Overview:

This paper presents the results of a narrative review of research related to predominantly non-visual exposures to 'nature' (broadly defined) and aspects of human health or well-being. It concisely introduces the topic and provides a justification as to why exploration of non-visual exposures to 'nature' may be useful for understanding unique contributions of nature to human health and well-being. The article then uses each section of the paper to describe research findings related to each sensory pathway (plus a section on non-sensory pathways). A short conclusion summarises the review and reemphasises the importance of sensory pathways of 'nature' and health or wellbeing.

Review:

The authors have clearly completed a vast amount of reading of disparate papers from a variety of disciplines in order to complete this review, and their goal to collate this research into one paper was ambitious and should be applauded. The main strength of the paper is its novelty; I cannot remember reading a review devoted specifically to the senses in this way, and I very much like the approach the paper is trying to take.

We thank the reviewer for this positive assessment.

However, I do believe the paper suffers from a number of related shortcomings throughout, which could be rectified with major revisions. Principally, I believe some extra explanation in the introduction could help direct what should and should not be discussed in the subsequent sections.

More detailed comments are below, but the principal limitations are:

1. I was expecting some form of systematisation in either the search for literature (systematic review) or in the synthesis of literature (scoping review). However, the article in fact comprises a narrative review with no systematic attempt at synthesis, nor critical appraisal of studies. I have no problem with this approach, indeed I believe synthesising such disparate literature would be extremely difficult, but the authors need to explain from the outset that this is a narrative review of literature and provide justifications as to why they did not undertake systematic searches (if indeed this was the case) or systematic syntheses, or critical appraisals. For example, the authors could state that the review is intended to provide a 'reference' for wider reading, rather than a critical review of evidence. This should help prepare the reader for what is to come.

We thank the reviewer for this insight, and we have dealt with this by (i) making clear in the methods section that this is a narrative review of disparate literature (line 84), (ii) stating that it is a reference for wider reading, and not a critical review (line 85), and (iii) adding a methods section after the end of the introduction, to explain our literature search strategy (lines 84-108).

1. The authors need to define what is and is not a health and/or well-being outcome at the outset of the paper. There are many times throughout the paper when I questioned whether the research being reviewed was health-related, or related to certain aspects of well-being. I go into detail for some of these below, but for example, are "freedom from noise," or memory recall or tastier food really health and well-being outcomes?

We have now defined what we mean by the terms “health” and “well-being” in the methods (lines 97-103).

1. The authors need also to define what 'nature' is for the purposes of the present review. Currently, it ranges from whole natural environments (e.g. scenes of urban green spaces), right down to microbiota; from indoor to outdoor; in laboratory simulations to field studies; from soil to air to dogs. If the remit was indeed intended to be this large, the authors need to be upfront about this, and say why it was necessary to be this broad (noting that such an approach is obviously going to limit the depth that they can go into). There are some instances where I dispute altogether that the research is about 'nature' e.g. "soft, cuddly animals."

Recognizing that nature is a term with many different definitions, we now provide our working definition of the term early in the piece to set out the landscape (lines 94-97).

1. There are many instances where health and well-being outcomes terminology are non-specific. For example, instances of the term "mood" are not explained. Does mood relate to specific measured mood states, domain-general measures of 'affect', transient emotions, or something else? I understand that narrative reviews are broad but I think you need to be specific when describing the results of research. I note some of these in my more detailed comments below.

We have addressed each specific comment in the responses below, and have checked the manuscript for additional occurences of this issue.

1. The authors need to avoid reporting research where it is obvious that a conflation between the senses you are studying and some other phenomenon overlap. The easiest example of this which comes to mind is where pet ownership is discussed in relation to touch - just because pet owners visit the doctor less, this does not mean that this has anything to do with the fact that they stroke their animal (and indeed is far more likely to result from things like increased exercise among dog owners). Again, instances of these are noted below in my detailed comments.

This particular case is addressed below in the detailed comments, but more generally we have reviewed the manuscript to ensure that correlational studies are clearly identified as such. We have retained correlational studies in the circumstances where there is a dearth of more robust studies on a particular topic (line 91-93).

1. There are elements of non-scientific language throughout which I do not necessarily pick out but nonetheless could be addressed.

We have edited the manuscript to eliminate non-scientific language.

