Spontaneously hypertensive rats manifest deficits in emotional response to 22-kHz and 50-kHz ultrasonic playback

Many symptoms used routinely for human psychiatric diagnosis cannot be directly observed in animals which cannot describe their internal states. However, the ultrasonic vocalizations (USV) rodents use to communicate their emotional states can be measured. USV have therefore become a particularly useful tool in brain disease models. Spontaneously hypertensive rats (SHR) are considered an animal model of attention deficit hyperactivity disorder (ADHD) and schizophrenia. However, the specifics of SHR ’ s behavior have not been fully described and there is very little data on their USV. Recently, we developed a communication model


Introduction
Many of the symptoms used to establish psychiatric diagnoses in humans (e.g. hallucinations, sadness, guilt) cannot be directly measured in animals, since they cannot verbally convey to humans what they "think" or "feel" (Hitzemann, 2000), and the correspondence between other human symptoms and rodent correlates is only approximate (Nestler and Hyman, 2010).
However, the premise that the structure of rats' ultrasonic vocalizations (USV) could be used to assess their internal states in vocal expressions of emotional arousal has been proposed (Brudzynski, 2021). 2019; Wöhr and Schwarting, 2013). Aversive and appetitive USV emitted by rats or played from speakers evoke physiological and emotional changes in conspecifics (Brudzynski, 2013).
SHR are the most studied animal model of essential hypertension (Pinto et al., 1998). This rat strain was obtained by breeding Wistar-Kyoto (WKY) rats with high blood pressure (Okamoto and Aoki, 1963). There has been an accumulation of reports on SHR exhibiting hyperactivity, impulsivity, poor sustained attention and deficient learning and memory (Heal et al., 2008;Meneses and Hong, 1998;Wells et al., 2010). SHR are presently the only strain that shows major behavioral symptoms of ADHD (Meneses et al., 2011a;Sagvolden and Xu, 2008). Furthermore, SHR, compared to Wistar rats, are considered a model of schizophrenia. Firstly, SHR have elevated locomotor activity (Calzavara et al., 2011a), which can correspond to positive symptoms of schizophrenia (Lipska and Weinberger, 2000). They also show behaviors analogous to negative symptoms of schizophrenia (Kapur, 2003) manifested as a lower drive for social interaction (O'Tuathaigh et al., 2010;Sams-Dodd, 1998). SHR also have deficits in startle prepulse inhibition (PPI; Levin et al., 2011a) reflecting a neural sensorimotor deficit present in many neuropsychiatric disorders (Kapur, 2003). The altered behaviors of SHR could be interpreted as cognitive impairments in some tests relating to aversive situations such as fear conditioning and latent inhibition (Calzavara et al., 2009). Finally, schizophrenia-like behavioral abnormalities in young SHR could be reversed by antipsychotic drugs (Niigaki et al., 2019).
Rats emit USV to express their internal states to their conspecifics (Brudzynski, 2013). Though USV are not considered a form of rat language, they could however reflect and model some aspects of human speech and its deficits (Konopka and Roberts, 2016). ADHD patients generally have difficulties regulating attention, inhibition, and maintaining information in working memory (for review, see Pievsky and McGrath, 2018). Notably, language impairment and ADHD are also often comorbid (Rielly et al., 1999;Tannock, 1996). Schizophrenia patients also have speech problems, defined as formal thought disorder (an impaired capacity to sustain coherent discourse) and alogia (poverty of speech, i.e. brief and concrete replies with restricted spontaneous speech, and poverty of content, i.e. abstract, repetitive, or stereotyped speech; Sajatovic, 2012). Even though SHR are models of two mental disorders, little is known about the nature of their vocalization. The only report (to the best of our knowledge) was done by Zhang-James et al. (2014), they measured female SHR's USV emitted to bedding coming from cages of unfamiliar males. SHR vocalized as often as WKY and less often than Sprague-Dawley rats. Describing SHR's vocalization would therefore be an important factor to confirm the use of these animals as a model of ADHD and schizophrenia.
Production of 50-kHz USV is significant in initiating and maintaining social contacts (for review, see Brudzynski, 2021). Since SHR present impaired social interactions (O'Tuathaigh et al., 2010;Sams-Dodd, 1998), their responsiveness to social signals like 50-kHz USV, could also be impaired. Therefore, we hypothesized that SHR will respond to 50-kHz playback with less USV than normotensive control rats, similar to results obtained by Zhang-James et al. (2014).
Recently, we discovered that previously fear-conditioned Wistar rats, when subsequently exposed to both 50-kHz and 22-kHz sound-playback, displayed features of hypervigilance (Olszyński et al., 2021), a symptom of PTSD in human patients (Martinez et al., 2016). Over-responsivity to environmental stimuli is also observed in ADHD patients (Lane and Reynolds, 2019). Furthermore, human ADHD subjects have deficits in the hippocampus and ventromedial prefrontal cortex functions, similar to those documented in PTSD subjects compared to traumatized controls without PTSD (Spencer et al., 2017). Spencer et al. (2017) hypothesize that ADHD may be an antecedent risk factor for PTSD and is thus associated with a neurobiological vulnerability for PTSD. We hypothesized that SHR's response to aversive 22-kHz sounds will be strengthened (compared to Wistar rats), and even more exaggerated in SHR preexposed to foot-shock.
HR is one of the most studied psychophysiological parameters in anxiety disorders. Particularly in humans, emotions influence physiologic responses like HR and blood pressure (Siegel et al., 2018). Both valence (positivity/negativity) and arousal (low/high) are important when considering how emotions are associated with physiological reactivity. In general, low arousal leads to little change in HR, whereas high arousal is associated with greater HR responsiveness (Gordon and Mendes, 2021). It has been proposed that evaluating a situation as threatening or stressful, combined with increases in HR, may consequently lead to greater feelings of anxiety (Jones et al., 2009;Mallorqui-Bague et al., 2016). Therefore, it would be important to measure HR changes to stimuli carrying an emotional charge  in SHR, rats with altered cardiovascular system.
To verify our hypotheses, we fear-conditioned SHR with one, six or ten shocks. One day after conditioning, animals were investigated in our recently-developed communication model, i.e. exposed to pre-recorded playbacks (22-kHz and 50-kHz calls) in home-cage-like conditions (Olszyński et al., 2020;Olszyński et al., 2021). Locomotor activity, USV emission and HR were assessed. All SHR experiments were conducted in parallel with control Wistar rats and their data were analyzed together. Data on control Wistar rats alone were previously published (Olszyński et al., 2021) and some of these results are reported again in the present manuscript for comparison.

