arxiv: 2605.07960 · v1 · submitted 2026-05-08 · 💻 cs.HC

Recognition: no theorem link

Exploring a Virtual Pet to Provide Context Notifications in a Tourism Recommender System: a Pilot Study

Patr\'icia Alves , Joana Neto , Ana Barreiro , Jorge Lima , Fausto Alves , Henish Balu , Lu\'is Concei\c{c}\~ao , Goreti Marreiros

Authors on Pith no claims yet

Pith reviewed 2026-05-11 03:01 UTC · model grok-4.3

classification 💻 cs.HC

keywords virtual petcontext-aware notificationstourism recommender systemsintrusivenessuser acceptancepilot studypersonalized recommendations

0 comments

The pith

A virtual pet can soften the intrusiveness of real-time context alerts in tourism recommender systems.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tests a proof-of-concept tourism recommender that uses a virtual pet to deliver personalized, context-aware notifications drawn from live environmental data such as air quality, noise, and weather. The pet acts as a social mediator that adds friendly, character-based justifications tied to each user's personality traits and preferences. A four-week within-subjects pilot with eleven participants compared the pet version against a baseline without contextual alerts. Early results indicate the pet makes safety-critical alerts feel more natural and welcome while making the reasons for each notification easier to understand.

Core claim

The virtual pet serves as a social mediator that delivers context-aware alerts generated by a multi-agent microservice, thereby reducing perceived intrusiveness of notifications and improving their clarity through character-mediated justifications in real-world tourism scenarios.

What carries the argument

Virtual pet as social mediator, integrated with a multi-agent microservice that personalizes recommendations from personality traits, user preferences, and real-time environmental data.

Load-bearing premise

A within-subjects pilot with only eleven participants over four weeks can supply reliable preliminary evidence of reduced intrusiveness without detailed statistical tests or controls for order effects.

What would settle it

A follow-up study with more participants that measures no difference in intrusiveness ratings or clarity scores between the pet-mediated version and the baseline version would undermine the preliminary claim.

Figures

Figures reproduced from arXiv: 2605.07960 by Ana Barreiro, Fausto Alves, Goreti Marreiros, Henish Balu, Joana Neto, Jorge Lima, Lu\'is Concei\c{c}\~ao, Patr\'icia Alves.

**Figure 6.** Figure 6: Aggregated responses (%, n = 11) to the UEQ-S questionnaire for the baseline (left) and pet-mediated (right) versions. Scale: 1 (negative pole) to 7 (positive pole). Q.1 Obstructive–Helpful, Q.2 Complicated–Easy, Q.3 Inefficient–Efficient, Q.4 Confusing–Clear, Q.5 Boring–Exciting, Q.6 Uninteresting–Interesting, Q.7 Conventional–Inventive, Q.8 Usual–Innovative. Items Q.1–Q.4 measure Pragmatic Quality (PQ); … view at source ↗

**Figure 7.** Figure 7: Aggregated responses to the Pet-mediated notifications scale (n = 11, 1–7). Q.1, Q.3, Q.5, Q.7, Q.9: Utility; Q.2, Q.6, Q.10: Acceptance; Q.4, Q.8, Q.11: Virtual Pet mediation. These qualitative results triangulate the UEQ-S findings: gamification effectively boosts the Hedonic Quality for experience-seekers but can hinder the Pragmatic Quality for efficiency-oriented users. This “conceptual mismatch” rep… view at source ↗

read the original abstract

While context-aware personalization has been widely explored in modern tourism Recommender Systems (RS), the delivery of real-time notifications remains a significant design challenge due to issues of intrusiveness and user fatigue. This paper presents a proof-of-concept for a tourism recommendation framework that utilizes a virtual pet as a social mediator for delivering context-aware alerts. The system integrates real-time environmental data - including air quality, noise levels, and weather forecasts - and proximity-based notifications with a Multi-Agent Microservice that generates personalized recommendations based on the user's personality traits and preferences. A within-subjects pilot study (n=11) was conducted to evaluate the feasibility and user acceptance of this pet-mediated approach. Participants interacted with two versions of the system - a baseline without contextual alerts and a version featuring pet-mediated notifications - over a four-week period (two weeks per version) in real-world scenarios. Quantitative and qualitative data were collected to assess engagement, perceived naturalness, notification utility, and acceptance. Preliminary results suggest that the virtual pet effectively can "soften" the perceived intrusiveness of system alerts, making safety-critical information feel more welcome and natural. Furthermore, the character-mediated justifications significantly improved the clarity of the notifications, effectively supporting users in their real-time travel decisions. These findings provide a foundation for using virtual pet companions to enhance the transparency and acceptance of context-aware communication in tourism RS.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a small pilot on a virtual pet mediating notifications in a tourism recommender, with an interesting interface idea but data too thin to support the reported benefits.