Points 1 to 3 above could be combined in some short methodology section before the results of each sense are discussed. This would also give the authors the chance to describe how they searched for literature (if a systematic search was undertaken).

This is an excellent suggestion, and we have added such a methods section after the introduction with definitions and the basis for the literature search (lines 84-108).

Detailed comments:

Abstract:

 Line 16: "Avenues of nature benefit delivery" is awkward terminology. Rewrite to something like, "Here we review evidence of lesser studied sensory pathways."

Updated as suggested (line 17).

 Line 25: "We argue for the need to explore these avenues much more deeply." There is actually very little content in the introduction and conclusion sections, certainly not enough to produce a fully-informed argument about why non-visual pathways are "fundamentally" important, nor any less or more important than visual pathways. I'd suggest returning to the abstract and its contents after revisions.

We have rewritten these lines to soften the language (line 24-27).

Introduction:

 Lines 32-33: "Relate the health benefits of countryside and greenspaces" does not make sense. Suggest rewording.

This sentence has been reworded (line 34).

 Lines 43-44: "The cure was common and simple: experience of pleasant rural scenery." I'm not sure if this is you speaking or the preceding reference speaking. If this is your wording, it is a speculative suggestion and should be omitted.

This has been changed to “the cure they suggested was” as it is a reference to the preceding reference (line 59).

 Line 46: "reviews ref": I'm unsure what this means.

This was a mistake and has been removed.

 Lines 46-47: "Yet herein lies the crux." I would suggest rewording for more scientific language.

This has been reworded (line 61).

 Line 47: "Domination of vision," I'm not sure what is meant by this.

This has been reworded to “as a result of vision being the dominant human sense” (line 62).

 Line 52: "Scratch the surface." Again please reword to be more scientific e.g. "The research base isolating sensory experiences of nature beyond vision is sparse" or something like that.

Updated as suggested (line 66-67).

 Line 58: "Synergistic beneficial effects." Do these articles actually suggest a synergistic benefit of multiple senses? i.e. greater than the sum of the benefit of each sense alone? I had a look and would argue that they only suggest "additive beneficial effects" i.e. one sense only adds to the benefit of another.

Indeed, this has been changed to “additive beneficial effects” (Line 73).

 Line 58: The three referenced papers are all conference proceedings - have any been turned into peer-reviewed articles? Otherwise this statement is contentious. Also only 1 refers to nature experiences; I would therefore suggest the other two to be unnecessary.

Two of the references have been removed, and it has been reworded less strongly to “may possibly lead to” instead of “can lead to” (line 73).

 Line 61: "The limitations of." I would argue that it is not a limitation as such to assume this, but that one might be naïve to think it is the strongest sensory pathway for predicting health effects.

This has been changed to “possible shortcomings” instead of “limitations” (line 76).

 After this section is where I think you could be very specific that you are undertaking a narrative review with no systematic search or systematic synthesis attempted (if indeed this was the case). If there were systematic searches used, you should be explicit about them.

We agree, and have added a method section here, incorporating the material suggested by the reviewer (lines 84-93).

 As mentioned above, you could also use this section to explain more fully what you consider a health and well-being outcome to be, and how broad your definition of "nature" is for the present purposes. I think it would be a good idea to be as specific as possible about these things, for example delimiting well-being to measures/findings which relate to traditional Aristotelian concepts (experiential, eudemonic, evaluative), or whatever you see well-being as. Equally does "nature" refer only to natural environments, biotic elements of natural environments (animals, plants), microbiotic as well (soil, bacteria) etc. Are we talking only about outdoor exposures? Or indoor/simulated? Are we talking only about one kind of study design or multiple? Do we consider animal research or only that which involves human subjects? Etc. etc.

This has been incorporated into the new methods section. We provide our working definitions of nature, health, and well-being (line 94-103).

Sight:

 Line 67: Three references are cited, but only one of these (number 18) refers to the visual sense exclusively.

The superfluous references have been removed.

 Line 67-68: It would be good if you could be more specific about the duration of these purported benefits - are they transient or long-lasting, for instance.

A statement has been added that the duration of benefits has not for the most part been measured (line 123-124).

 Line 69: What are "negative notes from nurses"?

We now describe these (line 126).

 Line 74-78: You are arguing that certain elements of landscapes might be more responsible for health benefits than other elements. Of course, it may be only that a combination of elements in a coherent scene confers benefits. This could be acknowledged.

