If you notice the skies above you becoming darker, your first thought might be to seek shelter. Experience will have taught you that darkening skies are often a sign of an approaching storm. Learning to recognise changes that occur prior to an unpleasant event can help us avoid danger. But this is not the only strategy people can use to predict when something bad is about to happen. Another option is to use the intensity, or salience, of sensory information. Soldiers fighting on the front line, for example, might rely on the loudness of enemy voices or vehicles to judge how close an advancing enemy is. This information will help them decide when to retreat.

Different brain processes are active when individuals use each of these two strategies to predict when an upcoming event will occur. One approach to study these processes is to use a technique called “SCS conditioning”. This involves exposing mice to two sounds, followed by a mild electric shock administered to the feet. The first sound is a pure tone; the second is a burst of white noise. After repeated trials, mice begin to show distinct responses to the two sounds. They freeze in response to the tone but run away upon hearing the white noise.

These responses parallel behaviors seen in the wild. When mice detect a distant predator, they freeze to avoid detection. But if the predator comes too close for the mice to avoid being spotted, they instead try to flee. Some have argued that in the SCS task, mice learn that the white noise predicts an imminent shock. The mice therefore flee as soon as they hear it. By contrast, they learn that the tone predicts a delayed shock and therefore choose to freeze instead.

However, by tweaking the SCS procedure, Hersman et al. now show that even if the white noise occurs before the tone, it is still more likely than the tone to trigger an escape response. In fact, mice are more reactive to white noise than tones even if the sounds are never paired with shocks. This suggests that mice find white noise naturally more noticeable than tones. Moreover, Hersman et al. show that tones can also trigger escape responses if they are sufficiently intense. Together these results suggest that mice use the intensity of the stimuli – rather than the length of time between each stimulus and the shock – to decide whether to freeze or flee.

People with anxiety disorders often show exaggerated responses to things that do not pose a genuine threat. At present the pathways in the brain that are responsible for these excessive reactions are unclear. The results of Hersman et al. will aid research into the brain circuits that detect, assess and respond to threats. Understanding these circuits could in the future lead to better treatments for anxiety disorders.

Learned temporal relationships between cue stimuli and aversive events allow individuals to avoid danger. For example, progressive darkening of clouds often precedes lightning storms, and dark skies prompt evacuation from exposed spaces. Other forms of threat prediction derive not from cue timing or sequencing but rather from the intensity or salience of a stimulus, such as an entrenched soldier who uses relative volume of auditory threat stimuli (e.g. foreign vehicles or voices) to gauge proximity of an advancing enemy and determine when to retreat. Human studies indicate that the specific neural circuits engaged during prediction of event occurrence depend on which cognitive strategy is used to solve a particular task (Breska and Ivry, 2018). Thus, determination of the neural mechanisms that regulate different forms of threat prediction, and the consequences when such mechanisms are dysfunctional, requires behavioral paradigms in which the cognitive processes engaged are clearly defined.

In SCS conditioning (Fadok et al., 2017), sequential presentation of two different auditory stimuli (pure tones followed by white noise, in that order) precedes delivery of an aversive unconditioned stimulus (US, footshock). Following repeated SCS-US presentations, mice exhibit distinct defensive behaviors to each SCS component: tones elicit freezing whereas white noise elicits flight. The paradigm thus appears to model natural behavioral shifts that occur as the perceived probability of directly encountering threat increases. As posited by ‘predatory imminence theory’, prey animals initially freeze (to avoid detection) when predators are present at a distance, but then switch to flight (escape) to avoid entrapment if a predator becomes close enough that avoiding detection is no longer possible (Blanchard and Blanchard, 1989; Blanchard et al., 1995; Bouton and Bolles, 1980; Fanselow, 1994; Fanselow and Lester, 1988; Perusini and Fanselow, 2015).

Given the presence of a specific, repeating sequence of auditory stimuli preceding shock during conditioning, defensive behaviors elicited by individual components of the SCS could in principle be driven by learned CS-US temporal relationships. However, the form of both appetitive and aversive conditioned responses is known to vary substantially according to the particular properties of a given conditioned stimulus, even when the same underlying construct has been learned (Holland, 1977; Holland, 1979; Holland, 1980). Therefore, differences in the intrinsic properties of tone and white noise stimuli themselves, rather than their temporal relationship to the US, could underlie the distinct behaviors these stimuli evoke during SCS conditioning.

To define the key factors responsible for the topography of behavioral responding in this paradigm, we systematically varied CS-US temporal relationships and properties of SCS component stimuli during or following conditioning. We found that when presented at equal sound pressure levels (SPL), white noise elicits greater active defensive behavior than tones, irrespective of stimulus order during conditioning. Following standard tone-white noise SCS conditioning, each stimulus was also capable on its own of evoking either conditioned freezing or flight, according to SPL. Furthermore, when presented at equivalent SPL, white noise promoted greater arousal and simple locomotor responding than pure tone stimuli, even in the absence of any prior conditioning. Together, these data argue that stimulus salience is the major factor determining the form of conditioned responses during SCS conditioning.

In conclusion, we found that audio frequency properties strongly influence the defensive behaviors elicited by SCS fear conditioned auditory stimuli. Escape behaviors were most potently triggered by stimuli that contain frequencies to which mouse hearing is most sensitive, an effect that was independent of the order in which auditory stimuli were presented during learning. In addition, pure tones that elicit freezing at typical experimental sound pressure levels can promote conditioned escape when presented at higher levels. These data argue that stimulus salience, not temporal proximity to the US, is the primary means by which mice assess imminence and engage appropriate defensive strategies in the SCS paradigm. This would appear to be similar mechanistically to how mice respond to innately threatening visual stimuli, where the probability and intensity of escape behaviors scale with visual stimulus salience (Evans et al., 2018).

