Judgments of American English male talkers who are perceived to sound gay or straight: Which personal attributes are associated with each group of talkers?

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice PLoS One . 2026 Apr 10;21(4):e0346897. doi: 10.1371/journal.pone.0346897 Search in PMC Search in PubMed View in NLM Catalog Add to search Judgments of American English male talkers who are perceived to sound gay or straight: Which personal attributes are associated with each group of talkers? Erik C Tracy Erik C Tracy 1 Department of Psychology, University of North Carolina at Pembroke, Pembroke, North Carolina, United States of America Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing Find articles by Erik C Tracy 1, * , Elizabeth D Young Elizabeth D Young 2 School of Behavioral and Brain Sciences, University of Texas at Dallas, Richardson, Texas, United States of America 3 Callier Center for Communication Disorders, Dallas, Texas, United States of America 4 Department of Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center, Dallas, Texas, United States of America Data curation, Formal analysis, Software, Validation, Visualization, Writing – review & editing Find articles by Elizabeth D Young 2, 3, 4 , Kelly A Charlton Kelly A Charlton 1 Department of Psychology, University of North Carolina at Pembroke, Pembroke, North Carolina, United States of America Conceptualization, Writing – original draft, Writing – review & editing Find articles by Kelly A Charlton 1 Editor: Laura Morett 5 Author information Article notes Copyright and License information 1 Department of Psychology, University of North Carolina at Pembroke, Pembroke, North Carolina, United States of America 2 School of Behavioral and Brain Sciences, University of Texas at Dallas, Richardson, Texas, United States of America 3 Callier Center for Communication Disorders, Dallas, Texas, United States of America 4 Department of Otolaryngology-Head and Neck Surgery, University of Texas Southwestern Medical Center, Dallas, Texas, United States of America 5 University of Missouri Columbia, UNITED STATES OF AMERICA ✉ * E-mail: [email protected] Competing Interests: The authors have declared that no competing interests exist. Roles Erik C Tracy : Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing Elizabeth D Young : Data curation, Formal analysis, Software, Validation, Visualization, Writing – review & editing Kelly A Charlton : Conceptualization, Writing – original draft, Writing – review & editing Laura Morett : Editor Received 2025 Sep 24; Accepted 2026 Mar 23; Collection date 2026. © 2026 Tracy et al This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice PMCID: PMC13068255  PMID: 41961808 Abstract Upon hearing a spoken utterance, listeners associate certain attributes (e.g., emotional) with self-identified gay male talkers and other attributes (e.g., reserved) with self-identified straight male talkers. In the current study, we explored whether listeners associated additional personal attributes with these types of talkers, and whether different contexts (e.g., listeners being informed of the talker’s sexual orientation) affected how strongly listeners associated personal attributes with talkers. Twenty-four talkers (twelve who self-identified as gay and twelve who self-identified as straight) from an established corpus were examined. Notably, previous work found that these talkers’ self-described sexual orientation (SO) did not always align with listener-perceived SO (i.e., a self-identified gay talker was perceived as straight sounding, and vice versa). Listeners evaluated these talkers for eight attributes (e.g., boring, confident, intelligent, mad, old, outgoing, sad, and stuck-up) in three contexts: talkers’ SO not referenced, talkers’ SO truthfully referenced (i.e., listeners were informed that a straight talker was straight), and talkers’ SO falsely referenced (i.e., listeners were informed that a straight talker was gay). Results suggested that self-identified gay and straight talkers whom listeners perceived as sounding gay were perceived as confident, mad, stuck-up, and outgoing; self-identified gay and straight talkers whom listeners perceived as sounding straight were perceived as sad and old. Furthermore, listeners’ judgments did not differ when the talkers’ SO was truthfully referenced, falsely referenced, or not referenced for all attributes except sad and stuck-up. The results indicate that perceived SO generally has the greatest effect on listeners’ perception of a talker’s attributes and, for most attributes examined, this is the case regardless of whether the listeners are informed (truthfully or falsely) of the talkers’ self-identified SO. Introduction Humans are decidedly social; we have numerous social interactions every day, and many of these interactions last for only a few seconds. Yet, even in these brief encounters, we quickly form impressions about others and decide how to react to them [ 1 ]. There is a wealth of research that investigates this process; some researchers focused on how viewers form impressions merely by looking at another individual’s face [ 1 ], while others have focused on how listeners form impressions by hearing a talker’s voice [ 2 ] Indeed, listeners do believe they can perceive a broad range of a talker’s personal attributes, such as their sexual orientation (SO) [ 3 ], age [ 4 ], emotional state [ 5 ], and personality traits [ 6 ]. To illustrate this, consider the following quote from The Golden Girls [ 7 ]. Sophia Petrillo (Estelle Getty): Do people treat you differently because you’re a lesbian? Blanche Devereux (Rue McClanahan): Well, most people don’t know. Sophia Petrillo: Really? I would have guessed right off. This exchange exemplifies how one believes that they can perceive another individual’s SO immediately, presumably through how they speak. Could Sophia have formed other impressions of Blanche, besides SO, just based on her voice? Might she think that Blanche is outgoing or intelligent? In fact, research has confirmed that listeners can quickly place talkers into a social category (e.g., gay man or lesbian woman) and associate various personal attributes with them [ 8 , 9 ]. What if Sophia already knew that Blanche was a lesbian woman? Would this label affect how she formed impressions of Blanche? For instance, if Sophia already knew Blanche’s SO, Sophia may believe that Blanche is even more outgoing and even more intelligent simply by hearing her voice. In the current study, we explored which personal attributes are associated with American English male talkers who were classified in a previous study [ 10 ] to sound either gay or straight. However, the self-identified SO of these talkers did not necessarily match the perception of their SO by listeners (e.g., talkers self-identified as gay, but were perceived as sounding straight, or vice versa). The first goal of the current investigation was largely exploratory; we investigated which attributes listeners associated with talkers who are perceived to sound either gay or straight, as well as whether actual or perceived SO is more important to listeners when associating attributes with talkers. Not knowing which attributes listeners may associate with gay and straight male talkers, we ran a pilot study in which participants freely described said gay and straight talkers [ 10 , 11 ]. The attributes used in the present study were those attributes that were identified most often by the listeners in the pilot study. The second goal was to determine if labeling talkers’ SO would change listeners’ perceptions of the talkers’ attributes [ 12 ]. To that end, we manipulated whether listeners were informed of the talkers’ self-identified SO in various experimental conditions; we either did not reference the talkers’ SO (Not Referenced), or the talkers’ SO was either truthfully or falsely reported (Truthfully Referenced and Falsely Referenced, respectively). It should be noted that there is a great need to investigate how listeners perceive other gender categories, such as women [ 13 ] and trans individuals [ 14 ], as well as other SOs [ 15 ]. However, this exploratory investigation is concerned with how listeners perceive self-identified gay cis males and straight cis males as a starting point to hopefully generate discussion and be used as a springboard for much-needed research into other populations. Perception of SO Many perceptual studies [ 2 , 3 , 16 – 18 ] have found that listeners can perceive a talker’s SO solely from the sound of their voice (often termed “auditory gaydar” in the literature). The perception of SO for men has been linked to various acoustic cues, such as pitch variability. Listeners, for instance, may rely on pitch variability to distinguish gay talkers from straight talkers [ 3 , 19 – 21 ]. The acoustic cues used to produce a stereotyped gay voice may not necessarily be the same acoustic cues that listeners employ when perceiving a talker’s self-reported SO [ 22 ]. Listeners are also able to form relatively accurate SO judgments about talkers’ self-identified SO upon being presented with either a relatively long speech stimulus (i.e., 15 seconds) [ 3 ] or a relatively short speech stimulus (i.e., a single phone) [ 10 ]. Perception of attributes As noted above, many personal attributes of an individual, such as their emotion and attitude, can be perceived from auditory cues alone. Here, we will use the term attributes as an overarching term to describe emotions, attitudes, and other characteristics, such as age, that don’t necessarily fall into the emotion or attitude categories. Emotions, such as anger and fear, are often intense and produced in response to a significant event [ 23 , 24 ]. As defined by others [ 24 , 25 ], attitudes, such as confidence and selfishness, are relatively long-lasting beliefs and predispositions that are less intense and more controlled compared to emotions. The pilot study (described more fully in the methods section) found that certain attributes were often associated with gay male talkers (confident, outgoing, stuck-up), other attributes were often associated with straight male talkers (old, sad, mad, and boring), and one attribute was associated with both groups of talkers (intelligent). Thus, a brief overview regarding these personal attributes specifically for male talkers is given below. We classified sad, mad, and boring as emotions [ 25 , 26 ]. The participants used the term “mad” to describe straight male talkers rather than “angry”; however, we presumed that mad and anger were closely linked in our participants’ minds and thus we will discuss the research results for anger. Additionally, we further explored the use of the terms “mad” and “anger” in Experiment 1a. Confident, outgoing, stuck-up, and intelligent were classified as attitudes [ 6 ]. Lastly, we will briefly discuss, when appropriate, both listeners’ accuracy when identifying the attributes, as well as the acoustic cues necessary to perceive the attributes [ 25 ]. The social voice space consists of two dimensions [ 6 , 12 ]. The first dimension is valence, which concerns whether an emotion is positive or negative [ 26 ]. All emotions studied here (i.e., sadness, mad, and boring) have a negative valence, while one of the attitudes (stuck-up) also has a negative valence. The remaining attitudes (i.e., confident, intelligent, and outgoing) have a positive valence. The second dimension is dominance, and one of our attitudes (e.g., confident) is seen as being highly dominant. In terms of acoustic cues, a higher average pitch corresponds to an increase in valence, while a decrease in average pitch was found to be more dominant [ 6 ]. Listeners can perceive a talker’s emotion through various acoustic cues [ 5 , 26 – 28 ]. For example, listeners may rely on a greater variability in fundamental frequency ( f o ) [ 29 ] and a faster rate of articulation [ 28 ] to perceive anger. Listeners may rely on less variability in f o and a relatively slower rate of articulation to perceive sadness [ 30 ]. Finally, listeners often confuse talkers to sound boring with talkers who sound depressed [ 31 ]; we presume that the auditory cues that indicate depression are likely to indicate talkers sound boring as well. It was found, for instance, that depressed talkers had less variability in f o and speak softly [ 32 ]. Listeners may rely on these acoustic cues to perceive a boring voice. Many attitudes are signaled acoustically by changes in prosodic contour and speech rate [ 25 ]. However, the acoustic or perceptual features of many individual attitudes, including stuck-up and outgoing have not, to our knowledge, been studied in any detail. With respect to sounding intelligent, it was discovered that gay men who produced both/s/-fronting and the variable – ing (e.g., walking), as opposed to – in (e.g., walkin’), were more likely to be labeled as effeminate and educated [ 16 ]. Here, we presume that the labels educated and intelligent are closely linked in listeners’ minds. Finally, utterances perceived as confident have a relatively large f o range and a steeper amplitude slope (i.e., the utterance started quiet and eventually got louder) [ 33 , 34 ]. Finally, talker age is perceived through various acoustic cues. For example, listeners reported that older talkers (i.e., over the age of 65) had a slower rate of articulation compared to younger talkers (i.e., under the age of 35) [ 4 ]; thus, listeners may rely on rate of articulation to perceive a male talker’s relative age [ 35 ]. SO and attributes Before further consideration of the link between which attributes are associated with gay and straight male talkers, it would be helpful to briefly consider which attributes are associated with gay and straight men in general. To start, there is a stereotype that gay men have attributes associated with women, and lesbian women have attributes associated with men [ 36 ]; this stereotype appears to be consistent across different periods of time [ 37 ]. Indeed, a group stereotype may affect one’s initial impression of another individual; for instance, women are believed to warmer than men, so a gay man might be perceived to be warmer than a straight man even if the gay man displays no indication of warmth [ 38 ]. As for specific attributes, gay men are perceived as talkative, gentle, fashionable, artistic and sensitive [ 11 ], whereas straight men are perceived as angry [ 39 ]. Upon reading a short description of a man, participants indicated that younger men were gay and older men were straight; it was proposed that young men are automatically processed as gay and that elderly men are automatically processed as heterosexual [ 40 ]. Gay men, as a subset of men, are viewed as being less competent than men in general [ 41 ]. However, gay men are not necessarily a homogenous category; different groups of gay men might have different attributes associated with them. For instance, a leather and biker gay man might be perceived as “macho” and likely to have tattoos, while a flamboyant gay man might be perceived as being loud and dramatic [ 42 ]. To our knowledge, there are fewer experiments that have investigated which attributes are associated with gay and straight male talkers, and talkers who are perceived to sound gay and straight (even if their self-identified SO does not match their perceived SO). Aligning with previously discussed research, it was found that listeners assigned gender typical stereotypes to straight male voices and assigned gender atypical stereotypes to gay male voices. For example, straight male talkers are perceived as aggressive, while gay male talkers are perceived as emotional [ 2 ]. With respect to the attributes in the current investigation, a stereotypical gay male voice was judged to be more intelligent compared to a stereotypical straight male voice [ 16 ], and a gay-sounding man was perceived to be as social, or presumably outgoing, as a straight-sounding man [ 17 ]. In a different study [ 8 ], gay-sounding and straight-sounding men were rated equally competent (where “competence” was described using three terms: competent, intelligent and skilled). Additionally, gay-sounding men were rated to be more social than heterosexual-sounding men (where “social” was described using the terms: likable, warm, and friendly). While the term outgoing was not specifically used here, it seems logical that these listeners might rate gay-sounding men as being more outgoing. To our knowledge, there are no studies that directly investigated the attributes sad, mad, boring, stuck-up, and confident. Understanding the influence of SO on attributes is of critical importance, particularly if there is an influence