When people express uncertainty in online discourse (hedging claims, acknowledging limitations, or questioning their own positions), does this shape how others perceive the conversation? And can LLMs detect these social signals the way humans do? We address these questions by comparing human (N = 122) and LLM assessments of identical AI-generated social media posts about Universal Basic Income. To enable direct paired comparison, we developed calibrated "mirror personas"that replicate individual participants' rating tendencies. Both rater types reliably distinguished polarized from moderate discourse across four constructs (uncertainty, emotionality, group salience, perceived polarization), though LLMs amplified condition differences by over sixfold in the most extreme cases, with the degree of amplification varying by construct and model family. Crucially, structural equation modeling revealed divergent processing architectures: humans exhibited integrative processing where perceived uncertainty directly reduced perceived polarization, even when messages varied in emotionality. This pattern, consistent with dual-process theories where uncertainty signals trigger deliberative evaluation, was absent in LLMs, which processed each construct independently without cross-dimensional mediation. Cross-validation confirmed this architectural difference. These findings suggest that expressed uncertainty serves a social function in human discourse comprehension, dampening polarization judgments in ways that current LLMs do not replicate. For AI-mediated communication systems, this gap implies that simply detecting uncertainty is insufficient; the challenge lies in modeling how uncertainty reshapes interpretation of the broader message.

The rise of social media has fundamentally transformed how people engage in public discourse and form opinions. While these platforms offer unprecedented opportunities for democratic engagement, they have been implicated in increasing social polarization and the formation of ideological echo chambers. Previous research has primarily relied on observational studies of social media data or theoretical modeling approaches, leaving a significant gap in our understanding of how individuals respond to and are influenced by polarized online environments. Here we present a novel experimental framework for investigating polarization dynamics that allows human users to interact with LLM-based artificial agents in a controlled social network simulation. Through a user study with 122 participants, we demonstrate that this approach can successfully reproduce key characteristics of polarized online discourse while enabling precise manipulation of environmental factors. Our results provide empirical validation of theoretical predictions about online polarization, showing that polarized environments significantly increase perceived emotionality and group identity salience while reducing expressed uncertainty. These findings extend previous observational and theoretical work by providing causal evidence for how specific features of online environments influence user perceptions and behaviors. More broadly, this research introduces a powerful new methodology for studying social media dynamics, offering researchers unprecedented control over experimental conditions while maintaining ecological validity.

We propose the concept of meta-intents which represent high-level user preferences related to the interaction and decision-making in conversational recommender systems (CRS) and present a questionnaire instrument for measuring meta-intents. We conducted a two-stage user study, an exploratory study with 212 participants on Prolific, and a confirmatory study with 394 participants on Prolific. We obtained a reliable and stable meta-intents questionnaire with 22 question items, corresponding to seven latent factors (concepts). These seven factors cover important interaction preferences and are closely related to users’ decision-making process. For example, the factor dialog-initiative reflects whether users prefer to follow the system’s guidance or ask their own questions in a CRS. We conducted statistical analyses of meta-intents in two domains (smartphones and hotels), and a general chatbot scenario. We also investigated the influence of additional factors (demography, decision-making style) on meta-intents through Structural Equation Modeling (SEM). Our results provide preliminary evidence that the proposed meta-intents are domain and demography (gender, age) independent. They can be linked to the general decision-making style and can thus be instrumental in translating general decision-making factors into more concrete design guidance for CRS and their potential personalization. Meta-intents also provide a basis for future analyses of interaction behavior in CRS and the development of a cognitively founded theoretical framework.

Echo chambers are social phenomena that amplify agreement and suppress opposing views in social media which may lead to fragmentation and polarization of the user population. In prior research, echo chambers have mainly been modeled as a result of social information diffusion. While most scientific work has framed echo chambers as a result of epistemic imbalances between polarized communities, we argue that members of echo chambers often actively discredit outside sources to maintain coherent world views. We therefore argue that two different types of echo chambers occur in social media contexts: Epistemic echo chambers create information gaps mainly through their structure whereas ideological echo chambers systematically exclude counter-attitudinal information. Diversifying recommendations by simply widening the scope of topics and viewpoints covered to counteract the echo chamber effect may be ineffective in such contexts. To investigate the characteristics of this dual echo chamber view and to assess the depolarizing effects of diversified recommendations, we apply an agent-based modeling approach. We rely on knowledge graph embedding techniques not only to generate recommendations, but also to show how to utilize logical graph queries in embedding spaces to diversify recommendations aimed at challenging polarization in online discussions. The results of our evaluation indicate that counteracting the two different types of echo chambers requires fundamentally different diversification strategies.