An implication of this work is the critical need to consider the behavioral sensitivity of experimental subjects to auditory stimuli of different frequencies. Psychophysical studies have demonstrated that all species have a particular range of frequencies that they hear well (i.e. which are audible at 10 dB); stimuli outside of this range may need to presented at substantially higher SPL in order to be efficiently detected. In addition, although most laboratory animals exhibit overlap in their hearing ranges, there can be significant differences in their sensitivity to particular frequencies, even among closely related species. For example, whereas the 10 dB threshold includes frequencies ranging from ~5–40 kHz in rats, this range is very narrow in mice and limited to frequencies close to 16 kHz (Heffner and Heffner, 2007). Differences can also exist across mouse strains and between different ages of the same strain. For example, C57BL/6J mice undergo hearing-loss induced plasticity that by 5 months of age results in loss of responsivity to high frequency tones (>20 kHz) with concomitantly enhanced behavioral sensitivity to middle (12–16 kHz) but not low (4–8 kHz) frequency stimuli (Carlson and Willott, 1996; Willott et al., 1994). Moreover, certain frequencies may be innately aversive in rodents: rats emit and respond defensively to alarm vocalizations near 20 kHz (Beckett et al., 1996; Blanchard et al., 1992; Cuomo et al., 1992), and 17–20 kHz ultrasonic sweeps can elicit robust freezing and flight behaviors in mice (Evans et al., 2018; Mongeau et al., 2003). White noise stimuli, which are both aperiodic and include 17–20 kHz frequencies, may thus be uniquely salient to mice under conditions of impending potential threats due to recruitment of dedicated defensive circuits tuned to innately threatening auditory stimuli. Indeed, in conventional fear conditioning to a simple CS composed of a single auditory stimulus, significantly more flight behavior was evoked by a white noise CS than a tone CS (Fadok et al., 2017).

Importantly, discrimination studies that employ multiple auditory cues could be complicated both by variations in the ability of subjects to perceive different frequencies as well as potential innate valence associated with certain stimuli. For example, aversively conditioning a high intensity US with a 5 kHz CS+ followed by a generalization test using a higher salience CS- such as white noise could yield misleading conclusions if subjects exhibit escape behaviors to the CS- and, as is common, freezing is the only metric used to assess cue responsivity. Such confounds may be best avoided by assaying discrimination using tasks which measure behavioral responses to distinct patterns of a single, constant intensity sensory stimulus (e.g. drifting visual gratings of different orientation [Burgess et al., 2016]). Interpretation of discrimination studies that employ auditory stimuli would benefit from counterbalancing assignment of CS+ and CS- stimuli (Sanford et al., 2017), and also from use of stimuli at frequencies and SPL that are detectable but do not trigger active fear behaviors.

Severe stress can result in persistent generalization or sensitization of threat responding, such that stimuli which normally elicit little to no response come to evoke robust defensive behaviors. For example, in the stress-enhanced fear learning (SEFL) model, exposure to inescapable shocks in the absence of auditory stimuli results in nonassociative freezing to white noise in a novel context on the following day (Perusini et al., 2016). Although we observed white noise-elicited escape behavior in group 3 (‘gap’, Figure 1F and O), these responses cannot be attributed directly to generalization as it remains possible that some CS-US association formed despite the 60 s gap (i.e. via trace conditioning). Thus, the extent to which white noise can nonassociatively elicit active defensive behavior will need to be determined in future experiments where training is performed with a US presented in the complete absence of CS stimuli, as done in the SEFL model.

Previous work provided behavioral and neurophysiological evidence that SCS fear conditioned tone and white noise stimuli acutely elicit distinct defensive states indicative of different points along the threat imminence continuum (Fadok et al., 2017). We have found that these defensive states track with the frequency and intensity of the conditioned stimuli, not order of CS presentation during learning. This argues that threat imminence in this model is determined primarily via the salience of threat-predictive auditory stimuli which, together with recent experience (Mongeau et al., 2003) and current fear levels, determines the threshold for switching from freezing to flight. Our results contrast with those of another study (Dong et al., 2019), which reported that training with a ‘reversed SCS’ (white noise-tone-US) reverses the behaviors elicited (i.e. mice freeze to the WN but exhibit flight to the tones). As the experimental procedures used in both studies were essentially the same, the explanation for the discrepant results is presently unclear. However, while the B6J mice used here were obtained from JAX, the mice used in Dong et al. were of undefined substrain (‘C57Bl/6’) and obtained from a different vendor. Therefore, it is possible that different mouse strains utilize distinct neural processes to assess threat imminence. Future work will be required to determine if this is indeed the case and if so, the mechanistic underpinnings of such differences.

Although reversing the order of the white noise and tone stimuli during training did not qualitatively alter the type of behaviors elicited by the CSs, this switch did have a quantitative effect. Specifically, white noise elicited significantly less escape behavior when it preceded rather than followed the tone during training (Figure 1). One potential explanation for this result is that compound stimuli which increase in salience from CS1 to CS2 are more naturalistic and produce higher arousal levels and greater learning than the reverse order. Indeed, tonal stimuli which sweep from low up to high frequencies are rated by human observers as more alarming than high to low sweeps (Catchpole et al., 2004). Similarly, frequency upsweeps are associated with elevation of attention and arousal, whereas downsweeps are thought to have a calming effect (Owren and Rendall, 2001). Use of compound stimuli that either increase or decrease in salience from CS1 to CS2 might thus have opposing influences on arousal, resulting in either optimal or suboptimal states for sensory signal processing and learning (Aston-Jones and Cohen, 2005; McGinley et al., 2015; Yerkes and Dodson, 1908).

Finally, we note that conditioned responses exhibited at the onset of a CS can differ qualitatively from those displayed near CS offset (Holland, 1980). It thus remains possible that temporal factors make some contribution to defensive responding in SCS conditioning. Given the potent influence of stimulus salience, resolution of this issue will likely require the use of a SCS comprised of distinct component stimuli that can be clearly discriminated and yet are also matched for salience.

All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for all figures in MS Excel format, with primary measurements in one file and statistical analyses in another file.