Animals and housing
All experiments were approved by the Second Local Ethics Committee for Animal Experimentation in Warsaw. Naïve adult male spontaneously hypertensive rats (SHR, 7 weeks of age, from Mossakowski Medical Research Institute, Polish Academy of Sciences, Warsaw, Poland) were housed in pairs in two separate rooms: one with control animals (no electric shock, 0-Trial, n = 31), and conditioned rats (receiving 1, 6 or 10 conditioning stimuli; 1-Trial, n = 17; 6-Trial, n = 17; and 10-Trial, n = 15, respectively) in the second room. Standard chow and water were provided ad libitum, a 12 h light-dark cycle and an ambient temperature of 22-25 • C. Fear conditioning was conducted between 15:00 and 24:00 h when the overall noise level in the animal house was low. All playback experiments were conducted during the light cycle (9:00-21:00 h) during weekends. There were four weeks between rat arrival and the start of the experiment. In the first week, the rats were habituated to the new facilities. All animals were handled once for 2 min every day for 12 days before the start of the experiment. The three experimenters each had at least 4 handling sessions with the rats. Surgery was performed on the third week. Results obtained from SHR were compared with analogous Wistar rats groups (i.e. 0-, 1-, 6-, 10-Trial). Experiments were conducted at the same time with randomized and counterbalanced number of rats in corresponding groups. Detailed analysis of Wistar rats alone was published before (Olszyński et al., 2021).

Surgical procedures
A radiotelemetric transmitter (HD-S10, Data Sciences International, St. Paul, MN, USA) was implanted into the abdominal aorta as previously described (Olszyński et al., 2020;Olszyński et al., 2021). Rats were given at least seven days post-surgery for recovery before the start of the experiment. During recovery, the animals were handled and habituated to the conditions of the playback experiment 4 times over the course of the week of recovery. The animals were all 12 weeks of age at the start of the experiment.

Experimental design and settings: Fear conditioning
The animals were transported individually to the fear-conditioning K.H. Olszyński et al. room and placed in a sound-attenuated fear conditioning apparatus (MED-VFC2-USB-R, Med Associates, Fairfax, VT, USA) with interior of 54.64 × 64.04 × 29.21 cm, lined with sound attenuating foam (2.54 cm width), with the inner fear conditioning chamber (VFC-008-LP, Med Associates, USA) with interior of 29.53 × 23.50 × 20.96 cm and a grid floor with spacing of 15.62 mm between 4.75-mm-thick metal bars. Each animal was habituated to the conditioning cage for 10 min with no light inside; rats' freezing, defined as the absence of movement for at least 1 s, was scored automatically by Med Associates Video Freeze software during the first 5 min. Freezing was calculated by linear method of scoring episodes where the animal movement, i.e. changes in pixels between frames, was under the set motion threshold of 18 (default settings) lasting for at least 30 video frames. The cage was cleaned between animals using detergent, wiped using 10% ethanol and allowed to dry. The next day, rats were placed in the conditioning cage with no light inside. After 5-min habituation, they received 1, 6 or 10 conditioning stimuli which consisted of a 20-s long white light, i.e. LED stimulus light behind a white diffuser (2.5 cm diameter), co-terminating with an electric foot-shock (1 s, 1 mA). The inter-trial interval (ITI) ranged from 180 to 300 s (mean, 240 s; comp. Lindquist et al., 2004). Therefore, the conditioning procedure differed in length between groups: 9 min 20 s for 1-Trial, 31 min for 6-Trial, and 48 min 20 s for 10-Trial groups. An equal-time-length control group (no shock) was used for each group. A playback experiment was given one day later (see the following section); the following day (two days after conditioning), rats were returned to the same context to measure freezing levels (Test). After 5-min habituation, rats were exposed to three blocks of 20 s of white light (CS) followed by 5 min of silence. Freezing was evaluated Fig. 1. Examples of various 50-kHz USV (A), short (B) and long 22-kHz USV (C), emitted by Wistar rats and SHR. Examples of 50-kHz and short 22-kHz USV were taken from responses to ultrasonic playback, whereas long 22-kHz USV were from responses to shocks. Both strains emitted several types of 50-kHz USV, e.g. modulated (0-, 1-Trial rats), flats (6-Trial), mixed (10-Trial); examples of the types are shown; please note, these are not representative to particular experimental groups. For USV's detailed characteristics see Table 2 (for 50-kHz USV) and Table 3 (for 22-kHz USV). In most cases, SHR emitted 50-kHz USV of lower MPF than Wistar rats. during the habituation and exposure to CS. The conditioning procedure was executed by an investigator not involved in the playback experiment.

Experimental design and settings: Playback experiment
One day after the conditioning, the rats were transferred into individual experimental cages, identical to home cages (plastic; 37 × 25 × 16 cm), and transported to the experimental room, where under white light, in the absence of the experimenter and other rats in the room, acoustic stimuli were presented through an ultrasonic speaker (Vifa, Avisoft Bioacoustics, Berlin, Germany), placed just above the shorter side of the cage, connected to an UltraSoundGate Player 116 (Avisoft Bioacoustics). USV emitted by the rat were recorded by a CM16/CMPA condenser microphone (UltraSoundGate, Avisoft Bioacoustics) suspended 33 cm above the center of the cage floor, 20 cm away from the speaker. In this configuration, calls from the speaker were still visible in the recording (monitoring of playback), but they were distinctively weaker than USV emitted from the cage. Both playback and recording were performed using Avisoft Recorder USGH software (Avisoft Bioacoustics 4.2.28). The locomotor activity of the animal was recorded with a camera (acA1300-60gc, Basler AG, Ahrensburg, Germany) mounted above the cage and EthoVision XT software (version 10, Noldus, Wageningen, Netherlands). Signals from the radiotelemetric transmitters were collected by receivers located under the cage floor and then recorded with Ponemah software (version 6.32, Data Sciences International, St. Paul, MN, USA).

Ultrasonic-playback presentation
Upon placing a rat into the experimental cage, a 10 min of silence with a turned-on speaker, i.e. background noise of 20.6 ± 0.2 dB, was followed by four 10-s-long sets of signals, separated by 5-min-long silence intervals (see Fig. 1 in Olszyński et al., 2020). Four sets of signals (playbacks) were presented in counterbalanced order to each rat: i. 50-kHz natural calls (referred to as "50-kHz USV"), total of 84 calls, in 3 repeats, frequency modulated and trill type calls, of 49.2 to 73.4 kHz frequency and 58.6 ± 0.7 kHz mean peak frequency, 28.4 ± 1.6 ms duration, 31.9 ± 0.6 dB sound pressure, 91.4 ± 1.4 ms average sound interval; these were recorded during rats' exploration of an empty cage with bedding soiled by a conspecific; the sounds from the original recording were cleaned from background noise; ii. 50-kHz softwaregenerated tones ("50-kHz tones"), 32.6 ± 0.7 dB; iii. 22-kHz natural calls ("22-kHz USV"), 24 calls in 8 repeats, 21.4-23.0 kHz, 22.1 ± 0.1 kHz, 375.3 ± 21.6 ms, 38.3 ± 1.2 dB, recorded during fear conditioning, with average sound interval of 47.3 ± 0.5 ms (Avisoft Bioacoustics, 2014); and iv. 22-kHz software-generated tones ("22-kHz tones"), 43.3 ± 3.0 dB. However, the sound playbacks of the same frequency range, i. e. 50-kHz USV and 50-kHz tones (or 22-kHz USV and 22-kHz tones), always followed each another. Artificial tones were generated based on natural ones (mean peak frequency, duration, pauses between tones in the set, but with no frequency modulations) using Avisoft SASLab Pro (Avisoft Bioacoustic). Calls were presented with a sampling rate of 200 kHz in 16-bit format. The sound pressure levels of the background noise and playback signals were assessed in the middle of the test cage's floor, at the height of the animals' typical head position, facing the speaker.

