Perceptual Awareness and Its Relationship with Consciousness: Hints from Perceptual Multistability

Round 1

Reviewer 1 Report

I consider this manuscript to be a review article that promotes a certain research direction for future studies on consciousness.  That is a good approach, and the article makes a good, original argument that studying perceptual multistability can reveal much about the neural mechanisms behind consciousness.  This argument comes across well in Sections 1, 2, and 5.  However, the literature studies that are meant to support the argument are not summarized in sufficient detail, so the paper does not give enough data to support its thesis.  Instead, it relies on repeating over and over that the approach is valuable without showing enough of why this is the case.  Again, the thesis is good, but I ask for more evidence.

Another issue is the writing style and use of English.  There are a fair number of mistakes in grammar and punctuation.  Many sentences are too long, are run-on, or contain puzzling asides.  They really should be simplified, and with less repetition. The ideas are successfully conveyed, but it is all so difficult to read. I suggest that the author ask for help from native speaker of English who has experience with scientific writing.

Here are my comments on particular sections, starting with those that could use more clarification. Lines 101-107: Is a new point being made here, or is this just repeating what was said but in a slightly different way? Lines 128-131: Can you elaborate about the statistics and gamma distribution?

Overall, I understand Section 3.1 on functional neuroimaging, but the initial Lines 158-168 have so many subclauses that I am confused. Is the point that some investigators say the frontal lobe is responsible but other investigators say the parietal lobe is also involved? 

Lines 169-179 are good, on how fMRI gives evidence for perceptual rivalry in different parts of the superior parietal lobule.  This is the kind of factual evidence that I ask for more of in the manuscript.

Lines 206-255. This section had valuable parts on MEG, EEG, and new uses of Integrated Information Theory to decode the recorded findings, and I learned a lot about recent technical advances in the field.  However, this section often wanders off-topic to discuss consciousness in general rather than evidence for multistability specifically. 

In fact, there were several other places where general consciousness was considered without focusing on multistability, as had been promised: Lines 163-167, 257-265, and 284-291. 

Section 4 summarized four major theories of consciousness. I found this summary to be valuable, but found it a bit odd that only one of these theories was considered further: Predictive Processing Theory.  Then, the paper did not explain how Predictive Processing applies to multistable phenomena, but just said it does apply. That is a noteworthy omission and more elaboration is needed.  Finally, the early sentence on Predictive Processing, Lines 266-270, is too complex and difficult to understand. 

I hope these suggestions will help the author to revise the paper, which stems from a good idea, to make the presentation more focused and evidence-based.

Author Response

I consider this manuscript to be a review article that promotes a certain research direction for future studies on consciousness.  That is a good approach, and the article makes a good, original argument that studying perceptual multistability can reveal much about the neural mechanisms behind consciousness.  This argument comes across well in Sections 1, 2, and 5.  However, the literature studies that are meant to support the argument are not summarized in sufficient detail, so the paper does not give enough data to support its thesis.  Instead, it relies on repeating over and over that the approach is valuable without showing enough of why this is the case.  Again, the thesis is good, but I ask for more evidence. 

>> Answer: Thank you for pointing this out. Following your observation, I added some useful “evidence” in some parts (text in red indicates the revisions throughout the paper), although reading the original cited works would be surely more interesting and compelling, since the idea behind this proposal has been visualized much before from other authors. 

Another issue is the writing style and use of English.  There are a fair number of mistakes in grammar and punctuation.  Many sentences are too long, are run-on, or contain puzzling asides.  They really should be simplified, and with less repetition. The ideas are successfully conveyed, but it is all so difficult to read. I suggest that the author ask for help from native speaker of English who has experience with scientific writing. 

>> Answer: thank you for the suggestion, I asked for a fast proof reading, I hope it is now more readable. I didn’t track all the changes with red text in the manuscript. 

Here are my comments on particular sections, starting with those that could use more clarification. Lines 101-107: Is a new point being made here, or is this just repeating what was said but in a slightly different way?  

>> Answer: it is a summarizing conclusion of the previous ideas introducing the reference to the “window” metaphore; I get the point that it seems a simple redundant repetition, but I wanted to close the introduction paragraph stressing the idea that this methodological approach has the potential to unveil more than just perceptual awareness, as stressed with the quote at the end of the paragraph. I tried to rephrase the sentence hoping that it's a bit clearer now (Lines 102-111, text in red). 

Lines 128-131: Can you elaborate about the statistics and gamma distribution? 

