Learning general-purpose user representations based on user behavioral logs is an increasingly popular user modeling approach. It benefits from easily available, privacy-friendly yet expressive data, and does not require extensive re-tuning of the upstream user model for different downstream tasks. While this approach has shown promise in search engines and e-commerce applications, its fit for instant messaging platforms, a cornerstone of modern digital communication, remains largely uncharted. We explore this research gap using Snapchat data as a case study. Specifically, we implement a Transformer-based user model with customized training objectives and show that the model can produce high-quality user representations across a broad range of evaluation tasks, among which we introduce three new downstream tasks that concern pivotal topics in user research: user safety, engagement and churn. We also tackle the challenge of efficient extrapolation of long sequences at inference time, by applying a novel positional encoding method.

Pictorial emojis and stickers are commonly used in online social communications. We analyzed social communications using Bitmoji stickers, which are expressive pictorial stickers made from avatars resembling actual users. We collect a large-scale dataset of 3 billion Bitmoji stickers’ metadata, shared among 300 million Snapchat users. We find that individual Bitmoji sticker usage patterns can be characterized jointly on dimensions of reciprocity and selectivity. Generally speaking, users are either both reciprocal and selective about whom they use Bitmoji stickers with or neither reciprocal nor selective. We additionally demonstrate network homophily by showing that friends use Bitmoji stickers at similar rates. Finally, using a quasi-experimental approach, we show that receiving Bitmoji stickers from a friend encourages future Bitmoji sticker usage and overall Snapchat engagement. Our work carries implications for a better understanding of online pictorial communication behaviors.

Culture shapes people’s behavior, both online and offline. Surprisingly, there is sparse research on how cultural context affects network formation and content consumption on social media. We analyzed the friendship networks and dyadic relations between content producers and consumers across 73 countries through a cultural lens in a closed-network setting. Closed networks allow for intimate bonds and self-expression, providing a natural setting to study cultural differences in behavior. We studied three theoretical frameworks of culture - individualism, relational mobility, and tightness. We found that friendship networks formed across different cultures differ in egocentricity, meaning the connectedness between a user’s friends. Individualism, mobility, and looseness also significantly negatively impact how tie strength affects content consumption. Our findings show how culture affects social media behavior, and we outline how researchers can incorporate this in their work. Our work has implications for content recommendations and can improve content engagement.

The present research was aimed at investigating whether human-robot interaction (HRI) can be improved by a robot's nonverbal warning signals. Ideally, when a robot signals that it cannot guarantee good performance, people could take preventive actions to ensure the successful completion of the robot's task. In two experiments, participants learned either that a robot's gestures predicted subsequent poor performance, or they did not. Participants evaluated a robot that uses predictive gestures as more trustworthy, understandable, and reliable compared to a robot that uses gestures that are not predictive of their performance. Finally, participants who learned the relation between gestures and performance improved collaboration with the robot through prevention behavior immediately after a predictive gesture. This limits the negative consequences of the robot's mistakes, thus improving the interaction.