Analysis of USV and locomotor activity
Recordings were transferred to SASLab Pro 5.2.14 (Avisoft Bioacoustics), and a fast Fourier transform was conducted (512 FFTlength, 100% frame, FlatTop window and 75% time window overlap), resulting in high resolution spectrograms (frequency resolution: 391 Hz; time resolution: 0.64 ms). An experienced user marked USV on the spectrogram. The minimum duration of a single marked element was 1 ms. For analysis, mean peak frequency and element duration were taken via values measured by the software. Based on these values, USV were assigned to one of three categories: 50-kHz (mean peak frequency, MPF >32 kHz), short 22-kHz (MPF of 18-32 kHz, <0.3 s duration), long 22-kHz (MPF of 18-32 kHz, >0.3 s duration). An automated video tracking system (Ethovision XT 10, Noldus, Wageningen, The Netherlands) was used to measure the total distance traveled (cm), a measure for general locomotor activity, and proximity to speaker, i.e. (%) time spent in the speaker's half of the cage. Center-point of each animal's shape was used as a reference point for measurements of locomotor activity thus registering only full-body movements, i.e. distance traveled by each rat.

Statistical analysis
All data were analyzed using non-parametric Friedman, Wilcoxon, Mann-Whitney, and Fisher's exact tests with GraphPad Prism 8.4.3 (GraphPad Software, San Diego, CA, USA); the p values are given, p < 0.05 as the minimal level of significance. Figures were prepared using GraphPad Prism 8.4.3 software and depict average values with a standard error of the mean (SEM). Based on preconditioning evaluation (Fig.  S1), two rats from the 6-Trial Wistar group were excluded as outliers (i. e., emitting exceptionally many USV, >3 x Standard Deviation) and subsequently removed from the analysis. However, every reported significant p value was verified to be present with the two outlying rats included (apart from a few exceptions within the supplementary tables marked appropriately). In certain comparisons, results from playbacks of natural USV and artificial tones of the same frequency band, i.e. 50-kHz USV and 50-kHz tones (or 22-kHz USV and 22-kHz tones), were pooled together for analysis and are presented as "sounds".

SHR vocalized in ultrasonic range and demonstrated all major USV types
In general, SHR emitted the same types of USV as our control Wistar rats and rats described before by others (Wöhr and Schwarting, 2010) to both aversive and appetitive conditions (Fig. 1). During conditioning sessions, SHR produced mainly long 22-kHz USV (Table S1) and some short 22-kHz USV (Table S2). During appetitive playback sessions, SHR vocalized mainly 50-kHz USV and occasionally short 22-kHz USV at the onset of the playback. During the session, only two of the 80 SHR tested emitted long 22-kHz USV.
Two days after conditioning, i.e. one day after the playback experiment, during the test session, all rats from 1-, 6-, and 10-Trial groups showed increased freezing (Fig. S1EG). SHR freezing levels correlated with the number of shocks received (note p = 0.0000 group effects, Kruskal-Wallis, Fig. S1F). SHR freezing levels to context, but not to context + cue were higher than in Wistar rats (Fig. S1DEH).
As a result of fear conditioning, Wistar control rats and some of the SHR emitted long 22-kHz USV (Table S1). Only 9 out of 49 SHR emitted long 22-kHz USV during the fear-conditioning session. The percentage of SHR emitting 22-kHz USV increased with the number of shocks delivered; it was 0% for 1-Trial rats, 18% for 6-Trial, and 40% for 10-Trial, but it was overall significantly lower than Wistar rats (44 of 57 in sum; 77%) in all three groups (p = 0.0001; p < 0.0001; and p = 0.0337 respectively, Fisher's exact test). Moreover, of the fear-conditioned rats, only one SHR emitted long 22-kHz USV during the testin comparison with 38 Wistar rats (p < 0.0001; Fisher's exact test for all fearconditioned rats). The long 22-kHz USV emitted by SHR during fear conditioning sessions were shorter (p = 0.0012 for 6-and 10-Trial rats combined) and had a longer latency (p < 0.0001 for 6-and 10-Trial rats combined; both Mann-Whitney).

Except for the periods of ultrasonic playback, SHR's behavior remained relatively constant
SHR's locomotor activity, measured as distance traveled, remained at a constant level during the 10-min-silence period at an average speed of 2.09 cm/s (Fig. S2AEIM). The activity slowed during the playback session to an average speed of 1.39 cm/s (p < 0.0001, Fig. 2). In particular, it was lower during control-time intervals, i.e. 1.53 cm/s (p < 0.0001) from − 120 to − 100 s and 1.52 cm/s (p < 0.0001) from − 30 to − 10 s (all rats analyzed, all Wilcoxon; compare our previous results: Olszyński et al., 2020;Olszyński et al., 2021). However, the distance traveled remained relatively constant within the control periods (Figs. 2-4, Table S3).
SHR did not have a strong preference for either side of the cage during the initial 10 min (Fig. S2BFJN); or during the playback session, e.g. during both control time-intervals before playback in all groups analyzed (Table S6; please note p < 0.05 only in two cases out of 60, all Wilcoxon). SHR had no cage-side preference before playback presentation, as also noted by values near 50% (dotted line) before each ultrasonic playback (Figs. 2-4), as well as the relative no change in preference within the control time-intervals (Table S7; please note p < 0.05 only in three of 60 cases, all Friedman). Similar results were recorded in Wistar rats ( Fig. S2BFJN; Figs. 1 and 2 in Olszyński et al., 2021).