read the letter

Hey, this paper tests a virtual pet as a way to deliver context notifications in a tourism recommender system. The setup pulls real-time air quality, noise, and weather data, combines it with personality-based recommendations through a multi-agent microservice, and has the pet character explain the alerts to users. That specific combination of environmental context, personalization, and a social mediator is new enough to stand out from standard notification work in mobile tourism apps. Running the thing in real-world conditions for four weeks is also a reasonable choice for a feasibility check rather than staying in the lab. The within-subjects pilot with 11 participants comparing a no-alert baseline to the pet version shows they tried to measure engagement, naturalness, and acceptance through both numbers and comments. Those elements give the work some practical grounding. The main weakness is the evaluation. The abstract and summary report no specific scales, no raw data or tables, no statistical tests, no effect sizes, and no mention of counterbalancing or checking for order effects in the crossover design. With n=11 over two weeks per condition, any difference in perceived intrusiveness or clarity could easily come from time on task or novelty instead of the pet itself. The claims that the pet softens alerts and improves clarity therefore rest on unsupported preliminary statements. This is the kind of short idea paper that might interest HCI researchers working on notification design, virtual agents, or context-aware mobile systems. It could spark follow-up experiments for someone already thinking about companion interfaces in recommenders. I would not cite the findings as evidence in their current form. The paper does not yet deserve peer review; the authors need to add the actual measures, analysis, and controls before it would be worth a referee's time.

Referee Report

3 major / 1 minor

Summary. The paper proposes a tourism recommender system that uses a virtual pet as a social mediator to deliver context-aware notifications based on real-time environmental data (air quality, noise, weather) and user personality traits via a Multi-Agent Microservice. It describes a within-subjects pilot study with n=11 participants who used a baseline system without alerts and a pet-mediated version over two 2-week periods each in real-world settings, collecting quantitative and qualitative data on engagement, naturalness, utility, and acceptance. Preliminary results are claimed to show that the virtual pet reduces perceived intrusiveness of alerts and improves notification clarity.

Significance. If substantiated with proper data, the work could provide a novel HCI contribution by demonstrating how anthropomorphic interfaces like virtual pets can improve acceptance of context-aware notifications in mobile tourism applications, addressing a known challenge of intrusiveness and fatigue. The technical integration of environmental sensing with personality-based recommendations is a positive element, but the small pilot nature limits broader impact without further validation.

major comments (3)

The manuscript reports 'preliminary results' suggesting reduced intrusiveness and improved clarity but provides no quantitative data tables, statistical tests (e.g., paired t-tests or Wilcoxon), p-values, effect sizes, or even summary statistics from the questionnaires. This absence makes it impossible to evaluate whether the within-subjects comparison supports the central claims, as noted in the study description.
The within-subjects design (two weeks baseline, two weeks pet-mediated) does not mention counterbalancing of condition order, carry-over effects, or novelty confounds. Without these controls or analysis, the observed differences cannot be reliably attributed to the virtual pet rather than time-on-task or sequence effects.
No details are given on the exact measurement instruments (e.g., specific scales for intrusiveness, naturalness, or acceptance), response rates, or qualitative coding procedures. This lack of methodological transparency undermines the feasibility claims even for a pilot study.

minor comments (1)

Abstract: The phrasing 'effectively can “soften”' is grammatically awkward; consider 'effectively softens' for clarity.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for their constructive feedback on our pilot study. We address each major comment point by point below, acknowledging limitations in the current manuscript and outlining revisions where the comments are valid.

read point-by-point responses

Referee: The manuscript reports 'preliminary results' suggesting reduced intrusiveness and improved clarity but provides no quantitative data tables, statistical tests (e.g., paired t-tests or Wilcoxon), p-values, effect sizes, or even summary statistics from the questionnaires. This absence makes it impossible to evaluate whether the within-subjects comparison supports the central claims, as noted in the study description.

Authors: We acknowledge that the submitted manuscript emphasizes qualitative insights and does not include quantitative tables or statistical tests. As this is a small pilot (n=11), our initial focus was on feasibility rather than inferential statistics. We will revise to include summary statistics (means and standard deviations) for key measures such as perceived intrusiveness and notification clarity. We will also report results from non-parametric tests (e.g., Wilcoxon signed-rank) with p-values and effect sizes where the data permit, to better substantiate the preliminary claims. revision: yes
Referee: The within-subjects design (two weeks baseline, two weeks pet-mediated) does not mention counterbalancing of condition order, carry-over effects, or novelty confounds. Without these controls or analysis, the observed differences cannot be reliably attributed to the virtual pet rather than time-on-task or sequence effects.

Authors: The referee correctly identifies that these design details and potential confounds are not discussed. The study used a fixed order without counterbalancing to enable progressive familiarization in this pilot. We will revise the methods and discussion sections to explicitly describe the design choice, acknowledge risks of novelty effects, carry-over, and time-on-task confounds, and note that no order-effect analysis was conducted due to sample size. This will be framed as a limitation for future work. revision: yes
Referee: No details are given on the exact measurement instruments (e.g., specific scales for intrusiveness, naturalness, or acceptance), response rates, or qualitative coding procedures. This lack of methodological transparency undermines the feasibility claims even for a pilot study.

