A case study of the American Sign Language patterns of a natively-exposed Deaf autistic signer

ABSTRACT

Research on the acquisition of American Sign Language (ASL) by deaf autistic children has documented similarities to the linguistic profile of hearing children on the autism spectrum and has identified sign-specific phenomena that could serve as clinical markers of autism in the deaf population. However, the acquisition of a signed language by deaf autistic individuals whose language use appears to be at age-level is not well documented. Here we present a case study of ASL use by a Deaf autistic adolescent exposed to ASL since birth. Data include two video recordings (collected at age 16;11 and 17;4 years) of his spontaneous ASL. Utterances were coded for repetitions, phonological form, fingerspelling, metalinguistic awareness, and facial expressions having grammatical functions. Other data include a writing sample (17;6) and background information obtained from parental interviews and formal psycho-educational evaluations (between ages 14 and 17). Results revealed this adolescent has strong ASL skills including sophisticated vocabulary and semantic content, advanced fingerspelling, and evidence of metalinguistic awareness. His distinctive signing patterns include abundant repetition of phrases, intrusion of non-ASL handshapes, articulation disfluency, reduced facial expressiveness, and a dissociation between fingerspelling and writing skills. The results of this study diversify and broaden discussions of the ways that autism may interact with language development, especially in the visual-gestural modality. Understanding language patterns in autistic signers more fully will improve identification of autism in the deaf population, promote acceptance of diverse signing patterns in the Deaf community, and lead to better support for this population.

In writing this case study addressing the intersectionality of autism and deafness, the four co-authors are aware of our positionality as nonautistic, nondeaf individuals. Based on important conversations about ableist language between the autistic community and autism researchers, we recognize that language is a “powerful means for shaping how people view autism” (Bottema-Beutel et al. 2021:19). Similarly, Deaf Studies scholars write about depathologizing deafness and instead framing the experience within a cultural-linguistic model and identifying patterns of “Deaf gain” (Bauman & Murray 2014, Padden 1980, Padden & Humphries 1988). More recently, ideologies about deaf identities, native signer status, and sign language typologies are being explored and contested (Cheng et al. 2021, Hou & de Vos 2021, Kusters et al. 2017) as well as the intersectionality of deafness and autism (Friedner & Block 2017). In this paper, we write about an adolescent who is both Deaf¹ and autistic, pseudonym Brent. As a signer of American Sign Language (ASL) since birth, Brent has taught us a great deal about the strengths he possesses in his signed communication. In our effort to provide a detailed description of the distinctiveness of Brent’s ASL use, some readers may feel the presentation is too clinical. Be sure to read the discussion and conclusion where rather than problematizing Brent’s linguistic profile in terms of deficits, we are careful to situate his linguistic patterns within a broader quality-of-life context. We acknowledge patterns that may affect social interactions, such as Brent’s abundant repetition of most utterances, but also recognize that certain distinctive linguistic features in Brent’s signing most likely would not be distracting to interlocutors. Brent does have some medical conditions that may be relevant to his language patterns and how he experiences the world; these issues are presented with as little value-laden language as possible. With this detailed linguistic analysis of Brent’s ASL, we hope that clearer strategies will emerge so that clinicians can provide support for his ASL communication that Brent might want (or not want) for himself. We use identity first language (“autistic” and “Deaf”) as recommended by autistic community and Deaf Studies movements to avoid a medical-model rhetoric.

1. Introduction

Autism may affect a greater proportion of the population of deaf and hard of hearing (DHH) children than in the general population. While estimates in the general population have hovered between 1-2% over the past two decades (Maenner et al. 2021), estimates have varied from 1.9% (Szymanski et al. 2012) to 5.8% of the DHH population (Kancherla et al. 2013). Thus, there is a significant clinical population of people identified as on the autism spectrum who are deaf. Until very recently, there were no validated instruments designed for screening or diagnosing DHH autistic children, although the gold-standard Autism Diagnostic Observation Schedule (ADOS) was recently adapted and validated for use with the DHH population in the United Kingdom (Allgar et al. 2021, Phillips et al. 2022). Still, various complications in the diagnostic process, especially a lack of dually-trained clinicians expert in both autism and deafness (Roper et al. 2003) and a lack of established best practices for assessing autism in DHH youth (McFayden et al. 2023), suggest that autism may still be under-identified or mis-identified in the DHH population.

