arXiv subjects

Daniela Fernandes

Publications and source records attributed to Daniela Fernandes.

7 recordsLinked to original sources

arXiv2026

Conversations in Space: Non-Linear LLM Interaction in Everyday Use

As LLM conversations grow, their histories capture alternative directions, decisions, and evolving lines of thought that can be difficult to navigate through chat alone. We investigate an interaction concept that represents the same conversation through two synchronized views: a familiar linear chat for ongoing dialogue and a spatial canvas for navigating its emerging structure. To investigate this interaction concept, we developed CanvasConvo, which allows conversations to branch into alternative paths that remain accessible across both views. In a five-day field deployment with 24 participants, we examined how people appropriated this parallel representation in self-directed knowledge work. Participants selectively moved between the two views rather than replacing chat with the canvas. Chat remained central to conversational interaction, while the canvas supported overview, revisitation, and exploration of alternatives. Adoption was uneven, revealing challenges around established chat habits, transitions between representations, and understanding branch context. Our findings inform the design of user interfaces for LLMs that combine linear and non-linear conversation representations.

cs.HC↗

arXiv2026

Available but Unclaimed: An Empirical Study of Human-AI Synergy

People increasingly reason with large language models (LLMs), yet complementary capabilities do not guarantee outperforming both components. In a between-subjects study, participants (N=535) solved a 40-item battery of matrix reasoning, mental rotation, syllogisms, and letter-string analogies, unaided or with GPT-5.6-Luna, Claude Opus 4.8, Gemini 3.6 Flash, or Kimi K3. Each assisted trial required consultation with the model. Each model answered every item alone 100 times under matched elicitation. The assisted-unaided accuracy difference increased with item-level LLM competence. Deference varied across tasks and increased with competence within tasks. Post-advice confidence distinguished correct from incorrect answers less strongly than unaided confidence. In a reference comparison, about half the increase in LLM accuracy carried through to assisted accuracy. How much of that accuracy gain reached participants differed across the models. These findings motivate evaluating LLMs in interaction with humans and designing support for selective deference that preserves independent reasoning.

cs.HC↗

arXiv2026

Beyond "ChatGPT Can Make Mistakes": Designing Interventions to Support Metacognitive Monitoring in AI-Assisted Work

AI assistance places a metacognitive demand on users, who must judge their own competence and the system's. Yet designers lack comparative evidence on which interventions to choose, where to place them, and how to tell whether they worked. We elicited 30 interventions from 11 experts and, with prior work, organized them into a design space of time (when an intervention acts), level (whose competence is judged), and source (who supplies the monitoring cue). A between-subjects experiment (N = 917; 12 planning-and-organizing problems) compared a per-task reliability card, contrasting replies, pause points, and post-problem reflection against a baseline LLM assistant. Reliability cards and contrasting replies reduced estimation error and overconfidence and increased aggregate confidence discrimination. No task-performance improvement or average within-item discrimination gain was established. We contribute a shared vocabulary, a design space, and evidence that measured monitoring and task performance are separable design targets.

cs.HC↗

arXiv2026

A Deep Learning Model of Mental Rotation Informed by Interactive VR Experiments

Mental rotation -- the ability to compare objects seen from different viewpoints -- is a fundamental example of mental simulation and spatial world modeling in humans. Here we propose a mechanistic model of human mental rotation, leveraging recent advances in deep, equivariant, and neuro-symbolic learning. Our model consists of three stacked components: (1) an equivariant neural encoder, producing 3D spatial representations of objects from images, (2) a neuro-symbolic object encoder, deriving symbolic objects descriptions from these spatial representations, and (3) a neural decision agent, comparing these symbolic descriptions to prescribe rotation simulations in 3D latent space via a recurrent pathway. Our model design is guided by the existing experimental literature on mental rotation, which we complemented with experiments in VR where participants could at times manipulate the objects to compare. Our model captures well the performance, response times and behavior of participants in our and others' experiments, and through ablation studies we demonstrate the necessity of each component. Our work adds to a recent collection of deep neural models of human spatial reasoning, further demonstrating the potency of integrating deep, equivariant, and symbolic representations to model the human mind.

q-bio.NC↗

arXiv2026

Explaining Too Much? Understanding How Large Language Model Reasoning Traces Influence Performance and Metacognition

Large Language Model interfaces are increasingly verbose, exposing intermediate reasoning traces alongside final answers. Traces are framed as transparency mechanisms, yet it is unclear how people use them to solve problems. We report a preregistered between-subjects study (N = 559) in which participants solved ten LSAT-style reasoning problems under one of three conditions: an Answer-only baseline, a Full-trace revealed before the answer, and a Summary-trace presented alongside the answer. Summaries preserved task performance at the no-trace baseline while significantly elevating trust and hedonic appeal, establishing that trace exposure shifts subjective appraisal of the interaction without bringing performance benefits. Under an open-weight reasoning model exposing verbose intermediate output, full traces additionally impaired performance relative to the answer-only baseline. Across all conditions, participants substantially overestimated their performance, and no trace format supported calibrated self-evaluation. Further analysis indicates that hedonic appeal, not trust, carries the indirect path to overestimation, consistent with a processing-fluency account. Reasoning traces are best understood as user-facing interface artifacts rather than transparent windows into model cognition, and calibration is unlikely to emerge from the traces themselves and may best be scaffolded by interactions that elicit users' own reasoning first.

cs.HC↗

arXiv2026

The AI Memory Gap: Users Misremember What They Created With AI or Without

As large language models (LLMs) become embedded in interactive text generation, disclosure of AI as a source depends on people remembering which ideas or texts came from themselves and which were created with AI. We investigate how accurately people remember the source of content when using AI. In a pre-registered experiment, 184 participants generated and elaborated on ideas both unaided and with an LLM-based chatbot. One week later, they were asked to identify the source (noAI vs withAI) of these ideas and texts. Our findings reveal a significant gap in memory: After AI use, the odds of correct attribution dropped, with the steepest decline in mixed human-AI workflows, where either the idea or elaboration was created with AI. We validated our results using a computational model of source memory. Discussing broader implications, we highlight the importance of considering source confusion in the design and use of interactive text generation technologies.

cs.HC↗

arXiv2025

Performance and Metacognition Disconnect when Reasoning in Human-AI Interaction

Optimizing human-AI interaction requires users to reflect on their own performance critically. Our paper examines whether people using AI to complete tasks can accurately monitor how well they perform. In Study 1, participants (N = 246) used AI to solve 20 logical problems from the Law School Admission Test. While their task performance improved by three points compared to a norm population, participants overestimated their performance by four points. Interestingly, higher AI literacy was linked to less accurate self-assessment. Participants with more technical knowledge of AI were more confident but less precise in judging their own performance. Using a computational model, we explored individual differences in metacognitive accuracy and found that the Dunning-Kruger effect, usually observed in this task, ceased to exist with AI. Study 2 (N = 452) replicates these findings. We discuss how AI levels metacognitive performance and consider consequences of performance overestimation for interactive AI systems enhancing cognition.

cs.HC↗