Narrative Cognition Transformer (NCT): An Explainable Symbol–Emotion Model for Mapping Resilience Patterns in Contemporary Fiction

Fiction resilience is manifested in the shifting emotional status, symbolic repetitions, narrative rhythm and the remaking of the inner world of the character. The current systems of NLP do not reflect this complexity since they presuppose narrative meaning to be generated by individual words or sentence-level emotion as opposed to long-range symbolic and emotional pattern. This paper presents the Narrative Cognition Transformer (NCT), a symbolic-emotional enhanced deep learning model that is capable of following the dynamics of resilience through the whole novel. The model integrates symbolic memory graphs, emotion-flow embeddings, and a reasoning engine based on transformers, which can track destabilization, transition, and stabilization phases. Applying four modern novels as the evaluation corpus, NCT reached accuracy at 92.6 percent, reduction of drift instability between 44 and 57 percent, and emotional energy variance 35.8 per cent. less than baseline models. Recent figures like the figures of symbolical FFT spectra, drift curves, emotional energy profiles, and reward convergence graphs are very persuasive of the cognitive interpretability of NCT. Tables of stability, drift and emotional costs also confirm the performance of the model. The results indicate that resilience can be calculated when the symbolic cognitive and emotional transitions are integrated into a single deep learning model.