# Semantic Memory Structure and Self-Evaluation of Creativity: Evidence Across Tasks and Dimensions

## TL;DR

The paper explores how memory structure and idea fluency affect creativity and self-assessment across different tasks.

## Contribution

It reveals that semantic memory and fluency predict originality, while usefulness and self-evaluation are more context-dependent.

## Key findings

- Ideation fluency and semantic memory integration consistently predict originality across tasks.
- Usefulness is less reliably predicted and shows task-specific associations mainly in CPS.
- Self-evaluations often differ from actual performance, indicating metacognitive biases.

## Abstract

Creativity involves generating ideas that are both original and useful, relying on intertwined cognitive and metacognitive processes. We examined how individual differences in semantic memory structure and ideation fluency predict creative performance and self-evaluations across two studies. In Study 1, participants completed a creative problem-solving (CPS) task, with semantic memory networks estimated from a relatedness judgment task. Creative output was assessed for originality and usefulness, alongside participants’ self-evaluations. In Study 2, a within-subjects design compared participants’ output and self-evaluation of their performance in a divergent thinking task (alternative uses task) and CPS. Results revealed that ideation fluency and semantic memory network integration consistently predicted originality across tasks. In contrast, usefulness was less reliably predicted, showing task-specific associations with semantic memory network properties primarily in CPS. Importantly, self-evaluations often diverged from objective outcomes, reflecting metacognitive biases shaped by heuristic cues. These findings highlight both stable and context-sensitive mechanisms in creative performance and self-evaluation.