摘要：The concept of emotion is a complex neural and psychological phenomenon, central to the organization of human social behavior. As the result of subjective experience, emotions involve bottom-up cognitive styles responsible for efficient adaptation of human behavior to the environment based on salient goals. Indeed, bottom-up cognitive processes are mandatory to discriminate between emotion-cognition interactions. In this regard, a huge number of studies and standardized affective stimuli databases have been developed (ie. IPS, GAPED, NAPS). However, these do not accurately reflect the complex neural system underlying emotional responses nor do they offer a comprehensive framework for researchers. The present article aims at providing a bottom-up validation of affective stimuli that are independent from cognitive processing and control mechanisms, are related to the implicit relevance and evolutionistic significance of stimuli. A subset of 360 images from the original NAPS, GAPED and IAPS datasets was selected in order to proportionally cover the whole dimensional affective space. Among these, using a two-step analysis strategy we identified three clusters of stimuli with similar cognitive responses profile that prompt good, bad, and high rates of false alarms and fast reaction times performance, respectively. Subsequently, MANOVA showed the three clusters differed in terms of items arousal scores and source database. No differences were found among the clusters in terms of valence, luminance scores as well as experimental conditions. The new database, with accompanying ratings and image parameters, allows researchers to select visual stimulus materials that are independent from dimensional/discrete-category theoretical background, and provides them information on the implicit effects triggered by such stimuli.
