ChatGPT and CLT: Investigating differences in multimodal processing
Drawing on construal level theory, recent studies have demonstrated that ChatGPT interprets text inputs from an abstract perspective. However, as ChatGPT has evolved into a multimodal tool, this research examines whether ChatGPT's abstraction bias extends to image-based prompts. In a pre-regist...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
KeAi Communications Co., Ltd.
2025-11-01
|
| Series: | Journal of Economy and Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2949948824000611 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832591812638801920 |
|---|---|
| author | Michael Cahalane Samuel N. Kirshner |
| author_facet | Michael Cahalane Samuel N. Kirshner |
| author_sort | Michael Cahalane |
| collection | DOAJ |
| description | Drawing on construal level theory, recent studies have demonstrated that ChatGPT interprets text inputs from an abstract perspective. However, as ChatGPT has evolved into a multimodal tool, this research examines whether ChatGPT's abstraction bias extends to image-based prompts. In a pre-registered study utilising hierarchical letters, ChatGPT predominantly associated these images with local rather than global letters, suggesting a concrete bias when analysing images. This starkly contrasts human participants who predominantly identified the same images with the global letters, indicating that humans and ChatGPT significantly diverge in image interpretations. Furthermore, while humans generally perceive ChatGPT to be more concrete in image processing, there is a notable discrepancy between this perception and the actual level of concreteness exhibited by ChatGPT in handling image-based tasks. These findings provide insights into the distinct cognitive behaviours of LLMs compared to humans, contributing to an emerging understanding of LLM cognition in the context of multimodal inputs. |
| format | Article |
| id | doaj-art-fc00fb723bdf4df9a2624d06fd72513b |
| institution | Kabale University |
| issn | 2949-9488 |
| language | English |
| publishDate | 2025-11-01 |
| publisher | KeAi Communications Co., Ltd. |
| record_format | Article |
| series | Journal of Economy and Technology |
| spelling | doaj-art-fc00fb723bdf4df9a2624d06fd72513b2025-01-22T05:44:44ZengKeAi Communications Co., Ltd.Journal of Economy and Technology2949-94882025-11-0131021ChatGPT and CLT: Investigating differences in multimodal processingMichael Cahalane0Samuel N. Kirshner1School of Information Systems and Technology Management, University of New South Wales Kensington, Sydney, NSW 2052, AustraliaCorresponding author.; School of Information Systems and Technology Management, University of New South Wales Kensington, Sydney, NSW 2052, AustraliaDrawing on construal level theory, recent studies have demonstrated that ChatGPT interprets text inputs from an abstract perspective. However, as ChatGPT has evolved into a multimodal tool, this research examines whether ChatGPT's abstraction bias extends to image-based prompts. In a pre-registered study utilising hierarchical letters, ChatGPT predominantly associated these images with local rather than global letters, suggesting a concrete bias when analysing images. This starkly contrasts human participants who predominantly identified the same images with the global letters, indicating that humans and ChatGPT significantly diverge in image interpretations. Furthermore, while humans generally perceive ChatGPT to be more concrete in image processing, there is a notable discrepancy between this perception and the actual level of concreteness exhibited by ChatGPT in handling image-based tasks. These findings provide insights into the distinct cognitive behaviours of LLMs compared to humans, contributing to an emerging understanding of LLM cognition in the context of multimodal inputs.http://www.sciencedirect.com/science/article/pii/S2949948824000611ChatGPTAI cognitionConstrual level theory |
| spellingShingle | Michael Cahalane Samuel N. Kirshner ChatGPT and CLT: Investigating differences in multimodal processing Journal of Economy and Technology ChatGPT AI cognition Construal level theory |
| title | ChatGPT and CLT: Investigating differences in multimodal processing |
| title_full | ChatGPT and CLT: Investigating differences in multimodal processing |
| title_fullStr | ChatGPT and CLT: Investigating differences in multimodal processing |
| title_full_unstemmed | ChatGPT and CLT: Investigating differences in multimodal processing |
| title_short | ChatGPT and CLT: Investigating differences in multimodal processing |
| title_sort | chatgpt and clt investigating differences in multimodal processing |
| topic | ChatGPT AI cognition Construal level theory |
| url | http://www.sciencedirect.com/science/article/pii/S2949948824000611 |
| work_keys_str_mv | AT michaelcahalane chatgptandcltinvestigatingdifferencesinmultimodalprocessing AT samuelnkirshner chatgptandcltinvestigatingdifferencesinmultimodalprocessing |