1. 1. Aston-Jones G
   2. Cohen JD

   (2005) An integrative theory of locus coeruleus-norepinephrine function: adaptive gain and optimal performance

   Annual Review of Neuroscience 28:403–450.

   https://doi.org/10.1146/annurev.neuro.28.061604.135709

   • PubMed
   • Google Scholar
2. 1. Beckett SR
   2. Aspley S
   3. Graham M
   4. Marsden CA

   (1996) Pharmacological manipulation of ultrasound induced defence behaviour in the rat

   Psychopharmacology 127:384–390.

   https://doi.org/10.1007/s002130050102

   • PubMed
   • Google Scholar
3. 1. Blanchard RJ
   2. Agullana R
   3. McGee L
   4. Weiss S
   5. Blanchard DC

   (1992) Sex differences in the incidence and sonographic characteristics of antipredator ultrasonic cries in the laboratory rat (Rattus norvegicus)

   Journal of Comparative Psychology 106:270–277.

   https://doi.org/10.1037/0735-7036.106.3.270

   • PubMed
   • Google Scholar
4. 1. Blanchard RJ
   2. Parmigiani S
   3. Bjornson C
   4. Masuda C
   5. Weiss SM
   6. Caroline Blanchard D

   (1995) Antipredator behavior of Swiss-Webster mice in a visible burrow system

   Aggressive Behavior 21:123–136.

   https://doi.org/10.1002/1098-2337(1995)21:2<123::AID-AB2480210205>3.0.CO;2-E

   • Google Scholar
5. 1. Blanchard RJ
   2. Blanchard DC

   (1989) Antipredator defensive behaviors in a visible burrow system

   Journal of Comparative Psychology 103:70–82.

   https://doi.org/10.1037/0735-7036.103.1.70

   • Google Scholar
6. 1. Bouton ME
   2. Bolles RC

   (1980) Conditioned fear assessed by freezing and by the suppression of three different baselines

   Animal Learning & Behavior 8:429–434.

   https://doi.org/10.3758/BF03199629

   • Google Scholar
7. 1. Breska A
   2. Ivry RB

   (2018) Double dissociation of single-interval and rhythmic temporal prediction in cerebellar degeneration and Parkinson's disease

   PNAS 115:12283–12288.

   https://doi.org/10.1073/pnas.1810596115

   • PubMed
   • Google Scholar
8. 1. Burgess CR
   2. Ramesh RN
   3. Sugden AU
   4. Levandowski KM
   5. Minnig MA
   6. Fenselau H
   7. Lowell BB
   8. Andermann ML

   (2016) Hunger-Dependent enhancement of food cue responses in mouse postrhinal cortex and lateral amygdala

   Neuron 91:1154–1169.

   https://doi.org/10.1016/j.neuron.2016.07.032

   • PubMed
   • Google Scholar
9. 1. Carlson S
   2. Willott JF

   (1996) The behavioral salience of tones as indicated by prepulse inhibition of the startle response: relationship to hearing loss and central neural plasticity in C57BL/6J mice

   Hearing Research 99:168–175.

   https://doi.org/10.1016/S0378-5955(96)00098-6

   • PubMed
   • Google Scholar
10. 1. Catchpole KR
    2. McKeown JD
    3. Withington DJ

    (2004) Localizable auditory warning pulses

    Ergonomics 47:748–771.

    https://doi.org/10.1080/00140130310001629739

    • PubMed
    • Google Scholar
11. 1. Cuomo V
    2. Cagiano R
    3. De Salvia MA
    4. Mazzoccoli M
    5. Persichella M
    6. Renna G

    (1992) Ultrasonic vocalization as an Indicator of emotional state during active avoidance learning in rats

    Life Sciences 50:1049–1055.

    https://doi.org/10.1016/0024-3205(92)90100-4

    • PubMed
    • Google Scholar
12. 1. Dong P
    2. Wang H
    3. Shen XF
    4. Jiang P
    5. Zhu XT
    6. Li Y
    7. Gao JH
    8. Lin S
    9. Huang Y
    10. He XB
    11. Xu FQ
    12. Duan S
    13. Lian H
    14. Wang H
    15. Chen J
    16. Li XM

    (2019) A novel cortico-intrathalamic circuit for flight behavior

    Nature Neuroscience 22:941–949.

    https://doi.org/10.1038/s41593-019-0391-6

    • PubMed
    • Google Scholar
13. 1. Evans DA
    2. Stempel AV
    3. Vale R
    4. Ruehle S
    5. Lefler Y
    6. Branco T

    (2018) A synaptic threshold mechanism for computing escape decisions

    Nature 558:590–594.

    https://doi.org/10.1038/s41586-018-0244-6

    • PubMed
    • Google Scholar
14. 1. Fadok JP
    2. Krabbe S
    3. Markovic M
    4. Courtin J
    5. Xu C
    6. Massi L
    7. Botta P
    8. Bylund K
    9. Müller C
    10. Kovacevic A
    11. Tovote P
    12. Lüthi A

    (2017) A competitive inhibitory circuit for selection of active and passive fear responses

    Nature 542:96–100.

    https://doi.org/10.1038/nature21047

    • PubMed
    • Google Scholar
15. 1. Fanselow MS

    (1994) Neural organization of the defensive behavior system responsible for fear

    Psychonomic Bulletin & Review 1:429–438.

    https://doi.org/10.3758/BF03210947

    • Google Scholar
16. Book

    1. Fanselow MS
    2. Lester LS

    (1988)

    A functional behavioristic approach to aversively motivated behavior: predatory imminence as a determinant of the topography of defensive behavior

    In: Bolles RC, Beecher MD, editors. Evolution and Learning. Hillsdale, NJ: Erlbaum. pp. 185–211.

    • Google Scholar
17. 1. Heffner HE
    2. Heffner RS

    (2007)

    Hearing ranges of laboratory animals

    Journal of the American Association for Laboratory Animal Science 46:20–22.