of even the impression of a certain SO [ 39 ]. In fact, there are many investigations that have found that LGBTQ+ individuals are at higher risk for discrimination compared to non-LGBTQ+ individuals [ 1 , 2 , 8 , 17 , 37 , 38 , 43 – 45 ]. For example, gay and straight men may face discrimination in the workplace [ 38 ]; straight men may be turned down for jobs that are typically associated with women, such as nurses, while gay men may not get a job typically associated with men, such as an engineer. Upon seeing photographs of gay and straight men, participants, including experienced hiring managers, were more likely to assign gay men to more caretaking jobs, such as pediatricians and English teachers, and assign straight men to more occupational roles, such as managers and surgeons [ 43 ]. Men with stereotypical gay voices may also experience negative stereotyping [ 45 ]. For example, in a study of music teachers, a stereotypical straight voice was rated as more mature, a better leader, and a stronger classroom manager; a stereotypical gay voice was rated as better organized and having higher standards. Finally, gay-sounding job applicants are perceived to be less competent and, thus, not considered for leadership jobs [ 17 ]. In sum, merely sounding gay might create a negative impression, and these negative impressions could lead to discrimination; thus, a deeper understanding is needed between which attributes are associated with gay and heterosexual sounding men. Knowledge of an individual’s self-identified SO There is one largely unanswered question within the auditory gaydar literature that requires further examination: does one’s knowledge of a talker’s self-identified SO influence which attributes are associated with said talker? In four different studies [ 2 , 8 , 17 , 45 ], listeners judged gay-sounding and straight-sounding male talkers on a variety of dimensions ranging from the effectiveness of the talkers’ teaching abilities [ 45 ] to the suitability of talkers for a job [ 17 ]. For each of these four studies, SO was not explicitly revealed; listeners were not told whether the talker was gay or straight. Even other related studies [ 43 ] did not reference an individual’s SO when listeners were asked to make a hiring decision. It seems reasonable to assume that this additional piece of knowledge may affect participants’ responses when it comes to teaching effectiveness or job suitability. Studies in related areas provide some insight. For example, knowledge of a speaker’s age affected how listeners perceived a vowel [ 46 ]. Additionally, if listeners believed that a talker was from Detroit, USA, then they classified vowels as produced by Detroit talkers; if listeners believed that a talker was from Canada, then they classified these same vowels as being produced by a Canadian talker [ 47 ]. In both studies, the label itself affected a listener’s perception of the utterance. Since an individual’s SO might not be readily apparent to others, they might not use this information when forming their initial impressions [ 1 ]. However, once an SO label is presented, individuals may use this information to form their impressions. Recently, researchers have begun to investigate both social vision [ 48 ] and social hearing [ 2 ]. It has been theorized that there are two pathways in which social hearing affects listeners’ impressions and behaviors toward speakers: a category-based process and a feature-based process. For the category-based process, the speech signal may lead to a social categorization, which in turn activates stereotypes and corresponding behaviors. For the feature-based process, the speech signal may activate stereotypes and corresponding behavior directly, essentially bypassing the categorization [ 2 ]. In the present study, the category-based process would predict that a listener would hear a voice and that voice would be perceived as gay; as a result, gay stereotypes (e.g., warmth) are activated in the listener’s mind. For the feature-based process, certain stereotypes (e.g., warmth) might be activated simply because the listener perceives the voice as feminine [ 2 ]. While the current study does not test this model directly, it may offer some insight. If the category-based process is dominant, then a talker being labeled as either gay or straight would, presumably, result in stronger attributes for that particular SO compared to a talker not being labeled at all. For instance, knowing that a talker is straight might result in stronger mad ratings compared to not knowing that the talker is straight. However, if the feature-based process is dominant, then a talker’s SO label, or lack of SO label, should result in equally strong stereotypes. For instance, mad ratings should be relatively equal regardless of the listener’s knowledge, or lack of knowledge, of the talker’s SO. Experimental objectives The first objective of Experiment 1 was largely exploratory. It was to ascertain which attributes listeners associated with talkers who are perceived to sound either gay or straight, and whether perceived or actual SO most strongly affected attribute perception. The pilot study described in the methods section was used to select several personal attributes for examination for this objective. It has previously been found that a talker’s self-identified SO did not always match how listeners perceived the talker’s SO [ 10 ], thus, the relationship between perceived and actual SO and attribute perception is of interest. As noted throughout the literature review above, there is limited research examining the various attributes associated with gay and straight talkers; however, as discussed, it is important to understand the potential differences between the perception of these groups of talkers as these may be drivers of discrimination. The second objective of Experiment 1 was to assess whether a listener’s knowledge of a talker’s SO would influence which attributes were associated with them. To that end, we manipulated whether listeners were informed of the talkers’ self-identified SO; we did not reference the talkers’ SO (Not Referenced), or the talkers’ SO was either truthfully or falsely reported (Truthfully Referenced and Falsely Referenced, respectively). As indicated previously, if the category-based process is dominant, then an SO label would result in stronger ratings of an attribute compared to no SO label; if the feature-based process is dominant, then attribute ratings should be similar regardless of the SO label, or lack of a label. Finally, Experiment 1a was included to explore listeners’ interpretation of the attribute mad. Given that this attribute has two possible interpretations (angry or crazy), the aim of this experiment was to determine whether the results of Experiment 1 would be largely replicated if the word “mad’ was replaced with the word “angry”. Experiment 1 Materials and methods This research was approved by the Institutional Review Board at the University of North Carolina at Pembroke via an expedited review, protocol number 14-08-004. Participant recruitment started on August 4, 2016 and ended on June 8, 2019. Participants were provided an informed consent form by the first author or research assistants. If the participant consented to participate in the study, they signed the consent form with either the first author or a research assistant as a witness. Participants. Two hundred and ten participants who were taking an introductory psychology course at a rural southeastern university participated in the study in exchange for course credit; participants indicated their consent by signing their name to an informed consent form. In a post-experiment questionnaire, we collected demographic information about the participants, such as their age and gender, but regrettably, we did not collect information related to participants’ race or ethnicity. However, the student population at said rural southeastern university is racially and ethnically diverse. For example, the student population was approximately 34% White, 30% Black or African American, 12% American Indian or Alaska Native, 10% Hispanic or Latino, 6% Two or More Races, and 2% Asian [ 49 ]. We presume that the relatively large participant sample was similar, in this respect, to the larger campus population. The participants’ mean age was 19.6 (Range=[18–53]); approximately 95% of participants were between 18 and 23 years old. All participants self-reported that American English was the first language that they learned and used, none reported a history of hearing or speech disorders [ 50 ]. These are common inclusion criteria for listeners in these types of perceptual experiments [ 51 ]. Stimuli. As detailed previously, utterances were provided by 24 male talkers (12 self-identified as straight and 12 self-identified as gay) from the Ohio State University campus and the surrounding Columbus, Ohio metropolitan area [ 10 ]. It should be noted that the talkers’ self-identified SO did not always align with how listeners perceived them based on results gathered in a previous study [ 10 ]. For instance, a self-identified straight talker was sometimes perceived as gay and vice versa. While talkers were required to be from Ohio, the researchers did not control for specific areas of the state and their corresponding regional dialects. The first author, who is a trained phonetician, did not perceive any noticeable differences among the talkers’ dialects; we presumed that listeners would not perceive any differences either. Yet, we cannot rule out the possibility that disparities between the listeners’ regional dialects and the talkers’ regional dialects influenced some listeners’ responses [ 52 ]. Based on the results of earlier research [ 10 ], we selected utterances produced by the six self-identified gay talkers whom listeners were most confident in labeling as gay and the six self-identified gay talkers whom listeners were most confident in labeling as straight [ 45 ]. A similar procedure for selecting utterances was used for the self-identified straight talkers. Here, confidence in a talker’s SO was operationalized by using a 7-point scale, where a ‘7’ rating indicated that the listener was very confident that the talker was gay and a ‘1’ rating indicated that the listener was very confident that the talker was straight. Therefore, for instance, we selected the six self-identified gay talkers that had the highest rating on the 7-point rating scale and the six self-identified gay talkers who had the lowest rating. This procedure resulted in utterances from four groups of six talkers: gay talkers who were perceived as gay-sounding, gay talkers who were perceived as straight sounding, straight talkers who were perceived as gay-sounding, and straight talkers who were perceived as straight-sounding. The utterances produced by these four groups of talkers were concatenated strings of three monosyllabic CVC words that contained two instances of/s/, in line with previous research that listeners relied on/s/ to distinguish between gay and straight male talkers of American English [ 53 ]. The two utterances were dose-wet-soap and sell-tone-niece [ 54 ] . Thus, we created 48 different stimuli. In line with others, we used read speech to minimize the possibility that listeners would use the content of spontaneous speech to form their attribute judgments [ 14 ]. Pilot study to identify attributes. The purpose of the pilot study was to explore the most common attributes associated with gay and straight talkers. To that end, we selected the seven self-identified gay talkers whom listeners were most confident in labeling as gay and the seven self-identified straight talkers whom listeners were most confident in labeling as straight [ 10 , 45 ]. The utterances produced by these talkers were concatenated strings of three monosyllabic CVC words (e.g., sad-vein-niece and dose-wet-soap ) [ 54 ]. Two utterances were created for each talker and there were 28 trials. Upon hearing the utterances, 21 participants, from a rural southeastern university, freely indicated anything that they wanted to about the talkers [ 11 ]. Talkers’ self-identified SO was not discussed during the study procedures. Based on the results, the most cited attributes were selected for examination in the current study. For instance, sad was indicated 137 times across both SOs, and mad was indicated 56 times across both SOs. While some of these attributes were listed more frequently for gay talkers compared to straight talkers, care was taken to choose attributes that were listed with relatively high frequency across both talker groups. Based on the pilot data, the selected attributes were sad, boring, confident, outgoing, mad, old, intelligent, and stuck-up. We used the attribute mad, instead of angry, because this was the term indicated by participants [ 27 ]. Even though there are multiple interpretations of mad (e.g., angry or crazed), we presumed that listeners would associate mad with angry; to confirm this, we examined whether listeners associated mad with angry in Experiment 1a. Design. To test all the stimuli under the experimental conditions, we created four experimental lists for each of the three groups of listeners (e.g., Not Referenced, Truthfully Referenced, and Falsely Referenced); the three groups of listeners were all presented with the same four lists. Moreover, four attributes (e.g., mad, intelligent, stuck-up, and boring) were assigned to the first two lists, while the other four attributes (e.g., sad, outgoing, old, and confident) were assigned to the remaining two lists. Within a single list, there were 192 trials. Each list was divided into four blocks, and each block corresponded to one of the four attributes (e.g., boring, intelligent) that the listeners would be evaluating; utterances were randomized within each block. However, if a listener evaluated an utterance to be boring and, in a subsequent block, they were tasked to evaluate the same utterances as being intelligent, their earlier boring judgments might influence their intelligent judgments. To account for these potential carryover effects, we counterbalanced the order of assessed attributes. For example, within one list, one group of listeners evaluated utterances on the boring attribute in Block 1 and evaluated utterances on the intelligent attribute in Block 4. In a second list, a second set of listeners evaluated utterances on the intelligent attribute in Block 1 and evaluated utterances on the boring attribute in Block 4. Approximately 15–18 listeners were assigned to each of the twelve lists. Procedure. Participants were brought into a sound-treated room and seated in front of a computer monitor and mouse. They were instructed to listen carefully to the stimuli over a set of headphones and, after hearing each utterance, to indicate via the mouse how strongly the talker was associated with a particular attribute. There was a column of seven squares on the left side of the computer screen, numbered from ‘7’ to ‘1’ starting at the top going down. The ‘7’ square was labeled Strongly associated with this trait and the ‘1’ square was labeled Weakly associated with this trait ; the rest of the squares were not labeled other than their corresponding number. Here, in the experiment, we used the term trait because, presumably, participants would be more familiar with this term rather than attribute . The attribute that the participant judged was written at the top of the computer screen; next to this attribute, for both the Truthfully Referenced and Falsely Referenced conditions, the talkers’ SO was displayed. Participants were encouraged to use the entire 7-point scale and to respond as quickly as possible; only participant ratings were collected. After several practice trials, participants were allowed to ask the researcher questions and, at the conclusion of the experiment, participants completed a demographic questionnaire and were debriefed. Statistical analyses. Likert data, such as that used in the current study, are frequently treated as continuous, normally distributed data for the purposes of statistical analyses. However, a seven-point equal-appearing-interval scale should more appropriately be treated as count data, as it increases only in discrete units and is bounded by zero. Moreover, as discussed previously, the talkers were purposefully selected in a way that allowed us to treat perceived SO as a binary; perception of SO is not a binary, but we chose to treat it as such because we were examining the talkers at the extreme ends of the spectrum. Thus, the data were analyzed with a series of mixed-effects Poisson regression models, using a separate model for each