SHR moved faster during 50-kHz ultrasonic presentations and slowed down after 22-kHz ultrasonic presentations
Rats from all SHR groups traveled significantly longer distances during the presentation of 50-kHz signals (Figs. 2 and 3; note also the p values for − 10-10 s time-intervals in Table S3), i.e. at 0 s time-interval vs. neighboring − 10 and 10 s time-intervals, which was usually significant for both USV-and tone-playbacks, and when all SHR were pooled together (p values of 0.0000-0.0498; and p < 0.0001 for all SHR pooled together in all three cases, Table S1, Wilcoxon). In the case of 22-kHz playback (Figs. 2 and 4; also note the p values for 0-30 s timeintervals in Table S3), a reduction in distance traveled appeared immediately after signal presentation, i.e. at 10 vs. 0 s time-interval, and was significantly reduced in all SHR groups, i.e. for 22-kHz USVplayback (all 4 cases, 0.0005-0.0302 p levels) and 22-kHz toneplayback (all 4 cases, 0.0001-0.0079 p levels, all Wilcoxon) as well as after both USV-and tone-playbacks in all SHR analyzed together (p < 0.0001 for both conditions and sound-playback).
As a result, when presented with 50-kHz playback, SHR traveled more than animals exposed to 22-kHz sounds ( Fig. 2 ACEG, Table S4). For USV-playback, the difference lasted throughout the 0-170 s timeinterval (seventeen points out of eighteen were significantly different, for all SHR analyzed together, with <0.0001-0.0198 p levels), and for tone-playback, the difference was shorter and lasted throughout 0-60 s time-interval (six points out of seven significantly different, for all SHR analyzed together, with <0.0001-0.0212 p levels). Finally, when results from USV and tone-playbacks were averaged for analysis, the difference lasted throughout 0 to 100 s time-interval (with <0.0001-0.0100 p levels, all Wilcoxon). An increase in locomotor activity during 50-kHz playback and prolonged decrease after 22-kHz playback were also observed in Wistar rats (Figs. 3 and 4;also Fig. 1ACEG and Table S1 in Olszyński et al., 2021).
However, unexpectedly during the 50-kHz sound-playback, Wistar rats traveled longer distances than SHR (Fig. 3 ACEG, Table S5). The effect was observed in all Wistar vs. all SHR rats pooled together for analysis (p = 0.0001; compare lower locomotor activity in Wistar rats at − 10 s time-interval, p = 0.0294) as well as in almost all groups (0.0190-0.0471 p levels); as this difference was not significant only in the 6-Trial group (p = 0.0951, Table S5, all Mann-Whitney). In summary, SHR initially demonstrated higher locomotor activity than Wistar rats at the start of the playback session but upon 50-kHz playback, Wistar rats, in turn, demonstrated higher locomotor activity.

Effects of conditioning on playback-evoked changes in locomotor activity recorded in Wistar rats not significant in SHR
Since conditioned Wistar rats showed higher locomotor activity during 50-kHz playback than 0-Trial Wistar controls (see both Fig. 2AC and Results 3.4 chapter in Olszyński et al., 2021), we examined whether a similar increase occurred in SHR. When distance traveled during 50-kHz USV and tone playbacks at 0 s time-interval, were averaged for analysis, the group comparisons did not reveal a difference between the 0-Trial rats (1.98 ± 0.18 cm/s) and all fear-conditioned rats combined (2.27 ± 0.16 cm/s, p = 0.2249). All conditioned groups moved faster than the 0-Trial, however the effects were not significant (i.e. 1-Trial, 2.41 ± 0.29 cm/s, p = 0.2445; 6-Trial, 2.28 ± 0.24 cm/s, p = 0.2813; 10-Trial, 2.11 ± 0.30 cm/s, p = 0.7851).

All SHR approached the speaker; it was more pronounced during and following 50-kHz playback
During and following playbacks, rats from all SHR groups approached the speaker for both 50 and 22-kHz playbacks (Figs. 2-4); note the p values for 10-30 and 10-60 s time-intervals compared with 50% chance levels (Table S6) as well as changes within 0-30 s timeinterval in all the groups (Table S7, 0.0000-0.0493 p values, Friedman), and especially the changes between before vs. after playback (Table S8, p < 0.0001 for both 50-and 22-kHz sound-playback in all SHR, Wilcoxon). However, SHR spent more time in the speaker's half of the cage when presented with 50-kHz playback than with 22-kHz sounds (Fig. 2 ACEG, Table S9), e.g. for the tone-playbackthe significant difference was observed during 0-50 s time-intervals (<0.0001-0.0353 Fig. 2. Analysis of changes in distance traveled, time spent in the speaker's half of the cage, heart rate (HR), and the number of emitted vocalizations during playback session in SHR exposed to both 50-kHz-vs. 22-kHz playback, with averaged effects of natural and artificial sounds. Gray sections correspond to the 10-s-long ultrasonic presentation. Graphs depict responses after previous exposure to: no shock (A, B), one shock (C, D), six shocks (E, F), and ten shocks (G, H). In the left column (A, C, E, G), distance traveled is presented as connected dots (cm, left Y axis), percentage of time spent in the speaker's half of the cageas unconnected dots (%, right Y axis). The dotted horizontal line marks a 50% chance value for time in a side of the cage. In the right column (B, D, F, H), HR is presented as connected dots (bpm; beats per minute, left Y axis); the number of USV is presented as unconnected dots (right Y axis). Each point is a mean for a 10-s-long time-interval with SEM. Playback of 50-kHz sounds results in a rise in locomotor activity, increased USV emissions and no clear change in HR. Playback of 22-kHz sounds is followed by a decrease in locomotor activity, a profound decrease in HR and no clear change in vocalization. Groups: 0-Trial, n = 31; 1-Trial, n = 17; 6-Trial, n = 17; 10-Trial, n = 15.  3. Effect of 50-kHz ultrasonic playback sounds on distance traveled, time spent in the speaker's half of the cage, heart rate (HR) and USV emission in SHR (black dots) compared to Wistar rats (white dots). Gray sections correspond to the 10-s-long ultrasonic presentation. Graphs depict responses after previous exposure to: no shock (A, B), one shock (C, D), six shocks (E, F), and ten shocks (G, H). In the left column (A, C, E, G), distance traveled is presented as connected dots (cm, left Y axis), percentage of time spent in the speaker's half of the cageas unconnected dots (%, right Y axis). The dotted horizontal line marks a 50% chance value for time in a side of the cage. In the right column (B, D, F, H), HR is presented as connected dots (bpm; beats per minute, left Y axis); the number of USV is presented as unconnected dots (right Y axis). Each point is a mean for a 10-s-long time-interval with SEM. Please note the shorter distances traveled by SHR vs. Wistar rats during the 50-kHz playback. In Wistar rats, we observed an HR increase after playback, which was absent in SHR. Wistar rats emitted more USV after 50-kHz sounds presentation than SHR. Detailed analysis of the Wistar group was previously published in Olszyński et al.,Brain Sci. 11(8), 2021. Groups (Wistar): 0-Trial, n = 37; 1-Trial, n = 16; 6-Trial, n = 20; 10-Trial, n = 19. Groups (SHR): 0-Trial, n = 31; 1-Trial, n = 17; 6-Trial, n = 17; 10-Trial, n = 15. Fig. 4. Effect of 22-kHz ultrasonic playback sounds on distance traveled, time spent in the speaker's half of the cage, heart rate (HR) and USV emission in SHR (black dots) compared to Wistar rats (white dots). Gray sections correspond to the 10-s-long ultrasonic presentation. Graphs depict responses after previous exposure to: no shock (A, B), one shock (C, D), six shocks (E, F), and ten shocks (G, H). In the left column (A, C, E, G), distance traveled is presented as connected dots (cm, left Y axis), percentage of time spent in the speaker's cage halfas unconnected dots (%, right Y axis). The dotted horizontal line marks a 50% chance value for time in a side of the cage. In the right column (B, D, F, H), HR is presented as connected dots (bpm; beats per minute, left Y axis); the number of USV is presented as unconnected dots (right Y axis). Each point is a mean for a 10-s-long time-interval with SEM. Please note, the prevailing HR difference between SHR and Wistar rats, which was diminished by the playback (except for 1-Trial groups, where HR in Wistar rats was exceptionally low) through higher HR decrease observed in all SHR groups. Note also, Wistar rats emit higher USV emissions to playback presentations (Table S18). Detailed analysis of the Wistar group was previously published in Olszyński et al.,Brain Sci. 11 (8) p levels, all Wilcoxon). Similar observations were made in Wistar rats ( Fig. 1 in Olszyński et al., 2021). There were no reliable differences between SHR and Wistar rats in the time spent in the speaker's half of the cage (please note "all rats" columns in Table S10).