    • PubMed
    • Google Scholar
18. 1. Holland PC

    (1977) Conditioned stimulus as a determinant of the form of the pavlovian conditioned response

    Journal of Experimental Psychology: Animal Behavior Processes 3:77–104.

    https://doi.org/10.1037/0097-7403.3.1.77

    • Google Scholar
19. 1. Holland PC

    (1979) The effects of qualitative and quantitative variation in the US on individual components of pavlovian appetitive conditioned behavior in rats

    Animal Learning & Behavior 7:424–432.

    https://doi.org/10.3758/BF03209696

    • Google Scholar
20. 1. Holland PC

    (1980) Influence of visual conditioned stimulus characteristics on the form of pavlovian appetitive conditioned responding in rats

    Journal of Experimental Psychology: Animal Behavior Processes 6:81–97.

    https://doi.org/10.1037/0097-7403.6.1.81

    • Google Scholar
21. 1. Koay G
    2. Heffner R
    3. Heffner H

    (2002) Behavioral audiograms of homozygous med(J) mutant mice with sodium channel deficiency and unaffected controls

    Hearing Research 171:111–118.

    https://doi.org/10.1016/S0378-5955(02)00492-6

    • PubMed
    • Google Scholar
22. 1. McGinley MJ
    2. David SV
    3. McCormick DA

    (2015) Cortical membrane potential signature of optimal states for sensory signal detection

    Neuron 87:179–192.

    https://doi.org/10.1016/j.neuron.2015.05.038

    • PubMed
    • Google Scholar
23. 1. Mongeau R
    2. Miller GA
    3. Chiang E
    4. Anderson DJ

    (2003) Neural correlates of competing fear behaviors evoked by an innately aversive stimulus

    The Journal of Neuroscience 23:3855–3868.

    https://doi.org/10.1523/JNEUROSCI.23-09-03855.2003

    • PubMed
    • Google Scholar
24. 1. Owren MJ
    2. Rendall D

    (2001) Sound on the rebound: bringing form and function back to the forefront in understanding nonhuman primate vocal signaling

    Evolutionary Anthropology: Issues, News, and Reviews 10:58–71.

    https://doi.org/10.1002/evan.1014

    • Google Scholar
25. 1. Perusini JN
    2. Meyer EM
    3. Long VA
    4. Rau V
    5. Nocera N
    6. Avershal J
    7. Maksymetz J
    8. Spigelman I
    9. Fanselow MS

    (2016) Induction and expression of fear sensitization caused by acute traumatic stress

    Neuropsychopharmacology 41:45–57.

    https://doi.org/10.1038/npp.2015.224

    • PubMed
    • Google Scholar
26. 1. Perusini JN
    2. Fanselow MS

    (2015) Neurobehavioral perspectives on the distinction between fear and anxiety

    Learning & Memory 22:417–425.

    https://doi.org/10.1101/lm.039180.115

    • PubMed
    • Google Scholar
27. 1. Sanford CA
    2. Soden ME
    3. Baird MA
    4. Miller SM
    5. Schulkin J
    6. Palmiter RD
    7. Clark M
    8. Zweifel LS

    (2017) A central amygdala CRF circuit facilitates learning about weak threats

    Neuron 93:164–178.

    https://doi.org/10.1016/j.neuron.2016.11.034

    • PubMed
    • Google Scholar
28. 1. Willott JF
    2. Carlson S
    3. Chen H

    (1994) Prepulse inhibition of the startle response in mice: relationship to hearing loss and auditory system plasticity

    Behavioral Neuroscience 108:703–713.

    https://doi.org/10.1037/0735-7044.108.4.703

    • PubMed
    • Google Scholar
29. 1. Yerkes RM
    2. Dodson JD

    (1908) The relation of strength of stimulus to rapidity of habit-formation

    Journal of Comparative Neurology and Psychology 18:459–482.

    https://doi.org/10.1002/cne.920180503

    • Google Scholar
30. 1. Zelikowsky M
    2. Bissiere S
    3. Hast TA
    4. Bennett RZ
    5. Abdipranoto A
    6. Vissel B
    7. Fanselow MS

    (2013) Prefrontal microcircuit underlies contextual learning after hippocampal loss

    PNAS 110:9938–9943.

    https://doi.org/10.1073/pnas.1301691110

    • PubMed
    • Google Scholar

Acceptance summary:

In this paper, the authors test assumptions about the basis of differential responding in a serial fear conditioning preparation, in which rodents are exposed to a tone->white noise->shock US. Normally rodents react to this by freezing to the tone and then exhibiting flight behavior to the white noise. Previous work has assumed that the changeover in behavior from freezing to flight is due to the temporal relationship to the US, with distal behaviors being directed at avoiding detection by a predator and proximal behaviors being directed at escape. The current study challenges this assumption, and instead shows that a substantial component of the changeover in behavior is instead driven by the inherently higher salience of the white noise that is usually used for the proximal cue. The reviewers are in agreement that this is an important study with regard to serial fear conditioning, and that it further makes an important general point for those of us interested in the intersection between behavior and neuroscience, which is that there are many factors that control the form of any behavioral response.

Decision letter after peer review:

Thank you for submitting your article "Stimulus Salience Determines Defensive Behaviors elicited by aversively conditioned serial compound auditory stimuli" for consideration by eLife. Your article has been reviewed by three peer reviewers, including Geoffrey Schoenbaum as the Reviewing Editor and Reviewer #1, and the evaluation has been Laura Colgin as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: Gavan McNally (Reviewer #2).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

Essential revisions:

The reviewers are in agreement that this is an important study with regard to serial fear conditioning, and that it further makes an important general point for those of us interested in the intersection between behavior and neuroscience, which is that there are many factors that control the form of any behavioral response. For this reason, counterbalancing and other procedures are very important. The current study is a case study of this. In our discussion, all three reviewers generally agreed that essential revisions could be dealt with in the text. All three appreciated the author's point regarding salience but felt that it would be better to acknowledge that many factors determine the "topography" of the response, instead of concluding that salience is the entire cause in all prior studies. As part of this, it was felt that putting this result in a larger context regarding the form of a behavior would be good. Peter Holland's work is particularly relevant, but there are many other examples. Possibly if the authors are unsure about this, they could let us know. There were also important concerns about the statistics raised by R2 (3-factor anova) and questions regarding whether longer intervals were used to rule out nonassociative responding, which are important to address in the revision. The reviews are included in their entirety below.

