key: cord-0684905-6lvhru4q
authors: Bent, Adam M.; Hedwig, Berthold
title: Tolerant pattern recognition: evidence from phonotactic responses in the cricket Gryllus bimaculatus (de Geer)
date: 2021-12-22
journal: Proceedings. Biological sciences
DOI: 10.1098/rspb.2021.1889
sha: 0fdec131c5a0b08fc5362cb5be4e9a0b58d7b2f5
doc_id: 684905
cord_uid: 6lvhru4q

When the amplitude modulation of species-specific acoustic signals is distorted in the transmission channel, signals become difficult to recognize by the receiver. Tolerant auditory pattern recognition systems, which after having perceived the correct species-specific signal transiently broaden their acceptance of signals, would be advantageous for animals as an adaptation to the constraints of the environment. Using a well-studied cricket species, Gryllus bimaculatus, we analysed tolerance in auditory steering responses to ‘Odd’ chirps, mimicking a signal distorted by the transmission channel, and control ‘Silent’ chirps by employing a fine-scale open-loop trackball system. Odd chirps on their own did not elicit a phonotactic response. However, when inserted into a calling song pattern with attractive Normal chirps, the females' phonotactic response toward these patterns was significantly larger than to patterns with Silent chirps. Moreover, females actively steered toward Odd chirps when these were presented within a sequence of attractive chirps. Our results suggest that crickets employ a tolerant pattern recognition system that, once activated, transiently allows responses to distorted sound patterns, as long as sufficient natural chirps are present. As pattern recognition modulates how crickets process non-attractive acoustic signals, the finding is also relevant for the interpretation of two-choice behavioural experiments.

Orthopterans are well-studied insects due to their use of conspicuous acoustic communication [1] [2] [3] [4] . Through the stridulation of their wings, male field crickets produce a species-specific calling song, which females detect and if they are ready to mate, move toward males in a behaviour known as phonotaxis. In the bi-spotted field cricket, Gryllus bimaculatus, female phonotactic behaviour is sharply tuned to the calling song chirp and pulse pattern of conspecific males [5] [6] [7] [8] [9] . However, the calling song of male crickets is a long-distance signal which may become distorted in the transmission channel before being received [10] [11] [12] . Due to reverberations, the calling song's pulse pattern may become degraded within a short distance from the signaller, and conspecifics no longer respond to such distorted signals [13] , as they become difficult to process by a pattern recognition system. A tolerant sensory system, which is adapted to the impact of the transmission channel and would accept distorted signals while processing the species-specific sound pattern, could be beneficial for females orienting in a complex environment. Here, we provide evidence for the presence of such a sensory system in female G. bimaculatus.

We investigated the propensity of crickets to respond to distorted nonattractive acoustic stimuli. To do so, experiments were designed analogous to 'oddball' paradigms used in psychological studies, where natural stimuli are presented with 'odd' portions inserted [14] . We exposed female crickets to different ratios of Normal and Odd chirps to infer the presence and timescale of tolerant sensory processes and examined the effect this may have on their phonotactic behaviour. We tethered crickets so that they walked on an air-suspended trackball while sound patterns were presented from a speaker at the left or right side [14] . The rotations of the trackball were measured and revealed the female phonotactic steering behaviour to sound sequences with attractive and distorted non-attractive chirps combined at different ratios.

Last instar female G. bimaculatus were isolated from a colony at the Department of Zoology, Cambridge. Eclosed females were individually housed in plastic containers, they had access to food and water ad libitum, and were kept at a temperature of 25-28°C. Only unmated individuals older than 7 days posteclosion were used as phonotactic responsiveness increases over the first week of adulthood [15] . One day prior to trials, the front wings of the test animals were cut, and an insect pin (approx. 32 mg) was waxed vertically onto the first abdominal tergite, close to the animal's centre of gravity. Females with the pin could move around freely and were returned to their containers prior to testing. Due to laboratory restrictions imposed by the COVID-19 pandemic, we used different cohorts of crickets over the course of the experiments.