DHH children are exposed to language in different modalities, including visual-gestural signed languages of the Deaf such as American Sign Language (ASL), vocal-auditory spoken languages such as English (with or without amplification), written representations of spoken languages, or a combination of other sign forms and speech (e.g., Total Communication). Regardless of the modality in which language is acquired, the intersection of autism and language acquisition can be complicated. Although language impairment is no longer a diagnostic feature of autism (American Psychiatric Association 2022), language abilities in autistic individuals range from fluent to nonspeaking (Tager-Flusberg & Kasari 2013) or nonsigning. Pragmatic difficulties are universally attested, but research has also uncovered differences from typical development in nearly all aspects of language, including phonology, morphosyntax, and lexicon. For children acquiring a visual-gestural language such as ASL, there are a number of skills critical for the acquisition of signed language that are known areas of difference for autistic individuals.

As a brief background, signed languages such as ASL (or British Sign Language or Japanese Sign Language) rely on the two hands, the torso, and the face to build utterances that utilize lexical signs and deictic pointing. Those utterances capitalize on the space in front of the signer to convey grammatical relations (e.g., spatial marking for verb agreement). Signers also use co-expressive facial expressions with their manual signs to signal grammatical information such as Wh-questions or Yes/No questions.

It can be argued that these grammatical features of ASL require cognitive skills that are potentially vulnerable among autistic individuals: for example, visual perspective-taking (the ability to understand what others see from their different vantage points; Hamilton et al. 2009) could be important for spatially-marked verbs in sign; the ability to imitate gestures (Smith & Bryson 2007) and to produce indexical points for various communicative functions (Baron-Cohen 1989) may be important for processing deictic pronouns in sign; the ability to understand the emotional facial expressions of others (Denmark et al. 2014), the ability to produce such expressions (Denmark et al. 2019), and the ability to engage in sustained and conversationally appropriate eye contact (Pelphrey et al. 2002) are foundational for nonmanual (facial expression) grammatical markers in sign. Thus, studying how DHH autistic children acquire signed languages provides an opportunity to document how differing abilities in these cognitive areas interact with language development.

In the last decade, a body of work has emerged that focuses specifically on the acquisition of ASL by DHH autistic children. For methodological reasons, most of this work has been dedicated to a small subset of the DHH population, children exposed to ASL from birth by their Deaf parents, to ensure that any observed language or cognitive differences cannot be attributed to early language deprivation.² The work of Shield & colleagues on small samples of native-signing autistic children has documented aspects of their sign language acquisition. In particular, they have documented that these autistic deaf children show many similarities to the linguistic and cognitive profile of hearing, speaking autistic children, including echolalia (Shield et al. 2017a), avoidance of personal pronouns (Shield et al. 2015), theory of mind and visual perspective-taking difficulties (Shield et al. 2016), and motor difficulties affecting the production of language (Bhat et al. 2018; Shield et al. 2017b); Shield et al. (2015) also report on a subset of autistic deaf children (6 of 23; 26%) who are nonsigning (and nonspeaking), despite full exposure to ASL from birth by their Deaf, signing parents; this proportion mirrors the 30% of hearing autistic children who are minimally expressive in speech (Tager-Flusberg & Kasari 2013).

Additionally, this work has identified sign-specific phenomena that could serve as clinical markers of autism in the DHH population. In particular, signing DHH autistic children have been shown to sometimes reverse the direction of their palm while signing (Shield & Meier 2012; Shield et al. 2020, 2022), a unique pattern which does not regularly appear in the typical acquisition of signed languages and which exhibits characteristics that are suggestively similar to the reversal of personal pronouns by hearing autistic children (Shield et al. 2022). However, these palm reversals appear inconsistently and infrequently and do not occur in all DHH autistic children, instead clustering in children with lower overall receptive language abilities (Shield et al. 2022).

Shield and colleagues (2015, 2016, 2017a, 2017b, 2020, 2022) have never included a language-matched comparison group of DHH nonautistic children in their work, with only a subset of the DHH autistic children falling in the average range of receptive language skills. It therefore appears likely that prior studies of natively-exposed signing autistic children have focused less on children with fluent ASL, probably due in part to difficulties with accurately identifying autism in such DHH children. Thus, there is currently a gap in the literature. While the full spectrum of language profiles in autism in hearing, speaking children and adults is well known, the acquisition of a signed language by autistic children with fluent language is not well documented or understood.

In this case study, we will discuss the ASL- and English-language patterns of “Brent” (pseudonym), a natively-exposed signing Deaf adolescent on the autism spectrum. We identify his many strong ASL skills alongside distinctive signing patterns that include repetition, intrusion of noncanonical hand configurations, reduced facial expression, and written English dysgraphia.