Reviewer #1:

In this paper, the authors test assumptions about the basis of differential responding in a serial fear conditioning preparation, in which rodents are exposed to a tone->white noise->shock US. Normally rodents react to this by freezing to the tone and then exhibiting flight behavior to the white noise. Previous work has assumed that the changeover in behavior from freezing to flight is due to the temporal relationship to the US, with distal behaviors being directed at avoiding detection by a predator and proximal behaviors being directed at escape. The current study challenges this assumption, and instead shows that a substantial component of the changeover in behavior is instead driven by the inherently higher salience of the white noise that is usually used for the proximal cue. Overall, the experiments are well done and provide convincing evidence supporting what is essentially a cautionary tale regarding the importance of careful, well-controlled behavioral designs. My criticism are entirely suggestions for caveating or softening the conclusions a bit.

There are two main areas I think could be made clearer and perhaps softened a bit. The first concerns the relationship between these findings and the general idea that behaviors to cues can differ based on temporal factors, salience, or even cue modality. I do not know the SCS field particularly, but even without a compound cue, it is well documented that both unconditioned and conditioned behaviors differ across time, particularly during a long (10, 20, 30s) cue. These behaviors and their relationships can also differ based on the amount of training, and the density of reward I believe. So, it seems to me, that there are many factors that can explain differential responding of the sort that seems to be dogmatically-highlighted in the SCS literature. I wonder if a more general introduction that acknowledges some of this complexity might be considered, versus what seems to be a dichotomy in the current introduction?

Related to this, I think the authors should soften their conclusions a bit. At present, it seems to me that they are saying the SCS effect is solely due to salience differences. While this may be the case, I think this is conclusion goes beyond what is necessary and the data. At best, what the authors demonstrate is that your salience can play a major role in determining the behavior. But I think this is not the same as staying the temporal relationship to the US is not important or never plays a role. Indeed, logic suggests, as the authors Introduction points out, that it surely should. The question just is what training procedures need to be used to demonstrate this conclusively – i.e. excluding salience and other factors. This I think is one of the main messages of this very nice study – that it is important to consider and control for these effects carefully. I think if the authors can make these points in a bit more nuanced way, it will improve the impact of the study.

Reviewer #2:

In an interesting series of experiments, Fadok et al., (2017) and Dong et al., (2019), reported that mice would show flight responses to auditory CSs that signalled an imminent shock US. These papers used a serial compound conditioning procedure whereby a tone CS was presented then a white noise CS then shock. Serial compound conditioning is, of course, a very old procedure. The novelty in these papers was the finding that mice would freeze to the distal tone CS then engage in active defense (escape, movement) to the proximal white noise CS. These effects were interpreted in terms a shift from passive to active defense as US imminence increased and were consistent with the important and influential predatory imminence models of Fanselow. Although Fadok et al., did not counterbalance the identity of the CSs (i.e. they used tone then white noise), Dong et al., showed the same effect with white noise then tone presentations (freezing to the white noise and flight the white noise). Fadok et al., also showed that the white noise itself was not aversive and did not elicit escape responses or flight behaviour in the absence of footshock (Extended data Figure 1H-J).

In the present manuscript, Hersman et al. provide a careful behavioral analysis of the topography of unconditioned and conditioned responses to white noise and tone CSs paired with shock. Their main claim is that physical properties of the CSs (frequency and sound pressure levels), not their temporal imminence, determines the topography of responding. These are very interesting experiments addressing an important question. On the one hand, if the field simply used appropriate counterbalancing of the identity of CSs in individual papers, we may not be having these discussions. On the other hand, the manuscript is a systematic investigation into the effects of auditory CS properties on the topography of defensive behaviour. The manuscript is well written, economical, and well presented.

I had the following three comments on the designs, analyses, and interpretation:

1) It is well established that multiple factors determine the topography of behaviour as conditioned responses. Imminence to the US is an important one, but so too are the physical properties of the CS. Holland (Holland, 1979; 1980a,b among others) has shown this convincingly for both appetitive and aversive Pavlovian conditioning. There are CS generated behaviours that can be increased across conditioning and there are US generated behaviours that can also increase across conditioning. So, any dichotomy between freezing and escape is not absolute if one is CS generated (escape) and the other is US generated (freezing).

I think the authors could embrace this complexity a little more. There are multiple determinants of the form of the CR. As conducted here, the physical properties of the CSs (frequency and sound pressure levels) are important, But, Dong et al., found the opposite. They showed robust escape responses to a tone when it was a proximal CS and freezing to the white noise when it was a distal CS. This difference is never really reconciled. Nor are the present data reconciled with the findings of Fadok et al. showing that the white noise CS in their experiments did not elicit escape in the absence of shock. This contrasts with Experiment 2/Figure 2 here.

2) Is there evidence for Pavlovian conditioning to the auditory CSs?

If the authors are seeking to argue that behaviour to the auditory CSs are conditioned responses, then they need to show evidence for conditioning.

I was struggling to understand the evidence the authors were invoking for Pavlovian conditioning to the CSs. The authors have three groups: two paired groups that receive a serial compound comprised of two auditory CS, white noise and tone, followed by a US. The groups differ simply in the order of the two CSs. The third group received tone then white noise with a 60 s interval between the offset of the CS and delivery of the US. This is an "unpaired" group. The inclusion of a control is to be commended. It is a conservative control, because it is really a trace conditioning, rather than unpaired, control. Regardless, from this kind of design, the evidence for conditioning to the CSs would be to show that freezing and escape responses were significantly greater in the two paired groups compared to the unpaired control (i.e., a 3-way ANOVA with a 3-way interaction driven by more responding in the two paired groups than the unpaired group). It is hard to tell from Figure 1 if this will come out. I suspect it will not for freezing but it may for average motion and escape score (see next point). I encourage the authors to consider this analysis. They need to persuade readers that they are studying learning before they persuade readers that CS salience determines topography of defensive behavior as conditioned responses. This same principle applies to Figure 1—figure supplement 1, as well as Figure 4. In fact, I really could not see any evidence that conditioning to the CSs occurred (if it is defined relative to the control unpaired group) in many key experiments in this paper.

A different solution could be to report pre-CS levels of freezing/escape for each CS presentation and show significant increases in freezing/escape during CS presentations relative to each 10s pre-CS period. Less ideal would be to show that there is more freezing/escape across CS presentations than in the pre-CS/baseline period.

The reason this is important is simply that one could interpret the data as showing contextual fear conditioning in each group upon which different unconditioned responses to the CSs are superimposed.