HR was higher in SHR than in Wistar rats
When average HR levels from the first 5 min of the 10-min-silence period were compared with those from the last 5 min of the playback session, there was a significant decline in SHR (− 96.0 ± 3.4 bpm for all SHR analyzed together, i.e. from 496.7 ± 2.9 to 400.7 ± 2.7, p < 0.0001, Wilcoxon). This is in line with previously reported changes in direction and scope of this transition for Wistar rats, i.e. − 97.6 ± 3.8 bpm; from 475.7 ± 3.6 to 378.1 ± 3.0 (p = 0.6776 for Wistar rats vs. SHR comparison, Mann-Whitney; see also Olszyński et al., 2021).
The dynamic of HR changes in SHR during the 10-min-silence period was distinct from Wistar rats (Figs. S2CGKO and S3). In the latter, HR was initially stable; the decline began at 180-200 s and continued throughout the 10 min. In contrast, SHR's HR gradually increased until 180-200 s, then stabilized from 280 to 300 s and gradually reduced for the remainder of the 10 min. HR reduction was also observed during the playback sessions (Figs. 2-4) but not during control intervals (Table  S11). SHR HR was predominantly higher than Wistar rats during the first half of the 10-min-silence period. The difference was significant starting from 190 s (0-Trial rats; 0.0000-0.0349 p levels during the period), 90 s (1-Trial; 0.0000-0.0362), 140 s (6-Trial; 0.0000-0.0475), and 200 s time-intervals (10-Trial; 0.0000-0.0281, all Mann-Whitney) until the end of the 10-min-silence period (Fig. S3). This difference in HR between SHR and Wistar rats was observed until the end of the whole experiment, i.e. throughout the playback sessions (Figs. 3 and 4). Notably, in the 1-Trial group, Wistar rat HR was significantly lower than SHR HR (Figs. 3D, 4D, see also p values in Table S14). For other groups, this strong difference was not observed following the playback (Table S14).
We observed no clear effect of the amount (number of shocks) of conditioning, nor the conditioning itself on HR levels in SHR (Figs. 2-4 Table 1). Unexpectedly, in SHR, 50-kHz sound-playback did not significantly increase HR and even caused HR decrease (especially in 0-, 6-, 10-Trial SHR, 0.0302-0.0353 p levels; please note HR "before" vs. average HR from 10 to 60 s timeintervals, Table S12), which resulted in an overall HR decrease in all SHR (p = 0.0002, Table S12, all Wilcoxon) of − 6.5 ± 1.7 bpm in comparison to an HR increase of 7.7 ± 1.8 bpm recorded in Wistar rats (Table 1).
In contrast, both SHR and Wistar rats showed a decrease in HR following 22-kHz sound-playback (comp. Fig. 1BDFH in Olszyński et al., 2021). However, the HR decrease between − 10 and 10 s time-intervals (p ≤ 0.0010 in all five cases, Table S12, Wilcoxon) in SHR was about twice as large (− 33.9 ± 2.4 bpm in all SHR) in comparison to Wistar rats (− 18.1 ± 2.2 bpm, p < 0.0001, Mann-Whitney, Table 1; Fig. 4 BDFH). As a result, HR was consistently higher following 50-kHz vs. 22-kHz playbacks, in all SHR groups (Figs. 2-4, Table S13) and HR values differed significantly throughout 0-110 s time-intervals for all SHR Table 1 Evaluation of changes in heart rate (HR) around the time of ultrasonic playback, i.e. between before (i.e. averaged data from − 30 to − 10 s time-intervals, upper panel) or immediately before (at − 10 s time-interval, lower panel) vs. after (i.e. average time spent from 10 to 60 s time-intervals) or immediately after (at 10 s time-interval) between Wistar rats and SHR; significant values (p < 0.05) marked in bold, all Mann-Whitney.
A. Change in heart rate (HR) between 10 s time-interval and average of − 30 to − 10 s time-intervals Values are means with SEM. Table presents  combined (Table S13, <0.0001-0.350 p levels, Wilcoxon).

SHR vocalized often during and following 50-kHz playback
We have previously demonstrated that the 50-kHz playback resulted in a substantial increase in the number of USV emitted by Wistar rats, while this increase was modest after 22-kHz sounds presentation ( Fig. 1 in Olszyński et al., 2021, also Figs. 3 and 4). This increase was also observed in SHR during and following 50-kHz playbacks but not during and after 22-kHz sounds presentation (Figs. 2-4, Tables S15 and S16). As a result, when the values of USV emissions following 50-vs. 22-kHz playback were compared in SHR (Fig. 2 BDFH), there was a clear and prolonged difference across all analyzed groups, i.e. more USV following 50-kHz sound-playback throughout 0-180 s time-intervals (Table S17, p   Table 2 Comparison of the number and selected characteristics of 50-kHz USV emitted during the whole experiment, first 10 min of silence, and during playback sessions, i.e. during the 10-s-long playback and 110 s afterwards in response to 50-and 22-kHz USV, tone, and sounds (averaged results) in Wistar vs. SHR. MPFmean peak frequency; 50-kHz USV is defined as a USV with MPF >32 kHz. * p < 0.05, ** p < 0.01, *** p < 0. 001; Mann-Whitney. A single USV was eliminated as a long outlier (with >6 x SD). Groups ( < 0.0001 for each time-interval of 0-110 s period). Moreover, Wistar rats tended to emit USV after 50-kHz soundplayback more copiously than SHR (Fig. 3 BDFH), which was significant in 0-and 10-Trial groups (Tables 2 and S18), in all conditioned rats combined (Table 2) and in all rats analyzed together (Table S18; p < 0.0001 for 0-20 s time-intervals, Mann-Whitney). Both strains responded less to 22-kHz USV playback, and were nearly absent in SHR (Fig. 4  BDFH). As a result, there were some significant differences between SHR and Wistar rats in the number of USV emitted in response to 22-kHz playback observed in 0-, 6-, 10-Trial rats, all conditioned rats (Table 2 and S18), and especially in all rats analyzed together (Table S18; p < 0.01 for 0-50 s time intervals, Mann-Whitney).
When overall responses to 50-kHz vs. 22-kHz playback were analyzed, SHR produced fewer USV compared to Wistar rats, specifically fewer 50-kHz USV in response to both 22-kHz USV and tones than to 50-kHz USV or tones respectively. Also, the SHR emitted 50-kHz USV following 22-kHz playbacks tended to be shorter (Table S19). A similar trend in Wistar rat USV responses was also observed. Moreover, analysis of the number and parameters of USV produced by rats of both strains (Tables 2 and 3), show that in most cases, SHR emitted 50-kHz USV with lower MPF than Wistar rats (Table S14 in Olszyński et al., 2021). Every SHR group also emitted more short 22-kHz USV during the first 10-minsilence period than their corresponding Wistar group (Tables 3 and S20). In contrast, short 22-kHz USV parameters were relatively comparable between the strains (Table 3).

