Searcharxiv⌕ Search

arXiv subjects

Oshrat Ayalon

Publications and source records attributed to Oshrat Ayalon.

7 recordsLinked to original sources

AI as Anchor: How Interview Sequencing Relates to the Appropriation of Synthetic Stakeholder Voices in Design Education

Designers increasingly use large language models to simulate stakeholder conversations. Yet it remains unclear how AI-Simulated Stakeholders (AIS) are taken up when used alongside human stakeholder engagement. We report a field study in a single-institution, single-semester design course in which 8 teams (N=21) constructed, evaluated, and peer-reviewed an initial AIS. Then, the teams interviewed both AIS and human stakeholders in two sequences: AIS-first or human-first. Drawing on interview transcripts, design artifacts, course-embedded reflections, and post-course qualitative sessions, we examine exploratory patterns associated with these sequencing conditions. We find that students did not describe AIS as a neutral addition to the process. Instead, the order in which AIS and human stakeholders were formally interviewed was associated with different ways of interpreting and using information from the two sources. When the AIS interview came first, it provided a broad but often general frame that students later reconsidered through human interviews. When the AIS interview followed human engagement, AIS was more often used to revisit and validate insights already grounded in lived experience. Students described AIS as an "anchor," but its reported role varied with its position in the formal interview sequence. In the final designs, a larger proportion of prominent human-stakeholder-derived concerns than AIS-derived concerns had corresponding design elements. We contribute: characterization of AIS as a reference point whose reported role varies with its position in the formal interview sequence; exploratory empirical evidence that the implemented interview-order conditions were associated with distinct patterns of interpreting and using stakeholder input; and implications for integrating AIS into design workflows in ways that preserve direct engagement with human stakeholders

cs.HC↗

Lilo: A Multi-Agent Conversational Sandbox for Children's Digital Communication

Children's peer conflicts increasingly play out in group chats, where tone and timing are stripped away, and a hurtful message is easier to send than to say aloud. We present Lilo, a multi-agent conversational system that lets children practice such conversations, so they are better prepared when they encounter them in real digital communication. In Lilo, the child joins two AI-driven peers mid-argument, coached by a mentor character, also named Lilo, through a Stop-Think-Act traffic light. Through an exploratory study with 28 children (aged 9-11), we examine how children engage with Lilo's simulated peer conflict and the mentor's traffic-light interventions. The study included an intervention with Lilo, surveys before and after the intervention, and exit interviews. Participants reported significantly higher self-efficacy for coping with unpleasant messages after their interaction with Lilo. They felt safe experimenting, knowing nothing they wrote would be held against them.

cs.HC↗

From Regulation to Engineering: Organizational Privacy Climate as Mediator

Data protection legislation aims to protect privacy rights by regulating how information systems store and process personal information. While such legislation defines legal obligations, translating regulatory requirements into engineering practice is far from straightforward. As new frameworks such as the EU AI Act emerge, understanding these mechanisms becomes increasingly relevant. To investigate them, we conducted an online survey of 156 U.S. information system developers, examining the role of organizational privacy climate in the association between developers' perceptions of the GDPR and their privacy engineering practices. Using structural equation modeling, we found that organizational privacy climate mediates the relation between developers' perceived GDPR importance and their privacy attitudes, which in turn are associated with their privacy engineering practices. The mediation chain - regulation, climate, attitudes, practices - complements the direct effect of regulation on practice, revealing the organizational and attitudinal pathway through which regulatory intent reaches engineering behavior. We discuss implications for the next generation of privacy and AI regulation.

cs.CY↗

Deep Persona: A Psychologically Grounded Architecture and Evaluation Framework for Role-Playing Agents and Simulations

Existing approaches to persona simulation with Large Language Models (LLMs) mostly rely on shallow character descriptions that fail to sustain coherent character behavior across extended interactions. We introduce Deep Persona, a psychologically grounded, three-layered architecture that organizes personas into hierarchical levels of observable expression, latent beliefs, and core motivational drives, for constructing highly convincing role-playing agents. Governed by the principles of scripted determinism and bounded agency, the architecture restricts the model to a reactive engine guided by a structured internal script. We further propose a reference-free evaluation framework that benchmarks dialogue naturalness against empirical human distributions using established psychological clinical instruments and adversarial stress-tests. Empirical evaluation reveals that while LLMs achieve high pragmatic fluency, they exhibit systematic limitations in emotional expression and joint attention. In addition, we present a case study of two Deep Personas and evaluate them using the proposed framework, demonstrating that structured personas can produce interactions that more closely align with human conversational behavior.

cs.CL↗

Analyzing User Engagement with TikTok's Short Format Video Recommendations using Data Donations

Short-format videos have exploded on platforms like TikTok, Instagram, and YouTube. Despite this, the research community lacks large-scale empirical studies into how people engage with short-format videos and the role of recommendation systems that offer endless streams of such content. In this work, we analyze user engagement on TikTok using data we collect via a data donation system that allows TikTok users to donate their data. We recruited 347 TikTok users and collected 9.2M TikTok video recommendations they received. By analyzing user engagement, we find that the average daily usage time increases over the users' lifetime while the user attention remains stable at around 45%. We also find that users like more videos uploaded by people they follow than those recommended by people they do not follow. Our study offers valuable insights into how users engage with short-format videos on TikTok and lessons learned from designing a data donation system.

cs.SI↗

Likes and Fragments: Examining Perceptions of Time Spent on TikTok

Researchers use information about the amount of time people spend on digital media for numerous purposes. While social media platforms commonly do not allow external access to measure the use time directly, a usual alternative method is to use participants' self-estimation. However, doubts were raised about the self-estimation's accuracy, posing questions regarding the cognitive factors that underline people's perceptions of the time they spend on social media. In this work, we build on prior studies and explore a novel social media platform in the context of use time: TikTok. We conduct platform-independent measurements of people's self-reported and server-logged TikTok usage (n=255) to understand how users' demographics and platform engagement influence their perceptions of the time they spend on the platform and their estimation accuracy. Our work adds to the body of work seeking to understand time estimations in different digital contexts and identifies new influential engagement factors.

cs.CY↗

How mass surveillance can crowd out installations of COVID-19 contact tracing apps

During the COVID-19 pandemic, many countries have developed and deployed contact tracing technologies to curb the spread of the disease by locating and isolating people who have been in contact with coronavirus carriers. Subsequently, understanding why people install and use contact tracing apps is becoming central to their effectiveness and impact. This paper analyzes situations where centralized mass surveillance technologies are deployed simultaneously with a voluntary contact tracing mobile app. We use this parallel deployment as a natural experiment that tests how attitudes toward mass deployments affect people's installation of the contact tracing app. Based on a representative survey of Israelis (n=519), our findings show that positive attitudes toward mass surveillance were related to a reduced likelihood of installing contact tracing apps and an increased likelihood of uninstalling them. These results also hold when controlling for privacy concerns about the contact tracing app, attitudes toward the app, trust in authorities, and demographic properties. Similar reasoning may also be relevant for crowding out voluntary participation in data collection systems.

cs.HC↗