Behavioral and neural correlates of domain-general object recognition ability

Abstract

Recent research has found evidence of a domain-general object recognition ability, called o, which is distinct from general intelligence and other cognitive and personality constructs. In a first study, we used latent variable modeling to examine how this ability relates to the ability to recognize familiar objects. We found that novel and familiar object recognition factors perfectly correlated, suggesting that o underlies recognition of both novel and real-world object categories. In a second study, we explored potential neural correlates of o using an adaptation approach with functional magnetic resonance imaging. We found that several regions in the temporal and parietal lobes, including three independently-localized object-selective and one face-selective region, positively correlated with o. Correlations between the neural signals in these regions suggested common variance in distributed neural correlates is predictive of domain-general object recognition ability. Together, these two studies establish o as a construct with real-world relevance and measurable neural correlates, laying the groundwork for future studies of o.