arXiv:2605.26801v1 Announce Type: new Abstract: Psychological constructs are often measured in separate instruments, datasets, and research traditions, which makes direct comparison difficult. This paper proposes a framework for making such constructs semantically commensurate by representing and comparing them as directions in a shared word-embedding space. Using Supervised Semantic Differential, we estimate construct-specific semantic gradients from text-outcome associations and project them onto theoretically motivated reference axes. As an initial test case, we use Valence, Arousal, and Do
The paper leverages recent advancements in large language models and word embeddings to bridge a long-standing gap in psychological research methodology.
This work proposes a novel and scalable method for quantitatively comparing complex psychological constructs, which can advance both AI development and behavioral science.
Psychological constructs, traditionally measured disparately, can now be directly compared and integrated into a shared semantic space, enabling new forms of analysis.
- · AI ethicists
- · Psychology researchers
- · AI developers
- · Social scientists
- · Traditional qualitative psychology research
- · Companies relying on siloed psychological data
It provides a quantifiable framework for understanding and integrating human psychological dimensions into AI systems.
This integration could lead to AI models with more nuanced 'understanding' of human behavior and emotional states.
The ability to semantically map psychological constructs might enable advanced AI agents to simulate or even develop 'personality' profiles based on human data.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL