Mapping mental representation of arachnophobia stimuli through similarity modeling

Understanding the mental representation of anxiety-related stimuli is essential for elucidating the mechanisms underlying anxiety disorders. Traditional approaches have relied on predetermined psychological dimensions to model emotional responses, which may constrain flexibility and overlook important facets of mental representations. In this study, we focused on arachnophobia and adopted a similarity-based approach to flexibly investigate underlying mental representations of spider-relevant images. Drawing on data from 705 individuals spanning the full spectrum of spider fear, we combined exploratory and confirmatory samples covering 314 spider-relevant images and 48516 unique pairwise comparisons. To our knowledge, this is the largest dataset of similarity judgments in the context of specific phobias. Using Multidimensional Scaling (MDS), we identified a five-dimensional solution that captures perceptual dimensions of spider images. Importantly, dimension 1, along which images with comparable spider size tended to cluster together, significantly predicted fear ratings in a regression analysis, indicating that size is a key determinant in the perception of spider-related fear. Images were also grouped along the other dimensions according to features such as environmental context, real versus artificial spiders, and the prominence of spider legs. Comparisons between individuals with high and low fear of spiders revealed highly similar mental representation structures across groups, with significant correlations observed across all dimensions. These findings suggest that mental representations of fear-relevant stimuli are structured along specific dimensions containing complex information. Our approach provides a scalable and flexible framework for studying anxiety-related representations, offering potential applications for refining exposure therapy and anxiety treatment strategies.