2. Method

In this case study we analyzed two pre-existing video recordings of Brent’s spontaneous ASL signing in order to describe key linguistic features and patterns. Interviews with parents and information from school records (e.g., individualized education programs [IEPs], psychological evaluations, test results) were reviewed to understand the history and prevalence of Brent’s linguistic and behavioral patterns. A writing sample was also examined. Brent’s parents gave their informed consent and shared all psycho-educational materials with the authors. The parents explained this research project to Brent while on a videophone call with some of the authors and he gave his assent to participate. This study was approved by, or was considered exempt by, the authors’ respective institutional review boards.

2.1. Participant

Brent is a natively-exposed ASL signing³ Deaf male adolescent on the autism spectrum; his age at the time of the collected language samples was 16;11 and 17;4. His sensorineural bilateral severe-to-profound deafness was confirmed at age 18 months. Brent is exposed to ASL every day at home and at school. Beginning at birth, he acquired ASL from two natively-exposed ASL signing hearing parents and four Deaf grandparents. He experienced delayed developmental milestones and has received special education services in ASL since age 5 months. While he attends a school for the deaf, his full day placement is within a special education classroom with ASL-using teachers alongside other deaf children who have various educational needs. Brent was formally diagnosed with autism based on evaluations at age 5. Brent’s autistic traits include focused interests (restaurants, stores, hotels), some repetitive body movements as a form of self-regulation (face rubbing, hand wringing, and pacing), reduced facial expression when communicating, and a preference for routines/sameness.

At age 12, Brent was diagnosed with Temple syndrome (Ioannides et al. 2014, Juriaans et al. 2022), a rare genetic condition that is associated with various medical issues, some of which Brent has, including diabetes, short stature, smaller hands, joint hypermobility, and clinodactyly (abnormally curved finger).

In terms of mental health, Brent has a diagnosis of generalized anxiety disorder and experienced a major depressive episode at age 16. At the time of the video recordings examined in this study (16;11 and 17;4), he was considered in partial remission for major depressive disorder; he continues to take medication for both anxiety and depression. According to his parents, Brent had shown considerable improvement in overall mood in the months preceding the video recordings.

Even with full immersion in a fluent ASL language environment from birth, Brent showed some early acquisition patterns that seemed delayed or distinct from other native ASL signing children. While he did produce single signs at 12 months, his parents noted difficulties in establishing connected eye gaze and maintaining his attention during signing while a toddler. They reported that it did not feel like they were in a conversation with him until he was closer to 2;0 or 2;6 years of age. School evaluations mention that Brent has repetitive signing, stereotyped responses (possibly some instances of echolalia⁴), a limited range of facial expressions, and “odd finger posturing.” While his ASL signing currently shows some distinctive patterns (as will be described in detail in the following sections), his level of fluency is much higher than most of the DHH autistic children studied in previous research, many of whom produced limited expressive ASL compared to what could be expected for their ages (Shield et al., 2015, 2016, 2017a, 2017b, 2022; Bhat et al. 2018). In terms of spoken English, Brent’s mother reports that he can vocalize or articulate “mamamama” and “byebyebye,” with both words typically being produced with much repetition.

2.2. Materials and procedures

In this study of Brent’s current language abilities, we relied on several resources to describe his linguistic history and psycho-educational profile.

2.2.1. Parental interviews

We conducted several videoconference interviews with his parents to ask clarification questions and questions about prevalence of observed patterns. The first hour-long interview included three authors and both parents. Four follow-up interviews with Brent’s mother were conducted by the lead author (JLS). All interviews took place when Brent was 17 to 18 years of age. The interviews were recorded and transcribed.

2.2.2. School and psychological evaluations

Brent’s mother provided us with IEPs and evaluation reports from ages 6;0, 12;3, 14;7, and 17;7. These reports include results from various language, cognitive, and educational testing, as well as notes documenting observations of Brent in his classroom and in other settings. In the results section, we cite language-related findings from these IEP reports when the report or testing results inform our analyses of the language samples.⁵

Brent’s Wechsler Full Scale IQ (WISC-V; Wechsler 2014) was evaluated as 78, a score considered in the low range (age 16;6). The evaluator noted that his strengths include rote and working memory and visual discrimination skills. Areas of noted difficulty, with standardized scores in low ranges, were visual-spatial skills, fluid reasoning, visual-motor integration and processing of complex, abstract visual-spatial patterns. Brent’s mother provided a real-world example of differences in his pattern knowledge: he easily memorizes details of many “members of a category” that all share some aspect of a pattern (e.g., major cities in U.S. states), but has difficulty situating such details within a visual-spatial organization (e.g., placing the cities within a map or diagram reflecting relational patterns between members).