3) Is salience, imminence, or both important?

A key conclusion from this paper is that salience determines the topography of conditioned fear responses when these auditory stimuli are used. For example, in Experiment 1 the key conclusion here is white noise CS elicits escape behavior regardless of whether it is proximal or distal to the US whereas tone CS elicits freezing and not escape. In Experiment 4 a similar conclusion applies to the 12kHz vs.3 kHz tone. This seems reasonable based on inspection of the figures. To be sure, there is rarely if any escape to the tone CS in Experiment 1/Figure 1. However, I am not sure it is the complete answer. The related, critical question is whether the topography of defensive behavior to the CSs depends significantly on their temporal relation to the US? In Experiment 1, does the white noise CS elicit more escape responses when it is proximal rather than distal to the US? This analysis, like the evidence for conditioning described above, simply requires a 3-way ANOVA, in this case comparing behaviors between the two Paired groups, and specifically testing a 3-way interaction (G1 vs. G2 or G4 vs. G5 x CS [noise vs. tone] or [3kHz vs. 12 kHz tone] x trials [1 – 10]). This would test whether the difference between each response to the two kinds of CSs is or is not significantly greater between the two paired groups. If identity of the CS, not its temporal order, is important then this interaction should not be significant. If temporal order is important, then this interaction will be significant. For example, is the difference between the motion (Figure 1J) or escape scores (Figure 1M) for white noise versus tone significantly greater in Group 1 versus Group 2? If this interaction is significant, it lends support to the claim that temporal imminence matters, at least in part. If it is not, it lends support to the claim that temporal imminence does not matter.

I realize the authors want to focus on the lack of any real escape responses to the tone CS in Experiment 1 or the 3 KHz tone in Experiment 4, and this is important as well as obvious from the data. However, the data appear more nuanced. The data as presented do appear to suggest that the extent of escape responses to the white noise are determined, at least in part, by the temporal relation of the CS to the US (imminence) as well on the specific physical properties of the CS (see point 1).

I am not recommending any additional data collection. The manuscript is interesting, challenging, and also instructive. However, I think:

1) Further analyses are needed, specifically the 3-way ANOVAs described above. One set of 3 ANOVAs asking whether each behavior is different across trials for the control vs. two paired groups. A second set asking whether there is a difference between the two paired groups for each behavior. The authors strategy of analyzing the groups separately undermines their key conclusions.

2) deeper consideration of the role of CS and US generated behaviors as conditioned responses is warranted as is further attempts to reconcile these findings with Fadok et al., and Dong et al.

Reviewer #3:

This study examines the recent claim that conditional stimuli (CSs) presented proximal to footshock elicit escape responses while distal CSs induce freezing behavior. The authors make a convincing case that escape responses are controlled by the salience of the CS rather than its proximity to shock. Previous work had used white noise as the cue that was presented immediately prior to shock. The current experiments counterbalanced this cue with a pure tone and found that white noise produced bursting whether it was located proximal or distal to footshock. In contrast, the pure tone induced freezing even when it was presented immediately prior to shock. However, if the salience of the pure tone was enhanced by increasing its intensity, then it was able to drive some escape behaviors. Therefore, in contrast to previous claims, defensive behaviors elicited by auditory CSs (in mice) are primarily controlled by stimulus salience. These results have important implications for studies that use serial compound cues to study proximal and distal threat responses.

One thing I would like the authors to address is that there appears to be more bursting to the white noise stimulus when it is presented as CS2 compared to CS1. This suggests there is an interaction between stimulus salience and proximity to threat (i.e. white noise produces more bursting than a pure tone and the size of this response is larger when the stimulus occurs proximal to footshock).

A second issue is the amount of nonassociative CRs that occurs in the unpaired groups. It is possible that with the strong shocks that are used, mice are able to associate the CS with the US. That is, the unpaired procedure is actually a trace conditioning procedure. Have the authors tried using a longer gap between the CS and shock? Or presenting the CS and shock on different training days? Does this reduce the amount of nonassociative responding?

Essential revisions:

Reviewer #1:

1) There are two main areas I think could be made clearer and perhaps softened a bit. The first concerns the relationship between these findings and the general idea that behaviors to cues can differ based on temporal factors, salience, or even cue modality. I do not know the SCS field particularly, but even without a compound cue, it is well documented that both unconditioned and conditioned behaviors differ across time, particularly during a long (10, 20, 30s) cue. These behaviors and their relationships can also differ based on the amount of training, and the density of reward I believe. So, it seems to me, that there are many factors that can explain differential responding of the sort that seems to be dogmatically-highlighted in the SCS literature. I wonder if a more general introduction that acknowledges some of this complexity might be considered, versus what seems to be a dichotomy in the current Introduction?

This is an important point, and a good suggestion. We have:

a) Added a new paragraph (Introduction) that acknowledges the complexity of behavioral responses to cue stimuli, along with references to classic work which demonstrated this.

b) Moved the section highlighting the difference between our results and those of Dong et al., to the Discussion section.

2) Related to this, I think the authors should soften their conclusions a bit. At present, it seems to me that they are saying the SCS effect is solely due to salience differences. While this may be the case, I think this is conclusion goes beyond what is necessary and the data.

Agreed. We have altered the language used to summarize our findings to avoid giving the impression that we believe salience is the sole explanation of behavioral responding in the SCS paradigm. Specifically, we:

a) State that stimulus salience is the ‘primary determinant’ (Abstract), or ‘major factor’ (Introduction) determining behaviour.

b) Conclude that ‘audio frequency properties strongly influence defensive behaviors elicited by SCS’ conditioned stimuli (Discussion section), and that salience is the “primary means by which mice assess imminence…” (Discussion section).

c) Added a new paragraph to acknowledge “it remains possible that temporal factors make some contribution to defensive responding in SCS conditioning” (Discussion section).

Reviewer #2:

3) It is well established that multiple factors determine the topography of behaviour as conditioned responses. Imminence to the US is an important one, but so too are the physical properties of the CS. Holland (Holland, 1979; 1980a,b among others) has shown this convincingly for both appetitive and aversive Pavlovian conditioning. There are CS generated behaviours that can be increased across conditioning and there are US generated behaviours that can also increase across conditioning. So, any dichotomy between freezing and escape is not absolute if one is CS generated (escape) and the other is US generated (freezing). I think the authors could embrace this complexity a little more.

