At the crossroad of lexical-semantic features, affect, subclinical depressive symptoms and rumination: a linear mixed-effects model of emotional priming in abstract and concrete words.

Objective: Emotional priming is modulated by word concreteness, yet the literature is inconsistent. This study investigates the effects of affect, lexical features, depressive symptoms, and rumination on emotional priming for abstract and concrete words.

Methods: Eighty-one healthy participants (48 female, age = 24.12 ± 8.56 years) completed a valence categorization task in which they were asked to decide whether a target word, presented for 500 ms following a 65 ms prime, was pleasant, neutral, or unpleasant. Priming effect (PE) was defined as the RT difference between incongruent and congruent conditions for pleasant and unpleasant primes. Linear mixed-effects models were used to assess priming effects and error rates, with fixed effects for prime valence, concreteness, subscales of Positive and Negative Affect Scale, Beck Depression Inventory score, subscales of Rumination Response Scale, semantic relatedness and Levenshtein distance (LD). Model selection was performed using the buildmer algorithm. Post-hoc analyses of interactions and continuous predictor trends across categorical levels were performed using emmeans and emtrends. All statistical analyses were conducted by R 4.4.3.

Results: PEs were stronger for concrete than abstract words, irrespective of prime valence. Positive affect predicted higher error rates for unpleasant targets and enhanced PEs for unpleasant primes, particularly in abstract words. Depressive symptoms were associated with fewer errors for unpleasant targets but did not predict PE. Brooding was associated with larger PEs for abstract words, independent of valence. Lastly, greater semantic relatedness amplified PEs for abstract items, whereas smaller LD both strengthened PEs and increased errors in valence‑incongruent trials.

Conclusion: This study highlights the role of emotional and clinical traits in word processing, showing that abstract word priming is driven by the interaction of affect and stimulus value. Future studies examining biases in abstract emotional word processing may guide the development of methods to identify individuals at risk for depression.