2.2.3. Video recordings

Our primary sources of data are two video recordings of Brent signing in ASL. Brent enjoys asking his mother to videorecord his short narratives, which she then posts on her social media. In the videos, he signs continuously without conversational turn taking with his mother, who is behind the video camera. She reports that, although she was an interlocutor for him, nodding her head, she did not converse or feed him suggestions. For both videos, she provides real-time voice interpreting of the content. Brent’s mother, a hearing ASL signer from birth, is also a certified sign language interpreter. The two videos are 5 months apart: the first was recorded at age 16;11 (10 m, 20s in length) and the second at age 17;4 (8 m, 1 s).

The topics of Brent’s video narratives are mostly consistent with his focused interests. He lists all his favorite restaurants and describes what he likes to eat at each. He then goes on to list his favorite home improvement stores and hotels. In terms of context, Brent seems aware that he is sharing his narrative with a social media audience because he asks whether they agree with one of his points; he also asks that they please inform him if they are fascinated with his favorite things. His mother said that he wants viewers to show their response by posting on his mother’s social media page.

A secondary source of video data is one we consulted to gather historical information regarding Brent’s fingerspelling abilities. This video was collected when Brent was 8;5 as part of the last author’s postdoctoral research (Bhat et al. 2018, Shield et al. 2022). In this 3 m, 16s video, Brent responded to a set of 15 printed English words presented on an iPad. The last author (AS), a hearing, ASL-fluent L2 learner of ASL, asked Brent to fingerspell each word.

Finally, Brent’s mother shared several additional videos of him signing that had been posted on social media. While we did not analyze these videos for the variables under investigation, we used them as confirmation that the phenomena we describe in this paper are representative of his overall signing abilities.

2.2.4. Video coding

For both primary source videos, we transcribed Brent’s mother’s voiced translation and added that verbatim into video annotation software (ELAN 6.2 M1). However, her real-time voiceover did not include Brent’s repetitions. Author JLS (a hearing, natively-exposed ASL signer), confirmed his mother’s translation and created ASL glosses of each utterance, including every repetition.⁶ The three other co-authors reviewed the ASL glosses for accuracy. Several utterances were revisited, and the glosses corrected after discussion. Using the glossed utterances of Video 1 (V1: total 126 utterances) and Video 2 (V2: total 217 utterances), the authors then coded for number of repetitions, atypical phonological forms, occurrence of facial expression or body movement to signal emphasis or grammatical function, use of fingerspelling, and evidence of metalinguistic awareness. Once our analyses were completed, we shared our observations with Brent’s parents (both hearing, natively-exposed ASL signers) who both concurred with our findings and interpretations. As a final verification step, the authors shared Brent’s video recordings (with parent permission) to a Deaf, ASL-fluent researcher who validated our reported patterns in Brent’s signing and provided feedback on our interpretations of the data. This consultant agreed, based on several indicators, that the way Brent signs places him firmly in L1 usage of ASL (compared to L2 usage).

2.2.5. Calculation of signing and fingerspelling pace

To estimate Brent’s signing pace, we calculated the average duration of his signed utterances in each video. We used ELAN to extract time codes from the start of each utterance to its end. When an utterance was repeated, we divided the duration time of the entire expression containing those repetitions by the number of repetitions to determine an average single-utterance duration time.

To calculate fingerspelling pace, we coded 126 utterances from the two videos (out of 343 total utterances; 36.7%) for fingerspelled words, extracted the duration of each fingerspelled word, then divided by the number of letters/handshapes to yield an average duration per letter. We then averaged across all fingerspelled words to obtain a total average handshape duration while fingerspelling. We excluded fingerspelled words that are considered lexicalized in ASL, such as #BUS or #PIZZA, as these forms may result in a faster fingerspelling pace.

2.2.6. Writing sample

Brent’s mother provided us with a video of him writing a “plan for the day” by hand with pen and paper (age 17;6). At the end of the writing activity, Brent looked at the document and signed what he had written; his mother provided real-time ASL-to-English translation as he signed. She then took the written document from Brent and put it within the camera’s view and read aloud what the writing meant with additional clarification. She reported that this writing sample is typical of what he tends to compose if asked and of how his compositions look in structure and format.

3. Results

3.1. Brent’s strengths in ASL

3.1.1. Overall narrative skills

We begin with a general picture of Brent’s strengths in his ASL communication. The subject matter of both video narratives reflects his focal interests (favorite restaurants, hotels, and home improvement stores). Brent includes other sophisticated semantic content such as a discussion about the impact of COVID-19, the beauty of a college campus he visited, and the geographical relationship among some of his favorite restaurants. He accurately uses what we, and our Deaf ASL expert, consider cognitively complex, phonologically marked, and low frequency ASL vocabulary including signs like wrecked, fascinated, embarrass, and opinion.⁷

Brent’s school evaluation at age 17;7 summarizes an ASL receptive skills evaluation that was administered at age 14;7. The report indicates that he had average scores on a sign comprehension task (watching a video-presented sign and choosing a signed definition from a set of choices). It reports that he had difficulty with “higher level nuanced language concepts such as synonyms, antonyms, syntax, analogies, declarative sentences, and pragmatic language items.” Similarly, an earlier evaluation conducted at age 6;0 indicated “good comprehension of ASL and [production] at [a] normal rate with some repetition & rephrasing, broad vocabulary knowledge.”