Agreed, this was also suggested by reviewer #1; please see our response on this point above (issues #1 and 2).

4) There are multiple determinants of the form of the CR. As conducted here, the physical properties of the CSs (frequency and sound pressure levels) are important, But, Dong et al., found the opposite. They showed robust escape responses to a tone when it was a proximal CS and freezing to the white noise when it was a distal CS. This difference is never really reconciled.

We made multiple attempts to contact the corresponding senior author of the Dong et al. study via e-mail in order to obtain additional details of their experiments that could help reconcile the differences with our results. Unfortunately, they did not respond to our e-mails, and so our ability to explain the discrepant results is limited to scrutinizing methodological details reported in their paper, which appear essentially the same as those that we employed. As the Dong et al. study did not specify the particular mouse substrain used in their experiments, we have proposed this as a potential explanation, and suggest that future work would be required to determine if different mouse substrains use distinct processes to determine threat imminence (Discussion section).

5) Nor are the present data reconciled with the findings of Fadok et al. showing that the white noise CS in their experiments did not elicit escape in the absence of shock. This contrasts with Experiment 2/Figure 2 here.

The results in Experiment 2/Figure 2 of our manuscript were done in unconditioned mice and did not examine escape behavior, but rather physiological (pupil dilation) and simple locomotor responses (movement of head-fixed animals on a running wheel) to tone and white noise stimuli to which the animals had not been previously exposed. Therefore, our data are not directly comparable to the experiments that Fadok et al. performed on conditioned animals. To clarify this for readers, we modified the text to read “simple locomotor responses on the running wheel” in the section detailing the results of Figure 2 (Results section).

This being said, it is important to note that supplemental figure 1 of the Fadok et al., study did in fact provide evidence that white noise is more salient and/or threatening to mice than tone stimuli. Specifically, when they performed conventional fear conditioning to a simple CS composed of just a single auditory stimulus, significantly more flight behavior was evoked by a white noise CS than a tone CS (Fadok et al., 2017, Extended Data Figure 1F). We have added explicit mention of this point to the manuscript, as well as other evidence that white noise may be uniquely salient if not threatening to mice (Discussion section).

6) If the authors are seeking to argue that behaviour to the auditory CSs are conditioned responses, then they need to show evidence for conditioning.

I was struggling to understand the evidence the authors were invoking for Pavlovian conditioning to the CSs. The authors have three groups: two paired groups that receive a serial compound comprised of two auditory CS, white noise and tone, followed by a US. The groups differ simply in the order of the two CSs. The third group received tone then white noise with a 60 s interval between the offset of the CS and delivery of the US. This is an "unpaired" group. The inclusion of a control is to be commended. It is a conservative control, because it is really a trace conditioning, rather than unpaired, control. Regardless, from this kind of design, the evidence for conditioning to the CSs would be to show that freezing and escape responses were significantly greater in the two paired groups compared to the unpaired control (i.e., a 3-way ANOVA with a 3-way interaction driven by more responding in the two paired groups than the unpaired group).

Thank you for raising this important point; we have addressed it as follows:

a) TN-WN SCS conditioning data (Figure 1) was analyzed using 3-way ANOVA:

i) Freezing to TN was higher in paired Group 1 (G1) than unpaired Group 3 (G3) (3-Way ANOVA on Day 2, G1 vs. G3, Main Effect of Stimulus (F(1,23) = 429.5, p<0.0001), Main Effect of Trial (F(4,92) = 5.083, p=0.001), Group X Stimulus Interaction, (F(1,23) = 27.51, p<0.0001); Follow-Up Two-Way RM ANOVA for freezing just to the tone stimulus, Main Effect of Trial (F(3.2, 75.9) = 3.79, p<0.05), Main Effect of Group (F(1,23) = 6.41, p<0.05)).

ii) Motion to WN was higher in G1 than G3 (3-Way ANOVA on Day 2, G1 vs. G3 Activity, Main Effect of Stimulus (F(1,23) = 69.89, p<0.0001), Main Effect of Group (F(1,23) = 11.75, p<0.01), Group X Stimulus Interaction (F(1,23) = 19.77, p<0.001); Follow-up Two-Way RM ANOVA for activity just to WN stimulus, Main Effect of Group (F(1,23) = 15.79, p<0.001)).

iii) Escape scores during WN were higher in G1 than G3 (3-Way ANOVA on Day 2, G1 vs. G3 Escape Score, Main Effect of Stimulus (F(1,23) = 67.85, p<0.0001), Main Effect of Group (F(1,23) = 15.98, p<0.001), Group X Stimulus Interaction (F(1,23) = 20.41, p<0.001); Follow-up Two-Way RM ANOVA for escape score just to WN, Main Effect of Group (F(1,23) = 18.25, p<0.001)).

iv) Paired Group 2 (G2) displayed significantly different behavioral responses to the two CS stimuli across all metrics (2-Way ANOVA results already reported) and in the same direction as G1 (i.e. more freezing during TN, more activity and escape during WN). However, G2 did not show significantly different magnitude of these responses compared with G3 (3-Way ANOVA on Day 2, G2 vs G3, no group differences or interactions for freezing, motion, or escape score). We hypothesize that this may be due to impaired learning when using a high-to-low salience SCS (see more on this issue below in point ‘3’).

b) To directly demonstrate an acute conditioned freezing response to the TN stimulus, we also performed new experiments in which a tone test in a novel context was performed following SCS conditioning of mice trained on group1 (paired TN-WN-US) or group 3 (gap TN-WN-gap-US) protocols. Whereas mice in the gap group (G3 protocol) did not show significantly increased freezing between baseline and tone onset (p>0.05), paired (G1 protocol) mice exhibited robust freezing upon tone onset (p<0.001)(Two-Way RM ANOVA, Main Effect of Stimulus (F(1,13) = 19.98, p<0.001), Stimulus X Group Interaction (F(1,13) = 5.492, p<0.05), Sidak’s comparisons to determine which group drives the Main Effect of Stimulus). This data has been added to a new figure supplement (Figure 1—figure supplement 1).

