The Effect of Cross-Modal Stroop Interference on Modality-Specific Selective Attention Processing

Abstract

Selective attention refers to the cognitive function of prioritizing goal-relevant information among simultaneously presented multisensory inputs. Prior Stroop-type research has largely been confined to a single sensory modality, typically vision, and few studies have systematically compared cross-modal interference among heterogeneous stimulus types. This study developed a cross-modal Stroop interference task comprising six tasks that paired four stimulus modalities—picture, text, voice, and non-speech sound (animal cries)—to examine whether reaction time and response accuracy differ by target modality and stimulus congruency (congruent vs. incongruent). Data from thirty healthy young adults were analyzed using a two-way repeated-measures ANOVA (task × congruency). Results: The main effect of task, the main effect of congruency, and the task × congruency interaction were all significant (all ps<.001). Simple-effects analyses showed significant interference (incongruent slower than congruent) for the picture-voice, picture-cry, and text-cry tasks, whereas no statistically significant interference was found for the picture-text and text-voice tasks. A significant hierarchy of reaction speed was observed across target modalities, text < picture < voice < cry (p<.001), with all six pairwise comparisons significant. A 12-condition (task × target-modality) analysis also showed significant differences among conditions (p<.001). Response accuracy showed a ceiling effect, remaining at or above 97.8% across all conditions. These findings indicate that interference emerges most prominently in tasks that combine a visual and an auditory stimulus with at least one non-linguistic component, supporting a hierarchy of processing automaticity across sensory modalities.