>> Answer: some more details on this topic (not very central for the discussion, indeed) have been added (lines 131-137, text in red) 

Overall, I understand Section 3.1 on functional neuroimaging, but the initial Lines 158-168 have so many subclauses that I am confused. Is the point that some investigators say the frontal lobe is responsible but other investigators say the parietal lobe is also involved?  

>> Answer: I rephrased the awkward sentences, I hope it will be easier to understand now. The point is that Frassle and colleagues (2014) claim that frontal and prefrontal activity typically found in neuroimaging studies in bistable phenomena are not due to perceptual alternations, but are actually related to the introspective report of the subjects which have to communicate to the researchers which percept are seeing at any time (of course this imply that they consider that only parietal areas are crucially involved in alternations); nevertheless, even using no-report paradigms (which are supposed to eliminate the subjective report issue mentioned by Frassle and colleagues), further studies confirmed the previous evidence of an involvement of frontal and prefrontal areas in perceptual switching; besides this, also parietal lobes (and particularly a structure such as SPL) have been reported to be involved, highlighting their connections with other brain areas in the fronto-parietal network (central in some of the most famous Theories of Consciousness). The traditional finding that frontal and prefrontal areas are involved in multistability has been for a long time considered an evidence of top-down direction for the explanation of the phenomena (for example, the long-discussed attentional involvement in the switching), but as I present earlier in other sections of the manuscript, many studies offered evidence for bottom-up processes explaining perceptual alternations, too. In that paragraph I tried to syntesize something that reflects a large discussion in literature showing that the conflict might be only apparent and that it can be solved with other theoretical approaches. 

Following your observations, I have anyway rephrased the whole part trying to give some more details on these findings to make it clearer (now they are in lines 164-188, text in red color). 

Lines 169-179 are good, on how fMRI gives evidence for perceptual rivalry in different parts of the superior parietal lobule.  This is the kind of factual evidence that I ask for more of in the manuscript. 

>> Answer: thanks for mentioning, I tried to add more details on the other areas’ involvement in general (other parts with red text in the 3.1 section).

Lines 206-255. This section had valuable parts on MEG, EEG, and new uses of Integrated Information Theory to decode the recorded findings, and I learned a lot about recent technical advances in the field.  However, this section often wanders off-topic to discuss consciousness in general rather than evidence for multistability specifically.  

In fact, there were several other places where general consciousness was considered without focusing on multistability, as had been promised: Lines 163-167, 257-265, and 284-291.  

>> Answer: maybe the number of the lines are wrong, but I cannot see where general consciousness is considered instead of multistability:  

• in 163-167 I talk about no-report paradigms used to physiologically track perceptual alternations without the need for the subjects to say what they’re saying (this has been done, as mentioned before, to check if Frassle and colleagues were right claiming that frontal activity has been found only because of that subjective report on alternations, and not on alternations themselves). 
• In lines 257-265 it’s only a general introduction to the main current ToCs as they are summarized by Seth (2022). This is supposed to introduce a perspective which has been applied to multistability (the Predictive Coding, or Predictive Processing), on which I briefly spend some words later, and that Seth includes in the “re-entry (Recurrent) and predictive Processing theories (RP)”, the authors proposing both kinds of theories are cited. For more information, I suggest to see the cited reviews, discussion of which exceeds the main topic of this manuscript, this is why I don’t give more space to that issue. 
• 284-291: this paragraph actually starts with perceptual inference, and ends wondering why should our sensory systems have evolved in such a way that they have to resolve perceptual ambiguity with a phenomenological multistability as a result. What is the reason (and the evolutive advantage) why, in presence of ambiguous sensorial information, as a result percepts start to alternate producing different stable but competing states? How do they enter consciousness in turn? 

So I am not sure the mentioned lines are correctly numbered. Please tell me if I misunderstood something. 

Section 4 summarized four major theories of consciousness. I found this summary to be valuable, but found it a bit odd that only one of these theories was considered further: Predictive Processing Theory.  Then, the paper did not explain how Predictive Processing applies to multistable phenomena, but just said it does apply. That is a noteworthy omission and more elaboration is needed.   

>> Answer: it’s true, I gave too few space to the proposal, being basically the only one trying to address bistable phenomena amongst the other theories. I took advantage of your observation to spend some words on how the predictive coding explains the phenomena (lines 334-354 text in red), but I added some other hint in the section 4 in general to make it easier to follow the point I make here. 

Finally, the early sentence on Predictive Processing, Lines 266-270, is too complex and difficult to understand.  

>> Answer: thank you for the advice, sentence rephrased (lines 317-322, text in red). 

I hope these suggestions will help the author to revise the paper, which stems from a good idea, to make the presentation more focused and evidence-based. 