An excellent point, and we now include this thought (line134-135).

 Line 78-81: I can see what the authors are trying to do here, but I question whether discussion of colour research is relevant - these papers are not about nature.

This seems to flow naturally now from the recognition of elements in a coherent scene from the previous statement, so we have retained it (line 135-138).

 Line 83-84: "human influence" - this contention is not supported by the research you cite in the paragraph afterwards.

The statement about human influence has been removed.

 Line 86-88: You cite two preference studies and then parallel these findings with "restorative benefits." Preferences are not the same as restorative benefits, and I question whether preferences could be conceived as health or well-being outcomes.

We agree with this, and recognize our treatment of the distinction between preference and benefit was superficial. The inclusion of some preference studies has been discussed in the new methods section (line 91-93), and we have made the distinction between preference and benefit where relevant.

 Line 89-90: This is repetition of lines 77-78.

This has been removed.

 Line 92: "Beneficial response" - can you be more specific about the health or well-being effect in this study?

This has been clarified (line 148-149).

Sound:

 Line 96: Do you have a reference for sound being the second most studied pathway or is this a supposition based on the amount of literature you found?

This was based on the amount and specificity of literature found and has been described as such (line 152).

 Line 105: You link place attachment to potential positive feelings about one's environment. I think this sort of contention warrants a reference that agrees.

We have added a reference on this point (line 162).

 Line 108-109: As earlier, I question whether a preference can be considered a health or well-being outcome (the reference in the next line about preferences being associated with restorative potential is much better).

We agree that preferences and benefits are distinct, and we now discuss the inclusion of some preference studies in the methods section (lines 91-93).

 Line 113: The paper cited by White et al (2010) has nothing to do with sounds or local river visits; it is a preference study of visual scenes. I assume this is a mistake.

Thanks for picking this up – the reference has been corrected (line 182).

 Line 118-119: Can rating sounds as "pleasant" be considered a health or well-being outcome?

We consider rating something as pleasant or enjoyable to indicate a well-being outcome in the sense we define well-being (lines 98-103).

 Line 121-130: "Higher ratings…decrease the ratings." Ratings of what?

We recognize this lack of clarity and have added more detail (line 191).

Again, if these are preference ratings, it may be irrelevant. Equally "appraisals," "perceived naturalness," "perceived crowding," "soundscape eventfulness" - are any of these health or well-being outcomes?

We consider rating something as pleasant or enjoyable to indicate a well-being outcome in the sense we define well-being (lines 98-103).

 Line 134-146: I accept that these paragraphs are an introduction to noise itself, so do not have to be related to health and well-being specifically, but I feel they take up too much space at present. Two thirds of this subsection is an introduction to noise - only the last paragraph focuses on health and well-being.

We consider rating something as pleasant or enjoyable to indicate a well-being outcome in the sense we define well-being (lines 98-103).

 Line 147-148: "escaping noise…visiting parks" - this is repetition of line 141-142.

The earlier line has been removed.

 Line 151: "A reason to protect…the natural quiet and nature sounds" - a compelling finding, but it is not about health or well-being.

We included this here because part of the aim of conserving natural quiet and sounds is with the purpose of deriving benefits from them (line 222).

 Line 151-152: I can just about see how ratings of "pleasing" sounds could be construed as a well-being outcome, but again this kind of thing could be much more easily accepted if there were some general delimitations of your conception of well-being in the introduction or a methodology section.

We now define well-being in the new methods section (line 98-103).

 Line 156: "urban environment" - is this meant to read "natural environment"?

No, it is meant to refer to why silence in an urban environment is not a positive experience like quiet in a natural environment is. We have clarified this to avoid confusion (line 227-228).

 Line 156-157: "'peace and quiet' is frequently…restoration" - is there a reference to support this contention?

This is based on references in the previous section and this is clarified (line 228).

 Line 166-170: Is this research on animals really relevant to this article? I'm not sure why introducing comparative psychology work is necessary here.

We see this research on animals as being relevant to the question of why silence is perceived negatively, and part of well-being is mastery of the environment (line 98-103).

Smell:

 Line 184: "The world we smell is all around us." I accept this is personal preference, but I would prefer more scientific phrasing here - "Smell is one of our weakest senses, yet odours are emitted from many objects we encounter each day" - or something like that.

This has been reworded to “yet the world around us emits all kinds of smells” (line 258).