Female phonotaxis to acoustic stimuli was measured through an open-loop trackball set-up. Females tethered to an insect pin were placed stationary on top of the lightweight air-suspended track ball, which they moved with their legs when walking. The forward-backward and left-right movements of the trackball were captured with an optoelectronic sensor and indicated the walking speed and direction of the specimen. The lateral steering component was analysed as a reliable indicator of phonotactic behaviour (see details in [14] ). Females that showed a steering response below 50 mm to a Normal chirp sequence over the course of a minute were not included in analyses.

Auditory stimuli were created with the software Cool Edit 2000 (Syntrillium, Phoenix, USA), with a sampling rate of 44.1 kHz, a 16-bit amplitude resolution and the sound intensity calibrated to 75 dB SPL. We created Normal calling song chirps similar to the natural song of G. bimaculatus [13] , with a carrier frequency of 4.8 kHz. Each chirp consisted of five pulses with 20 ms pulse duration, including a 2 ms rise and fall, and a 40 ms pulse period, followed by a 175 ms chirp interval, resulting in a 375 ms chirp period. All acoustic stimuli were delivered via two speakers (Neo 13S, Sinus Live, Conrad Electronics, Hirschau, Germany) positioned 57 cm in front of the cricket at the left and right side at 45°to its long axis.

(d) Experiment 1: identifying a distorted non-attractive Odd chirp

Crickets do not respond to conspecific calling songs with the pulse structure of chirps distorted by reverberations [13] . In order to create a distorted non-attractive chirp pattern, we designed artificial chirp stimuli with different amplitude envelopes that maintained the same chirp period, duration and frequency as the Normal chirps, but had no pulse structure. The envelope shapes covered one chirp-long sound and were 'Rectangle', 'Oval', 'Crescendo' and 'Decrescendo' (figure 1a), they did not contain the pulsed pattern of normal chirps. Chirps with these envelopes were presented sequentially, for 30 s from the left then 30 s from the right, with 15 s silence between each, followed by a Normal chirp song pattern (figure 1a). A total of seven females were tested to select the Odd chirp with the lowest phonotactic response. The oval envelope chirp was the least attractive ( figure 1b) tested a different cohort of crickets (N = 27) to analyse the propensity of crickets to respond to Odd chirps when Normal chirps are more frequent. The ratios presented here were 1 : 3 (one Normal chirp followed by three Odd chirps), 1 : 2, 1 : 1, 2 : 1 and 3 : 1. In both sets of experiments, an All Odd and All Normal chirp sequence were presented at the beginning or end of the tests. As a control, we created test patterns with the same temporal organization, but we replaced the Odd chirps with a period of silence for the same duration (figure 1c), hereafter referred to as Silent chirps. This resulted in a corresponding number of stimulus sequences with increasing numbers of Normal chirps, with the addition of an All Silent chirp and an All Normal chirp sequence. Sound patterns were always presented from least to most occurrences of Normal chirps. Females were tested once with the Odd chirp and Silent chirp paradigms, with at least 24 h between each test. The paradigm which was tested first was selected pseudo-randomly. In all experiments the overall lateral deviation of the females toward the different acoustic test patterns was analysed.

(f ) Experiment 3: phonotactic steering velocity in response to Normal chirps and Odd chirps

To understand the dynamics of the behaviour, we analysed the lateral steering velocity of the phonotactic responses. Triggered by the start of the Normal chirps, we averaged the lateral steering velocity in response to the Normal, Odd and Silent chirps in certain ratios. The steering velocity provides the dynamics of the female behaviour at high temporal resolution and allowed us to compare the amplitude of the steering responses. Subsequently, we tested a final group of females (N = 30) to analyse if processing of Normal chirps would affect the response to Odd chirps presented from the opposite side. We used a ratio of Normal to Odd chirps at 1 : 1. For comparison, we also tested females on a sequence with Normal chirps presented from alternating sides, and an All Odd sequence not presented from alternating sides.