Natural and artificial ultrasounds produced similar results, with some remaining differences
In some cases, natural playbacks evoked stronger responses than artificial tones. The 6-Trial SHR group traveled more during natural 50-kHz playback than artificial 50-kHz tone-playback (p = 0.0232 in 0 s time-interval; Wilcoxon). Also, the 10-Trial SHR group traveled less during natural 22-kHz playback than artificial 22-kHz tone-playback (p = 0.0215 in 0 s time-interval; Wilcoxon). Moreover, the HR decrease after 22-kHz USV-playback in 1-Trial SHR was larger than after 22-kHz tone-playback (the difference lasted over the 10-100 s time-intervals with seven of the ten time-intervals significantly different, 0.0007-0.0361 p levels; Wilcoxon). These differences were subtle and diminished when several SHR groups were pooled together. There was no difference in the time spent in the speaker's half and in the number of USV emitted.

Previously-shocked SHR did not vocalize more
In SHR, there were some significant differences in number, duration and frequency of emitted USV between the no-shocked and 6-Trial groups. The SHR 6-Trial group emitted more 50-kHz USV in response to 50-kHz USV-playback (31.0 ± 6.0 vs. 22.1 ± 4.7; p = 0.0456) and the USV produced were longer than the 0-Trial animals in all experiments (29.6 ± 2.4 vs. 23.6 ± 1.2 ms; p = 0.0341), particularly after 50-kHz USV playback (34.9 ± 2.4 vs. 27.8 ± 2.0 ms; p = 0.0286). Moreover, during the first 10-min period, the 50-kHz USV emitted by 6-Trial SHR was lower in frequency than the 0-Trial SHR group (47.3 ± 1.6 vs. 51.6 ± 1.3 kHz; p = 0.0491; all Mann-Whitney). These three variables (number, duration and frequency) differed significantly within Wistar groups ( Table 1 in Olszyński et al., 2021), such that all conditioned rat groups vocalized more, their 50-kHz USV were longer and of higher frequency than those of the 0-Trial group.
Only a single SHR emitted long 22-kHz USV in response to 22-kHz USV playback (Table S19). The majority of SHR and Wistar rats emitted short 22-kHz USV in response to 22-kHz USB playback (79 out of 80 SHR animals (98.75%) and 79 out of 92 (85.87%) Wistar rats).

Table 3
Comparison of the number and selected characteristics of short 22-kHz USV emitted by Wistar and SHR during the whole experiment, first 10 min of silence and during playback sessions, i.e. during the 10-s-long playback and 110 s afterwards in response to 50-vs. 22-kHz USV, tones and sounds (averaged results) in control and fear-conditioned rats.

Complex response to ultrasonic playback is a general phenomenon; more features of SHR's behavior confirm their usefulness as models of ADHD and schizophrenia
Using our behavioral model (based on rats' exposure to pre-recorded playbacks in home-cage-like conditions, Olszyński et al., 2020;Olszyński et al., 2021) with the additional impact of prior fear conditioning, we investigated SHR's cardiovascular, locomotor and vocal responses. The results were obtained from a series of experiments in which control Wistar rats and SHR were analyzed alongside. Results of experiments conducted on Wistar rats alone were previously published (Olszyński et al., 2021).
SHR's USV have not yet been investigated much (see Zhang-James et al., 2014 study of female vocalization). In our hands, these rats generally behaved similarly to Wistar control rats, which is in line with previous publications (Olszyński et al., 2020;Olszyński et al., 2021; the behaviors are listed in Table S21A), e.g. SHR vocalized all commonly recognized types of USV, i.e. 50-kHz being the most abundant, short and long 22-kHz USV (Fig. 1). These behaviors, observed in both SHR and Wistar rats, point to a general phenomenon of differential locomotor, vocal and cardiac responses to 50-vs. 22-kHz playbacks.
However, some SHR's behaviors differed from those of Wistar rats (listed in Table S21B). In most cases, the 50-kHz USV from SHR had lower MPF than Wistar rats. Every SHR group also emitted more short 22-kHz USV during the first 10-min-silence period compared to their corresponding Wistar groups, however, the short 22-kHz USV parameters did not differ between the strains (Table 3). Moreover, fewer fear-conditioned SHR emitted long 22-kHz USV. Finally, in line with our main hypothesis, SHR emitted less USV than Wistar ratsin response to 50-kHz playback. These differences, deficits and characteristic features of SHR's USV could be used to model both mechanisms and treatments of ADHD and schizophrenia. Altogether these features of SHR support their use as an animal model for both diseases.
We also hypothesized that SHR's response to aversive 22-kHz sounds will be stronger in comparison with Wistar rats, and even more pronounced in SHR pre-exposed to foot-shock. 22-kHz playback indeed resulted in profound HR decrease in SHR. However, this decrease was not larger in previously-shocked SHR.

