Spatial frequencies affect cuteness perception of infant faces.

Abstract

Cuteness perception is a basic function in social interactions. Most studies focus on the impact of facial elemental features on cuteness ratings, but there are many factors that affect cuteness perception. Spatial frequency (SF) is one of the most important parameters in studies on faces. However, few studies have investigated the impact of SFs on cuteness perception. In this study, 16 images of infant faces with four cuteness levels were selected by a prerating experiment. Using a 7-point Likert scale paradigm, participants were asked to rate the cuteness of infant faces, including one version of broad unfiltered faces and four versions of filtered faces. The results showed that filtered SFs reduced cuteness ratings and that the impact of SFs was related to the cuteness levels of faces. Specifically, faces with low SFs got the lowest cuteness ratings. The ratings of faces with low SFs in neutral cuteness had a greater reduction than that in positive cuteness. In comparison, faces with medium and high SFs obtained relatively high cuteness ratings. However, the ratings in medium SFs were higher than that in high SFs if the cuteness of faces exceeded a certain level. Interestingly, their ratings reduction size increased with the improvement of cuteness levels. These results extend our understanding of the cuteness mechanism from an SF processing perspective.

The construct of cuteness: A validity study for measuring content and evoked emotions on social media

“…Despite some of these models having relatively low accuracy rates (e.g., Makula et al, 2017), research investigating the level of cuteness in images of faces with altered attributes is becoming increasingly popular and its findings have been used to inform product design and consumer research, where specific parameters of cuteness may affect product popularity (e.g., Marcus et al, 2017). Metrics for judging the level of cuteness in images of human and animal faces include self-report scales (e.g., Nittono et al, 2022;Zhou et al, 2022) and discrimination tasks (Nittono et al, 2022). Kindchenschema can even be extended to inanimate, abstract objects such as geometric shapes.…”