 Line 188: "affect us" - it would be better to say "affect our health or well-being"

Updated as suggested (lines 262-263).

 Line 191: "feelings of pleasure" - is there a reference for this contention?

No, this is based on the authors’ experience and now reads “might” (line 265-266).

 Line 197: Is "spiced apple" a naturally occurring smell?

Perhaps not in combination, but the two elements are naturally occurring.

 Line 198: Is "brain activity" or "cognition" a health or well-being outcome? At the least, what brain activity and what aspect of cognition should at least be noted.

We now include more detail on this study (line 273-274).

 Line 200: "All the positive feelings we have about nature." This either needs a reference for support or needs omitting as it sounds assumptive; many people may not have such feelings.

This has been reworded to “any of the positive feelings” instead of “all of” (line 276).

 Line 200-201: "Nature smells can…deliver nature benefits by proxy" - this statement needs a supporting reference.

This is a conclusional statement from the preceding reference and the language has been softened (line 277).

 Line 216-226: As the author admits, there is a paucity of work on smells of nature and associated benefits to health or well-being - therefore I don't know whether this paragraph is necessary really.

We included these to highlight possible future studies on natural smells and to look at the ideas brought up by smell in general.

 Line 228: "Increased signal detection" - this is not a health or well-being outcome and therefore talk of this study should be omitted.

The outcome variables in this study related to the cognitive function component of well-being, defined in the new methods section (line 98-103).

 Line 229-235: Vocalization in dogs, nor memory recall in humans can really be considered health or well-being outcomes. This whole paragraph could be omitted then.

We have removed the material on dogs, but retain the material on memory as a cognitive function, and as such, part of our definition of well-being (lines 98-103).

 Line 243: Is "memory activation" a health or well-being outcome?

We define well-being to include cognitive functions such as memory (lines 98-103).

 Line 249-250: "The treatment of elderly dementia" - what outcomes are the treatments targeted at? They must obviously be health or well-being related. In the context of dementia, if they are to do with memory, I suppose this could be relevant, since memory in dementia patients could be considered an aspect of functional well-being.

The authors agree and the types of symptoms are clarified (line 349).

 Line 251-252: I'm not sure why talk of hypertension accompanies talk of dementia in the same paragraph.

These are two review studies that have been put together to support smell as a benefit.

 Line 254-255: "The use of lavender oil" - for what health or well-being outcome?

We now describe the outcomes (line 354).

 Line 260: "have an effect" - on what outcome?

We now describe the outcome (line 360).

 Line 267: Field experiments on essential oils - what would such an experiment look like? I'm not sure this limitation is relevant.

We now make clear that such experiments would use the plants themselves rather than the oils (line 384-385).

 Line 272-274: Comparing differently valenced smells to indicate preference - as with elsewhere this assumes that preference is somehow synonymous with a health or well-being outcome when it is not. Similarly, do "memories of preferred smells" provide health and well-being insights?

We agree with this and recognize our treatment of the distinction between preference and benefit was superficial. The inclusion of some preference studies has been discussed in the new methods section (lines 91-93), and it is presumed preference studies can provide indications into benefits that can then be explored further.

Taste:

 Again, as this whole section is about food, it would be useful to define 'nature' much earlier in the article - 'nature' obviously now comprises naturally occurring foodstuffs, and is quite a departure from the landscapes talked about under the vision section.

Recognizing that nature is a term with many different definitions, we now provide our working definition of the term in the methods section to set out the landscape (lines 94-97).

 Line 300-304: The study about choosing healthy food when feeling positive emotions seems to be irrelevant; the rest of the review is assuming a nature to well-being directionality and not the other way around.

Removed as suggested.

 Line 310-315: The talk preceding about emotional appeal seems relevant, as does the proceeding talk about nutritional value, but I don't seen how these lines on "better taste" constitute health and well-being outcomes unless they are specifically related to nutritional value.

Comfort in the environment is part of wellbeing as defined in the methods section (Lines 98-103).

 Line 317-320: With no further detail, the Norwegian study cited could easily be conflated with socio-economic status or a number of other confounds. I know this is a scoping review so only surface detail of studies is necessary, but, not knowing the referenced paper, I would be vigilant about including studies' findings where the chances of confounding are high and overt to the reader of this review.

We have now explicitly stated that this particular study was correlational (line 450).

