Exploring the Effects of Additional Vibration on the Perceived Quality of an Electric Cello

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This original paper reports the impact of artificially introduced haptic vibrations on how cellists perceive the quality of an electric cello. Helpfully, some nuanced results indicate that these vibrations allow players to enhance “feel” and “liveliness”, and significantly increasing “perceived power”, amidst somewhat noisy uncertainty in the statistics.

 

Importantly, this paper contributes towards understanding how vibrotactile feedback complements a musician’s audition when performing an instrument, in this case specifically the cello. This study also highlights the practical difficulties in exploring these links, in light of limited instruments and musicians used. The methodology is generally well thought through and comprehensive, except for a glaring concern (raised below, Issue 1). The writing quality in this paper is otherwise excellent (no issues of English or clarity).

 

Overall, despite the subtle/limited outcomes reported, this paper has scholarly value and adds to the wider discussion on haptics and emerging music technology.

 

However, this paper can be improved by addressing the following concerns:

 

1.       A crucial and significant fundamental flaw/oversight in this study is the matter of “latency” (microsecond to millisecond delays) between the cellist’s playing (signals collected at the bridge) and the vibrotactile signal introduced, precisely because of the intervening signal processing involved and responsiveness of transducers used. This concern arises at two time-scales: First, is whether the introduced signal constructively or destructively interferes with particular notes played (associated with the fundamental frequency + harmonics + sub-harmonics); Second, is whether the “latency” is significantly noticeable to the player, such that the onset of the vibrotactile signal may affect shorter duration notes than longer notes played (for example, on touch-screen devices, response delays of >80-100 milliseconds become noticeable to users, and become annoying). Latency issues also profoundly affect dynamic performance aspects such as vibrato. In the field of speech, the mechanism of “Delayed Auditory Feedback” (a user speaks into a microphone and then hears their voice in headphones milliseconds later) https://en.wikipedia.org/wiki/Delayed_auditory_feedback has been shown to induce mental stress and significantly compromise verbal fluency (“speech jammer” https://www.clicktorelease.com/code/speech-jammer/). These latency effects together will certainly contribute to subjective perceptions of player’s Preference, Power, Liveliness and Feel. At the very least, this latency must be measured and reported, and discussion must be included to convince readers whether this latency ought to be ignored or not. This issue of Latency MUST be addressed.

 

2.       Line 265: “…the importance of coherent vibrotactile and auditory channels…” In light of Concern 1 above, with signal latency, the vibrotactile and auditory channels will specifically NOT be coherent at the microsecond to millisecond timescales!

 

3.       No explanation as to why only one silent cello was used; would similar results be obtained on other e-cello models/makes? There was also no consideration of vibrotactile feedback from the cellist’s bowing hand? (in fact, cellists were allowed to choose the bow?)

 

4.       Lines 260-263: “Considering that the silent cello… comparisons were not made against a non-vibrating setup but one vibrating with a lower amplitude.” Indeed, do provide actual amplitude measurements comparing the relative magnitudes of vibrations for the various settings considered. Only then is this paragraph’s discussion actually meaningful.

 

5.       Lines 284-286: “…this study contributes valuable knowledge…” “…we shed light on the complex interplay…” How so? What specific knowledge and light was accrued here? This final paragraph makes broad sweeping statements that conflate and inflate the value of the current study’s findings. Please revise this paragraph to be more circumspect with the actual modest findings reported.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper presents a very nicely performed study about haptics when playing an e-cello. Although the outcome is not what one would expect, and the authors really make an effort to justify this outcome, I find the study worth publishing. There are few minor corrections to be made; and I would like to know more about these issues:

- It would be nice to know more about the input from the expert cello players (lines 78-80). At this point there was the subjective comparison between electric and acoustic. Could you tell how did that influence the rest of the study? 

- You then say that “we could acquire compare and match vibration produce by the two instruments” - how was that used? I would like to see the comparison of the two instruments (Figure 2). 

- why was the actuator placed at the top of the e-cello? Why not at the chest? more insights into that decision would help understand the outcome

- on Figure 4, the player does not touch the sides of the cello? Is there a reason for that? Was it like that for all participants? 

- The sound used in the in-ear phones was that of the bridge signal and a convolved IR of an acoustic cello. Is this a common procedure? Do you think this was a good decision?

- Was there the thought to let participants compare between an acoustic and an electric cello? Why was this idea discarded? 

- Given the fact that the participants did not know what options A and B were doing to the system, I suggest one could call these experiments blind tests(?). Unless you think they could immediately tell that one option was activating something and the other was not. Do you think that participants could tell what was going on (line 153: They were not aware of the specific changes in the two setups”)? were they asked to give some feedback afterwards?

- Figure 5: how many data points are considered for these plots? 

 

In-line comments:

 

Line 69: this should probably be mm and not cm

Page 2, footnotes: add “accessed on…”

Line 98: “From out tests, that location resulted…” this sentence is not clear

Line 131: “Labelled A and B” does not need quotes 

Figure 6: x-label could be “vibration type”

Line 241: “...Frobenium norm.” (punctuation missing)

Line 259: in those studies (I would call “those” the past, and “these” the present results)

Reference 19: 286-387

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have responded satisfactorily to the concerns raised - thank you - and I commend the authors on their candour in their response and commitment to scholarly rigor. The quality of this paper is now of merit.