Trackball movements were recorded using software designed in LabView v. 5.01 (National Instruments, London, UK). Off-line data analysis was done with Neurolab [16] . Each chirp type was presented for 30 s from the left and right. The steering velocity of the crickets was calculated based on the output of the optical sensor and the lateral deviation over each presentation was obtained by integrating the velocity data for the left and right sound presentation. These were pooled to obtain a measure for the phonotactic response over a 1 min window [17] . All statistical analyses were conducted using R Studio [18, 19] and the packages 'dunn.test' [20] and 'PMCMR' [21] . Where necessary, data were tested for normality using Shapiro-Wilk tests, and non-parametric tests were used where appropriate. Graphs detailing results of the experiment were either exported from Neurolab, or created in R using the package ggplot2 [22] .

The steering responses to chirps with different envelopes and Normal chirps (Experiment 1) were tested using a Friedman rank sum test paired with Conover post hoc tests with Bonferroni corrections. The same procedure was also used to test for differences in steering responses between chirp ratios in Experiment 2. Where appropriate, we used multiple Wilcoxon signed-rank tests and paired t-tests to test for within chirp ratio differences between Odd and Silent chirp conditions and between the same ratios presented in different paradigms. In Experiment 3, we used Kruskal-Wallis tests paired with a Dunn post hoc test with Bonferroni corrections, and Friedman rank sum tests paired with Conover post hoc tests with Bonferroni corrections, to test for differences in the average change in steering velocity.

(a) Experiment 1: identifying a distorted non-attractive Odd chirp

We tested the phonotactic response of females to different chirp envelopes (figure 1a, top) to select the least attractive stimulus. A typical recording reveals that this female did not show a pronounced steering response to any of the chirp envelopes; however, she responded strongly to the Normal chirp pattern, and walked to the left or right when the corresponding speaker was activated (figure 1a). The pooled data show that all chirp envelopes induced some minor steering, as values were not centred on zero, but responses were significantly different overall (Friedman rank sum: x 2 4 ¼ 23:12, N = 7, p = <0.001; figure 1b ). Each type of envelope chirp elicited a significantly lower average steering response when compared to the Normal chirp (Rectangle: z = 4.9, p = <0.001; Oval: z = 8.16, p = <0.001; Crescendo: z = 7.35, p = <0.001; Decrescendo: z = 6.8, p = <0.001). As the Oval envelope chirp had the lowest response ( x ¼ 12:45 mm), it was selected as the Odd chirp and used for designing test sequences with different ratios of Normal and Odd chirps (see Methods). For the Silent chirp conditions, the temporal organization of sequences was the same as for the Odd chirp condition (figure 1c).

We tested the responses of females to sequences with different ratios of Normal chirps combined with Odd or Silent chirps. The steering response of a female to a sequence of All Odd chirps (figure 2a, left) shows no obvious steering response, and changes in the animal's walking direction were not coupled to the active speaker. When combinations of Normal and Odd chirps were presented, the cricket showed no steering behaviour when the ratio between Normal and Odd chirps was 1 : 9, or lower. Steering toward the active speaker started at a ratio of 1 : 7, it increased with the fraction of Normal chirps increasing, and in this case for the 1 : 1 ratio the response was about as strong as the response to All Normal chirps (figure 2a, right). We found significant differences between steering responses to different ratios (Friedman rank sum:

.001), with increased directional steering responses when the relative number of Normal chirps increased (figure 2c, blue bars, table 1). Compared to the sequence with All Odd chirps, the sequences with ratios of 1 : 14 and all ratios from 1 : 7 to 1 : 1 elicited a significant stronger phonotactic response (figure 2c and table 1). This suggests that a steering response can be elicited when a Normal chirp is present in a sequence of Odd chirps once every 5.6 s (as in the 1 : 14 ratio) or less. Although individuals showed a very strong response to the 1 : 1 ratio, the mean response to the sequence with the 1 : 1 ratio was 59% of the mean response to an All Normal chirp sequence, and significantly lower.