attribute as the dependent variable. Both research aims, or objectives, were answered using separate statistical tests. The first aim, to investigate how strongly participants associated eight attributes (i.e., boring, confident, intelligent, mad, old, outgoing, sad, and stuck-up) with utterances produced by self-identified gay and straight male talkers who sounded both gay and straight, were analyzed using a mixed-effects Poisson regression. To further probe this question, we investigated whether a talker’s self-identified SO, or their perceived SO, was the most important factor when listeners formed their judgments about the talkers. The second aim (i.e., whether there were differences in the listeners’ attribute ratings depending on whether truthfully or falsely informed of the talker’s SO) was likewise examined in two ways. First, it was examined using the fixed effect of condition in the mixed-effects Poisson regression models described above. It was also examined using a second model for each attribute containing two interaction effects, using the same referent categories described above; the first interaction was between condition and perceived SO, and the second interaction was between condition and actual, or self-identified, SO. Random structures for all mixed-effects models were determined separately, where each model included all considered random effects so long as each a) improved model fit as measured by a −2 log-likelihood test compared to the previous best-fitting model, and b) the data continued to support the more complex model, as evidenced by visual examination of variance components and model estimates. Random effects considered for each model included the following, in order of consideration: a) random intercept of listener; b) random by-listener slope of listener gender (so long as a random intercept of listener improved model fit); c) random intercept of stimulus; d) random intercept of list. All statistical analyses were completed in R (version 4.3.3). Mixed-effect Poisson regression models were completed using the glmmTMB package (version 1.1.9) [ 55 ]. Significant interactions were decomposed using the emmeans package (version 1.10.0) [ 56 ]. We determined that a statistical test reached significance if p < 0.05. Finally, it should be noted that given the large number of statistical analyses that were performed, for the most part, we will only be reporting and discussing those analyses that reached statistical significance. All the data sets and R files, containing statistical analyses, are publicly available at the following link: https://osf.io/bc2wm/overview Those readers who wish to discuss all statistical analyses contained in the R files, such as those that did not reach statistical significance, should contact the first author. Results Aim 1. Strength of Association of attributes with actual and perceived sexual orientation. As a reminder, the first aim was to determine how strongly participants associated eight attributes with utterances produced by self-identified gay and straight male talkers who sounded both gay and straight. Descriptive statistics of each attribute by talker actual SO, perceived SO, and condition are displayed in Table 1 . An examination of this table reveals relatively small differences in ratings between the utterances produced by gay and straight talkers (i.e., actual SO) for all attributes. This was confirmed in the results for the fixed effects of actual SO as examined in the mixed-effects Poisson regression, which are summarized in Table 2 . As can be seen in Table 2 and Fig 1 , none of the fixed effects of actual SO reached statistical significance. In contrast, Table 1 reveals several attributes with substantial differences in ratings between utterances produced by talkers who were perceived as sounding straight and talkers who were perceived as sounding gay, regardless of the talker’s actual SO, which can be seen in Fig 2 . For example, the largest magnitude difference was for stuck-up, where talkers perceived as sounding straight were given an average rating of 3.21 and talkers perceived as gay were given an average rating of 4.44 (difference = 1.23 Likert scale points). This was again confirmed with the regression models; as seen in Table 2 , there was a significant effect of perceived SO for the following attributes: sad, confident, outgoing, old, mad, and stuck-up. The effect was positive (i.e., the attribute was more likely to be perceived for talkers who were perceived as gay) for confident, outgoing, mad, and stuck-up. The effect was negative (i.e., the attribute was less likely to be perceived for talkers who were perceived as gay) for sad and old. As for the two remaining attributes, boring approached significance and the effect was negative, whereas intelligent was not significant. Furthermore, consistent with the descriptive statistics, the magnitude of the effect was largest for stuck-up, where the rate ratio (RR) predicts an increase of 1.38 Likert scale points in this attribute as perceived SO changed from straight to gay. This is followed closely by outgoing, where the RR predicts an increase of 1.37 Likert scale points as perceived SO changed from straight to gay. Thus, the results of Aim 1 indicated that none of the attributes were significantly associated with a talker’s actual SO; certain attributes (i.e., confident, outgoing, mad, and stuck-up) were significantly associated with utterances produced by talkers perceived to sound gay, and other attributes (i.e., sad and old) were significantly associated with utterances produced by talkers perceived to sound straight. Table 1. Descriptive statistics for each attribute. Sad Mad Mean SD Mean SD Actual SO Gay 3.69 2.02 Actual SO Gay 3.47 1.89 Straight 3.50 1.83 Straight 3.48 1.92 Perceived SO Gay 3.18 1.79 Perceived SO Gay 3.86 1.93 Straight 4.01 1.97 Straight 3.08 1.80 Experiment Not Referenced 3.30 1.98 Experiment Not Referenced 3.48 1.91 Truthfully Referenced 3.74 1.89 Truthfully Referenced 3.41 1.87 Falsely Referenced 3.71 1.89 Falsely Referenced 3.54 1.94 Confident Intelligent Actual SO Gay 4.16 1.96 Actual SO Gay 3.99 1.74 Straight 4.29 1.81 Straight 4.3 1.70 Perceived SO Gay 4.67 1.77 Perceived SO Gay 4.06 1.69 Straight 3.78 1.89 Straight 4.23 1.77 Experiment Not Referenced 4.16 1.9 Experiment Not Referenced 4.02 1.75 Truthfully Referenced 4.18 1.89 Truthfully Referenced 4.21 1.67 Falsely Referenced 4.32 1.86 Falsely Referenced 4.19 1.77 Outgoing Stuck-up Actual SO Gay 3.88 1.95 Actual SO Gay 4.07 2.08 Straight 3.82 1.77 Straight 3.59 1.95 Perceived SO Gay 4.43 1.81 Perceived SO Gay 4.44 1.96 Straight 3.28 1.74 Straight 3.21 1.90 Experiment Not Referenced 3.85 1.88 Experiment Not Referenced 3.71 2.10 Truthfully Referenced 3.85 1.86 Truthfully Referenced 3.87 1.93 Falsely Referenced 3.85 1.86 Falsely Referenced 3.88 2.07 Old Boring Actual SO Gay 3.44 1.62 Actual SO Gay 4.20 2.02 Straight 3.80 1.71 Straight 3.99 1.82 Perceived SO Gay 3.09 1.56 Perceived SO Gay 3.74 1.85 Straight 4.14 1.61 Straight 4.45 1.93 Experiment Not Referenced 3.48 1.69 Experiment Not Referenced 4.05 1.99 Truthfully Referenced 3.69 1.65 Truthfully Referenced 4.11 1.81 Falsely Referenced 3.67 1.67 Falsely Referenced 4.12 1.99 Open in a new tab Notes: SO = sexual orientation. SD = standard deviation. Table 2. Summary of mixed-effects Poisson regression fixed effects models for each attribute. Attribute Fixed effect Estimate RR RR 95% CI z p Sad Actual SO 1 −0.03 0.97 [0.78, 1.22] −0.23 .822 Perceived SO 2 −0.23 0.79 [0.63, 0.99] −2.02 .044 Truthfully Referenced 3 0.13 1.14 [1.04, 1.24] 2.98 .003 Falsely Referenced 3 0.12 1.13 [1.04, 1.23] 2.95 .003 Confident Actual SO 1 0.07 1.07 [0.87, 1.32] 0.61 .540 Perceived SO 2 0.24 1.27 [1.03, 1.56] 2.22 .026 Truthfully Referenced 3 0.00 1.00 [0.95, 1.06] 0.16 .874 Falsely Referenced 3 0.04 1.04 [0.99, 1.09] 1.54 .124 Outgoing Actual SO 1 0.01 1.01 [0.81, 1.27] 0.11 .910 Perceived SO 2 0.31 1.37 [1.09, 1.71] 2.73 <.001 Truthfully Referenced 3 0.00 1.00 [0.94, 1.06] 0.01 .992 Falsely Referenced 3 0.00 1.00 [0.94, 1.06] 0.00 .998 Old Actual SO 1 0.10 1.10 [0.95, 1.27] 1.32 .186 Perceived SO 2 −0.30 0.74 [0.64, 0.85] −4.11 <.001 Truthfully Referenced 3 0.07 1.07 [0.99, 1.16] 1.66 .097 Falsely Referenced 3 0.06 1.06 [0.98, 1.15] 1.49 .137 Mad Actual SO 1 0.00 1.00 [0.87, 1.14] −0.05 .962 Perceived SO 2 0.22 1.25 [1.09, 1.43] 3.13 .002 Truthfully Referenced 3 −0.01 0.99 [0.89, 1.1] −0.20 .840 Falsely Referenced 3 0.02 1.02 [0.92, 1.14] 0.44 .661 Intelligent Actual SO 1 0.09 1.09 [0.94, 1.27] 1.16 .247 Perceived SO 2 −0.03 0.97 [0.84, 1.13] −0.33 .741 Truthfully Referenced 3 0.05 1.05 [1.0, 1.1] 1.84 .066 Falsely Referenced 3 0.04 1.04 [0.99, 1.1] 1.58 .114 Stuck-Up Actual SO 1 −0.12 0.89 [0.79, 1.01] −1.84 .066 Perceived SO 2 0.32 1.38 [1.22, 1.56] 5.12 <.001 Truthfully Referenced 3 0.05 1.05 [0.97, 1.14] 1.20 .232 Falsely Referenced 3 0.05 1.05 [0.97, 1.15] 1.19 .234 Boring Actual SO 1 −0.03 0.97 [0.81, 1.15] −0.39 .695 Perceived SO 2 −0.17 0.85 [0.71, 1.0] −1.90 .057 Truthfully Referenced 3 0.02 1.02 [0.97, 1.06] 0.69 .490 Falsely Referenced 3 0.02 1.02 [0.97, 1.07] 0.77 .440 Open in a new tab Notes: SO = Sexual Orientation. RR = Rate Ratio. CI = Confidence Interval. 1 Reference category = gay. 2 Reference category = perceived as sounding straight. 3 Reference category = Not Referenced condition. Bolded p values indicate the value was < .05. Fig 1. Average listener ratings of each attribute by talker self-identified SO. Open in a new tab Listeners rated each attribute using a 7-point Likert scale, where 7 represented “Strongly associated with this trait ” and 1 represented “Weakly associated with this trait”. Solid lines within the boxplots represent median values, while diamonds represent mean values. Fig 2. Average listener ratings of each personal attribute by talker perceived SO. Open in a new tab Listeners rated each attribute using a 7-point Likert scale, where 7 represented “Strongly associated with this trait ” and 1 represented “Weakly associated with this trait”. Solid lines within the boxplots represent median values, while diamonds represent mean values. * = significant at p < .05, ** = significant at p < .01, *** = significant at p < .001. Aim 2. Differences in attribute ratings by listening condition. The second aim was to determine whether there were differences in the listeners’ attribute ratings depending on if they were truthfully or falsely informed of the talker’s SO. This was accomplished, first, by examining the fixed effect of condition in the mixed-effects Poisson regression models, which is described in Aim 1 and displayed in Table 2 . Moreover, we used an additional model that contained two interaction effects; the first interaction was between condition and perceived SO, and the second interaction was between condition and actual SO. With respect to the model described in Table 2 , the fixed effect of condition was significant for the sad attribute only. This effect for sad was significant for both the Truthfully Referenced and Falsely Referenced conditions, and was positive in both cases, indicating that, compared to the Not Referenced condition, listeners were more likely to associate the attribute sad with utterances produced by talkers when listeners were both truthfully and falsely told of the talkers’ SO. None of the other seven attributes’ fixed effects of condition were significant. With respect to all the attributes except sad, this result suggests that the speech signal itself was driving participants’ attribute perceptions rather than their knowledge of the talkers’ SO. The interaction between actual SO and condition was significant for confident in the Truthfully Referenced condition (β = −0.07, RR = 0.93, z (5031) = −2.02, p = .044), stuck-up in the Falsely Referenced condition (β = −0.13, RR = 0.88, z (5031) = −3.57, p < .001), and boring for both conditions (β > 0.06, RR > 1.07, z (5031) >2.03, p < .042). However, after correcting for multiple comparisons using the Sidak method, none of the follow-up contrasts for these interactions reached statistical significance, indicating that the magnitude of these interaction effects was likely small [ 57 ]. Finally, the interaction between perceived SO and condition was significant for sad for both conditions (β > 0.09, RR > 1.04, z (5031) >2.43, p < .015) and stuck-up in the in the Falsely Referenced Condition (β = 0.09, RR = 1.10, z (5031) = 2.47, p = .014). Significant follow-up contrasts are shown in Table 3 , with row numbers labeled and referenced throughout the following two paragraphs. To briefly summarize these results, there was a significant decrease in ratings of sad for listeners who heard utterances produced by talkers perceived as gay in the Not Referenced condition compared to all other talkers and all other conditions (e.g., Rows 6–9). These results suggest that, in the absence of any information regarding SO, listeners perceived utterances produced by talkers who sounded gay to be less sad compared to utterances from all other talker groups when information regarding SO was provided; these results align with the findings from the model described in Table 2 . Further, these findings indicate that listeners rated utterances produced by talkers perceived as sounding gay as significantly more sad when they had information regarding their SO, regardless of whether that information was true or false. This interaction effect, as well as the significant contrasts, are displayed in Fig 3 . Table 3. Follow-up pairwise contrasts for significant perceived SO x Condition interaction effects; also illustrated in Figs 3 and 4. Sad Stuck-Up Difference z p Difference z p 1. Straight – Not Referenced// Gay – Not Referenced 1.36 2.59 .099 0.74 −4.22 <.001 2. Straight – Not Referenced// Straight – Truthfully Referenced 0.93 −1.70 .531 0.94 −1.34 .762 3. Straight – Not Referenced// Gay – Truthfully Referenced 1.12 0.93 .938 0.71 −4.24 <.001 4. Straight – Not Referenced// Straight – Falsely Referenced 0.92 −1.87 .419 1.00 0.04 1.000 5. Straight – Not Referenced// Gay – Falsely Referenced 1.14 1.06 .898 0.68 −4.81 <.001 6. Gay – Not Referenced// Straight – Truthfully Referenced 0.68 −3.10 . 024 1.27 2.95 .038 7. Gay – Not Referenced// Gay – Truthfully Referenced 0.83 −3.99 . 001 0.96 −0.92 .943 8. Gay – Not Referenced// Straight – Falsely Referenced 0.68 −3.16 . 020 1.35 3.72 .003 9. Gay – Not Referenced// Gay – Falsely Referenced 0.84 −3.73 . 