 Line 335-336: Again, I read the Church study as being easily confounded by wealth. Can people's happiness be directly attributed to their growing of their own food in the Church study? If not, I would consider omitting.

We now discuss this (line 468-469).

 Line 339: "Better tasting food" - is this a health/well-being outcome?

Comfort is part of how we defined well-being in the methods section (lines 98-103).

 Line 341: "skill development" and "connection to nature" are these health and well-being outcomes?

Mastery of the environment and comfort are part of well-being as defined in the methods section (lines 98-103).

 Line 342: "economic savings" - is this a well-being outcome?

Comfort is part how we defined well-being in the methods section (lines 98-103).

 Line 343-344: "racial equality" - obviously important, but again, is this a health outcome?

Comfort is part how we defined well-being in the methods section (lines 98-103).

 Line 346-350: Are these things related to health or well-being? A link between community cohesion and individual well-being is plausible, but I question whether things like gender equality, social togetherness etc., though admirable, are health and well-being outcomes.

Comfort and happiness are part how we defined well-being in the methods section (lines 98-103).

 Line 358: "Cognitive effects of diet." Nice suggestion, but is cognition related to health and well-being?

Cognition is part how we defined well-being in the methods section (lines 98-103).

Touch:

 Line 362: "Positive health effects and enhanced well-being." I know this is introductory, but some specificity on what health effects and what type of well-being would be useful (related to the reference).

We agree and have added specificity here (line 493-495).

 Line 364-365: The Harlow studies are about maternal deprivation and touch per se.

They are used in the intro to touch to emphasize the importance of touch.

 Line 367: "positive benefits" - be specific about what benefits you are referring to.

We have now specified what these benefits are (line 524-526).

 Line 369-370: The reference is about pet ownership and not touch per se. Suggest omitting.

This is made clear now (lines 529-531).

 Because this section is largely about pets, you need to be clear in the introduction or methodology section that 'nature' includes domesticated animals.

Recognizing that nature is a term with many different definitions, we now provide our working definition of the term to set out the landscape (lines 94-97).

 Line 377-378: "The touch of nurses" - whilst the preceding and proceeding sentences are not about nature, I can see their relevance. This study though seems far removed from touching 'nature' so I would suggest omitting.

Omitted as suggested.

 Line 381: You are not talking about classical conditioning here (associative learning), but about operant conditioning.

This has been changed (line 549).

 Line 383-385: The study about bird ownership is not about touch per se.

Omitted as suggested.

 Line 386: I don't think plush toys can be considered 'nature.'

We have clarified that this is about rabbits, not plush toys (line 552-553).

 Line 390-394: None of these studies are about touch, they are merely about pet ownership. The health effects could be because of numerous other factors than merely the petting of an animal, such as increased rates of walking and physical activity among dog owners. This section needs considerable revision as to the studies which are explicitly about touching animals (relevant) and those about pet ownership more generally (irrelevant).

The correlational nature of these studies and the reason for their inclusion has now been clarified (line 560-564).

 Line 395-402: This is much better because the reader can identify that the findings are explicitly about touch, and a mechanism is proposed (oxytocin).

Thank you.

 Line 403-405: These studies are not about touch explicitly.

Companion and therapeutic animals are used with patients to give companionship and tactile comfort. The patients talk to and pet the animals (line 587).

 Line 406-408: These studies are also not about touch.

One study has been removed and the animal-assisted intervention uses animals for touch and interaction.

 Line 413: Typo - "health and well-being and health effects."

Corrected (line 597).

 Line 415: "wind on your face" could be considered as the sense of thermoception and not as touch.

We have added a statement that this could be considered thermoception (line 599).

 Line416-417: I'm not sure why the blood pressure effects of lying in the grass would be interesting to readers of this topic - lying in a supine position would of course reduce blood pressure as you are not being physically active. These sorts of suggestions warrant, at the least, a comparator (e.g. compared to lying in bed), but again, it is hard to see any public health relevance to such a suggestion.

This could have an effect on relaxation and anxiety in addition to the effects of just lying down and this has now been noted(line 601-602).

 Line 419-421: I would be wary about concluding much considering the paucity of studies and high chance of confounding.

We have carefully moderated what we say here, in light of the patchy evidence and this has been moved to the future research section (line 852-856).

Non-sensory pathways:

 This first subsection on the whole is much stronger and clearly the author's strength.