The response of a female to combinations of Normal and Silent chirps (figure 2b) reveals that the insect did not show pronounced steering to any sequence when the ratio between Normal and Silent chirps was lower than 1 : 1. We found a difference in steering responses to different ratios (Friedman rank sum: and table 1 ). The mean response to the sequence with the 1 : 1 ratio was 28% of the mean response to an All Normal chirp sequence, and significantly lower.

In comparison, sequences containing Odd chirps elicited significantly stronger steering responses than corresponding Silent chirp sequences at the ratios 1 : 14, 1 : 7, 1 : 3 and 1 : 1 ( figure 2c and table 2 ). This demonstrates that the Odd chirps make a significant contribution to the females steering behaviour, and that the response to the Odd chirps was altered when these were embedded in a sequence of Normal chirps. Even sparsely presented Normal chirps like in the 1 : 7 or the 1 : 3 sequences, had a substantial impact on phonotactic responses to subsequent Odd chirps.

We analysed further ratios of Normal to Odd or Silent chirps to see where full phonotactic responsiveness would occur. Females in this paradigm were tested with ratios of 1 : 3, 1 : 2, 1 : 1, 2 : 1 and 3 : 1 (Normal to Odd or Silent chirps), along with an All Odd or All Silent chirp and All When comparing corresponding responses between the two cohorts (figure 2c), we found that lateral steering in this cohort was significantly higher at the ratio of 1 : 3, for both the Odd and Silent chirp sequences (table 3) , as in the previously cohort tested.

We found a significant difference in steering among all ratios in both the Odd (Friedman rank sum: The velocity data of the trackball system revealed the fast steering responses toward individual chirps. We averaged the steering velocity for all ratios that showed significantly higher steering than the All Odd or All Silent chirp sequences (these were 1 : 14, 1 : 7, 1 : 3, 1 : 1) within Experiment 2. Velocity responses to a given chirp were calculated as the difference between the velocity after the occurrence of the second pulse (or after 60 ms, for Odd and Silent chirps) and the peak velocity (figure 3a). In the 1 : 1 ratio, the steering response toward a Normal chirp followed by a Silent chirp reaches a maximum velocity at around 250 ms and then gradually declines over the next 500 ms to zero as there is no response to the Silent chirp (figure 3a, black trace). The steering response to the Normal chirp is still obvious for the 1 : 3 ratio, but it becomes smaller with increasing number of Silent chirps included in the paradigm. The averaged signals also become noisier, as fewer chirps can be evaluated. Averaging the response to Normal chirps followed by Odd chirps for the 1 : 1 ratio reveals again an increased steering velocity in response to the Normal chirp but also a clear response to the Odd chirp (figure 3a, blue trace). It reaches the same peak velocity as the response to the Normal chirp but comes with a faster decay. As females steered toward the Normal and the Odd chirps, the steering velocity signal does not decline to zero. The change in velocity in response to the Normal and Odd chirp was still obvious for the 1 : 3 condition and then became weaker with overall decreasing numbers of Normal chirps presented.

The change in steering velocity in response to Normal chirps in the Odd and Silent condition reached 4.4 cm s −1 and 4.3 cm s −1 , respectively. The response to the Odd chirps reached 3.6 cm s −1 , and the response to the Silent chirps Table 1 . Differences in lateral deviations between each ratio in the Odd and Silent acoustic paradigms. P-values from post hoc Conover tests for pairwise comparisons of directional steering responses between each ratio. Results for both the Odd chirp (top-right) and Silent chirp (bottom-left) conditions from the first cohort of crickets are shown. P-values were calculated with Bonferroni adjusted methods. Italic p-values indicate a significant result. figure 3b) . Changes in average steering velocity differed significantly between the chirp types measured (Kruskal-Wallis: x 2 4 ¼ 67:016, N = 54, p = >0.001). Females showed a significant change in steering velocity when presented with Normal chirps. They also showed a significant response to Odd chirps following a Normal chirp, but not to Silent chirps following Normal chirps (table 6). These data reveal that Odd chirps do not just change an overall phonotactic steering bias, when Odd chirps are preceded by a Normal chirp, they rather elicit a fast steering response similar to the Normal chirps.