3.1.2. Signing pace

Overall, Brent is quite fluent in his signing; yet our initial impression was that his pace felt rapid at times. This impression was corroborated by the IEP at age 14;7 indicating that Brent “signs too fast at times.” We draw from several cited examples in the literature on signing pace. Based on 12 signers, Wilbur (2009) provides estimates for fast (2.43 signs per second), average (1.95 signs/sec), and slow (1.55 signs/sec) paces of ASL signing. Based on three Deaf natively-exposed signers, Bellugi et al. (1979) reported signing paces ranging from 1.7-2.1 signs/sec (however, they excluded pausing from their analysis). Brent’s signing pace across the two videos was V1: 1.7 signs per sec (SD 4.64), and V2: 1.88 signs per sec (SD 2.25). Based on these means, Brent’s pace did not appear to diverge from reported average signing pace in prior research; the differences in standard deviations between Videos 1 and 2 are likely explained by the fact that Brent consistently produced utterances twice in Video 2, and the wider range of number of repetitions occurring in Video 1 (see Table 1). Further analysis into signing pace that considers the extent of repetition or extent of fingerspelling may reveal a more nuanced picture of Brent’s signing pace.

Table 1. Summary of repetition patterns in videos 1 and 2.

Pattern	Video 1 Unique Utterances (n = 126)	Percentage of Total	Video 2 Unique Utterances (n = 217)	Percentage of Total
Utterance not repeated	8	6.3	62	28.5
2 repetitions	19	15.0	145	66.8
3 repetitions	45	35.7	8	3.6
4 repetitions	29	23.0	2	0.9
5 repetitions	21	16.6	0	0
6 repetitions	0	0	0	0
7 repetitions	1	0.7	0	0
Other	3	2.3	0	0

Note: Bolded numbers indicate the most frequent pattern in each column.

3.1.3. Metalinguistic awareness

In the video narratives, we documented several examples of Brent’s metalinguistic awareness. Once, he caught and corrected his own semantic error regarding a restaurant’s location. Upon signing a certain name sign for a city, he stopped and commented that the sign (which is produced on the nose) is embarrassingly close to a sign for “a private area.” During one interaction with the authors and his mother, Brent corrected her sign autism, showing her a new sign neurodivergent.⁸ He shows audience awareness by asking on camera “Are you fascinated with my list of favorite things?”

The school evaluations and observations also provided examples of metalinguistic awareness. In a classroom observation at age 17;7, Brent was engaged in an ASL class activity in which the students were asked to come up with a sign that has the X handshape (he correctly responded wise) and a sign that has the s handshape (he correctly responded work). That he was able to generate these signs without support shows that Brent possesses the ability to decompose a sign into its phonological parts.

A second classroom observation, also age 17;7, highlighted that Brent joked and used sarcasm in his interactions with (mainly) adults; the same report also indicated that he struggled to read social cues and the intentions of conversational partners. These observations were echoed in the parent interviews. Brent’s mother commented that Brent likes to tease others and make funny comments in conversations. We thought it important to identify this strength inasmuch as sarcasm and humor are considered to be advanced cognitive-linguistic abilities that relate to Theory of Mind. There are mixed findings in the autism literature with regard to sarcasm; some autistic individuals can readily detect sarcasm or irony in other speakers (Pexman et al. 2011) but others show difficulty inferring the mental states of others, as is necessary for the detection of sarcasm (Kaland et al. 2002). O’Reilly & colleagues (2014) found that while young deaf natively-exposed ASL signers were significantly poorer than their hearing counterparts in understanding sarcasm, they continued to develop this skill from pre-adolescence into adulthood at which their demonstrated sarcasm abilities were on par with hearing adults.

