1 SCIENTIFIC DATA | (2021) 8:272 | https://doi.org/10.1038/s41597-021-01053-z www.nature.com/scientificdata Updating the Chieti Afective Action Videos database with older adults Pasquale La Malva 1,5 , Irene Ceccato 2,3,5 , Adolfo Di Crosta 2 , Anna Marin 4 , Mirco Fasolo 2 , Riccardo Palumbo 2,3 , Nicola Mammarella 1 , Rocco Palumbo 1 ✉ & Alberto Di Domenico 1 Validation of the Chieti Afective Action Videos (CAAV) database was replicated with a sample of older adults (age range 65–93). When designing experimental studies of emotions, it is crucial to take into consideration the diferences in emotional processing between young and older adults. Therefore, the main goal of the present study was to provide an appropriate dataset for the use of CAAV in aging research. For this reason, the CAAV administration and the data collection methodology was faithfully replicated in a sample of 302 older adults. All the 360 standardized stimuli were evaluated on the emotional dimensions of valence and arousal. The CAAV validation in an older adults’ population increases the potential use of this innovative tool. The present validation supports the use of the CAAV database in future experimental studies on cognitive functions in healthy and pathological aging. Background & Summary Te Chieti Afective Action Videos (CAAV) 1 is an innovative database of movie clips developed specifcally for experimental research in psychology. Te CAAV comprises a large number of emotional action videos rated for valence and arousal. Te action videos are homogeneous in terms of length (15 seconds), brightness, and camera angle. All the videos include everyday life actions. Crucially, the innovative aspect of this tool consists in controlling two factors: a) the gender of the actors performing the action; and b) the point of view (POV) through which the action is carried out. In particular, an actor and an actress performed the same 90 actions both in frst-person POV and in third-person POV, for a total of 360 emotional action videos. For each stimulus, the CAAV provides an emotional rating based on arousal and valence scores. Indeed, the database validation is based on the Dimensional Model of Emotions 2,3 and specifcally on the the circumplex model of afect 4 . Briefy, this model postulates that emotions can be identifed based on their location along two dimensions: valence and arousal. Te dimension of valence (i.e., pleasantness) diferentiates positive (pleasant) from negative (unpleasant) emotional states. Te dimension of arousal (i.e., activation) diferentiates highly exciting and arousing states from calm and relaxing states. Te rating of the CAAV database is based on this dimensional approach and provides a continuous and balanced distribution of the stimuli across valence and arousal. As a result, the CAAV allows the identifcation of action videos with intermediate scores on both dimensions, which can be classifed as emo- tionally neutral. Te CAAV has innovative features when compared to previous emotional databases 5–7 . Several databases explored the emotional dimensions with static and dynamic stimuli (e.g., words 8 , pictures 9 , sounds 10 , faces 11 and movie clips 12 ). Among those using dynamic emotional stimuli, many of them consists of a collection of stimuli extracted from movie scenes, which reduces the possibility to standardize (or experimentally control) specifc features (e.g., duration, brightness, and camera angle). On the contrary, the CAAV’s movie clips were tailored to maintain constant these features. In addition, each action of the CAAV was performed by a female and a male actor in both frst and third person POVs. Terefore, the CAAV provides a gender and POV balanced material. Tese features make the CAAV expressly well-suited to analyze the role of POV on individuals’ emo- tional response 13 . Furthermore, the CAAV may be especially useful to explore and avoid evaluation bias due to the actor’s gender 14,15 . Tese characteristics make the CAAV a highly ecological and immersive tool. In the previous validation of the CAAV, we included only young adults. Terefore, we neglected age-related diferences in valence and arousal ratings of the emotional stimuli 16,17 . Indeed, studies showed that older adults 1 Department of Psychological, Health and Territorial Sciences (DiSPUTer), University G. d’Annunzio - Via dei Vestini, 31 - 66100, Chieti, Italy. 2 Department of Neuroscience, Imaging and Clinical Science, University G. d’Annunzio - Via dei Vestini, 31 - 66100, Chieti, Italy. 3 Behavioral Economics and Neuroeconomics, Center of Advanced Studies and Technology (CAST), G. d’Annunzio University of Chieti-Pescara, Chieti, 66100, Italy. 4 Department of Neurology, Boston University, 150 South Huntington Avenue, Boston, MA, 02130, USA. 5 These authors contributed equally: Pasquale La Malva, Irene Ceccato. ✉ e-mail: rocco.palumbo@unich.it DATA DeSCRIPToR oPeN