The previous experiments did not reveal if processing of Normal chirps had a bilateral effect. We, therefore, tested females with a 1 : 1 Normal to Odd ratio where the Odd chirps were presented from the opposite side to the Normal chirps. Additionally, females were presented with an All Normal chirp sequence where each chirp was presented from alternating sides. Individuals showed their ability to rapidly adjust their velocity in response to Normal chirps presented from alternating sides (figure 4a, black trace) [17] . The same pattern of response was observed to Odd chirps as well (figure 4a, blue trace), albeit a slightly weaker response to both types of chirps.

We found a significant difference in the change to steering velocity when comparing responses to Normal and Odd chirps presented from alternating sides, and Odd chirps presented in the 30 s control (Friedman rank sum: X 2 2 ¼ 48:48, N = 30, p = <0.001; figure 4b ). There was no significant difference between the response to a Normal chirp and a following Odd chirp presented from the opposite speaker (z = 1.23, p = 0.327). Both responses were significantly higher than the change in steering in the All Odd control (Normal chirp z = 6.55, p = <0.001; Odd chirp z = −5.32, p = <0.001). These data reveal that the processing of Normal chirps presented from one side of the animal also has an impact on the response to Odd chirps presented from the opposite side of the animal.

The experiments presented here were based around an oddball paradigm used in psychological studies [14] , where humans are observed to see how easily they can detect anomalous stimuli. We used this paradigm to test whether female crickets would discriminate against Odd chirps or if they would show responses similar to Normal chirps. Our results revealed phonotactic steering to distorted non-attractive Odd chirps when these were presented in combination with attractive Normal chirps. This is evidence that G. bimaculatus did not discriminate against the Odd chirps but rather employs a pattern recognition system that transiently becomes tolerant to distorted stimuli [23] . Our data reveal that Odd chirps, which do not elicit a phonotactic response by their own, will contribute to the phonotactic behaviour when they are interspersed in a sequence of Normal chirps. This may allow females during a phonotactic approach in nature to use all chirps of a conspecific calling song for orientation even when the pulse pattern does not meet criteria for pattern recognition [8, 9] , although the chirp rhythm may still provide a species-specific clue for orientation [24] . Our data are in line with previous studies, demonstrating that non-attractive pulses would transiently elicit phonotactic responses when presented after an ongoing sequence of calling song [25] and that the recognition of the species-specific calling song, at least transiently, alters the female phonotactic responsiveness. Here we show that responses to distorted signals not only occur after a sequence of calling song, but even when the Odd chirps are interleaved with the Normal chirps, if a sufficient number of Normal chirps are still perceived. This is a situation females may likely encounter during phonotactic walking under natural conditions which imposes a more challenging situation on orientation than laboratory-based experiments [26] . It is not clear if the modulation happens at the level of pattern recognition processing or at the motor control of phonotactic steering. As all chirp envelopes tested elicited some weak steering responses, a low-level auditory-to-motor pathway may be present in the cricket nervous system that is upregulated by the pattern recognition process. Steering to Normal and Odd chirp patterns presented from alternating sides indicates that the processing of Normal chirps comes with a bilateral effect on the steering behaviour. This highlights that the response to Odd chirps does not represent a general bias of the females to orient toward sound, but that females actively steer toward Odd chirps occurring during a sequence of normal song. This had not been observed before [25, 27] and may require an additional feature to the well-described concepts of how pattern recognition and directional steering may interact [27] . Besides the implications for pattern recognition tolerant auditory processing has ecological implications. G. bimaculatus may have evolved a tolerant auditory system to overcome distortions of acoustic communication signals due to environmental factors. Acoustic signals are prone to degradation due to attenuation and scattering caused by the physical characteristics of the environment [28] [29] [30] . Male G. bimaculatus signal naturally from burrows [31] , and sounds emitted close to the ground are particularly susceptible to distortions [32] [33] [34] . The pattern of pulses may become highly distorted [13] and as a result, acoustic signals become degraded before reaching the receiver [10] . By using a tolerant auditory system that is activated by the normal species-specific signal, a receiver may be royalsocietypublishing.org/journal/rspb Proc. R. Soc. B 288: 20211889 able to orient more reliably to a conspecific signal even if it has been distorted by environmental interference. In our study, the presented Odd chirps may be considered a distorted and degraded signal, which maintains the same frequency and duration as a chirp but does not feature intra-chirp temporal pulse structures as observed by Simmons [13] for cricket songs measured at a distance to the sender. However, females still responded to these distorted signals as long as a sufficient number of natural chirps were present, highlighting the benefit of a tolerant sensory pathway. Receivers may also be detecting acoustic signals from other sources simultaneously, such as other signalling individuals [35] or anthropogenic noise [36] , which may influence their ability to approach a conspecific signals. In the acridid grasshopper Chorthippus biguttulus, broadband background noise presented with the conspecific signals reduces the receivers' ability to respond and orient toward the mate's acoustic signal [37] . In crickets and bushcrickets adaptions of receiver auditory systems prevent such interference by noisy background sound sources from disrupting the neural response to the dominant sender signal. For example, crickets show a gain control mechanism [38, 39] that cancels the response to low-level background signals [40] .

