Affective schemas: Acquisition, updating, and inference.

Schematized knowledge structures have been extensively studied in the cognitive domain, and yet the nature of affective schemas remains an uncharted area, with experimental work virtually nonexistent. Here, we examined how affective schemas are acquired, updated, and used for inference-making using three novel experimental paradigms. We show that affective schemas emerge by abstracting a common affective value from a distribution of unique affective associations. This common abstracted affective value semanticizes from the discrete exemplars into complex, valenced schemas (negative, positive, neutral), which consolidates across a 24-hr period. Valenced schemas (negative/positive) form faster than neutral schemas, resist affective reversals more strongly, and facilitate rapid learning and memory for related emotional information. Negative-valenced schemas, in particular, are most prioritized for learning, show greater resilience to change, and are more effective in supporting generalized (gist-based) inferences. This work defines key features of affective schemas, moving the study of emotional learning and memory systems from the conditioning of specific associations to the abstraction and consolidation of complex emotional knowledge. (PsycInfo Database Record (c) 2025 APA, all rights reserved).