003 0.91 −1.94 .376 10. Straight – Truthfully Referenced// Gay – Truthfully Referenced 1.21 1.65 .567 0.76 −3.98 .001 11. Straight – Truthfully Referenced// Straight – Falsely Referenced 0.99 −0.14 1.000 1.07 1.43 .710 12. Straight – Truthfully Referenced// Gay – Falsely Referenced 1.23 1.70 .533 0.72 −4.12 .001 13. Gay – Truthfully Referenced// Straight – Falsely Referenced 0.82 −1.62 .583 1.41 4.32 <.001 14. Gay – Truthfully Referenced// Gay – Falsely Referenced 1.02 0.34 .999 0.95 −1.11 .876 15. Straight – Falsely Referenced// Gay – Falsely Referenced 1.24 1.83 .444 0.68 −5.56 <.001 Open in a new tab Note: “Gay” and “Straight” throughout the above table refer to perceived SO, regardless of the talker’s self-identified SO. Bolded p values indicate the value was < .05. Fig 3. Interaction effects between perceived SO and perception of the “sad” attribute. Open in a new tab Listeners rated the sad attribute using a 7-point Likert scale, where 7 represented “Strongly associated with this trait ” and 1 represented “Weakly associated with this trait”. In the three conditions, listeners either received no information about the talkers’ SO (Not Referenced), accurate information about the talkers’ SO (Truthfully Referenced), or inaccurate information about the talker’s SO (Falsely Referenced). * = significant at p < .05, ** = significant at p < .01, *** = significant at p < .001. As illustrated in Fig 4 , the pattern of significant contrasts was more complex for stuck-up. The difference between this attribute was significant between talkers who were perceived as sounding straight and gay in all three conditions (Rows 1, 10, and 15), signifying a strong effect of perceived SO on the perception of being stuck-up, regardless of condition. All other contrasts reveal that the perception of the attribute was higher for utterances produced by talkers who were perceived as gay than for utterances produced by talkers who were perceived as straight (Rows 3, 5, 6, 8, 12, and 13). None of the contrasts indicating a change in the information listeners had regarding SO altered their perception of who sounded stuck-up were significant (i.e., the contrasts that compared a single perceived orientation across conditions, such as contrasts 2, 4, 7, 9, or 11). Overall, therefore, these results suggest that, regardless of the listening condition, listeners perceived utterances produced by talkers who were perceived as gay as being more stuck-up compared with talkers who were perceived as straight. Fig 4. Interaction effects between perceived SO and perception of the “stuck-up” attribute. Open in a new tab Listeners rated the stuck-up attribute using a 7-point Likert scale, where 7 represented “Strongly associated with this trait ” and 1 represented “Weakly associated with this trait”. In the three conditions, listeners either received no information about the talkers’ SO (Not Referenced), accurate information about the talkers’ SO (Truthfully Referenced), or inaccurate information about the talker’s SO (Falsely Referenced). * = significant at p < .05, ** = significant at p < .01, *** = significant at p < .001. Experiment 1a The purpose of this experiment was to determine if participants interpreted mad to mean angry, rather than alternate interpretations such as crazy. We reran a portion of Experiment 1 and substituted the word angry for mad. We hypothesized that if the averages for mad, from Experiment 1, were like the averages for angry, then participants in Experiment 1 interpreted mad as angry. Materials and methods Participants. A new sample of twenty-five participants, who participated in the experiment in exchange for course credit, had similar characteristics to the sample of participants from Experiment 1. Stimuli. The two stimuli used in this experiment, dose-wet-soap and sell-tone-niece , were the same stimuli used in Experiment 1. Design. We created one experimental list for participants in which the talkers’ SO was not referenced. We assigned four of the eight attributes (e.g., angry, intelligent, stuck-up, and boring) to this list. Within this list, there were 192 trials. The list was divided into four blocks, and these blocks corresponded to which attribute (e.g., angry, intelligent, stuck-up, and boring, respectively) the listeners judged. Within each block, we randomized the utterances. Procedure. The procedure for Experiment 1a was identical to the procedure for Experiment 1. Results We analyzed the data to determine if participants interpreted mad as angry. To accomplish this, the angry ratings in the present experiment were compared against the mad ratings from Experiment 1. The average ratings for angry were 3.59 ( SD = 1.98) and the average ratings for mad were 3.48 ( SD = 1.91). The difference between the ratings was not significant, t (2686) = −1.54, p = 0.12. The average ratings for angry and mad were also compared across actual and perceived SO for gay and straight talkers, as displayed in Table 4 below. None of these contrasts were significant ( t (29) ≤ 1.06, p ≥ 0.296). Based on these results, we concluded that mad was likely to have similar interpretation as angry. Table 4. Two-sample t-tests comparing the attributes mad and angry across perceived and actual SO. Mad Angry t (df) p Mean (SD) Mean (SD) Actual SO Gay 3.47 (1.89) 3.49 (2.00) 0.05 (29) 0.962 Straight 3.48 (1.92) 3.70 (1.95) 0.53 (29) 0.598 Perceived SO Gay 3.86 (1.93) 3.65 (1.94) 0.51 (29) 0.613 Straight 3.08 (1.80) 3.53 (2.02) 1.06 (28) 0.296 Open in a new tab Discussion The present experiment had two aims. The first aim, which was largely exploratory, was to determine how strongly listeners associated certain attributes (i.e., boring, confident, intelligent, mad, old, outgoing, sad, and stuck-up) with utterances produced by self-identified gay and straight male talkers who were perceived as sounding gay or straight. Second, we investigated how listeners’ perceptions of these attributes might change based on their knowledge of the talkers’ SO (i.e., Not Referenced, Truthfully Referenced, or Falsely Referenced). There are three important results. One, as detailed in Aim 1, some attributes were more associated with one group of talkers compared to the other group of talkers. For example, the attributes confident, outgoing, mad, and stuck-up were significantly more associated with utterances produced by talkers perceived to sound gay, while the attributes sad and old were significantly more associated with utterances produced by talkers perceived to sound straight. Of the two remaining attributes, boring approached significance (i.e., listeners trended towards associating this attribute with utterances produced by talkers perceived as straight) and intelligence was not significant (i.e., listeners did not associate this attribute with utterances produced by either group of talkers). Two, the talkers’ perceived SO, and not their self-identified SO, appeared to be the more important factor for how listeners formed their judgments for all of the studied attributes except boring and intelligent. While boring approached significance (i.e., listeners’ ratings trended towards showing differences in this attribute based on perceived SO), intelligent was not significant (i.e., listeners’ ratings did not differ based on perceived SO). Stated differently, listeners trended towards perceiving straight-sounding talkers as being more boring compared to gay-sounding talkers; listeners found both gay- and straight-sounding talkers to be equally intelligent. One possible reason for this finding is that listeners were deliberately avoiding high or low ratings for certain attributes, or certain talkers, to avoid appearing biased. For instance, listeners might want to be accepting of different populations, and thus they would not describe a certain group of people as being less intelligent than another group. This possibility is further discussed in the Implications for the Workplace section. Three, for the attributes confident, outgoing, mad, old, boring, intelligent, and stuck-up, listeners’ ratings did not change based on the listening condition (i.e., whether the talker’s SO was referenced). In terms of the social hearing model [ 2 ], there are two pathways in which social hearing can impact listeners’ impressions of talkers. The first pathway is a category-based process in which the speech signal leads to social categorization, which in turn activates stereotypes. The second pathway activates stereotypes directly and bypasses the categorization. We predicted that if listeners rely on the categorization pathway, then knowledge of the talkers’ SO might result in stronger ratings. For example, if listeners have a stereotype that straight men are old, then listeners should indicate, via the rating scale, that they are more confident in their old ratings compared to when the listeners had no knowledge of the talkers’ SO. For seven of the eight attributes, this was not the finding. Rather, there were no differences in ratings if listeners had knowledge, or no knowledge, of the talkers’ SO. In fact, even when we disguised the talkers’ SO (i.e., Falsely Referenced Condition), listeners’ ratings did not change compared to when listeners had no knowledge of the talkers’ SO (i.e., No Referenced Condition). Therefore, our results suggest that, for the most part, the speech signal itself can directly activate stereotypes. The results for sad, in contrast, did not align with the other attributes. In the absence of any SO labels, listeners rated gay-sounding listeners as being significantly less sad compared to other listening conditions, such as straight-sounding talkers with no SO label and all talkers with a SO label, whether it be true or false. One interpretation of this result is that, in the absence of any information, a gay-sounding voice is not perceived as sounding sad compared to straight-sounding voices. However, there are also several alternate interpretations. First, attributes may work along different pathways in social hearing model [ 2 ]. Perhaps when listeners are asked to judge some attributes (e.g., outgoing), the voice itself could activate the stereotypes; however, when asked to judge other attributes (e.g., sad), the voice activates a SO categorization, which in turn, activates the stereotypes. Hence, when SO is made apparent, this SO label activates the sad stereotype. One way to test this hypothesis is to ask participants, in the absence of any vocal cues, whether they associate sadness with gay or straight men. Another possibility is that there is a stimulus-specific effect occurring for this experiment. A concatenated string of three monosyllabic words (i.e., the stimuli used here) presumably does not contain a lot of pitch variability, and less pitch variability is an acoustic feature that cues sadness [ 30 ]. However, it should be noted that, within the pilot study, which also used three monosyllabic words as stimuli, listeners freely associated our talkers with the attribute sad. Despite this, it is possible that different speech stimuli would lead to different results. Thus, the most likely explanation given the current results is that, in the absence of any identifying information, listeners perceive utterances produced by gay-sounding talkers as being less sad compared to utterances produced by straight-sounding talkers. Implications for the workplace There are social implications for the results of the current study. As mentioned previously, LGBTQ+ individuals, compared to non-LGBTQ+ individuals, are at a higher risk for discrimination during hiring practices [ 43 ] and gay men may also be viewed negatively in the classroom [ 45 ]. Indeed, this type of discrimination appears to be quite common in the workplace [ 58 , 59 ]. The evidence suggests that this discrimination appears in both Western and non-Western societies, and, across both these societies, gay men are disliked more than lesbian women [ 60 ]. An overarching finding is that gender, sexual, and heterosexist harassment are linked together [ 58 , 59 ]. Sexual harassment is defined as unwanted sex-behavior at work that the intended target perceives as both threatening and offensive; heterosexist harassment is defined as behaviors and/or thoughts that deny, malign, and stigmatize any non-heterosexual behaviors, identities, or relationships [ 58 ]. Indeed, the link between gender and heterosexist discrimination should not be surprising given that across many domains (e.g., legal, historical, cultural, psychological), people view lesbian women, gay men, and bisexual individuals (LGB) as failing to conform to traditional gender expectations [ 44 ]; gender inequality and a lack of LGB rights align with one another [ 58 , 61 ]. Often, sexual and/or gender harassment can affect an employee’s wellbeing in many ways, including increased job dissatisfaction, procrastination, decreased loyalty to employers, and taking time off work [ 59 ]. With respect to the current study, current or future employees who are perceived as gay, regardless of their self-identified SO, may experience harassment at work. For example, as part of the interview process, an employer may conduct a phone interview with potential employees, who may be perceived as gay and, therefore, stuck-up. This might not be a personality attribute that is desirable to an employer, and the employee thus may not be hired based solely on how their voice is perceived. Since SO is not always an obvious identity [ 1 , 39 ], this harassment may extend to individuals who do not self-identify as LGBTQ + . For instance, a self-identified straight man, who is perceived as sounding gay, may be considered less competent and not considered for a leadership role compared to other straight men who are perceived as sounding straight [ 17 , 38 ]. Providing employers and supervisors additional education regarding how their thoughts and behaviors could be guided by the way that their employees look and sound [ 1 ], then it may aid in reducing workplace discrimination. Additional considerations and interpretations We want to be mindful of the conclusions that we draw from our data. One consideration is whether the talkers themselves would self-identify with these attributes. For example, utterances produced by talkers perceived as straight were rated as sad, and utterances produced by talkers perceived as gay were rated as stuck-up – but do these talkers self-identify as being sad or stuck-up? Do they perceive their own speech as sounding sad or stuck-up? Earlier research suggested that labels assigned to talkers might not align with how talkers describe themselves [ 62 ]. For example, after rating talkers on a variety of attributes, listeners were often in agreement with one another; if one listener rated a talker as aggressive, then other listeners also rated this same talker as aggressive. However, listeners’ ratings did not align with how the talkers described themselves; talkers attributed as aggressive did not describe themselves as aggressive. We should therefore presume that listeners’ perceptions of the talkers’ speech may not be reflective of how the talkers self-identify. A second consideration is that many of the attributes used in the current study are what are known as “polar adjectives”, which are terms that tend to come in pairs and contain opposite meanings (e.g., sad and happy) [ 63 ]. However, just because a single polar adjective was included in the study (e.g., old) does not necessarily mean that talkers would be rated in the opposite direction for the other polar adjective not included in the study (e.g., young). For instance, we found that utterances produced by talkers attributed as straight are perceived as sounding more sad compared with utterances produced by talkers attributed as gay, but this does not necessarily mean that talkers attributed as gay are perceived as sounding more happy. While it may be true that the utterances produced by talkers attributed as gay could be perceived as sounding more happy, the current data does not directly support this assertion. Additional investigations are needed to make further determinations for all our attributes; thus, we urge caution in over-interpreting the results. Limitations and future directions There are methodological limitations to the current study that, if corrected, could lead to further insight. First, we do not wish to overgeneralize our findings. For example, our stimuli were utterances consisting of a concatenated string of three monosyllabic words produced by talkers who were perceived to sound gay or straight; the talkers produced these utterances in a careful manner. Thus, the stimuli themselves do not consider the full range of gay and straight speech; we do not wish to presume that all gay men, for instance, would be perceived as stuck-up or all straight men would be perceived as old. However, even considering these constraints, we did discover relatively consistent impressions of the talkers’ identity. Future experiments could further investigate other types of gay and straight speech, either from a wider range of talkers and/or other types of speech (e.g., read paragraphs, spontaneous speech, etc.), and how these additional speech samples could correspond to various attributes. Moreover, while we presume that our listener population was racially and ethnically diverse [ 49 ], the age of our listener population was not diverse. Given the homogeneous age of our relatively young listener sample (i.e., 95% of listeners in Experiment 1 were between 18 and 23 years old), it was not possible to draw any conclusions about how older participants associate attributes with SO. It is possible that older and younger individuals have different interactions with gay and straight individuals (e.g., social interactions and media exposure), and these different encounters could lead to different attributes associated with each group. For example, a younger individual might see gay individuals at college parties, and assume that gay individuals are outgoing, whereas an older individual might only see gay individuals at work and assume that they are not as outgoing. There are also potential generational differences between listeners in their overall perception of the LGBTQ+ community, which have come alongside the general cultural shift towards increased public support and acceptance and the passage of LGBTQ+ affirming legislation in the United States where the study took place [ 64 ]. Alongside this cultural shift, there may be increased internal and external pressure for individuals to be observant of their own behaviors towards the LGBTQ+ community to avoid offense. Thus, particularly in the listening conditions where listeners were informed of the talkers’ SO (either truthfully or falsely), listener’s own perception of the LGBTQ+ community, and their desires to minimize stigma or bias, may have influenced the results. Future work could investigate this further by repeating the experiment with listeners across