I spy with my little eye: An eye-tracking study examining adolescents' attention to alcoholic and non-alcoholic beverages in Instagram stories

Social media users are frequently exposed to alcohol images on Instagram, which in turn influences their own alcohol behaviors. Yet, it is unclear what factors drive attention to alcohol cues. In an eye-tracking study (N = 108; Mage = 16.54), we examined adolescents' attention to Instagram Stories depending on: (a) the type of beverage depicted (beer vs water), (b) the character-product interaction portrayed (CPI: peers in images shown consuming [high CPI] vs holding beverages [low CPI]) and, (c) participant's own susceptibility (high vs low-risk alcohol drinker). Our results illustrated that adolescents allocated an equal amount of attention to beer and water depicted in Instagram images. Furthermore, they devoted more attention to Instagram images wherein peers were shown consuming water and beer (high CPI) compared to those wherein peers were holding these beverages (low CPI). Surprisingly, high-risk alcohol drinkers were more responsive to both beer and water cues than low-risk drinkers. This was particularly the case for Instagram images with high CPI. These findings have implications for how health cues on Instagram are attended to and processed.

Evaluating the use of Instagram images color histograms and hashtags sets for automatic image annotation

Color similarity has been a key feature for content-based image retrieval by contemporary search engines, such as Google. In this study, we compare the visual content information of images, obtained through color histograms, with their corresponding hashtag sets in the case of Instagram posts. In previous studies, we had concluded that less than 25% of Instagram hashtags are related to the actual visual content of the image they accompany. Thus, the use of Instagram images' corresponding hashtags for automatic image annotation is questionable. In this study, we are answering this question through the computational comparison of images' low-level characteristics with the semantic and syntactic information of their corresponding hashtags. The main conclusion of our study on 26 different subjects (concepts) is that color histograms and filtered hashtag sets, although related, should be better seen as a complementary source for image retrieval and automatic image annotation.