An Empirical Study on Pointing Gestures Used in Communication in Household Settings

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper aims to develop a gesture-understanding system that allows to better interpret human instructions in household robotics settings. To further improve the quality of the paper, the following suggestions are put forward:

1. Under different races, cultural environments and physiological conditions, the gestures commonly used by people vary greatly. Although the 34 volunteers in this paper took into account different genders and ages, they were basically volunteers in the same campus who used the same native language. Therefore, the experiments and data in the paper lack comparisons across cultural and physiological characteristics. It is suggested that the author, where conditions permit, add tests for people of different nationalities and cultures (such as international students on campus) in the paper. Furthermore, tests for people with hearing and speech impairments (the deaf and mute) are also necessary because they use gestures more frequently.
2. There are no figures in the paper indicating the gestures displayed by the test subjects, which makes the analysis and result presentation of the paper seem very unintuitive

3. The color matching and process structure display effect of Figure 3 in the paper is poor. It is suggested to optimize it and provide specific explanations in the paper.

Author Response

1. Cultural and Physiological Diversity in Participants:
   We appreciate your insightful suggestion regarding the inclusion of participants from diverse cultural and physiological backgrounds. While our current study focused on native Polish speakers to control for cultural variability, we fully acknowledge the importance of cross-cultural comparisons. We agree that hearing impared groups are important but deaf participants require seperate specialized protocols. Due to logistical constraints of a single-site study, we prioritized foundational insights for typical users. In the revised manuscript, we have expanded the Limitations and Future Research section to highlight this gap and propose future studies involving international participants. 

2. Visual Representation of Gestures:
   Thank you for pointing out the lack of visual examples. In response, we added Figure 7, which showcases exemplar gestures from the dataset with their corresponding codes (e.g., one-handed pointing with the index finger). 

3. Optimization of Figure 3:
   We have revised Figure 3 to improve clarity, using distinct colors and a clearer layout to represent the four task phases. The caption now includes a detailed explanation.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear Authors,

Please find below my comments:

The manuscript is well-structured and supported by a relevant theoretical foundation, combining computational analyses and manual annotations to characterize participants' gestural behavior.

Although the article presents substantial merits, there are aspects that warrant constructive criticism aimed at scientific improvement. Below, I highlight some weaknesses that do not compromise the overall validity of the study but reduce its impact or clarity:

Firstly, the sample of 34 participants, while reasonable for exploratory studies, is limited in terms of representativeness and generalizability. Trend analyses based on gender and age, for instance, should be interpreted with caution given the sample imbalance (only 7 men) and lack of cultural diversity. This limitation is acknowledged by the authors, but its implications could be more critically addressed in the discussion.

Secondly, the fact that participants performed the tasks in English, despite being all native Polish speakers, may have introduced an uncontrolled linguistic variable. Although they were required to pass a proficiency test, there is no analysis of how language proficiency might have affected gestural or verbal expressiveness—an important consideration in multimodal studies.

The analysis of gesture informativeness based on information theory, though innovative, lacks a deeper interpretation of its pragmatic significance. Shannon’s metric is treated as a useful quantitative index, but there is insufficient articulation between the numerical values obtained and their communicative or cognitive implications.

Another point to consider is that the cross-validation between automatic and manual analysis is only superficial. Although both approaches indicate stability in gestural styles, discrepancies, quantitative convergences, or more analytical complementarities between the two methods are not explored.

Finally, the results section presents excessive data in certain parts, with extensive tables and figures that could be summarized or placed in appendices. This affects the readability and dilutes the focus on the main findings.

Therefore, although the article is solid, these limitations highlight real opportunities for improvement, which justify the recommendation of acceptance with minor revisions, especially focused on deepening interpretation and editorial refinements.

Author Response

1. Sample Size and Representativeness:
   We agree that the gender imbalance (7 men vs. 27 women) limits generalizability. In the revised manuscript, we added a justification for the sample size in the Participants section, citing prior gesture research with similar samples. We also critically discuss this in the Limitations section, and we revised the statement of gender-based analysis in the Results section.