>> Answer: yes, thank you for the suggestions, I hope the manuscript has improved and I managed to address the observations in a satisfactory way. 

Reviewer 2 Report

In the article, the author makes a clear point on perceptual awareness and its relationship with consciousness, identifying the interactions between distinct neurocognitive systems as a crucial point in the subjective experience of the phenomenological world. Such a narrative is accompanied by a concise recapitulation of neural correlates.

The article is well written, and I believe it can be published in the journal pending (very) minor revisions.

I really appreciated the fact that the author never lost her mandate in delineating the gnoseological perimeters of the discipline in a balanced manner.

Here I will limit myself to highlight how the “integrated” perspective the author endorses (i.e., summa of cognitive processes that concur in perceptual experience) is something that have been recently proposed also in the cognitive neuroscience of perception. Specifically, the concept of “action reappraisal” has been proposed in perception-for-action literature to identify a multidimensional cognitive mechanism trough which individuals may “integrate” semantic, mechanical and sensorimotor knowledge (i.e., a large fronto-parietal network) when perceiving the phenomenological world. Therefore, the author may find useful to consider the following literature:

1.     Federico, G., Osiurak, F., Reynaud, E., & Brandimonte, M. A. (2021). Semantic congruency effects of prime words on tool visual exploration. Brain and Cognition, 152, 105758.

2.     Federico, G., Osiurak, F., Brandimonte, M. A., Salvatore, M., & Cavaliere, C. (2022). The visual encoding of graspable unfamiliar objects. Psychological Research, 1-10.

Additionally, by following the idea of multiple-information-integration in a slightly different gnoseological context (i.e., memory and language), the author may find helpful the work of Wurm & Caramazza (2019; Nature Communications), as well as the review by Lambon-Ralph et al. (2017; Nature Reviews Neuroscience) to construct a transversal discussion on “convergences” between neural networks and cognitive processes.

Minor points:

1) Abstract: they’re happening -> they are happening. 

Author Response

Thank you a lot for providing useful integrative insights to support the proposed view. 

I was trying to figure out in which part of the manuscript I could mention these other views (the fact that they're integrative doesn't help, being these theoretical perspectives related to object/action representations and not related to HOW these representations enter the consciousness, I mean, the mechanisms allowing the processed information to be considered aware), but I honestly couldn't find an appropriate section to add this kind of discussion without going off-topic. In a previous study [Di Pace, E., & Saracini, C. (2014). Action imitation changes perceptual alternations in binocular rivalry. PloS one, 9(5), e98305.] I think that we could have definitely benefited from the action reappraisal perspective, but unfortunately that study was too old and this new perspective is more recent.

I have read the works you suggested and it seems interesting to develop and extend also to other perceptual phenomena such as multistability, although the "multiple-information-integration" in the multistable phenomena would be only related to different neural populations fighting to trespass the consciousness threshold, while they lack the other specific multi-dimensional characteristics (context, concept, semantic, objects, tools, action, which are quite central issues in the newly proposed perspective).

Reviewer 3 Report

 The manuscript presented from Chiara Saracini, entitled "Perceptual awareness and its relationship with consciousness: hints from perceptual multistability" is interesting and original. The text is well written and the discussion of the results is remarkably thorough.

However I would suggest the author to insert a figure or drawing scheme, that summarizes the methodologies described in the manuscript. This should improve the quality of this original manuscript.

Author Response

Thank you for the suggestion: I was indeed thinking that the paper is a bit too much words and no figures.

I added two figures taken from other eminent papers in the field, with some examples of reversible (or ambiguous) figures (Figure 1) and Binocular Rivalry stimuli (Figure 2).

Round 2

Reviewer 1 Report

This draft of the manuscript is significantly improved.  I see now how it all relates to the multistability theme, and it ties everything together. It convincingly and cleverly makes its case about the value of multistability for consciousness studies. The new information on additional studies is helpful as is the further information on the studies reported in the first draft. The author addressed my concerns well in the Response to Reviews as well as in the revision, so I am recommending publication.

However, even though the English is more correct now, the manuscript still has some wordiness and repetition, and a few parts are difficult to understand. I have suggested many ways to improve all these things by writing my suggested changes in pencil on a scanned version of the printed manuscript, which I submit as a file with this review, and ask that the Editor forward it to the author.  I offer these changes to be of assistance, and the author can adopt those she wishes at her own discretion.

Comments for author File: Comments.pdf

Author Response

I am really grateful to Reviewer #1 for the incredible help in improving the manuscript. I would have never expected such a detailed report, written by hand. I value this so much!

Reviewer 3 Report

The author satisfied all my concerns.

Author Response

Thank you!