistically significant differences in readability scores, with the textbook achieving 14.5 (SD = 2.5) compared to GPT-4's 17.3 (SD = 1.9), indicating that GPT-4's explanations are generally more complex (p < 0.001). Using the Flesch Reading Ease Score, 86% of GPT-4's explanations fell into the 'Very difficult' category, significantly higher than the textbook's 58% (p = 0.006). GPT-4 successfully demonstrated adaptability by reducing the mean readability score of the top-nine most complex explanations, maintaining the word count. Regarding reliability, GPT-3.5 and GPT-4 scored 84% and 96% respectively, with GPT-4 outperforming GPT-3.5 (p = 0.046). Conclusions: Our results confirmed GPT-4's potential in medical education by providing highly accurate yet often complex explanations for NTSCI, which were successfully simplified without losing accuracy. [ABSTRACT FROM AUTHOR]
Copyright of Medical Teacher is the property of Taylor & Francis Ltd 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: Education Research Complete
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: ehh
DbLabel: Education Research Complete
An: 186774714
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Assessing readability of explanations and reliability of answers by GPT-3.5 and GPT-4 in non-traumatic spinal cord injury education.
– Name: Author
  Label: Authors
  Group: Au
  Data: &lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Garc&#237;a-Rudolph%2C+Alejandro%22&quot;&gt;Garc&#237;a-Rudolph, Alejandro&lt;/searchLink&gt;&lt;relatesTo&gt;1,2,3&lt;/relatesTo&gt;&lt;i&gt; alejandropablogarcia@gmail.com&lt;/i&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Sanchez-Pinsach%2C+David%22&quot;&gt;Sanchez-Pinsach, David&lt;/searchLink&gt;&lt;relatesTo&gt;1,2,3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Wright%2C+Mark+Andrew%22&quot;&gt;Wright, Mark Andrew&lt;/searchLink&gt;&lt;relatesTo&gt;1,2,3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Opisso%2C+Eloy%22&quot;&gt;Opisso, Eloy&lt;/searchLink&gt;&lt;relatesTo&gt;1,2,3&lt;/relatesTo&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;AR&quot; term=&quot;%22Vidal%2C+Joan%22&quot;&gt;Vidal, Joan&lt;/searchLink&gt;&lt;relatesTo&gt;1,2,3&lt;/relatesTo&gt;
– Name: TitleSource
  Label: Source
  Group: Src
  Data: &lt;searchLink fieldCode=&quot;JN&quot; term=&quot;%22Medical+Teacher%22&quot;&gt;Medical Teacher&lt;/searchLink&gt;. Aug2025, Vol. 47 Issue 8, p1336-1343. 8p.
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: *&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Educational+tests+%26+measurements%22&quot;&gt;Educational tests &amp; measurements&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Evaluation%22&quot;&gt;Evaluation&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Generative+artificial+intelligence%22&quot;&gt;Generative artificial intelligence&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Medical+education%22&quot;&gt;Medical education&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Readability+%28Literary+style%29%22&quot;&gt;Readability (Literary style)&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Textbooks%22&quot;&gt;Textbooks&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Certification%22&quot;&gt;Certification&lt;/searchLink&gt;&lt;br /&gt;*&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Educational+technology%22&quot;&gt;Educational technology&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Success%22&quot;&gt;Success&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Research+evaluation%22&quot;&gt;Research evaluation&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Spinal+cord+injuries%22&quot;&gt;Spinal cord injuries&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Descriptive+statistics%22&quot;&gt;Descriptive statistics&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Reliability+%28Personality+trait%29%22&quot;&gt;Reliability (Personality trait)&lt;/searchLink&gt;&lt;br /&gt;&lt;searchLink fieldCode=&quot;DE&quot; term=&quot;%22Chatbots%22&quot;&gt;Chatbots&lt;/searchLink&gt;
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Purpose: Our study aimed to: i) Assess the readability of textbook explanations using established indexes; ii) Compare these with GPT-4&#39;s default explanations, ensuring similar word counts for direct comparisons; iii) Evaluate GPT-4&#39;s adaptability by simplifying high-complexity explanations; iv) Determine the reliability of GPT-3.5 and GPT-4 in providing accurate answers. Material and methods: We utilized a textbook designed for ABPMR certification. Our analysis covered 50 multiple-choice questions, each with a detailed explanation, focusing on non-traumatic spinal cord injury (NTSCI). Results: Our analysis revealed statistically significant differences in readability scores, with the textbook achieving 14.5 (SD = 2.5) compared to GPT-4&#39;s 17.3 (SD = 1.9), indicating that GPT-4&#39;s explanations are generally more complex (p &lt; 0.001). Using the Flesch Reading Ease Score, 86% of GPT-4&#39;s explanations fell into the &#39;Very difficult&#39; category, significantly higher than the textbook&#39;s 58% (p = 0.006). GPT-4 successfully demonstrated adaptability by reducing the mean readability score of the top-nine most complex explanations, maintaining the word count. Regarding reliability, GPT-3.5 and GPT-4 scored 84% and 96% respectively, with GPT-4 outperforming GPT-3.5 (p = 0.046). Conclusions: Our results confirmed GPT-4&#39;s potential in medical education by providing highly accurate yet often complex explanations for NTSCI, which were successfully simplified without losing accuracy. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: &lt;i&gt;Copyright of Medical Teacher is the property of Taylor &amp; Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder&#39;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.&lt;/i&gt; (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=ehh&AN=186774714
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/0142159X.2024.2430365
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 8
        StartPage: 1336
    Subjects:
      – SubjectFull: Educational tests & measurements
        Type: general
      – SubjectFull: Evaluation
        Type: general
      – SubjectFull: Generative artificial intelligence
        Type: general
      – SubjectFull: Medical education
        Type: general
      – SubjectFull: Readability (Literary style)
        Type: general
      – SubjectFull: Textbooks
        Type: general
      – SubjectFull: Certification
        Type: general
      – SubjectFull: Educational technology
        Type: general
      – SubjectFull: Success
        Type: general
      – SubjectFull: Research evaluation
        Type: general
      – SubjectFull: Spinal cord injuries
        Type: general
      – SubjectFull: Descriptive statistics
        Type: general
      – SubjectFull: Reliability (Personality trait)
        Type: general
      – SubjectFull: Chatbots
        Type: general
    Titles:
      – TitleFull: Assessing readability of explanations and reliability of answers by GPT-3.5 and GPT-4 in non-traumatic spinal cord injury education.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: García-Rudolph, Alejandro
      – PersonEntity:
          Name:
            NameFull: Sanchez-Pinsach, David
      – PersonEntity:
          Name:
            NameFull: Wright, Mark Andrew
      – PersonEntity:
          Name:
            NameFull: Opisso, Eloy
      – PersonEntity:
          Name:
            NameFull: Vidal, Joan
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 08
              Text: Aug2025
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 0142159X
          Numbering:
            – Type: volume
              Value: 47
            – Type: issue
              Value: 8
          Titles:
            – TitleFull: Medical Teacher
              Type: main
ResultId 1