It was hypothesized that effect size in therapy outcome research would correlate positively with congruence effects. Congruence was defined as the degree to which what had been practiced in treatment was scored as improvement when outcome was measured. Additionally, it was hypothesized that correcting effect sizes for estimated nongeneralizable change attributable to congruence (i.e., representativeness reduction) would significantly reduce the average magnitude of effect.

<p dir="ltr">Many tasks require processing, filtering, and responding to information from multiple sensory modalities. Crossmodal interactions are common and visual dominance often arises with incongruent sensory information. Past studies have shown that visual dominance tends to be strong in spatial tasks. Experiments in a crossmodal attention switching paradigm with physical-spatial stimuli (e.g., stimuli in left and right locations) have demonstrated a robust visual dominance congruence pattern with conflicting visual-spatial information impairing responses to auditory-spatial stimuli, but conflicting auditory-spatial information having less impact on visual-spatial processing. Strikingly, this pattern does not occur with verbal-spatial stimuli (e.g., the words LEFT and RIGHT as stimuli). In the present study, experiments were conducted to systematically examine the occurrence and underlying basis of this distinction. Participants were presented with either verbal-spatial or physical-spatial stimuli, simultaneously in the visual and auditory modalities, and were to selectively attend and respond to the location of the cued modality. An initial experiment replicated previously reported effects, with similar patterns of crossmodal congruence effects for visual and auditory verbal-spatial stimuli. Three further experiments directly compared crossmodal congruence patterns for physical-spatial and verbal-spatial stimuli across varying attentional conditions. Intermixing verbal and physical spatial stimulus sets did not meaningfully alter the distinct congruence patterns compared to when the sets were blocked, and biasing attention to verbal-spatial processing amplified the modality-congruence interaction for physical-spatial stimuli. Together, the consistent findings of the modality-congruence interaction showing visual dominance for physical-spatial stimuli but not for verbal-spatial stimuli suggests that the effect is driven by the particular spatial sets based on their sensory properties rather than endogenous attentional mechanisms.</p>