The article contributed by Yury Shevchenko, Jakob Heisenstein, and Ulf-Dietrich Reips [32] titled “How do avatars make a positive impression: The effect of facial cues on avatar evaluation” investigates the influence of specific facial features, namely, eye size, jaw shape, and hairstyles on human participants’ perception of static avatars.

Following the lead of previous human face and avatar perception literature, the authors are interested in participants’ perception of attractiveness, trustworthiness, and personality characteristics, such as perceived sympathy, hostility, and dominance. The rationale behind this investigation is that better understanding and quantifying users’ perception of features like “trustworthiness” would be relevant to the design of automated systems and would inform developing technologies concerning the willingness of users to interact with avatars in, for example, user interfaces created for AI systems in financial banking or insurance sectors; virtual applications in which the sharing of confidential information is required; and other virtual e-commerce and gaming environments. Thus, the studies investigate an optimization challenge that is central to the effective design and use of automated systems.

The work systematically manipulates facial feature parameters by varying the size of avatar eyes and the shape of avatar jawlines as well as altering hairstyles on a dimension of plain (“neat”) to stylish (“spiky”). Avatar gender is also assessed as avatars were evaluated in both male and female forms, and interactions between assessed factors are examined. Results based on participants’ interactions with avatars whose features were systematically manipulated support three hypotheses proposed by the authors. Namely, participant data revealed that large eyes yield a more positive reaction, square jawlines yield higher impressions of dominance, and stylish hair is linked to higher impressions of creativity.

The implications of this research inform the creation and use of avatars, and make significant steps towards forming “best practices” for a wide range of user interfaces that aim to employ avatars in virtual environments in which human users engage with AI assistants for the purposes of information gathering and collaboration.

The article contributed by Dietmar Saupe and Simon Hviid Del Pin [31] titled “Uncovering cultural influences on perceptual image and video quality assessment through adaptive quantized metric models” delves into the study of national differences in perceptual image and video quality assessments using advanced statistical models on large-scale crowdsourced datasets. The article presents a compelling exploration of how national differences influence perceptual assessments of image and video quality by using large-scale crowdsourced datasets and advanced statistical models, which combine into a significant contribution to the understanding of global perception variations in media quality.

The research employs methodological rigor, in the form of both frequentist and Bayesian approaches, to uncover country-specific nuances in quality evaluations. The study highlights significant cultural variations, particularly contrasting responses from Japan and the United States, underscoring the importance of cultural context in interpreting rating scales like Absolute Category Rating and Degradation Category Rating. By integrating complex statistical models and considering factors such as lapse rates and quantized metrics, the work contributes a novel perspective to cross-cultural multimedia studies.

The innovative findings presented not only enhance the understanding of global perception variations but also suggest practical implications for improving quality assessment methodologies in industries like video streaming. The article is notable for its substantial methodological and substantive contributions, advancing the discourse on cultural influences in media quality assessments.

The article contributed by Maria Rosa Miccoli and Ulf-Dietrich Reips [16] titled “Exact versus conceptual replication: Internet-based research investigating the replicability of cognitive effects” investigates the usefulness of Internet-based experiments as replication studies, for both laboratory-based and online research, by comparing empirical research results from studies conducted through remote empirical research platforms. The article presents analyses that distinguish the subtleties between exact and conceptual replications using experiments that focus on examining the mental accounting effect [8]. In the experiments presented, they directly investigate exact empirical replication (Experiment 1, N: 410) and assess chronological cultural shifts while examining conceptual replications (Experiments 2 and 3, Ns respectively: 270, 365).

Conceptual replications address the need to adapt to historical changes, which happen even to very basic and economical facts like the value of money. Failing to do so (as required in an exact replication to the letter) results in a much smaller effect, as they show in Experiment 1. The authors then adapt the $10 price used in the original experiment to the current value of money, which is more than fourfold. The two conceptual replications show effects of increased size that are more comparable to the original findings.

Through this research, the authors further identify, unpack, and examine important features inherent in the premises that underlie both replication and comparison of behavioral research findings from laboratory settings with those derived from remote assessment methods. The findings of this research raise important general implications for Internet-based research investigations and their comparability to lab-based research results.

In their article, “How Internet habits influence social cognitive research: Theoretical and methodological considerations in the era of social media,” Jonathan D. Kim, Pascal Gygax, and Ute Gabriel [9] explore the profound ways Internet use has reshaped social cognition and its implications for research. They argue that the Internet’s central role in communication and information processing has redefined cognitive processes like memory, attention, and social interaction. Social media algorithms, they note, amplify biases and create “filter bubbles,” influencing users’ beliefs and behaviors while fostering polarization. The authors also examine the potential of virtual reality as a tool for controlled social experiments, highlighting both its opportunities and risks. They emphasize that these shifts challenge traditional distinctions between “digital natives” and “digital immigrants” [19], as individual use patterns increasingly shape cognitive profiles.

Methodologically, the article underscores the need for balancing external (“ecological”) validity with experimental control in Internet-based research (see also [22, 23]). Although online platforms provide naturalistic settings, they introduce variability due to differences in participants’ devices and multitasking behaviors. The authors recommend strategies like purposive sampling, controlling for algorithmic engagement, and leveraging digital laboratories or AI avatars to enhance data quality. They also address challenges in studying group membership online, such as anonymity and misrepresentation, particularly in minority groups. Ultimately, the article provides actionable guidance for researchers to navigate these complexities while advancing the remote study of social cognition in a digitally connected world.