General problem of an appropriate control rat strain for SHR
There is a general problem regarding the choice of an appropriate control rat strain for SHR. As it is not possible to use siblings, rats from other strains are utilized. Although Wistar-Kyoto rats (WKY) are a frequently used normotensive strain-matched control for SHR, their use is in question as they show remarkable behavioral differences to other normotensive strains (Clements and Wainwright, 2006;Diana, 2002;Drolet et al., 2002;Ferguson et al., 2007;Grundt et al., 2009;Hard et al., 1985;Pare, 1994). Increased locomotor activity in SHR was found to be a consequence of the well-demonstrated hypoactivity of WKY (Alsop, 2007;Hard et al., 1985;McCarty and Kirby, 1982;Rosecrans and Adams, 1976;Tilson et al., 1977). In addition, WKY rats have been indicated to display high levels of anxiety and increased depressive behaviors in several behavioral models of depression (Braw et al., 2008;Dugovic et al., 2000;Pare, 1994;Redei et al., 2001). Moreover, SHR and WKY, despite their common origin, are genetically disparate; such that SHR have the same genetic profiles irrespective of source, while some WKY colonies have high genetic diversity (H'Doubler Jr. et al., 1991).
As a consequence, discrepancies emerged regarding SHR's behavioral and physiological results depending on the control strain used. For example, while some publications present SHR's memory deficits in some tasks (i.e. water maze, radial maze, open field, passive avoidance) using Wistar-Kyoto as controls (Hernandez et al., 2003;Meneses et al., 2011a), and in fear conditioning using Wistar rats, others present SHR without deficits in discriminative avoidance task (Calzavara et al., 2004) or locomotor habituation in open-field (Calzavara et al., 2011a) when compared to Wistar rats.
This issue of a suitable control strain for SHR studies could explain the disparity in our fear conditioning results (Fig. S1) compared to that in the existing literature. This discrepancy could be viewed as a limitation of the study but alsoas a strong signal that the consensus on SHR's fear conditioning results remains to be determined. Therefore, we proceed below with an analysis of the differences we observed between Wistar and SHR.

SHR demonstrated higher freezing than Wistar rats in fear conditioning
Differences in freezing between SHR and Wistar rats (as described in chapter 3.2) were already observed at the baseline period before conditioning and also during the context test (Fig. S1H). The higher freezing in fear conditioning in SHR and their susceptibility to stress could be an effect of higher sympathetic nerve activity (Judy et al., 1976), increased vascular sensitivity to noradrenaline (Smith et al., 1984), and greater sympathetic adrenal-medullary reactivity to a variety of stressors   (McCarty, 1983). However, others have reported contrary observations, i.e. SHR froze less than normotensive controls like Wistar (Calzavara et al., 2009;Calzavara et al., 2011b;Peres et al., 2018a;Peres et al., 2018b), Wistar-Kyoto and Sprague-Dawley rats in fear conditioning (Ledoux et al., 1983). The lower freezing in SHR reported were compared to Wistar rats from one particular colony (Calzavara et al., 2009;Calzavara et al., 2011b;Peres et al., 2018a;Peres et al., 2018b). Finally, these discrepancies may arise from the problem of an ideal control-strain stated above.

SHR had generally elevated HR
There are contradictory reports on the basal-resting and stimulationaltered HR of SHR and normotensive control rats. The resting HR of SHR was shown to be higher than Wistar-Kyoto rats (Friques et al., 2015), lower than Wistar "normotensive control rats" (Hallback and Folkow, 1974), and similar to Wistar-Kyoto rats (McDougall et al., 2000). Results from studies using a stressor to measure HR changes are more consistent.
To an air-puff startle stimulus, SHR exhibited greater tachycardia than La Jolla Wistar-Kyoto, Sprague-Dawley and Brown-Norway rats (Jaworski et al., 2002). Loud noises and vibrations evoked a greater HR increase in SHR than in Wistar "normotensive control rats" (Hallback and Folkow, 1974). Restraint-stress induced a significantly greater HR increase in SHR than in Wistar rats (Bott-Flugel et al., 2011). However, in another experiment, restraint-stress induced a similar HR increase in SHR and Wistar-Kyoto, but this change was longer-lasting in SHR (McDougall et al., 2000). In our hands, SHR HR was similar or even lower than Wistar rats (Fig. S3), and subsequently increased and remained elevated throughout the experiment (Fig. 3). This elevated level may be the cause for the lack of 50-kHz-playback-induced HR increase and profound 22-kHz-playback-induced HR decrease in SHR (Figs. 3-4).

Weaker behavioral and cardiovascular response of SHR to appetitive 50-kHz playback may be due to impaired social interactions
SHR response to appetitive 50-kHz playback was weaker, such that they emitted less 50-kHz USV than normotensive rats (Fig. 3 BDFH, Table S18). SHR were previously reported to vocalize as copiously as two WKY strains but less than Sprague-Dawley normotensive control rats (Zhang-James et al., 2014), similar to what was observed here to Wistar rats. This again highlights the issue of a proper control rat strain for SHR. Also, the appetitive playback-induced increase in locomotor activity was less pronounced in SHR and they traveled significantly less distance even though their basal locomotor activity was higher than Wistar rats (Fig. 3 ACEG, Results chapter 3.5). Finally, the appetitive playback-induced increase in HR response in Wistar rats was absent in SHR (Fig. 3 BDFH) and their HR even decreased in some measures (Table 1).
SHR are known to present impaired social interactions which mimic negative symptoms of ADHD (Gauthier et al., 2015) and schizophrenia (O'Tuathaigh et al., 2010;Sams-Dodd, 1998). Notably, the amount of time SHR spent engaging in social interaction was shorter or equal to the amount exhibited by Brown Norway, Wistar Furth and Lewis rats (Ramos et al., 1997). SHR also spend less time engaging in both passive (animals lying next to each other) and active (sniffing, following) social interactions (Calzavara et al., 2011a). As 50-kHz USV are thought to communicate positive emotional states and induce appetitive arousal in recipient rats (Burgdorf et al., 2008), it is therefore possible that our SHR displayed a reduced social response to the appetitive social signaling conveyed by the 50-kHz playback.
Similarly, the lack of HR increase in SHR to appetitive 50-kHz playback could be attributed to the antisocial features observed in this strain. Hallback (1975) reported that SHR reared in groups, and thus exposed to socioemotional stimuli, developed higher blood pressure than SHR reared in isolation. The isolated SHR, but not isolated normotensive control rats, had significantly lower blood pressure than their group-housed controls. It is therefore possible that our SHR had generally increased HR compared to Wistar rats, as a result of being reared with conspecifics, i.e. two rats per cage. From this, we speculate that a 50-kHz-playback may increase the HR of singly-reared SHR. We have shown before the effects of social context in our model, such that singly-reared Wistar rats have a stronger response to playback than pairreared ones. However, in contrast to SHR (Hallback, 1975), singlyreared Wistar rats displayed higher HR throughout the experiment (Olszyński et al., 2020).
Therefore, we propose that the impairment in social functioning explains the overall lower response in distance traveled, USV emission and HR increase to appetitive 50-kHz playback in SHR compared to Wistar rats. 4.6. SHR showed profound HR decrease when exposed to aversive stimulus , while behavioral responses were weaker After a 22-kHz playback, the HR decrease in SHR was about twice that of Wistar rats (Fig. 4 BDFH, Table 1), while the changes in distance traveled (Fig. 4ACEG, Results chapter 3.6) and USV production ( Fig. 4 BDFH, Table S18) were much smaller. LeDoux et al. (1982LeDoux et al. ( , 1983 similarly observed an increase in cardiovascular correlates of fear (blood pressure) in fear conditioning, while the behavioral manifestations of fear (drink suppression test) were reduced in SHR. The authors concluded that these findings oppose the view that emotional arousal is directly related to the peripheral autonomic output. Li et al. (1997) observed bradycardia in SHR during the presentation of a conditioned stimulus associated with pain, which was similar to the − 50 bpm observed in our experiment. Bradycardia was observed despite a large, concomitant increase in sympathetic nerve activity, while a corresponding HR decrease was not observed in Sprague-Dawley or Wistar-Kyoto rats.
Some studies have demonstrated that SHR show decreased fear/ anxiety-related behavioral responses compared to normotensive strains in the open-field, elevated plus-maze, social interaction and conditioned fear tasks (Gentsch et al., 1987;Goto et al., 1993;Ramos et al., 1997;Ramos et al., 2002). The measures recorded were locomotor frequency, immobility duration, time spent in the open arms, frequency of entries into the open area of the open field, transitions between black and white box, and defecation scores, all of which are external measures.