We thank the reviewer for this positive assessment.

 Line 449-450: "potential causes for the benefits of natural places." Please clarify what this means and what benefits you are referring to.

This has been changed to “potential mechanisms ” and the benefits have been clarified (line 625-626).

 Line 453: "reduced well-being." Please be clear about what aspect of well-being you are referring to.

This has been clarified (line 630).

 Line 466: You say that seashores have high air ion concentrations, but in line 462, you say that coastal sites only have moderate amounts?

This has been changed to “have higher air ion concentrations” (line 670).

 Line 468: Be specific about the flaws - are they so flawed that we shouldn't read anything into the studies?

We have described the flaws more specifically, and clarified what should or shouldn’t be read into the studies (line 674-676).

 This section now considers "the air" to be 'nature.' Again this should be made specific in the introduction or methodology.

Recognizing that nature is a term with many different definitions, we now provide our working definition of the term to set out the landscape (lines 94-97).

 Line 484: "performance" - does this mean exercise performance? Please clarify.

This has been clarified (line 692).

 Line 488: "improvement" - presumably in some measure of affect or mood? Please clarify what the improvement is in.

This has been clarified (line 697-698).

 Line 488-489: "the first 30 minutes of administration" - what was being administered exactly?

This has been clarified (line 698).

 Line 490-491: "affect alpha brain wave activity." Can you say how it affects brain activity? Otherwise I cannot judge whether it is related to health and well-being.

This has been clarified (line 700-701).

 Line 499-501: The link between air ions and longevity seems as though it could be confounded by many other factors - I would suggest omitting this reference if it cannot be clearly determined.

The correlational nature of this conjecture is stated (line 710).

 Line 520-521: "effect on mood." Please be more specific, were these self-rated mood states, measures of affective valence or arousal, ratings of certain emotions etc.?

This reference has been removed as suggested in the comment below.

 Line 520-530: I think this paragraph is unfortunately, quite flawed. The argument is that the effects of gardening on "mood" could be related to direct contact with the soil. Firstly this is surely highly contentious, it could be the physical activity, it could be the social aspects of gardening, it could be the good weather etc. Secondly, this assumes that people in the aforementioned study were getting their hands in the soil which may not always have been the case. Secondly, the paragraph then turns its attention to gut microbiota, but presumably soil is not ingested during gardening?! The conclusion is that exposure to soil microbes could boost people's mood. This is a highly tenuous link. I would suggest omitting this paragraph, unless you can find research linking the feeling of soil with some well-being benefit (and if so, this would surely come under the 'touch' section anyway).

The references to the gardening study have been removed.

 Line 531-539: Now "nature" is defined as microorganisms. Surely this is beyond your original intended definition of nature?

Microorganisms are squarely within our definition of nature (lines 94-97).

 Line 548: I don't think you can say there is a strong link between gut biodiversity and mental health for the reasons I state above.

There is a strong experimental literature on this point and it is referenced in this section.

 Line 549-557: Apart from the last sentence, we seem to be very removed from a traditional conceptualisation of "nature" now (unless 'nature' includes gut microbe diversity, in which case this needs to be explained in the introduction or in a separate methodology section).

Microorganisms are squarely within our definition of nature (lines 94-97).

 Line 561-562: "Increase emotional health" - can you clarify what is meant by this?

This has been clarified (line 790).

 Line 562: "cognitive function" - is this a health or well-being outcome?

This is a well-being outcome as now described in our method section (lines 98-103).

 Line 567: "maze run times" and "mistakes" - again it is hard to see these as health and well-being behaviours, unless your definition of well-being includes functional behaviours like looking after oneself. Even so, you need to make clear that animal and rodent research (comparative research) is included within the remit of this review (since it primarily about human health) much earlier.

We have clarified these points in the methods section (lines 87-88).

 Line 573-575: I think there is some repetition here that needs amending.

Repetition has been removed.

Conclusion:

 The conclusion could obviously change in light of the suggestions I have made. Generally speaking, I would like to see a conclusion that, as well as summarising the review's findings, highlighted one or two key limitations. These could include, the non-systematic approach, and the focus on benefits rather than risks.

The authors could then suggest a more systematic exploration of sensory exposures to nature with regards to particular (perhaps neglected) health and well-being outcomes.

We have now highlighted the limitations of the underlying studies and of the review process itself (line 893-899).