Our results also have implications for the use of choice experiments to study behavioural responses. Experimental design of mate choice experiments can influence the observed behavioural responses [41] . Two-way choice experiments present animals with signals from two sources and allow the experimental animal to choose between them and to indicate a preference. Such experimental designs can result in stronger mating preferences when compared to no-choice experimental designs [42] . Our data suggest that such a modulation of mating preference could be the result of a tolerant and adaptive sensory system, which might use the information from either stimuli to transiently adjust the attractiveness of the signals, resulting in a biased choice decision. Thus, potential underlying sensory biases need be considered when designing choice experiments [41] . Even no-choice experimental designs could lead to short-term sensory tolerances if stimuli are presented sequentially, without sufficient gaps to allow the modulatory effect of the natural signal to desist. In Teleogryllus oceanicus, females may even use remembered acoustic information to shape their phonotactic responses [43] .

Due to the COVID-19 pandemic, the crickets used in this study were from different generational cohorts. When we analysed steering responses between cohorts, the second cohort showed overall stronger auditory responses to the same acoustic stimuli. Nevertheless, the pattern of response remained the same between different cohorts, showing that Odd chirps presented after a Normal chirp contribute to positive phonotactic responses. This was also apparent in other cohorts used in preliminary tests (unpublished data). Overall, our data are evidence in support of an auditory system in G. bimaculatus that becomes tolerant to distorted signals when processing a conspecific calling song. This is a robust adaptation to the conditions of signal transmission in the field that results in the phonotactic steering to distorted stimuli as long as a Normal chirp is presented at least every 5.6 s. Our results have theoretical implications on the neural process of pattern recognition and the resulting control of phonotactic walking behaviour, as pattern recognition shows some form of short-term plasticity.

Ethics. As there are no legal requirements for studies involving orthopteran research subjects in the United Kingdom, no permits or licences were required to house, breed or experiment on G. bimaculatus.

Data accessibility. Analyses reported in this article can be reproduced using the data available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.vq83bk3th [44] . The data are provided in electronic supplementary material [45] .

