An official website of the United States government
Metaphor is crucial in human cognition and creativity, facilitating abstract thinking, analogical reasoning, and idea generation. Typically, human raters manually score the originality of responses to creative thinking tasks – a laborious and error-prone process. Previous research sought to remedy these risks by scoring creativity tasks automatically using semantic distance and large language models (LLMs). Here, we extend research on automatic creativity scoring to metaphor generation – the ability to creatively describe episodes and concepts using nonliteral language. Metaphor is arguably more abstract and naturalistic than prior targets of automated creativity assessment. We collected 4,589 responses from 1,546 participants to various metaphor prompts and corresponding human creativity ratings. We fine-tuned two open-source LLMs (RoBERTa and GPT-2) – effectively “teaching” them to score metaphors like humans – before testing their ability to accurately assess the creativity of new metaphors. Results showed both models reliably predicted new human creativity ratings (RoBERTa r = .72, GPT-2 r = .70), significantly more strongly than semantic distance (r = .42). Importantly, the fine-tuned models generalized accurately to metaphor prompts they had not been trained on (RoBERTa r = .68, GPT-2 r = .63). We provide open access to the fine-tuned models, allowing researchers to assess metaphor creativity in a reproducible and timely manner.
more »
« less
Automated Scoring of Scientific Creativity in German
ABSTRACT Automated scoring is a current hot topic in creativity research. However, most research has focused on the English language and popular verbal creative thinking tasks, such as the alternate uses task. Therefore, in this study, we present a large language model approach for automated scoring of a scientific creative thinking task that assesses divergent ideation in experimental tasks in the German language. Participants are required to generate alternative explanations for an empirical observation. This work analyzed a total of 13,423 unique responses. To predict human ratings of originality, we used XLM‐RoBERTa (Cross‐lingual Language Model‐RoBERTa), a large, multilingual model. The prediction model was trained on 9,400 responses. Results showed a strong correlation between model predictions and human ratings in a held‐out test set (n = 2,682;r = 0.80; CI‐95% [0.79, 0.81]). These promising findings underscore the potential of large language models for automated scoring of scientific creative thinking in the German language. We encourage researchers to further investigate automated scoring of other domain‐specific creative thinking tasks.
more »
« less
- PAR ID:
- 10526031
- Publisher / Repository:
- Wiley
- Date Published:
- Journal Name:
- The Journal of Creative Behavior
- ISSN:
- 0022-0175
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Creativity is increasingly recognized as a core competency for the 21st century, making its development a priority in education, research, and industry. To effectively cultivate creativity, researchers and educators need reliable and accessible assessment tools. Recent software developments have significantly enhanced the administration and scoring of creativity measures; however, existing software often requires expertise in experiment design and computer programming, limiting its accessibility to many educators and researchers. In the current work, we introduce CAP—the Creativity Assessment Platform—a free web application for building creativity assessments, collecting data, and automatically scoring responses (cap.ist.psu.edu). CAP allows users to create custom creativity assessments in ten languages using a simple, point-and-click interface, selecting from tasks such as the Short Story Task, Drawing Task, and Scientific Creative Thinking Test. Users can automatically score task responses using machine learning models trained to match human creativity ratings—with multilingual capabilities, including the new Cross-Lingual Alternate Uses Scoring (CLAUS), a large language model achieving strong prediction of human creativity ratings in ten languages. CAP also provides a centralized dashboard to monitor data collection, score assessments, and automatically generate text for a Methods section based on the study’s tasks, metrics, and instructions—with a single click—promoting transparency and reproducibility in creativity assessment. Designed for ease of use, CAP aims to democratize creativity measurement for researchers, educators, and everyone in between.more » « less
-
ABSTRACT The PISA assessment 2022 of creative thinking was a moonshot effort that introduced significant advancements over existing creativity tests, including a broad range of domains (written, visual, social, and scientific), implementation in many languages, and sophisticated scoring methods. PISA 2022 demonstrated the general feasibility of assessing creative thinking ability comprehensively at an international scale. However, the complexity of its assessment approach—such as time‐consuming scoring requiring human raters—implies the risk that it may not be easily applied by the scientific community and practitioners. In this commentary, we outline important next steps building on the PISA assessment to further enhance future assessments of creative thinking. Crucial future directions include 1) determining what tasks and scorings ensure high psychometric quality including content validity, 2) enabling efficient, objective scoring by applying AI methods such as Large Language Models (LLMs), 3) ensuring high language accessibility via multilingual tests, 4) targeting a broader age group, and 5) facilitating standardized, reproducible assessments via an open online testing platform. In sum, these developments would lead to an efficient, validated multilingual test of creative thinking, which enhances the accessibility of effective creative thinking assessments and thereby supports the democratization and reproducibility of creativity research.more » « less
-
Abstract Argumentation, a key scientific practice presented in theFramework for K-12 Science Education, requires students to construct and critique arguments, but timely evaluation of arguments in large-scale classrooms is challenging. Recent work has shown the potential of automated scoring systems for open response assessments, leveraging machine learning (ML) and artificial intelligence (AI) to aid the scoring of written arguments in complex assessments. Moreover, research has amplified that the features (i.e., complexity, diversity, and structure) of assessment construct are critical to ML scoring accuracy, yet how the assessment construct may be associated with machine scoring accuracy remains unknown. This study investigated how the features associated with the assessment construct of a scientific argumentation assessment item affected machine scoring performance. Specifically, we conceptualized the construct in three dimensions: complexity, diversity, and structure. We employed human experts to code characteristics of the assessment tasks and score middle school student responses to 17 argumentation tasks aligned to three levels of a validated learning progression of scientific argumentation. We randomly selected 361 responses to use as training sets to build machine-learning scoring models for each item. The scoring models yielded a range of agreements with human consensus scores, measured by Cohen’s kappa (mean = 0.60; range 0.38 − 0.89), indicating good to almost perfect performance. We found that higher levels ofComplexityandDiversity of the assessment task were associated with decreased model performance, similarly the relationship between levels ofStructureand model performance showed a somewhat negative linear trend. These findings highlight the importance of considering these construct characteristics when developing ML models for scoring assessments, particularly for higher complexity items and multidimensional assessments.more » « less
-
null (Ed.)Abstract Creativity research requires assessing the quality of ideas and products. In practice, conducting creativity research often involves asking several human raters to judge participants’ responses to creativity tasks, such as judging the novelty of ideas from the alternate uses task (AUT). Although such subjective scoring methods have proved useful, they have two inherent limitations—labor cost (raters typically code thousands of responses) and subjectivity (raters vary on their perceptions and preferences)—raising classic psychometric threats to reliability and validity. We sought to address the limitations of subjective scoring by capitalizing on recent developments in automated scoring of verbal creativity via semantic distance, a computational method that uses natural language processing to quantify the semantic relatedness of texts. In five studies, we compare the top performing semantic models (e.g., GloVe, continuous bag of words) previously shown to have the highest correspondence to human relatedness judgements. We assessed these semantic models in relation to human creativity ratings from a canonical verbal creativity task (AUT; Studies 1–3) and novelty/creativity ratings from two word association tasks (Studies 4–5). We find that a latent semantic distance factor—comprised of the common variance from five semantic models—reliably and strongly predicts human creativity and novelty ratings across a range of creativity tasks. We also replicate an established experimental effect in the creativity literature (i.e., the serial order effect) and show that semantic distance correlates with other creativity measures, demonstrating convergent validity. We provide an open platform to efficiently compute semantic distance, including tutorials and documentation ( https://osf.io/gz4fc/ ).more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1002/jocb.658