multiple age groups who complete questionnaires examining their interactions with and perceptions of the LGBTQ+ community. Another consideration for recruiting primarily young adult listeners is the differences in how older and younger listeners perceive emotion from voice. Older individuals (60–84 years), compared to younger individuals (17–29 years), were relatively inaccurate at identifying sadness and anger [ 65 ]. While one possible reason for this is due to age-related changes in certain parts of the brain, other literature has discussed the presence of a “positivity effect” in older listeners where the intensity of negative emotions is reduced due to their shift of motivation and attention towards positive events rather than negative events [ 66 ]. Thus, our participants may have been more attuned to judgments of emotion, particularly negative emotions, simply because they were younger. Additionally, other age-related changes such as hearing loss or changes in cognitive processing could contribute to differences in the perception of many attributes discussed here [ 67 , 68 ]. Future studies could investigate whether older individuals demonstrate the same results as younger individuals. Another methodological limitation lies within the experimental design itself and subsequent statistical design. As noted above, we chose an extreme-groups approach to this experiment (i.e., selecting the talkers that were most consistently perceived as sounding gay and straight), which allowed us to treat perceived SO as a binary variable during statistical analyses. This was done to simplify analyses given the primary focus of the current study was how listeners reacted to perceived vs actual SO given various amounts of information about the talkers’ actual SO. However, the perception of SO is not as simple as a gay-to-straight binary [ 15 , 50 , 69 ]. Thus, these analyses reflect how participants were primed to respond in the experiment itself (i.e., treating perceived SO as a binary), and therefore may have limited participants’ responses. Regrettably, we did not provide participants with an opportunity to describe how their own views and lived experiences may have affected their responses [ 50 ]; future experiments could provide these opportunities. Further, repeating the current experiment with talkers whose perceived SO fell on various points of perceived sexuality spectrum, rather than on the extremes, could lend further insight into the current findings. Since this study was largely exploratory, in terms of identifying attributes associated with gay and straight men, future experiments could include additional specific attributes combined with the attributes listed by listeners [ 11 ]. To fully investigate which attributes are associated with gay and straight male talkers, future investigations could include a greater balance of positively valenced attributes and negatively valenced attributes [ 26 ], as well as more high dominance attributes and low dominance attributes [ 6 ]. The current study included more negative valence attributes (sad, mad, boring, and stuck-up) than positive valence attitudes (confident, intelligent, and outgoing) and only included one high dominance attribute. As a result, only negatively valenced attributes were associated with talkers who sounded straight (sad and old), although an equal number of negatively valenced attributes were associated with talkers who sounded gay (mad and stuck-up). Additionally, the only positively valanced attributes were associated with talkers who sounded gay (confident and outgoing), as was the only high dominance attribute (confident). Thus, we have a limited ability to draw conclusions about how listeners attribute features such as valence and dominance to talkers who sound gay and straight. Using a balanced design in future studies would rectify this shortcoming. Finally, as mentioned previously [ 42 ], there are different attributes associated with different stereotypic groups of gay men; leather and biker gay men, for instance, are associated with tattoos and being “macho” , whereas flamboyant gay men are associated with being loud and dramatic. The current study investigated gay men, in general, and presumed that certain attributes (e.g., stuck-up) would be applied to all gay men. However, future studies could investigate how listeners perceive different groups of gay men. For example, instead of just including a label of gay or straight , which was done in the current study, more specific labels could be used. It may be the case that listeners associate certain attributes with certain types of gay men and not others. A similar experimental manipulation could also be made with straight men. We are, however, unaware of research findings that detailed which attributes are associated with different types of straight men given that most research specifically studied gay men [ 11 , 42 ]. Conclusion The current study demonstrated that perceived SO, for the most part, led to significant differences in listeners’ perceptions of a variety of personal attributes between gay and straight male talkers, regardless of the talkers’ actual SO. Additionally, giving listeners information about talkers’ self-identified SO, whether accurate or inaccurate, significantly interacted with listeners perception of some attributes (i.e., sad and stuck-up), but not others. Together, these findings indicate that a talker’s self-identified SO, whether hidden, disclosed, or falsely disclosed, has limited influence on a listener’s perception of many attributes; rather, for the most part, it is the listener’s perception of SO that has the greatest influence on the attributes examined in the current study. Thus, the results have potential implications for talkers perceived as gay (regardless of self-identified SO) in their day-to-day life, in relationships, and in the workplace. Acknowledgments The authors wish to thank Ciara Allsup, Matthew Bradford, Dalton Davis, Jonathon Godwin, Kalie Hagedorn, Caesar Palacios, Adam Roby, and the late Jason Shuping for their help in running participants. We also wish to thank Keith Johnson, Ashley Allen, Abby Nance, Brooke Merritt, Sarah Hargus Ferguson, Shilpa Regan, Xin (Cynthia) Zhang, Gaia Gould, and Tonyetta Perry for their helpful feedback. Data Availability The data are contained in a public repository. The link is below: https://osf.io/bc2wm/overview . Funding Statement The author(s) received no specific funding for this work. References 1. Rule NO. Perceptions of Sexual Orientation From Minimal Cues. Arch Sex Behav. 2017;46(1):129–39. doi: 10.1007/s10508-016-0779-2 [ DOI ] [ PubMed ] [ Google Scholar ] 2. Fasoli F, Maass A, Paladino MP, Sulpizio S. Gay- and Lesbian-Sounding Auditory Cues Elicit Stereotyping and Discrimination. Arch Sex Behav. 2017;46(5):1261–77. doi: 10.1007/s10508-017-0962-0 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Gaudio RP. Sounding Gay: Pitch Properties in the Speech of Gay and Straight Men. American Speech. 1994;69(1):30. doi: 10.2307/455948 [ DOI ] [ Google Scholar ] 4. Ptacek PH, Sander EK. Age recognition from voice. J Speech Hear Res. 1966;9(2):273–7. doi: 10.1044/jshr.0902.273 [ DOI ] [ PubMed ] [ Google Scholar ] 5. Bachorowski J, Owren MJ. Vocal expression of emotion: Acoustic properties of speech are associated with emotional intensity and context. Psychol Sci. 1995;6:219–24. [ Google Scholar ] 6. McAleer P, Todorov A, Belin P. How do you say “hello”? Personality impressions from brief novel voices. PLoS One. 2014;9(3):e90779. doi: 10.1371/journal.pone.0090779 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Parent G, Vallely J, Passaris L. The Golden Girls. Los Angeles: Touchstone Pictures; 1992. [ Google Scholar ] 8. Fasoli F, Dragojevic M, Rakić T, Johnson S. Voice matters: Social categorization and stereotyping of speakers based on sexual orientation and nationality categories. Lang Commun. 2023;90:114–28. [ Google Scholar ] 9. Painter D, Fasoli F, Sulpizio S. The Impact of Stimuli Length and Analytic Method on Auditory “Gaydar” Research. J Voice. 2024;38(1):246.e1-246.e14. doi: 10.1016/j.jvoice.2021.08.016 [ DOI ] [ PubMed ] [ Google Scholar ] 10. Tracy EC, Bainter SA, Satariano NP. Judgments of self-identified gay and heterosexual male speakers: Which phonemes are most salient in determining sexual orientation? J Phon. 2015;52:13–25. doi: 10.1016/j.wocn.2015.04.001 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Madon S. What do people believe about gay males? A study of stereotype content and strength. Sex Roles. 1997;37(9–10):663–85. doi: 10.1007/bf02936334 [ DOI ] [ Google Scholar ] 12. Baus C, McAleer P, Marcoux K, Belin P, Costa A. Forming social impressions from voices in native and foreign languages. Sci Rep. 2019;9(1):414. doi: 10.1038/s41598-018-36518-6 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Sulpizio S, Fasoli F, Antonio R, Eyssel F, Paladino MP, Diehl C. Auditory Gaydar: Perception of Sexual Orientation Based on Female Voice. Lang Speech. 2020;63(1):184–206. doi: 10.1177/0023830919828201 [ DOI ] [ PubMed ] [ Google Scholar ] 14. Zimman L. Hegemonic masculinity and the variability of gay-sounding speech: The perceived sexuality of transgender men. J Lang Sex. 2013;2:1–39. [ Google Scholar ] 15. Coombs Fine J. Performing graysexuality: A segmental and prosodic analysis of three voices employed in the construction of the graysexual self. J Lang Sex. 2019;8:1–29. [ Google Scholar ] 16. Campbell-Kibler K. Intersecting Variables and Perceived Sexual Orientation in Men. American Speech. 2011;86(1):52–68. doi: 10.1215/00031283-1277510 [ DOI ] [ Google Scholar ] 17. Fasoli F, Hegarty P. A Leader Doesn’t Sound Lesbian!: The Impact of Sexual Orientation Vocal Cues on Heterosexual Persons’ First Impression and Hiring Decision. Psychology of Women Quarterly. 2019;44(2):234–55. doi: 10.1177/0361684319891168 [ DOI ] [ Google Scholar ] 18. Levon E. Hearing “gay”: prosody, interpretation, and the affective judgments of men’s speech. American Speech. 2006;81(1):56–78. doi: 10.1215/00031283-2006-003 [ DOI ] [ Google Scholar ] 19. Baeck H, Corthals P, Van Borsel J. Pitch characteristics of homosexual males. J Voice. 2011;25(5):e211-4. doi: 10.1016/j.jvoice.2010.10.019 [ DOI ] [ PubMed ] [ Google Scholar ] 20. Levon E. Sexuality in context: Variation and the sociolinguistic perception of identity. Lang Soc. 2007;36:533–54. [ Google Scholar ] 21. Suire A, Tognetti A, Durand V, Raymond M, Barkat-Defradas M. Speech Acoustic Features: A Comparison of Gay Men, Heterosexual Men, and Heterosexual Women. Arch Sex Behav. 2020;49(7):2575–83. doi: 10.1007/s10508-020-01665-3 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Crist S. Duration of onset consonants in gay male stereotyped speech. U Penn Working Papers in Linguistics. 1997;4:53–69. [ Google Scholar ] 23. Scherer KR, Banse R, Wallbott HG. Emotion Inferences from Vocal Expression Correlate Across Languages and Cultures. Journal of Cross-Cultural Psychology. 2001;32(1):76–92. doi: 10.1177/0022022101032001009 [ DOI ] [ Google Scholar ] 24. Scherer K. Vocal communication of emotion: A review of research paradigms. Speech Communication. 2003;40(1–2):227–56. doi: 10.1016/s0167-6393(02)00084-5 [ DOI ] [ Google Scholar ] 25. Grichkovtsova I, Morel M, Lacheret A. The role of voice quality and prosodic contour in affective speech perception. Speech Communication. 2012;54(3):414–29. doi: 10.1016/j.specom.2011.10.005 [ DOI ] [ Google Scholar ] 26. Banse R, Scherer KR. Acoustic profiles in vocal emotion expression. J Pers Soc Psychol. 1996;70(3):614–36. doi: 10.1037//0022-3514.70.3.614 [ DOI ] [ PubMed ] [ Google Scholar ] 27. Bachorowski J. Vocal expression and perception of emotion. Curr Dir Psychol Sci. 1999;8:53–7. [ Google Scholar ] 28. Scherer KR, Banse R, Wallbott HG, Goldbeck T. Vocal cues in emotion encoding and decoding. Motiv Emot. 1991;15:123–48. [ Google Scholar ] 29. Titze IR. Some Consensus has been Reached on the Labeling of Harmonics, Formants, and Resonances. J Voice. 2016;30(2):129. doi: 10.1016/j.jvoice.2015.09.014 [ DOI ] [ PubMed ] [ Google Scholar ] 30. Abelin Å. Anger or fear? Cross-cultural multimodal interpretations of emotional expression. In: Izdebski K, editor. Emotions in the Human Voice Volume I: Foundations. San Diego: Plural Publishing, Inc.; 2008. p. 65–73. [ Google Scholar ] 31. González GM, Ramos AL. Assessing voice characteristics of depression among English- and Spanish-speaking populations. In: Izdebski K, editor. Emotions in the Human Voice Volume I: Foundations. San Diego: Plural Publishing, Inc.; 2008. p. 49–65. [ Google Scholar ] 32. Hargreaves WA, Starkweather JA. Voice Quality Changes in Depression. Lang Speech. 1964;7(2):84–8. doi: 10.1177/002383096400700203 [ DOI ] [ Google Scholar ] 33. Jiang X, Pell MD. The sound of confidence and doubt. Speech Communication. 2017;88:106–26. doi: 10.1016/j.specom.2017.01.011 [ DOI ] [ Google Scholar ] 34. Roche JM, Morgan SD, Fisk S. Gender stereotypes drive perceptual differences of vocal confidence. J Acoust Soc Am. 2022;151(5):3031. doi: 10.1121/10.0010382 [ DOI ] [ PubMed ] [ Google Scholar ] 35. Ryan WJ, Burk KW. Perceptual and acoustic correlates of aging in the speech of males. J Commun Disord. 1974;7(2):181–92. doi: 10.1016/0021-9924(74)90030-6 [ DOI ] [ PubMed ] [ Google Scholar ] 36. Kite ME, Deaux K. Gender Belief Systems: Homosexuality and the Implicit Inversion Theory. Psychology of Women Quarterly. 1987;11(1):83–96. doi: 10.1111/j.1471-6402.1987.tb00776.x [ DOI ] [ Google Scholar ] 37. Blashill AJ, Powlishta KK. Gay Stereotypes: The Use of Sexual Orientation as a Cue for Gender-Related Attributes. Sex Roles. 2009;61(11–12):783–93. doi: 10.1007/s11199-009-9684-7 [ DOI ] [ Google Scholar ] 38. Steffans MC, Niedlich C, Kachel S, Methner N. Impression formation of applicants differing in sexual orientation: An attempt to integrate theoretical models and a review of the empirical evidence. In: Earley F, editor. Sexual Orientation: Perceptions, Discrimination, and Acceptance. New York, NY: Nova Science Publishers, Inc.; 2016. p. 51–79. [ Google Scholar ] 39. Tskhay KO, Rule NO. Emotions facilitate the communication of ambiguous group memberships. Emotion. 2015;15(6):812–26. doi: 10.1037/emo0000077 [ DOI ] [ PubMed ] [ Google Scholar ] 40. Carnaghi A, Rusconi P, Bianchi M, Fasoli F, Coladonato R, Hegarty P. No country for old gay men: Age and sexuality category intersection renders older gay men invisible. Group Processes & Intergroup Relations. 2021;25(4):964–89. doi: 10.1177/1368430220987606 [ DOI ] [ Google Scholar ] 41. Fiske ST, Cuddy AJC, Glick P, Xu J. A model of (often mixed) stereotype content: competence and warmth respectively follow from perceived status and competition. J Pers Soc Psychol. 2002;82(6):878–902. doi: 10.1037/0022-3514.82.6.878 [ DOI ] [ PubMed ] [ Google Scholar ] 42. Clausell E, Fiske ST. When do subgroup parts add up to the stereotypic whole? Mixed stereotype content for gay male subgroups explains overall ratings. Soc Cogn. 2005;23:161–81. [ Google Scholar ] 43. Rule NO, Bjornsdottir RT, Tskhay KO, Ambady N. Subtle perceptions of male sexual orientation influence occupational opportunities. J Appl Psychol. 2016;101(12):1687–704. doi: 10.1037/apl0000148 [ DOI ] [ PubMed ] [ Google Scholar ] 44. Gowen CW, Britt TW. The interactive effects of homosexual speech and sexual orientation on the stigmatization of men: Evidence for expectancy violation theory. J Lang Soc Psychol. 2006;25:437–56. [ Google Scholar ] 45. Taylor DM, Raadt JS. Gay- and Straight-Sounding Auditory Cues Elicit Stereotyping About Teaching Effectiveness. Journal of Research in Music Education. 2020;69(1):62–84. doi: 10.1177/0022429420948229 [ DOI ] [ Google Scholar ] 46. Drager K. From bad to bed: The relationship between perceived age and vowel perception in New Zealand English. Te Reo. 2006;48:55–68. [ Google Scholar ] 47. Niedzielski N. The effect of social information on the perception of linguistic variables. J Lang Soc Psychol. 1999;18:62–85. [ Google Scholar ] 48. Johnson KL, Lick DJ, Carpinella CM. Emergent Research in Social Vision: An Integrated Approach to the Determinants and Consequences of Social Categorization. Social & Personality Psych. 2015;9(1):15–30. doi: 10.1111/spc3.12147 [ DOI ] [ Google Scholar ] 49. University of North Carolina at Pembroke. UNC Pembroke Fast Facts [Internet]. Pembroke, NC: Public Tableau; 2025. [cited 2025 Nov 13]. Available from: https://public.tableau.com/app/profile/uncp.ir/viz/FactBook_v3/FastFact [ Google Scholar ] 50. Tripp A, Munson B. Acknowledging language variation and its power: Keys to justice and equity in applied psycholinguistics. Appl Psycholinguist. 2023;44:495–513. [ Google Scholar ] 51. Munson B, McDonald EC, DeBoe NL, White AR. The acoustic and perceptual basis of judgments of women and men’s sexual orientation from read speech. J Phon. 2006;34:202–40. [ Google Scholar ] 52. Mack S, Munson B. The influence of/s/ quality on ratings of men’s sexual orientation: Explicit and implicit measures of the ‘gay lisp’ stereotype. J Phon. 2012;40:198–212. [ Google Scholar ] 53. Linville SE. Acoustic correlates of perceived versus actual sexual orientation in men’s speech. Folia Phoniatr Logop. 1998;50(1):35–48. doi: 10.1159/000021447 [ DOI ] [ PubMed ] [ Google Scholar ] 54. Laukkanen A, Vilkman E, Alku P, Oksanen H. On the perception of emotions in speech: The role of voice quality. Logopedics Phoniatrics Vocology. 1997;22:157–68. [ Google Scholar ] 55. Brooks ME, Kristensen K, van Benthem KJ, Magnusson A, Berg CW, Nielsen A. glmmTMB balances speed and flexibility among packages for zero-inflated generalized linear mixed modeling. R J. 2017;9:378–400. [ Google Scholar ] 56. Lenth R. emmeans: Estimated marginal means, aka least-squares means [Internet]. 2025. Available from: https://rvlenth.github.io/emmeans/ [ Google Scholar ] 57. Šidák Z. Rectangular Confidence Regions for the Means of Multivariate Normal Distributions. Journal of the American Statistical Association. 1967;62(318):626–33. doi: 10.1080/01621459.1967.10482935 [ DOI ] [ Google Scholar ] 58. Konik J, Cortina LM. Policing Gender at Work: Intersections of Harassment Based on Sex and Sexuality. Soc Just Res. 2008;21(3):313–37. doi: 10.1007/s11211-008-0074-z [ DOI ] [ Google Scholar ] 59. Rabelo VC, Cortina LM. Two sides of the same coin: gender harassment and heterosexist harassment in LGBQ work lives. Law Hum Behav. 2014;38(4):378–91. doi: 10.1037/lhb0000087 [ DOI ] [ PubMed ] [ Google Scholar ] 60. Bettinsoli ML, Suppes A, Napier J. Predictors of attitudes toward gay men and lesbian women in 23 countries. Soc Psychol Personal Sci. 2020;11:697–708. [ Google Scholar ] 61. Henry PJ, Steiger RL. Sexual Orientation as Gendered to the Everyday Perceiver. Sex Roles. 2022;87(3–4):117–38. doi: 10.1007/s11199-022-01313-1 [ DOI ] [ Google Scholar ] 62. Welch B, van Mersbergen MR, Helou LB. Listeners’ Perceptions of Speaker Personality Traits Based on Speech. J Speech Lang Hear Res. 2021;64(12):4762–71. doi: 10.1044/2021_JSLHR-20-00582 [ DOI ] [ PubMed ] [ Google Scholar ] 63. Hawkins DI, Albaum G, Best R. Stapel scale or semantic differential in marketing research. J Mark Res. 1974:318–22. [ Google Scholar ] 64. Witeck B. Cultural change in acceptance of LGBT people: lessons from social marketing. Am J Orthopsychiatry. 2014;84(1):19–22. doi: 10.1037/h0098945 [ DOI ] [ PubMed ] [ Google Scholar ] 65. Ryan M, Murray J, Ruffman T. Aging and the perception of emotion: processing vocal expressions alone and with faces. Exp Aging Res. 2010;36(1):1–22. doi: 10.1080/03610730903418372 [ DOI ] [ PubMed ] [ Google Scholar ] 66. Carstensen LL, Mikels JA. At the intersection of emotion and cognition: aging and the positivity effect. Curr Dir Psychol Sci. 2005;14:117–21. [ Google Scholar ] 67. Lin FR, Niparko JK, Ferrucci L. Hearing loss prevalence in the United States. Arch Intern Med. 2011;171(20):1851–2. doi: 10.1001/archinternmed.2011.506 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 68. Murman DL. The Impact of Age on Cognition. Semin Hear. 2015;36(3):111–21. doi: 10.1055/s-0035-1555115 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Lugones M. Toward a decolonial feminism. Hypatia. 2010;25(4):742–59. [ Google Scholar ] PLoS One. doi: 10.1371/journal.pone.0346897.r001 Decision Letter 0 Laura Morett Laura Morett Academic Editor Find articles by Laura Morett Author information Copyright and License information Roles Laura Morett : Academic Editor © 2026 Laura MorettLaura MorettLaura MorettLaura Morett This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 24 Oct 2025 PONE-D-25-51754-->-->Judgments of American English male talkers who are perceived to sound gay or straight:  Which personal attributes are associated with each group of talkers?-->-->PLOS ONE?> Dear Dr. Tracy, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Although the topic of the study reported in this manuscript is novel and intriguing, both reviewers raise substantial concerns needing to be addressed for this manuscript to be further considered for publication. Specifically, both R1 and R2 raise concerns about the characterization of the demographics of the participants, which may have affected their judgments in the experimental task. Moreover, R1 highlights the need for more coherent framing in the Introduction and a power analysis to justify the sample size and offers several suggestions concerning presentation of results and the focus and framing of the Discussion, and R2 points out the need for clear and consistent use of terminology. I encourage the authors to prepare a revision responsive to the reviewers' feedback, and if the authors do so, I will attempt to send the revised manuscript to the same reviewers for feedback on the extent to which their concerns are addressed. Please submit your revised manuscript by Dec 06 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .... We look forward to receiving your revised manuscript. Kind regards, Laura Morett Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information . 3. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? Reviewer #1: Partly Reviewer #2: Yes ********** 2. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: Yes ********** 3. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: No Reviewer #2: Yes ********** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ********** Reviewer #1: The manuscript titled ‘Judgments of American English male talkers who are perceived to sound gay or straight: Which personal attributes are associated with each group of talkers?’ presents a study testing how speakers’ self-identified sexual orientation, perception of sexual orientation, and ‘disclosed’ sexual orientation influence attribution of personality traits, emotions, and social categorisation (age). I believe the research is interesting, especially the context/disclosure part that, to my knowledge, has not been tested in this way before and can expand the literature on first impressions based on actual, perceived, or disclosed sexual orientation. The topic is also of interest to scholars in different fields (e.g., linguistics, sociolinguistics, psychology, communication) due to its interdisciplinary nature. Since this is the case, the manuscript should be revised to better account for this multidisciplinarity. Below I report my concerns and suggestions, which I hope the authors will find useful for revising the manuscript and for future research. Introduction The introduction needs to provide a better contextualisation of this work within the current multidisciplinary literature. I appreciate that each field has looked at this topic from a different perspective, but a clearer explanation would add value to the work and show how it expands the field. When referring to the method, a clearer rationale is needed for the decision to focus on attributes. At the moment, the study is presented as an expansion of a pilot study by the authors, but this limits the contribution of the work. It is important to show how this study is embedded in the existing literature and what its contribution is, beyond expanding a single previous study. The authors describe how the traits are relevant, but they discuss them separately and some get more attention than others in the introduction. The study focuses on a range of different attributes (i.e., sad, boring, confident, outgoing, mad, old, intelligent, and stuck-up; and subsequently anger is considered). The choice of these traits is based on a previous study (called the “pilot study”), but what is missing is an explanation of why these, and not other traits, were tested. These should be better contextualised in the multidisciplinary existing literature. The rationale for this selection is not fully explained. For instance, intelligent and confident belong to the same dimension of agency/competence and this has been tested before. So, the decision to consider them separately rather than on dimensions is needed. Moreover, one might expect the study to include traits stereotypically associated with gay and straight men, as well as neutral traits, to test whether these are differently attributed depending on actual, perceived, or disclosed sexual orientation (see McAleer et al., 2014, for how social traits were selected based on previous research; see also Madon or Blashill for work on trait attribution based on explicit sexual orientation information). Similarly, it is unclear why the attributes include these personality traits, but only a few negative emotions (sad, bored) and only one social category (age). If the goal was to test emotion and social category attribution, multiple emotions and social categories should have been included. Otherwise, the results remain quite limited, and it is unclear why sadness and age were considered more relevant than others. Finally, the authors have not considered the valence of the attributes, which can influence how they are attributed to social groups. In relation to multidisciplinarity, attention to definitions should be improved for clarity (for instance, confidence in psychology is a trait, not an emotion). It is fine to use definitions from different fields, but this should be acknowledged somewhere in the text. I believe the authors (who may disagree) could better introduce the study as impression formation, justify why they focused on certain attributes and not others, and clarify how these attributions are relevant for social consequences. While they do refer to social consequences, the narrative could be stronger. For instance, they could mention that intelligence and confidence may matter in educational contexts (see Taylor & Raadt on stereotyping and teaching effectiveness). As I mentioned before, I liked the idea of testing context (or disclosure of identity). I do not believe this has been directly studied before, but there is some indirect evidence that could help build the rationale. For instance, Sulpizio et al. (2015) told participants they would hear half self-identified gay and half self-identified straight speakers, but this information did not change how speakers were categorised by voice. This suggests that first impressions may be more strongly influenced by perceived sexual orientation than by explicit information. However, Gowen and Britt (2006) found that the interplay between perceived (voice-based) and disclosed sexual orientation affected impressions and decisions, with speakers introduced as straight but sounding gay being rated more negatively. This suggests that the interaction between perceived and disclosed sexual orientation could also influence first impressions. These studies could help the authors build a stronger narrative for studying these variables together. The section on intersectionality, when referring to age, needs further elaboration. For instance, it is not fully explained why Carnaghi et al. found that straight men were perceived as older than gay men in their work on the intersection between age and sexual orientation. This is important, as it relates to prototypes and stereotypes that are closely linked to trait attributions (see also work on voice-based intersections of sexual orientation and nationality for a similar rationale). I would also encourage the authors to consider Sven Kachel’s work on voice-based sexual orientation categorisation and attribution of masculinity/femininity, since he considered both speaker and listener perspectives, or mention of the “straight categorisation bias,” (Lick & Johnson, 2016; Fasoli et al., 2023) which is key in distinguishing between self-identified and perceived sexual orientation. These studies could be integrated to strengthen the rationale and justify the study design and variables. The pilot study informing the current study should be explained in more detail. This would make it clearer whether and how the results are replicated and expanded. The study should also be explicitly presented as exploratory, as no hypotheses are put forward. However, the literature on impression formation based on sexual orientation (as well as general impression formation based on voice and personality; see McAleer et al., 2014) does provide enough background to make predictions, in case the authors want to suggest some possible directions while keeping the study as exploratory. Method - The manuscript does not indicate whether participants’ sexual orientation was recorded. This is very important, as categorisation and perception can vary among listeners of different sexual orientations. Participants’ age and gender are also not reported. These are key demographics given the scope of the study. For instance, age should be similar across gay and straight speakers. - Please report the percentage of correct sexual orientation categorisation for the speakers, so it is clear how perceived sexual orientation was operationalised. It would also have been good to assess perceived sexual orientation in this study to ensure consistency between participants’ categorisation and operationalisation of the variable. - Please add a power analysis to justify the sample size and effect size for the analyses performed. - The lack of control for regional accents is problematic, as non-standard accents trigger different first impressions (see language attitude literature) and interact with perceived sexual orientation. - The valence of the traits should be considered to test whether it influences the attributions. - From a psychological perspective, mad and angry are different, one is a trait and the other an emotion, this brings us back to the definitions and the different nature of the attributes considered here. - It would be useful to clarify whether reaction times were recorded, as participants were asked to respond quickly. If so, it would be interesting to test differences in response speed depending on actual, perceived, or disclosed sexual orientation. - I wonder whether it would be clearer to present the pairwise comparisons by keeping actual sexual orientation constant and varying the context, for instance, if a gay speaker is perceived as more stuck-up when truthfully described as gay, falsely described as straight, or when no information is provided. I think this would make it easier for the reader, though I appreciate the authors’ reporting of all comparisons. - The section on anger attribution is unclear, as no tests of main effects or interactions are reported for anger. Without this, the results cannot be properly compared descriptively. It would also have been better to test mad and angry in the same sample for direct comparison. - I have not found any link to a repository where the data are available. My understanding is that this is a requirement for PLOS One. If I have missed it, please ignore my comment. Discussion - The conclusion should elaborate more on the significant interaction between condition/context and perceived sexual orientation. At times, the text suggests that perceived sexual orientation is the only factor that matters, but for some traits, the interaction was significant. - I would like to see more integration of how these results contribute to the multidisciplinary field of voice-based sexual orientation and impression formation. - While I appreciated the reference to acoustic correlates in the introduction, this may be less relevant in the discussion, as no data on speakers’ acoustic features are reported. - The explanation of the results concerning sadness needs further elaboration. - There is a mention of studies on lesbian-sounding speakers; however, the current study focuses only on male speakers, so this seems less relevant here. The authors should elaborate on this as a limitation and on the importance of testing female speakers in future work. - Some of the suggestions could be better supported with literature. For instance, the point about young people expecting gay men to be more outgoing based on personal experiences could be linked to the concept of familiarity tested in relation to gaydar (see Brambilla et al., 2011). Minor points: - Page 3, line 57: I suggest changing the sentence to clarify this is a belief: “that an individual can BELIEVE that they can perceive…” - Across the manuscript, references should be revised, as many are not appropriate. For instance, Painter et al. (page 3, line 60) is cited when discussing attribution of different traits, but that paper focuses on stimulus length and categorisation, and only on masculinity/femininity (not tested here) not on other trait attribution. Other studies on stereotyping and impression formation would be more suitable. Similarly, on page 10, reference 15 (Mack & Munson) may not best support the statement. There are several instances of misreferencing throughout. Reviewer #2: The manuscript is sound, legible and has data availability. Just be sure to have consistent formatting throughout the manuscript i.e. having the condition label Referenced capitalized, "SO" vs. sexual orientation, and etc. ********** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our Privacy Policy ..