Research Symposium

BIO

Rafael Merdinger is a freshman pursuing a Bachelor of Science in Biological Sciences with a minor in Chemistry at Florida State University on a pre-medical track. Originally from Fort Lauderdale, Florida, he is an International Baccalaureate diploma recipient and Dean's List honoree. Through the Undergraduate Research Opportunity Program (UROP), he investigates divergent thinking and creativity differences between humans and artificial intelligence under the mentorship of Nelufar D. Radpour. Beyond his coursework, Rafael is authoring an independent book exploring neuroscience and human behavior. His research interests lie at the intersection of cognition, neuroscience, and human perception, and he aspires to a career in cardiovascular or neurological surgery.

Divergent naming task and creativity in humans vs. artificial intelligence

Abstract

Artificial intelligence (AI) systems increasingly perform tasks that resemble human perceptual processing, yet differences remain in how visual information is interpreted. Prior research suggests that humans rely primarily on global shape features when categorizing objects, whereas Al systems often rely more heavily on texture-based features.

The present study examines differences between human participants and Al outputs in the interpretation of abstract visual stimuli lacking recognizable real-world meaning. Abstract two-dimensional images were presented to college-aged participants via a Qualtrics-based survey platform. Participants generated novel names and perceptual ratings for each image. Al systems separately produced names for the same stimuli using structured prompts. Responses will be evaluated using quantitative creativity scoring and statistical comparisons. This study aims to provide insight into perceptual and generative differences between humans and AI, specifically in cases of zero and few-shot learning (in which prior training data is sparse or non-existent)

Keywords: AI Artificial Intelligence Human Cognitive Psychology