No clear dependence of SHR's responses to playback on the number of shocks received during previous conditioning
The effect of prior conditioning on increasing the distance traveled during the ultrasonic playback in Wistar rats was negligible in SHR (Figs. 2ACEG and 3ACEG). Similarly, while all conditioned Wistar groups vocalized more and their 50-kHz USV were longer and more frequent than the unconditioned 0-Trial group ( Table 1 in (Olszyński et al., 2021), this effect was only observed in the SHR 6-Trial group (Table  S19). We suspect the reasons for these effects could be due to the issue of a suitable control rat strain for SHR as they also differ in fear/anxietyrelated perception and processing. Both issues are discussed above.

More short 22-kHz USV and fewer rats with long 22-kHz USV among SHR
SHR demonstrated higher emissions of short 22-kHz USV during the preconditioning period (Fig. S1BH) and 10-min period of silence before playback (Table 3). Moreover, 0-Trial SHR emitted more short 22-kHz USV than 0-Trial Wistar rats also at time intervals following playback (Table 3). It should be noted that the basic characteristics of these USV (duration and mean peak frequency) were similar in both strains (Table 3).
It has been suggested that both short and long 22-kHz USV are associated with aversion (Barker et al., 2010) and may be used to express different aversive states. Less is known about the short 22-kHz calls. These could signal internal distress (negative emotional state without an external danger or threat), whereas the long 22-kHz calls are responses evoked by external aversive stimuli (Brudzynski et al., 1993). Robakiewicz et al. found that 10-20 ms long (i.e. similar duration to SHR in our experiment, Table 3) short 22-kHz USV were emitted during a nosepoke task which has an element of novelty and exploration (Robakiewicz et al., 2019). This suggests that short 22-kHz calls are associated with investigative functions and novelty seeking. Increased locomotor activity is a characteristic response in rats to a novel environment (Blanchard et al., 2009;Dellu et al., 1996). Therefore, both the higher locomotor activity and higher number of short 22-kHz USV calls in SHR than Wistar rats during the first 10 min of silence (Fig. S2), could be interpreted as responses to novelty. However, preconditioned SHR had lower locomotor activity (higher freezing, Fig.  S1H) and emitted more short 22-kHz USV than Wistar rats, therefore, the putative investigative function of short 22-kHz USV calls cannot explain these observations. Rather, we propose that these short 22-kHz USV, emitted by SHR, signal a basal internal negative emotional state in this strain.
Please note that the short 22-kHz USV recorded in our experiment were 25.4 ± 0.9 ms (for all SHR) and 26.6 ± 0.9 ms (for all Wistar rats) in duration. The shortest USV recorded in our experiments was 1 ms in contrast to 20 ms (Brudzynski et al., 1993) or 30 ms (Kim et al., 2010), in other studies. Therefore, many short 22-kHz USV may be omitted in previous research.
In contrast, long 22-kHz USV were less prevalent in SHR than in Wistar rats. It has been shown that SHR's cholinergic input of the central nervous system is decreased, i.e. nicotinic receptors are fewer in numbers and are hypo-responsive to chronic nicotine administration . Additionally, the cholinergic input into the mediobasal forebrain play a role in the initiation and/or emission of 22 kHz calls in rats (Brudzynski, 1994). Therefore, the disrupted cholinergic transmission in SHR may alter the distribution of short vs. long 22-kHz USV to favor shorter calls, though the neurobiology of short 22-kHz USV is still unclear.

50-kHz USV from SHR had lower MPF than Wistar rats
The analysis of the number and parameters of USV produced by rats of both strains (Tables 2 and 3), shows that in most cases, SHR emitted less 50-kHz USV overall and those emitted were of lower MPF, in comparison with Wistar rats. It has been demonstrated that frequency is a crucial property for how rats interpret USV (Saito et al., 2019). However, the meaning and cause of such change in SHRwhether it results from different emotional processing or physiological differences between strains is unknown.
Hypertension has been shown to impair axon growth, affecting mainly the small myelinated fibers, probably by the alterations of the endoneural blood vessels. This alteration is common in the peripheral fibers in SHR and is also present in the recurrent laryngeal nerve (RLN; da Silva et al., 2016), which is crucial for USV formation in rodents (Asai et al., 2020). Notably, for adult male house mice (Mus musculus) and deer mice (Peromyscus maniculatus bairdi), unilateral transections of RLN significantly reduces the number of males emitting USV and the number of USV emitted (Nunez et al., 1985). Additionally, efferent axons in RLN arise from the cells of the dorsal formation of the nucleus ambiguus (Wetzel et al., 1980). Significant motoneuron loss in the nucleus ambiguus of aged rats is suggested to be a contributing factor in decreasing complexity and quality of USV, i.e. reduced bandwidth, intensity, and peak frequency (Basken et al., 2012). Therefore, neural pathways between the nucleus ambiguus and laryngeal muscle affected by hypertension may be, at least in part, responsible for lower numbers of USV and lower peak frequency of 50-kHz USV in SHR (Table 2).

Conclusion
Overall, we have shown that SHR respond to appetitive and aversive playback like other rats. However, unique to SHR were changes in HR. These changes, most likely determined by overall elevated HR levels, were the lack of 50-kHz playback-induced HR increase and the presence of profound 22-kHz playback-induced HR decrease. SHR show deficits in emotional processing. Some of the phenomena observed in SHR could result from deficits in emotional perception and processing which may manifest as (a) more short 22-kHz USV, (b) less 50-kHz USV, (c) weaker behavioral and cardiovascular responses to appetitive 50-kHz playback, and (d) no clear effect of the shock intensity in fear conditioning on their subsequent responses to playback. The described features of SHR's vocalization further support the use of these animals as a model of ADHD and schizophrenia.

Data sharing policy
The data that support the findings of this study are openly available in Mendeley Data at https://data.mendeley.com/datasets/zyyfbffcc6.
The authors declare no conflict of interest.

Ethical statement
Procedures involving rats were conducted in accordance with the Declaration of Helsinki, and approved by the Second Warsaw Local Ethics Committee for Animal Experimentation (WAW2/093/2019). Efforts were made to minimize animal suffering and to reduce the number of animals used. All the authors declare no conflicts of interest.