3.1.4. Fingerspelling strength

Brent’s signing includes frequent and impressive use of fingerspelling. Fingerspelling is a common means of conveying the letters of the English alphabet using a sequence of individual handshapes. Padden & Gunsauls (2003) estimate that most signers fingerspell about 10-15% of the time while signing ASL, with approximately 70% of the fingerspelled words being proper or common nouns, typically ones for which there is no ASL lexical sign. In V1 Brent produced 227 fingerspelled words (22.6%), including lexicalized fingerspellings, compared to 776 signs (77.4%). This frequency appears higher than the 10-15% average reported by Padden & Gunsauls (2003). However, this rate may be elevated due to Brent’s focused interest in restaurants, stores, and hotels; his many references to restaurant names and menu items likely inflated his usage of fingerspelled forms.

We documented lengthy words like fs-flatbread, fs-signature, and fs-bonelesswings. We did not detect any incorrectly spelled words. Like other signers, Brent’s fingerspelling consists mostly of common nouns (e.g., fs-tub) or proper nouns, such as restaurant names (e.g., fs-hardrockcafe).⁹ Brent correctly includes an apostrophe-s when fingerspelling the store name fs-lowe’s but also applied it at the end of fs-smore’s, whereas it should occur after the initial S of “s’mores.” In ASL fingerspelling, the convention used to convey an apostrophe-s is a twisting movement added to the letter s.

Brent’s inventory of fingerspelled words in the two videos includes several lexicalized fingerspelled forms (#pizza, #all, #ha-ha-ha, #ok). In ASL, a common fingerspelled word can become lexicalized (Battison 1978, Padden & Gunsauls 2003) when the form takes on a sign-like prosody (e.g., #all incorporates a sideways gliding movement as the letters A and L are expressed) or a conventional form that routinely omits certain letters (e.g., #pizza,¹⁰ which omits the P and I, and uses two extended fingers to execute the double letter Z, followed by an A). Brent also correctly uses common fingerspelled abbreviations such as ill for Illinois, ky for Kentucky, and rb for root beer.

3.2. Brent’s fingerspelling patterns: A more complex story

While fingerspelling is one of Brent’s strong language skills, a detailed analysis revealed some distinctive patterns in his fingerspelling, including pace, articulatory accuracy, contexts of use, and its relationship to English spelling knowledge.

3.2.1. Fingerspelling pace

Our general impression of Brent’s fingerspelling is that it is fast paced. His speed and accuracy strongly suggest that his fine-motor skills for producing rapidly sequenced letters are strong. In V1, Brent produced a total of 1,076 individual letter handshapes; each handshape lasted an average of 0.15 sec (SD = 0.03 sec), or approximately 6.6 handshapes/sec. Mayberry & Waters (1991) conducted a study of fingerspelling rate that included 8 deaf native signing adolescents. For the 13 to 15-year-olds, average fingerspelling production rate was approximately 0.75 seconds per three-letter word versus 1.1 seconds per five-letter word. For nonlexicalized fingerspelling, we identified 31 instances from Brent’s video of three-letter fingerspelled words and found the average was 0.42 sec (SD = 0.09 sec); the average for 22 instances of five-letter words was 0.71 sec (SD = 0.11 sec). Thus, based on this sample Brent’s fingerspelling is approximately 50% faster than has been documented for typical native signing adolescents. For example, when introducing a novel proper noun through fingerspelling, signers often use a clearer, more slowly executed “canonical” or “neutral” form of fingerspelling (Mulrooney 2002, Stone et al. 2015). In such contexts on the videos, Brent never appears to slow down his fingerspelled words. Brent’s mother shared that if he fingerspells too fast and she asks him to repeat, he often repeats the form at the same fast pace, possibly indicating that he does not recognize that slowing down would benefit his interlocutor.

3.2.2. Articulatory accuracy

In checking for the articulatory accuracy of Brent’s fingerspelling we noted that there was distinctiveness in his productions of the letters R (fs-factory, V2 1:38) with uncrossed fingers, and several instances where the thumb is extended rather than tucked in: P (fs-pool, V2 4:52), K (#ok, V2 6:17), and Q (fs-bbq, V1 2:06). As will be discussed below, Brent’s signs sometimes include pinky extension (PE) and/or ring finger retraction (RFR); these phonological intrusions are also observed in some of Brent’s fingerspelled words (e.g., the R handshape sometimes has the pinky finger extended).

3.2.3. Abbreviations

Brent often uses expected abbreviations in his fingerspelling, for example, ccf for Cheesecake Factory and bww for Buffalo Wild Wings. Less typical is his creation of fingerspelled shortcuts for signed phrases such as ffm for “favor for me.” Brent’s mother reports that he enjoys making up abbreviations such as was (“wait and see”) and vgp (“very good point”). These three examples indicate that Brent is aware of at least the first letters of words in the English translations of some ASL phrases that he uses.