--> Reviewer #1: No Reviewer #2: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation . NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications. Attachment Submitted filename: Review_PONE-D-25-51754.pdf pone.0346897.s001.pdf (74.7KB, pdf) PLoS One. 2026 Apr 10;21(4):e0346897. doi: 10.1371/journal.pone.0346897.r002 Author response to Decision Letter 1 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 9 Dec 2025 We would like to thank the editor and the reviewers for their helpful suggestions and comments regarding our manuscript. The manuscript, we believe, improved with these insightful revisions. We have included most, if not all, of the indicated revisions, and the new revisions are in red font in the file “Revised Manuscript with Track Changes.” Below, we respond to each of the suggestions made by the editor and reviewers. We have also provided line numbers, so that the editor and reviewers can find these suggestions within the revised manuscript. We attempted to be as accurate as possible with the line numbers, so please excuse any mistakes. Response to the Editor 1. We have uploaded seven files for further review: the rebuttal letter, the manuscript with track changes, the manuscript with no track changes, and the four figures. Please note that we believe that the figures should stand on their own; therefore, the figures use the term ‘sexual orientation’ instead of the SO acronym that was used in the manuscript. If the editor and/or reviewers would like us to change this, so that the figures contain the SO acronym, we are happy to oblige. 2. We have formatted the manuscript and figures in accordance with the information contained within the email that was sent to us on October 24. Response to Reviewer #1 1. The Introduction needs to better contextualize the work within the multidisciplinary field. The Introduction was extensively revised to meet this requirement. For instance, while we kept the discussion on perceiving sexual orientation (SO) [Lines 103-112] and attributes [Lines 113-160], we revised the section on SO and attributes [Lines 161-210]. This revised section includes a discussion on both stereotypes associated with gay and straight men [Lines 162 – 176] and discrimination in the workplace [Lines 193-210]. Additionally, we added a new section on the implications of knowing an individual’s SO [Lines 211-248]. This section includes a discussion on studies that did not reveal participants’ SO [Lines 212-221] and the emerging fields of social vision and social hearing [Lines 230-248]. 2. Why were these attributes considered and not other attributes? We further discussed the pilot study, which was used to identify the attributes in this manuscript. Briefly, we used those attributes that were identified most often by the listeners. The pilot study is discussed throughout the manuscript, and we hope that this discussion satisfies the reviewer’s question [Lines 91-92; 121-124; 323-342]. 3. Why consider intelligence and confidence separately and not on a dimension? These two attributes were both indicated by the listeners in the pilot study; the listeners differentiated between these two attributes. Here, we took a bottom-up approach and included those attributes most cited by participants in the pilot study. More information on the pilot study can be found throughout the manuscript [Lines 91-92; Lines 121-124; Lines 323-342]. 4. Why not use attributes more stereotypically associated with gay and straight men? We used the attributes that listeners indicated in the pilot study, and more information on the pilot study can be found throughout the manuscript [Lines 91-92; Lines 121-124; Lines 323-342]. We do agree with the reviewer that there is a necessity to further study these stereotypical attributes. To that end, we explained, in the Discussion, that future research studies should incorporate attributes stereotypically associated with gay and straight men [Lines 753-764]. 5. Why not use neutral attributes? As discussed previously, we used a bottom-up approach for this study by first asking listeners to identify those attributes most associated with various talkers. Unfortunately, neutral attributes were not indicated, at a high rate, by listeners; thus, we did not include them. However, in the Discussion, we do discuss further attributes that researchers could use in future studies. More information on the pilot study can be found throughout the manuscript [Lines 91-92; Lines 121-124; Lines 323-342], and more information on further attributes can be found throughout the Discussion [Lines 745-764]. 6. The reviewer asked to incorporate the work of Blashill, Madon, and McAleer in the manuscript. We found value in the work from these authors and included them within the manuscript. Blashill’s work is referenced in Line 166. Madon’s work is referenced in Lines 91, 170, 331, 747, and 764. McAleer’s work is referenced in Lines 68, 93, 129, 132, 139, and 750. 7. Why not include more social and emotion categories? We used a bottom-up approach for this study, and we only included those attributes indicated most often by the listeners in the pilot study [Lines 91-92; Lines 121-124; Lines 323-342]. Furthermore, we did note that future studies should include more attributes, such as low valence attributes and high valence attributes [Lines 745-764]. 8. Why are sadness and age more relevant than others? Here, we were unclear what ‘others’ meant. For example, we were unclear if ‘others’ meant the other attributes investigated in this study or the other attributes (e.g., social and emotion) that the reviewer suggested? It was our intention to discuss each of the eight attributes equally. However, the rating results of the ‘sad’ attribute required more explanation, and we added this discussion in Lines 616-633. 9. The researchers haven’t considered the valence of the attributes. Thank-you for the suggestion, and we did include a discussion on the valence of the attributes in both the Introduction [Lines 132-139; 745-752]. Additionally, we included a short discussion on another dimension: dominance [Lines 136-137; 749]. 10. Terminology should be improved. Thank-you, again, for the suggestion. We have strived to improve our terminology usage. For instance, we defined the terms attributes, attitudes, and emotions in the Introduction [Lines 114-120]. 11. Possibly introduce the study as impression formation. We agree with the reviewer and incorporated this term throughout the manuscript. For instance, it can first be found in Lines 61-69. 12. Justify why certain terms were used and why certain terms were not used. We hoped to satisfy the reviewer’s suggestion with a greater description of the pilot study, as indicated above. Please refer to #2-5 and #7 above for a more detailed explanation. 13. Clarify why attributes are relevant to social consequences. There is a relatively large section within the Introduction that details the social consequences of these attributes [Lines 177-210]. Here, we discussed stereotypes associated with gay and straight men, and the resulting consequences in the workplace and classroom. Additionally, there is a revised section in the Discussion that also discusses these social consequences [Lines 634-663]. 14. Intelligence and confidence may matter in educational situations. We have included Taylor and Raadt’s (2021) work on how stereotypical gay and straight male voices are perceived in an education setting [Lines 202-203; 214-216; 637-638]. 15. Strengthen the rationale for testing the disclosure of the talkers’ SO. Thank-you for this suggestion. We have included an additional section within the manuscript, ‘Knowledge of an individual’s self-identified SO’, to hopefully satisfy the reviewer’s concern [Lines 211-248]. Within this section, we discuss various auditory gaydar studies that did not disclose the talkers’ SO. Next, we discuss two studies in which knowledge of the talkers’ attributes changes how listeners perceive the talkers. Finally, we discuss both social vision and social hearing, and how these two fields provide motivation for our study. 16. The authors should consider the work on Gowen and Britt (2006) and Sulpizio et al. (2015) in terms of how the disclosure of a talker’s SO could influence the listener’s perception of them. While we did include the work on Gowen and Britt within the paper [Lines 195-196; 647-648], we did not include the work of Sulpizio et al. (2015). However, we did include other references that we hope satisfies the reviewer’s concern. As mentioned above, we included a new section on how knowledge of an individual’s SO can affect a listener’s perception [Lines 211-248] and included within this section, we have included numerous references that hopefully strengthen our argument (e.g., Taylor & Raadt, 2021; Johnson et al., 2015). 17. The section on intersectionality needs more clarification. Why did Carnaghi et al. (2022) find that straight men were perceived as older? We included more information about this study. For example, we included the statement that gay men were processed, by default, as young and that elderly men were processed, by default, as heterosexual [Lines 171-174]. 18. Consider Sven Kachel’s work, as well as Lick and Johnson (2016) and Fasoli et al. (2023). We did include one Kachel citation within the manuscript (Steffans, Niedlich, Kachel, & Methner, 2016) [Lines 166-169; 196-201; 660-662]. Here, we discussed Kachel’s work in terms of how a stereotype might affect one’s initial impression of another individual, and how gay and straight men might experience discrimination in the workplace. Furthermore, we included one work of Lick and Johnson within the manuscript (Johnson, Lick, & Carpinella, 2015) [Lines 232]; here we discussed their theory of social vision. Finally, we included the work of Fasoli et al. (2023) [Lines 78, 188, 198, and 216]; here we discussed how listeners place talkers in a social category and then associate various attributes with them. We also discussed how gay and straight men were perceived as equally competent and how LGBTQ+ individuals face discrimination within the workplace. 19. The pilot study needs to be explained in more detail. We hoped to satisfy the reviewer’s comment by further discussing the pilot study. As outlined above, there is a new section within the Materials and Methods section that details the pilot study [Lines 325-344]. 20. Present the study as exploratory since we are not putting forth direct hypotheses. We agree with the reviewer in that we did not specifically predict which attributes would be associated with gay and straight talkers. Therefore, we presented the study as exploratory, and this is referenced throughout the manuscript [Lines 86, 100, 252, 574, 747]. 21. Based on McAleer et al. (2014), we can offer possible directions while still keeping the study exploratory. We included two works by McAleer: McAleer et al. (2014) [Lines 68, 132, 139, 752] and Baus et al. (2019) [Lines 93, 132,]. We discussed how listeners perceived personal attributes from talkers and the different dimensions of the social voice space. 22. Did we collect information on listener’s SO, age, and gender? Yes, we did collect information about age and gender, but regrettably, did not collect information on SO [Lines 279-293]. We further discussed the age of our listeners both in the Methods and Materials section [Lines 279-293] and in the General Discussion [723-732], where we discussed some of the limitations of our sample. Within the manuscript, we did not report the listeners’ SO and, if memory serves correctly, there were few LGBTQ+ individuals within the listener sample. We also did not further discuss the listeners’ gender because, within the context of the manuscript, it seemed relatively less relevant, whereas age seemed more relevant and worthy of additional discussion. 23. Report the percentage of correct SO decisions, so that we know how SO was operationalized. We discussed how the talkers were selected for the experiment [Lines 306-317]. We discussed the results of Tracy et al. (2015) in which each talker was rated by how confident listeners were of the talker’s SO. Next, we selected those talkers who fell at the extremes for inclusion in the present experiment. By extremes, we mean those self-identified gay talkers whom listeners labeled gay most often, those self-identified gay talkers whom listeners labeled straight most often, those self-identified straight talkers whom listeners labeled gay most often, and those self-identified straight talkers whom listeners labeled straight most often. 24. Please add a power analysis to justify the sample size and effect size for the analyses performed. Please elaborate on the effect sizes. A sample size calculation for a poisson regression was completed using G*Power 3.1. Based on a predicted medium effect size, a beta of 0.8, and alpha of 0.05, the sample size of listeners needed for this experiment was calculated to be 164. However, the use of a nested data structure (i.e., stimuli nested within talkers nested within listeners) increases the standard error of a dataset. We estimated the interclass correlation between talkers and listeners to be approximately 0.1, which introduces approximately an 8-10% margin of error for the proposed sample of 24 talkers (Gelman & Hill, 2007); thus an additional ~15 participants were needed to account for this error. Our target sample size was thus ~180. The recruitment method for this study was to offer course credit to undergraduate classes with a size ranging anywhere from 20-40 students. Based on the power analysis, we determined to offer this experiment to students in 9 classes (9 x 20 = 180) to ensure adequate power. Given that all students in each class were given the option to participate in the experiment, we expected this sampling method to result in our minimum recruitment target (i.e., 180), but anticipated that it may give us participants beyond this number. The rate ratios (RR) provided throughout the manuscript were provided as measures of effect size as demonstrate the magnitude of the effect in a way that is directly comparable between effects and studies. Please let us know if you feel another effect size would be more appropriate. Citation: Gelman, A. & Hill, J. (2007). Data Analysis Using Regression and Multilevel/Hierarchical Models. Cambridge University Press. 25. Lack of control for regional accents is problematic as non-standard accents can affect first impressions. We addressed the reviewer’s concern within the Materials and Methods section [Lines 295-305]. Talkers were required to be from the state of Ohio, but we did not control for regional accents within the state. However, the first author, who is a trained phonetician, did not notice any differences in regional accents of the talkers and, presumably, the listeners would not notice this either. Moreover, it was difficult to recruit gay talkers for the original study (e.g., Tracy et al., 2015), and those authors did not have the luxury of only sampling gay talkers who grew up in the Columbus, Ohio metropolitan area; therefore, the authors needed to expand their geographic region to the entire state of Ohio to recruit a large enough sample of gay talkers. We hope that these explanations satisfy the reviewer’s concerns. 26. Valence of the attributes should be considered as to whether it influences the attributions. Within the Introduction, we included a more detailed section on the dimensions of attributes [Lines 132-139] and discussed where our chosen attributes fall within these dimensions. We admit that our attributes were not balanced along these dimensions. For example, we did not include equal numbers of high valence attributes and low valence attributes. We also did not include equal numbers of high dominance attributes and low dominance attributes. This is a shortcoming in the current exploratory study, and we discussed including additional attributes in future experiments within the General Discussion [Lines 747-754]. 