Keywords
Aesthetics, Objective, MML, Holistics, Evaluation

Technologies
LLMS, CLIP, Nexus, MML, Qualtrics

The project involved the development of an AI-driven application that integrates a multimodal model (CLIP) with a neural probe to analyse images and provide structured feedback on formal qualities such as lighting, composition, and visual clarity. The system was designed to deliver consistent, objective feedback on measurable visual features, enabling users to refine images through iterative interaction.

The psychological component of the project focused on examining how engagement with such a system influences aesthetic evaluation. A theoretical framework grounded in cognitive and empirical aesthetics was developed to investigate the relationship between formal image properties, emotional response, and holistic judgment. A user experience study was conducted to assess perceptions of the system and inform interface refinement, followed by the design of a controlled experimental methodology. This included the development of photographic tasks, a structured rating framework, and a statistical analysis strategy to test whether prioritising formal image qualities translates into differences in emotional and holistic evaluation.