c) Additional evidence that conditioning occurred to both TN and WN stimuli comes from the experiments detailed in Figure 3:

i) Mice in paired G1 showed significantly higher motion and escape scores than gap G3 mice to TN stimuli in the tone SPL step test (Figure 3F-I).

ii) Mice in paired G1 showed significantly higher freezing than gap group G3 in response to WN stimuli at low SPL, and G1>G3 for escape score in response to WN at high SPL (Figure 3M-O).

iii) To clarify this for readers, we modified the text to indicate that “group 1 responses are at least in part influenced by perceived threat levels which are a function of conditioned fear” (Results section).

d) For the two-tone SCS experiments (Figure 4):

i) Motion to 12 kHz tone was higher in G4 than G6: 2-Way RM ANOVA, G4 vs. G6, Motion to 12 kHz tone: Main Effect of Trial (F(2.5, 45.7) = 3.31, p<0.05), Main Effect of Group (F(1,18) = 8.64, p<0.01).

ii) Escape score to 12 kHz tone was higher in G4 than G6: 2-Way RM ANOVA, G4 vs. G6; Main Effect of Trial (F(2.7, 48.0) = 4.36, p<0.05), Main Effect of Group (F(1,18) = 10.1, p<0.01).

iii) Escape score to 12 kHz tone was higher in G5 than G6: 2-Way RM ANOVA, G5 vs G6, Escape score to 12 kHz tone: Main Effect of Group (F(1,18) = 4.49, p<0.05).

iv) Freezing to 3 kHz tone was higher in G4 than G6 in a tone test: Though groups did not differ in freezing behavior to the 3 kHz tone during conditioning, this difference was revealed in novel context tone test (Figure 4N-P).

In sum, these analyses indicate that conditioning to the auditory stimuli occurred in groups 1 and 4, and that the active and passive behaviors elicited by these stimuli are at least in part a function of Pavlovian conditioning. Moreover, reversing the order of TN and WN (or 3 and 12 kHz tones) during conditioning does not reverse the behaviors elicited by these stimuli. Details of these analyses have been added to the text (Results section), a new figure has been added (Figure 1—figure supplement 1), and the specific statistical tests performed appended to the source data sheets.

7) The related, critical question is whether the topography of defensive behavior to the CSs depends significantly on their temporal relation to the US? In Experiment 1, does the white noise CS elicit more escape responses when it is proximal rather than distal to the US?

Reversing order of SCS component stimuli did not reverse behaviors elicited (i.e. both G1 and G2 froze more during TN presentations and exhibited higher motion and escape in response to WN). Therefore, temporal relationship of a CS to the US is not the key factor determining whether mice execute an active or passive behavior. However, the data do clearly indicate that SCS order has an effect of the magnitude of behavioral responding; this was revealed in the 3-way ANOVA analyses, which showed that G2 (paired) active and passive behaviors are not significantly different from G3 (gap). One explanation for these data is that the reversed WN-TN SCS may impair learning, and we have added a new section that explicitly highlights these quantitative (but not qualitative) differences between G1 and G2 behavior, as well as propose potential explanations for why a high-to-low salience SCS might impair learning (Discussion section). In addition, although salience appears to be the major explanation for differential responding in SCS conditioning, we acknowledge that some contribution of temporal association cannot be excluded, and so added a new statement to this effect (Discussion section).

Reviewer #3:

8) One thing I would like the authors to address is that there appears to be more bursting to the white noise stimulus when it is presented as CS2 compared to CS1. This suggests there is an interaction between stimulus salience and proximity to threat (i.e. white noise produces more bursting than a pure tone and the size of this response is larger when the stimulus occurs proximal to footshock).

Yes, this is an important observation also noted by reviewer #2. Please see our response on this point above (issue #7).

9) A second issue is the amount of nonassociative CRs that occurs in the unpaired groups. It is possible that with the strong shocks that are used, mice are able to associate the CS with the US. That is, the unpaired procedure is actually a trace conditioning procedure. Have the authors tried using a longer gap between the CS and shock? Or presenting the CS and shock on different training days? Does this reduce the amount of nonassociative responding?

Agreed, this is an important point which we have addressed as follows:

a) We changed the name of Groups 3 and 6 from ‘unpaired’ to ‘gap’ to indicate that these groups are not fully ‘unpaired’, and to reflect the possibility that some CS-US association may have developed despite the 60s gap between the SCS and US.

b) We added a new paragraph to explicitly acknowledge that the gap group may have undergone trace conditioning, and that future work would be needed to determine the extent to which the behaviors exhibited by the gap groups are nonassociative (Discussion section).

Author details

1. Sarah Hersman

   F.M. Kirby Neurobiology Center, Boston Children's Hospital and Harvard Medical School, Boston, United States

   Contribution

   Formal analysis, Investigation, Writing - original draft

   Competing interests

   No competing interests declared

2. David Allen

   F.M. Kirby Neurobiology Center, Boston Children's Hospital and Harvard Medical School, Boston, United States

   Contribution

   Software, Investigation, Methodology

   Competing interests

   No competing interests declared

3. Mariko Hashimoto

   F.M. Kirby Neurobiology Center, Boston Children's Hospital and Harvard Medical School, Boston, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

4. Salvador Ignacio Brito

   1. F.M. Kirby Neurobiology Center, Boston Children's Hospital and Harvard Medical School, Boston, United States
   2. Program in Neuroscience, Harvard Medical School, Boston, United States

   Contribution

   Investigation

   Competing interests

   No competing interests declared

5. Todd E Anthony

   1. F.M. Kirby Neurobiology Center, Boston Children's Hospital and Harvard Medical School, Boston, United States
   2. Program in Neuroscience, Harvard Medical School, Boston, United States
   3. Departments of Psychiatry and Neurology, Boston Children's Hospital, Boston, United States

   Contribution

   Conceptualization, Formal analysis, Writing - original draft, Writing - review and editing

   For correspondence

   todd.anthony@childrens.harvard.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-7284-7556

• 2,543

  Page views

• 332

  Downloads

• 17

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.