27. Mad and angry are different. Bring us back to definitions and different natures of attributes. We took care to use consistent and clear definitions of both emotions and attitudes [Lines 114-120]. Furthermore, we also noted that the listeners in the pilot study described the talkers as mad (i.e., they did not use the term angry), and we wanted to utilize the terminology used by the listeners [Lines 126- Attachment Submitted filename: Response to Reviewers.docx pone.0346897.s002.docx (37.2KB, docx) PLoS One. doi: 10.1371/journal.pone.0346897.r003 Decision Letter 1 Laura Morett Laura Morett Academic Editor Find articles by Laura Morett Author information Copyright and License information Roles Laura Morett : Academic Editor © 2026 Laura MorettLaura MorettLaura MorettLaura Morett This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 17 Feb 2026 PONE-D-25-51754R1-->-->Judgments of American English male talkers who are perceived to sound gay or straight:  Which personal attributes are associated with each group of talkers?-->-->PLOS One?> Dear Dr. Tracy, ================================= Please submit your revised manuscript by Apr 03 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at [email protected] . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .... We look forward to receiving your revised manuscript. Kind regards, Laura Morett Academic Editor PLOS One Journal Requirements: 1. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. 2. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #2: All comments have been addressed Reviewer #3: All comments have been addressed ********** 2. Is the manuscript technically sound, and do the data support the conclusions??> Reviewer #2: Yes Reviewer #3: Yes ********** 3. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #2: Yes Reviewer #3: Yes ********** 4. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #2: Yes Reviewer #3: No ********** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #2: Yes Reviewer #3: Yes ********** Reviewer #2: Thank you for the revised manuscript. This revision further contextualizes the study with relevant theory and implications (e.g., social hearing, workplace/classroom social consequences) and has substantially improved from the prior draft. I reviewed the authors’ response letter alongside the revised manuscript and found my major concerns to be addressed. I also reviewed the authors’ response to Reviewer 1’s many insightful feedback points. The authors report substantive revisions that address Reviewer 1’s major concerns (e.g., stronger conceptual framing, expanded pilot study description/rationale, and added reporting items like the repository link and power analysis). I did not see anything in Reviewer 1’s list that would, on its face, change my recommendation, though I defer to the editor on any remaining points requiring additional verification. The remaining issues are copyedits/consistency only and do not affect the interpretation of the results. Thus, I believe the manuscript is suitable for publication pending minor editorial/copyediting revisions. Comment R2-1, 5, 6, 7 The manuscript is in noticeably better shape after the terminology updates and grammatical clean-up. I appreciate the thoroughness of these revisions and am satisfied with the changes. I did catch a few remaining consistency items that are easy to fix (e.g., standardizing “outgoing” vs. “out-going,” and ensuring the manuscript consistently uses “boring” rather than “bored” as the label). Comment R2-2 I appreciate the authors’ effort to engage the literature more directly here. The added context is reassuring with respect to the study’s design, and I agree with the authors’ decision to prioritize brevity given the aims of the paper. I am satisfied with the current state of this section. Comment R2-3 I appreciate the inclusion of Niedzielski (1999) and the broader contextualization around listener perception and labeling. While I would have liked a slightly more explicit tie back to how this literature informs the present study’s predictions and interpretation, the addition strengthens the conceptual grounding, and I am satisfied with it in its current form. Comment R2-4 I am appreciative of the additional detail provided here. The discussion of Tracy & Johnson (2015) (i.e., similar judgments across sites) and the inclusion of campus demographic context serve as a reasonable proxy for addressing the concern. I am satisfied that appropriate precautions and contextual framing have been considered. [48] bored -> boring (consistency) [106] Add “as” in between “such” and “pitch” [178] Not sure what [eo] means here, assuming it’s a typo [201] Delete “a” between “seeing” and “photographs” [289] participants -> participants’ [515] False Referenced -> Falsely Referenced [536] False Referenced -> Falsely Referenced [603-604] transpose “was” and “SO” [621] of -> or [631] word -> words (?, I think the plural was meant here) [773] talkers’ -> talker’s Reviewer #3: This is a highly interesting study in which the authors explored whether listeners associated additional personal attributes with these types of talkers and whether different contexts (e.g., listeners being informed of the talker’s sexual orientation) influenced the strength of such attribute associations made by listeners. The findings revealed that sexual orientation per se did not affect attribute perception, yet informing listeners of the talker’s sexual orientation (whether the information was accurate or not) exerted an impact on their perceptions. This finding is of great significance for an in-depth understanding of the mechanisms underlying the formation of listener impressions. The authors have responded positively to the comments raised by the reviewers in the first round of review and made corresponding revisions accordingly. In my opinion, this paper has basically met the publication requirements and is acceptable for publication upon completion of minor revisions. Minor Revision Requirements 1. The manuscript requires a thorough proofread, as several errors may impede readers’ comprehension. For instance, Line 289: The participants mean age was 19.6 (grammatical error); Line 351: 24 talkers X 2 utterances X 4 attributes (using "X" as a multiplication sign, which is non-standard in academic writing). 2. The participants recruited for the experiment are relatively young (with a mean age of 19.6 years). Perceptual differences may exist among participants of different age groups and with varying social experience, and this limitation should be explicitly addressed in the Limitations section. 3. Ensure terminology consistency across all tables and figures. The authors noted in their response to reviewers that they have addressed consistency issues (e.g., using "stuck-up" instead of "stuck up"). A final careful proofread of all tables and figures is therefore recommended, as consistent terminology in the final version will enhance the paper’s professionalism. 4. Verify the accessibility of the materials available at the link: https://osf.io/bc2wm/overview?view_only=8da8397267a04e21a34f1c367f626a9b . Currently, the page loads as a blank screen. 5. The subheadings of the paper can be further refined to improve readability. For example, the subheading Results: Aim 1 could be revised to include a concise description of Aim 1 as the heading. 6. Line 428: seen in Table 2 and Fig 1 – "Fig" should be spelled out in full as "Figure". 7. The paper’s structure lacks clarity and appears disorganized. For example, Experiment 1a includes a combined Results and Discussion section, while Experiment 1 only has a standalone Results section; additionally, the sequencing of Experiment 1 followed by Experiment 1a is confusing and requires clarification. 8. Lines 747–754: Although the study did not balance high- and low-valence attributes, the authors may still provide a relevant discussion of high- and low-valence attributes here. 9. The expanded discussion on the "sad" attribute, as requested by the reviewers, is valuable. The explanation that different attributes may activate distinct pathways in the social hearing model is particularly insightful. To further improve readability, the conclusion of this paragraph should be more succinct, concisely summarizing the key finding as to why the "sad" attribute exhibited different patterns from other attributes; this will create a clearer logical bridge to the subsequent discussion points. ********** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our Privacy Policy ..--> Reviewer #2: No Reviewer #3: No ********** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation . NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications. PLoS One. 2026 Apr 10;21(4):e0346897. doi: 10.1371/journal.pone.0346897.r004 Author response to Decision Letter 2 Article notes Copyright and License information Collection date 2026. PMC Copyright notice 3 Mar 2026 We would like to thank everyone for their insightful comments that helped us to improve our manuscript. The new revisions are in red font in the file “Revised Manuscript with Track Changes.” Below, we’ve responded to each of the suggestions made by the reviewers. Additionally, we’ve attempted to provide line numbers, so that the editor and reviewers can easily find the changes. We apologize if there are any discrepancies with the line numbers. Response to Reviewer #1 1. Please be consistent with out-going versus outgoing. We have changed all instances of “out-going” to “outgoing”; this is consistent throughout the manuscript. 2. Please be consistent with boring versus bored. We have changed all instances of “bored” to “boring”; this is consistent throughout the manuscript. 3. The reviewer provided several small grammatical changes to make, and we have fixed all these small grammatical changes, as well as any other typos or grammatical changes we noticed during our read-through. Specifically, the reviewer noted they thought that [eo] was a typo. After checking this, we discovered that it was not a typo, but rather a missing reference. We have changed this to the correct reference [Line 178]. Response to Reviewer #2 1. Line 289, The participants mean age was 19.6 and this was not grammatical. We assumed that the reviewer meant that we should include an apostrophe after "participants”, and we included this. 2. Line 351: 24 talkers X 2 utterances X 4 attributes (using "X" as a multiplication sign, which is non-standard in academic writing) We deleted these sections (e.g., 24 talkers X 2 utterances X 4 attributes) in both Experiment 1 and Experiment 1a. Instead, both sections state the number of stimuli, trials, or lists [Lines 324, 332, 351, 361, 555]. 3. The participants recruited for the experiment are relatively young (with a mean age of 19.6 years). Perceptual differences may exist among participants of different age groups and with varying social experience, and this limitation should be explicitly addressed in the Limitations section. We agree that perceptual differences between age groups likely exist; we addressed differences in the social perception of LGBTQ+ communities in lines 708 – 728, as well as age-related differences in emotion perception in lines 729 – 738. In response to this comment, we further clarified that our listener sample is relatively young [Lines 707-708], and added information regarding other factors that may affect perceptual differences between age groups [Lines 736-738] 4. The data has not been made publicly available / Page is blank. We apologize that the link was not working for the reviewer; it is possible that the link may have erroneously included an extra period at the end. All three authors independently verified that the supplied link is now working correctly. If you click on the link, it will take you to a page with two folders – Data and Statistical Analyses. Files are included in each of the two folders, even though the folders are indicated as 0 B. If you click on the Data folder, there are two publicly available data sets. If you click on the Statistical Analyses folder, there are four publicly available data files that contain all the statistical analyses for the current project. 5. The subheadings of the paper can be further refined to improve readability; Results Aim 1 changed to a concise description of Aim 1. We changed the subheadings to align with the reviewer’s suggestion [Lines 419 – 420; 469]. 6. Line 428: seen in Table 2 and Fig 1 – "Fig" should be spelled out in full as "Figure". We changed all instances of the word “Fig” to the word to “Figure.” 7. The paper’s structure lacks clarity and appears disorganized; Experiment 1a includes a combined Results and Discussion section, while Experiment 1 only has a standalone Results section. We deleted the “Results and Discussion” from Experiment 1a. It now reads “Results.” [Line 560]. As a result, the subheadings in Experiment 1 and Experiment 1a align more clearly with one another. 8. The sequencing of Experiment 1 and Experiment 1a is confusing and needs clarification. To address this, we clarified earlier in the manuscript that anger and mad will be explored in Experiment 1a [Line 129]. Furthermore, we clarified the objectives of both Experiment 1 and Experiment 1a [Lines 252, 262, 270-271]. 9. Lines 747–754: Although the study did not balance high- and low-valence attributes, the authors may still provide a relevant discussion of high- and low-valence attributes here. We now include a discussion of the distribution of positive/negative valence in the attributes that were associated with gay and straight talkers and expand upon how a balanced design in future studies could further elucidate how listeners use features such as valence and dominance when making judgements of listener attributes [Lines 761-768], 10. Discussion on Sad Attribute; Conclusion of paragraph needs to be succinct; Summarize the key findings as to why sad demonstrated a different pattern. We summarized the discussion on “sad” by highlighting the simplest, and perhaps, most likely explanation [Lines 634-637]. Attachment Submitted filename: Response_to_Reviewers_auresp_2.docx pone.0346897.s003.docx (21.1KB, docx) PLoS One. doi: 10.1371/journal.pone.0346897.r005 Decision Letter 2 Laura Morett Laura Morett Academic Editor Find articles by Laura Morett Author information Copyright and License information Roles Laura Morett : Academic Editor © 2026 Laura MorettLaura MorettLaura MorettLaura Morett This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice 26 Mar 2026 Judgments of American English male talkers who are perceived to sound gay or straight:  Which personal attributes are associated with each group of talkers? PONE-D-25-51754R2 Dear Dr. Tracy, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact  and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact  and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact  and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support .... If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact [email protected]. Kind regards, Laura Morett Academic Editor PLOS One Additional Editor Comments (optional): I thank the authors for addressing the remaining minor comments raised by the reviewers. I am now pleased to recommend acceptance of the manuscript for publication in PLOS One. Reviewers' comments: PLoS One. doi: 10.1371/journal.pone.0346897.r006 Acceptance letter Laura Morett Laura Morett Academic Editor Find articles by Laura Morett Author information Copyright and License information Roles Laura Morett : Academic Editor © 2026 Laura MorettLaura MorettLaura MorettLaura Morett This is an open access article distributed under the terms of the Creative Commons Attribution License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited., which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. PMC Copyright notice PONE-D-25-51754R2 PLOS One Dear Dr. Tracy, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact [email protected]. You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing . If we can help with anything else, please email us at [email protected]. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Laura Morett Academic Editor PLOS One Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Attachment Submitted filename: Review_PONE-D-25-51754.pdf pone.0346897.s001.pdf (74.7KB, pdf) Attachment Submitted filename: Response to Reviewers.docx pone.0346897.s002.docx (37.2KB, docx) Attachment Submitted filename: Response_to_Reviewers_auresp_2.docx pone.0346897.s003.docx (21.1KB, docx) Data Availability Statement The data are contained in a public repository. The link is below: https://osf.io/bc2wm/overview . Articles from PLOS One are provided here courtesy of PLOS ACTIONS View on publisher site PDF (1.2 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top