2. Language Proficiency:
   We now address the potential impact of using English (L2) in the Limitations section. We have also included the results of the A1 English test, which given the close to perfect results precluded from further investigating any effects.
   
   

3. Pragmatic Interpretation of Shannon’s Metric:
   To clarify the significance of information values, we added a pragmatic interpretation in Section 4.1.1:
   "From a pragmatic perspective, high-I(x) gestures function as signals that convey novel or disambiguating information, while low-I(x) gestures represent predictable signals that only reinforce existing context."

4. Cross-Validation of Methods:
   We deepened the discussion of computational vs. manual analysis in Conclusions, emphasizing their complementary results. 

5. Data Presentation:
   We streamlined and removed excessive figures from 4.2.2.

Reviewer 3 Report

Comments and Suggestions for Authors

This study empirically explored the pointing gestures used in communication in household settings, which contributes to effective human-robot interaction. The reviewer has following questions for the authors to address in the manuscript.

1. How the four tasks were determined?
2. Please proofread the manuscript and avoid typos, like “\text{tsfresh}”.
3. Please discuss why the sample size is sufficient for conducting the statistical analysis?
4. Please discuss what is the implication for human-robot communication base on the findings/conclusions from this study?

 

Author Response

1. Task Design Rationale:
   Thank you for your question regarding the task design. To provide further clarity, we have added a justification in Section 3.4 of the revised manuscript, which highlits that the tasks were selected to invoke a natural houshold setting, which could simulate possible human robot interactions.

2. Typographical Errors:
   We sincerely appreciate your attention to detail. We hope all typographical errors have been carefully corrected in the revised manuscript.

3. Sample Size Sufficiency:
   We appreciate your concern regarding sample size. In response, we have added the following justification in the Participants section, supported by prior literature:
   "Given the exploratory nature of our study and that prior gesture-research (Reference to Choi et al. 2014) has reported meaningful findings with similar samples, we considered our sample size (N=34) sufficient to identify stable individual patterns and trends relevant to our research questions."

4. Implications for Human-Robot Interaction (HRI):
   Thank you for prompting us to elaborate on the practical implications of our findings. We have expanded the Conclusion to emphasize the relevance for HRI design.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have carefully modified the content of the paper and the analysis of the results as suggested previously.

 

Author Response

We sincerely thank the reviewer for acknowledging our revisions. We confirm that all suggested modifications from previous rounds have been fully implemented in the manuscript. This response is formally submitted to fulfill the journal's procedural requirements for manuscript processing.

Should any further minor adjustments be needed, we remain available to address them promptly.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have addressed most of the reviewer's comments, the reviewer has some follow-up questions below.

1. In Section 3.4, please add the clarification about "the tasks were selected to invoke a natural household setting, which could simulate possible human robot interactions".
2. Please discuss the sample size issue in the limitation section.

Author Response

Comment 1:
"In Section 3.4, please add the clarification about 'the tasks were selected to invoke a natural household setting, which could simulate possible human robot interactions.'"

Response:
We thank the reviewer for this valuable suggestion. As requested, we have added the specified clarification to Section 3.3 (Tasks) on page 6 (lines 138–139):

"During the experiment, participants were asked to perform four different tasks designed to enforce gesturing and to invoke a natural household setting, which could simulate possible human-robot interactions."

 

Reviewer Comment 2:
"Please discuss the sample size issue in the limitation section."

Response:
We appreciate the reviewer’s emphasis on methodological transparency. We have expanded the Limitations section (Section 6) to explicitly address sample size constraints (page 21, lines 491–495):

"Third, the sample size of 34 participants was relatively small and may limit the generalizability of the findings. This is particularly true for subgroup analyses, such as the gender differences we observed (where 79.4% of participants were females). Future studies should explore if our observed consistency and trends behave similarly across cultures and in more generalizable cohorts."

We are grateful for the reviewer’s feedback