Recommender Systems have seen substantial progress in terms of algorithmic sophistication recently. Yet, the systems mostly act as black boxes and are limited in their capacity to explain why an item is recommended. In many cases recommendations methods are employed in scenarios where users not only rate items, but also convey their opinion on various relevant aspects, for instance by the means of textual reviews. Such user-generated content can serve as a useful source for deriving explanatory information to increase system intelligibility and, thereby, the user's understanding.

Trust in a Recommender System (RS) is crucial for its overall success. However, it remains underexplored whether users trust personal recommendation sources (i.e. other humans) more than impersonal sources (i.e. conventional RS), and, if they do, whether the perceived quality of explanation provided account for the difference. We conducted an empirical study in which we compared these two sources of recommendations and explanations. Human advisors were asked to explain movies they recommended in short texts while the RS created explanations based on item similarity. Our experiment comprised two rounds of recommending. Over both rounds the quality of explanations provided by users was assessed higher than the quality of the system's explanations. Moreover, explanation quality significantly influenced perceived recommendation quality as well as trust in the recommendation source. Consequently, we suggest that RS should provide richer explanations in order to increase their perceived recommendation quality and trustworthiness.

User studies are increasingly considered important in research on recommender systems. Although participants typically cannot consume any of the recommended items, they are often asked to assess the quality of recommendations and of other aspects related to user experience by means of questionnaires. Not being able to listen to recommended songs or to watch suggested movies, might however limit the validity of the obtained results. Consequently, we have investigated the effect of consuming suggested items. In two user studies conducted in different domains, we showed that consumption may lead to differences in the assessment of recommendations and in questionnaire answers. Apparently, adequately measuring user experience is in some cases not possible without allowing users to consume items. On the other hand, participants sometimes seem to approximate the actual value of recommendations reasonably well depending on domain and provided information.

In user studies of recommender systems, participants typically cannot consume the recommended items. Still, they are asked to assess recommendation quality and other aspects related to user experience by means of questionnaires. Without having listened to recommended songs or watched suggested movies, however, this might be an error-prone task, possibly limiting validity of results obtained in these studies. In this paper, we investigate the effect of actually consuming the recommended items. We present two user studies conducted in different domains showing that in some cases, differences in the assessment of recommendations and in questionnaire results occur. Apparently, it is not always possible to adequately measure user experience without allowing users to consume items. On the other hand, depending on domain and provided information, participants sometimes seem to approximate the actual value of recommendations reasonably well.

Recurrent Neural Networks are powerful tools for modeling sequences. They are flexibly extensible and can incorporate various kinds of information including temporal order. These properties make them well suited for generating sequential recommendations. In this paper, we extend Recurrent Neural Networks by considering unique characteristics of the Recommender Systems domain. One of these characteristics is the explicit notion of the user recommendations are specifically generated for. We show how individual users can be represented in addition to sequences of consumed items in a new type of Gated Recurrent Unit to effectively produce personalized next item recommendations. Offline experiments on two real-world datasets indicate that our extensions clearly improve objective performance when compared to state-of-the-art recommender algorithms and to a conventional Recurrent Neural Network.

To increase transparency and interactive control in Recommender Systems, we extended the Matrix Factorization technique widely used in Collaborative Filtering by learning an integrated model of user-generated tags and latent factors derived from user ratings. Our approach enables users to manipulate their preference profile expressed implicitly in the (intransparent) factor space through explicitly presented tags. Furthermore, it seems helpful in cold-start situations since user preferences can be elicited via meaningful tags instead of ratings. We evaluate this approach and present a user study that to our knowledge is the most extensive empirical study of tag-enhanced recommending to date. Among other findings, we obtained promising results in terms of recommendation quality and perceived transparency, as well as regarding user experience, which we analyzed by Structural Equation Modeling.