3.2.4. Substitution of fingerspelling for a known ASL sign

We noted several instances of Brent using fingerspelled forms in contexts where the sign equivalent is known and would be expected. Padden & Gunsauls (2003) explain that there are occasions when ASL signers fingerspell a known sign when they are using fingerspelling itself as a signifier to emphasize or to make a contrast. In V2, Brent fingerspells my, shirt, and love, all of which have common conventional signs in ASL. Brent asks his audience to look at “fs-my fs-shirt,” and later he says, “ME fs-love restaurants.” Our impression was that these fingerspelled expressions were not being used as signifiers; that is, they did not appear to be establishing emphasis or contrast. We also noted that Brent’s IEP from age 14;7 reports a classroom observation when Brent was observed “fingerspelling words that do have a sign in ASL.”

3.2.5. Chaining signs and fingerspelling

On some occasions, the fingerspelled form that Brent uses is part of a chaining sequence that includes both a sign and its fingerspelled equivalent. In the examples from V2 shown in the following, we observed chaining sequences in various orders that were then repeated: (a) he fingerspells, then signs the equivalent; repeats the sequence; (b) he signs, then fingerspells the equivalent; repeats the sequence; and (c) signs twice, then fingerspells the equivalent twice.

Table

(a)	fs-factory factory
	fs-factory factory
(b)	look, fs-look
	look, fs-look
	look
(c)	signature, signature
	fs-signature, fs-signature

Chaining is typical in an ASL narrative when a signer is providing clarification to an interlocutor who might not know the sign or the fingerspelled form being expressed; it is also a common practice in bilingual deaf education in which teachers express an ASL sign followed by the fingerspelled translation of that sign, and then sometimes produce the sign again (see “chaining” and “sandwiching” in Humphries & MacDougall 2000, Padden & Ramsey 2000). Brent’s chaining pattern may have been learned from his exposure to this educational practice, but it also may indicate that he has some awareness of needing to clarify for his audience. For example, the sign glossed here as the noun signature can also be glossed as the verb sign-a-document. In example (c), Brent could thus be indicating that he intends a particular interpretation of this sign consistent with the name of the restaurant he is discussing. However, this account is countered by sequences in which Brent chains a common ASL sign like look with its fingerspelling translation; the ASL lexical sign look should be known to any signing interlocutor. Thus, it remains unclear whether Brent’s chaining pattern reflects audience awareness (i.e., possibly demonstrates Theory of Mind) or is a product of a signing pattern he often would see at school.

3.2.6. Fingerspelling and its relationship to spelling knowledge

An important question is whether Brent’s fingerspelling reflects knowledge of printed English through a direct mapping between printed letters and handshape forms. We take this up in a later section where we compare his writing sample to his signing, but here we share findings about Brent’s fingerspelling that came up in an IEP and a parent interview.

At age 14;7, Brent’s IEP states that on the Kaufman Test of Educational Achievement-3 he scored in the average range on a task involving identifying letters. This suggested that he could process letters quickly and automatically. The IEP adds that Brent was able to identify sight words and read simple or concrete sentences but suggests that he needed to continue developing his vocabulary knowledge. However, Brent scored in the very low range on a spelling task in which he was asked to write the English translation equivalent of an ASL lexical sign being signed to him; the report indicates that he typically wrote a different word–albeit one that was correctly spelled–suggesting that he had difficulty with accessing spelling knowledge in real time and then translating that knowledge into printed form. This finding also suggests Brent has some memorized written forms.

In an interview with Brent’s mother, she shared that she occasionally asks him whether he knows how to fingerspell the word corresponding to a certain ASL sign. If he says no, she spells it out slowly for him. He then practices it on his hand a few times and fingerspells it correctly going forward. He does not use an intermediate step of writing the new word on paper, nor does his mother write it down for him; instead, the fingerspelling of the new word seems to be learned “through the air.” She also reports that he has strong fingerspelling comprehension skills. We note that Brent’s IEP indicates he was unable to complete a speechreading assessment and, as indicated earlier, he articulates only “mamama” and “byebyebye” in the speech modality. Thus, we do not believe his internal representation of spelling is connected to an internal representation of speech sounds.

3.3. Brent’s signing distinctiveness

Alongside Brent’s clear strengths in his ASL signing, we documented several patterns that are not considered canonical ASL: numerous repetitions of phrases, intrusion of unexpected handshapes in some of his signs, some sign and fingerspelling articulation disfluency, and reduced facial expressiveness.

3.3.1. Phrasal repetition

The most dominant pattern observed in Brent’s signing is the abundance of repeated phrases. In both videos, Brent consistently repeats his utterances; however, the number of repetitions was considerably higher and more variable in the first video than in the second, which was recorded 5 months later.