Evolutionary change in cricket acoustical communication

Acoustic communication in insects and anurans

The neurobiology of an insect brain

Cricket behavior and neurobiology

Phonotactic behavior of crickets

Auditory interneurons in the prothoracic ganglion of the cricket, Gryllus bimaculatus deGeer

The large segmental auditory neuron (LSAN)

An auditory feature detection circuit for sound pattern recognition

Song pattern recognition in crickets based on a delayline and coincidence-detector mechanism

Critical song features for auditory pattern recognition in crickets

From sender to receiver: propagation and environmental effects on acoustic signals

Environmental motion delays the detection of movement-based signals

Vibratory noise in anthropogenic habitats and its effect on prey detection in a web-building spider

The calling song of the field cricket, Gryllus bimaculatus (de Geer): constraints on transmission and its role in intermale competition and female choice

Event-related potentials in an auditory semantic oddball task in humans

2021 Does the choosiness of female crickets change as they age?

NEUROLAB, a PCprogram for the processing of neurobiological data

Mechanisms underlying phonotactic steering in the cricket Gryllus bimaculatus revealed with a fast trackball system

RStudio: integrated development for R

2021 R: a language and environment for statistical computing

dunn.test: Dunn's Test of Multiple Comparisons Using Rank Sums

The Pairwise Multiple Comparison of Mean Ranks Package (PMCMR)

Ggplot2: elegant graphics for data analysis

Song recognition and localization in the phonotaxis behavior of the field cricket, Gryllus bimaculatus (Orthoptera: Gryllidae)

Auditory processing at two time scales by the cricket Gryllus bimaculatus

Auditory orientation in crickets: pattern recognition controls reactive steering

Selective phonotaxis of female crickets under natural outdoor conditions

Decision making and preferences for acoustic signals in choice situations by female crickets

Reverberations and amplitude fluctuations in the propagation of sound in a forest: implications for animal communication

Sound localisation in a habitat: an analytical approach to quantifying the degradation of directional cues

The sensory ecology of acoustic communication in insects

Intermale competition and mating success in the field cricket, Gryllus bimaculatus (de Geer)

Strategies for acoustic communication in complex environments. In royalsocietypublishing.org/journal/rspb

Insect hearing in the field

Sound transmission and directional hearing in field crickets: neurophysiological studies outdoors

The 'cocktail party problem': what is it? How can it be solved? And why should animal behaviorists study it?

Anthropogenic noise and conservation

Influence of amplitude modulated noise on the recognition of communication signals in the grasshopper Chorthippus biguttulus

Selective attention in an insect auditory neuron

A gain-control mechanism for processing of chorus sounds in the afferent auditory pathway of the bushcricket Tettigonia viridissima (Orthoptera; Tettigoniidae)

Contralateral inhibition as a sensory bias: the neural basis for a female preference in a synchronously calling bushcricket, Mecopoda elongata

Designing mate choice experiments

The effect of experimental design on the measurement of mate choice: a meta-analysis

Field crickets change mating preferences using remembered social information

2021 Data from: Tolerant pattern recognition: evidence from phonotactic responses in the cricket Gryllus bimaculatus (de Geer)

2021 Data from: Tolerant pattern recognition: evidence from phonotactic responses in the cricket Gryllus bimaculatus (de Geer). Figshare. royalsocietypublishing.org/journal/rspb

Acknowledgements. We thank two final year undergraduate project students, Oliver Bock-Brown and Jack Stockdale, who conducted preliminary experiments that formed the basis of this research. We are grateful to two anonymous reviewers and the manuscripts handling editors for their constructive feedback. Thank you also to members of the Neurobiology of Acoustic Communication Group, who assisted in aspects of experimental design, analysis and interpretation of the results.

Authors' contributions. A.B.: formal analysis, investigation, resources, visualization, writing-original draft, writing-review and editing; B.H.: conceptualization, formal analysis, investigation, methodology, supervision, writing-original draft, writing-review and editing. All authors gave final approval for publication and agreed to be held accountable for the work performed therein.Competing interests. We have no competing interests. Funding. Equipment used in this study was funded by the Royal Society and the BBSRC.