Authors: We agree that the manuscript lacks sufficient methodological transparency on these points. The quantitative instruments were adapted Likert scales from prior HCI literature on notifications and acceptance, and qualitative data underwent thematic analysis. In revision, we will expand the methods section to list the specific scales and sample items, report the 100% response rate from all 11 participants, and detail the qualitative coding procedure including how themes were identified. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical pilot study with direct observation

full rationale

The paper describes a within-subjects pilot study collecting quantitative and qualitative data on user acceptance of pet-mediated notifications. No equations, fitted parameters, predictions derived from prior fits, or self-citations appear in the provided text. Claims rest on participant feedback over four weeks rather than any reduction of results to inputs by construction, self-definition, or imported uniqueness theorems. The derivation chain is therefore self-contained and non-circular.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The paper relies on standard HCI evaluation assumptions rather than new mathematical constructs. No free parameters are fitted. The virtual pet is introduced as a design element without independent falsifiable evidence beyond the pilot itself.

axioms (1)

domain assumption Self-reported questionnaires and qualitative feedback reliably capture perceived intrusiveness, naturalness, and acceptance in a within-subjects design
The evaluation depends on participant ratings collected after each two-week period.

invented entities (1)

Virtual pet as social mediator for notifications no independent evidence
purpose: To deliver context-aware alerts in a less intrusive and more natural manner
The pet is postulated as the key interface innovation that softens alerts; no external validation outside the pilot is provided.

pith-pipeline@v0.9.0 · 5580 in / 1405 out tokens · 59224 ms · 2026-05-11T03:01:00.216352+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

7 extracted references · 7 canonical work pages

[1]

Context -Aware Recom - mender Systems

[Adomavicius and Tuzhilin 2011] Adomavicius, G., Tuzhilin, A.: “Context -Aware Recom - mender Systems”; Recommender Systems Handbook, Springer (2011), 217–253. [Al Hakim et al. 2024] Al Hakim, V. G., Yang, S. -H., Wang, J. -H., Lin, H. -H., Chen, G. -D.: “Digital twins of pet robots to prolong interdependent relationships and effects on student learning p...

work page doi:10.2139/ssrn.6487258 2011
[2]

The World of Coordinated Play: Same- and Cross-Sex Friend- ship in Young Children

[Gottman 1986] Gottman, J. M.: “The World of Coordinated Play: Same- and Cross-Sex Friend- ship in Young Children”; in: Conversations of Friends: Speculations on Affective Development, Cambridge University Press, New York, NY (1986), 139–191. [Granic et al. 2014] Granic, I., Lobel, A., Engels, R. C. M. E.: “The Benefits of Playing Video Games”; Am. Psycho...

work page 1986
[3]

Interactive and context-aware systems in tourism

[Jannach and Zanker 2020] Jannach, D., Zanker, M.: “Interactive and context-aware systems in tourism”; in: Handbook of e-Tourism, Springer (2020), 1–22. [Junglas and Watson 2008] Junglas, I. A., Watson, R. T.: “Location-based Services”; Commun. ACM 51, 3 (2008), 65–69. [Kim et al. 2016] Kim, S.-K., Kim, S.-Y., Kang, H.-B.: “An analysis of the effects of s...

work page 2020
[4]

Mixed reality-based interaction between human and virtual cat for mental stress management

[Na et al. 2022] Na, H., Park, S., Dong, S.-Y.: “Mixed reality-based interaction between human and virtual cat for mental stress management”; Sensors 22, 3 (2022),

work page 2022
[5]

Advances in real time smart monitoring of environmental parame- ters using IoT and sensors

3015 Alves et al.: A Virtual Pet to Provide Context Notifications in a Tourism RS [Narayana et al. 2024] Narayana, T. L., Venkatesh, C., Kiran, A., Kumar, A., Khan, S. B., Al - musharraf, A., Quasim, M. T.: “Advances in real time smart monitoring of environmental parame- ters using IoT and sensors”; Heliyon 10, 7 (2024). [Nguyen and Nguyen 2026] Nguyen Th...

work page 2024
[6]

An In -situ Study of Mobile Phone Notifications

[Pielot et al. 2014] Pielot, M., Church, K., de Oliveira, R.: “An In -situ Study of Mobile Phone Notifications”; Proc. 16th International Conference on Human-Computer Interaction with Mobile Devices & Services (MobileHCI ’14), ACM (2014), 233–242. [Pitic and Pitic 2022] Pitic, A., Pitic, A.: “The positive effects of video games on the human mind”; Int. J....

work page 2014
[7]

La ética y los animales de compañía: El hogar más allá de la especie

[Sánchez-Moreno 2025] Sánchez-Moreno, I.: “La ética y los animales de compañía: El hogar más allá de la especie”; Metode Sci. Stud. J. 15, 5 (2025), e28372. [Sahami et al. 2014] Sahami Shirazi, A., Henze, N., Dingler, T., Pielot, M., Weber, D., Schmidt, A.: “Large -scale Assessment of Mobile Notifications”; Proc. SIGCHI Conference on Human Factors in Comp...

work page arXiv 2025