In example (d), we provide the ASL gloss of an excerpt of one of Brent’s narrative segments, followed by its English translation. This example shows that he can keep his intended message in mind across the many repeated utterances by which he encodes that message.

Table

(d)	English translation of narrative segment (without repetitions):
	“I feel 2021 is a better year than 2020, right? 2020 was wrecked, right? Because of COVID. 2021 is a better year. Do you agree?”
	ASL gloss of signed utterances, including repetitions (V1 00:47):^11
	feel 2021, feel 2021, feel 2021, feel 2021
	better year, better year, better year, better year
	than 2020, than 2020, than 2020, question
	2020 wrecked, 2020 wrecked, 2020 wrecked, question
	because fs-covid, because fs-covid, because fs-covid
	2021 better year, 2021 better year
	agree question, agree question

We documented the number of Brent’s unique utterances in each video and how often each utterance was repeated. These results are summarized in Table 1. In the first video, only 6.3% of his 126 unique utterances were produced just a single time. Rather, his utterances were repeated 2 (15%), 3 or 4 (58.7%), or 5+ times (19.8%). This variation indicates that in V1 Brent was not adhering to a routine calling for a fixed number of repetitions. In V2, Brent produced 217 unique utterances, of which 28.5% were produced a single time. It is striking that in this video Brent is more consistent with 66.8% of his utterances being repeated exactly twice. Only 4.6% of his phrases are repeated 3-4 times.

What linguistic structures does Brent repeat? We next analyzed what kinds of linguistic structures Brent repeated. Looking across the 118 utterances in V1 that included repetition, 31 (26%) were single-sign utterances. Of these 31 signs, 25 are fingerspelled food or place names. Other tokens are fs-love, delicious, and hello. In V2, for the 155 utterances with repetition, only 21 (13.5%) involved single signs. Of these 21 signs, 14 are fingerspelled food or place names. Other tokens are delicious, restaurant, several, shut-down, and different++. The majority of Brent’s utterances in both narratives are multi-sign strings that are wholly repeated. Table 2 shows the types of repeating structures that we documented, representing a broad range of grammatical structures including noun phrases, verb phrases, prepositional phrases, and complete sentences.

Table 2. Grammatical categories of Brent’s repeated phrases.

Nouns & NPs	Fingerspelled names: fs-cheese cake factory
	Compound nouns: chocolate milk
	NPs: every family friend (“all family and friends”)
Prepositional Phrases	in city-name
Verb Phrase	like restaurant
Sentences	fs-bbq yummy fs-ribs bone-hold-rib-eat-f-cl (“Yummy BBQ, (I) gnaw on the ribs”) 2021 better year (“2021 is a better year”)
Other sequences	hmmm different++
Phrase fragment	(omitted subject) feel 2021 (“I feel 2021”)

We explored whether any of Brent’s repetition patterns carried into repeating the internal sub-components of a sign. For example, Meier et al. (2008) report a child language form by Noel (age 1;6) who produced the single-cycle ASL sign black with 6 internal movement cycles (akin to the movement one sees in the sign for toothbrush). In the few cases in which Brent repeated just a single sign (e.g., delicious, restaurant, next), it was difficult to determine whether the repetition would be considered a multiplication of the internal movement cycle within the sign, or if it was one intact sign repeated several times. We did observe one example, store, which was produced with seemingly 8 internal movement cycles (typically it would have 2 movement cycles). But, in this case, we felt that Brent seemed like he was “thinking while signing” and extending the sign’s duration while his eye gaze went off to the side. There were three single signs that showed repetition of a different type: two seemed to express enthusiasm by adding a repeating circular movement (hello!! and i-love-you!!). In the third instance, different++, Brent added a spatial-number marker that means “many different”; here, the sign different is appropriately repeated, while moving across the signing space. In sum, we found little or no evidence that the repetitions in Brent’s signing were anything other than fully executed repetitions of the whole sign, partial phrase, or whole phrase; it did not appear that he was repeating sign-internal movement cycles as a general pattern.

Finally, we examined what linguistic structures did Brent not repeat. In V1, 6 of the 8 (75%) nonrepeated utterances were phrases with only one sign. For V2, 57 of the 62 (92%) nonrepeated utterances were phrases with only one sign. In both videos, the majority of these nonrepeated single-sign expressions were related to the larger discourse structure (but, and, next) or were side comments (shh, yummy, hmmm, well). This finding suggests that Brent maintains an overall discourse structure and planned narrative and that his repetitions tend to not include these connectors or side comments.

Another example of Brent’s adherence to discourse structure is his frequent use of listing structures in ASL referred to as list buoys (Davidson 2013, Liddell 2003). For example, Brent uses list buoys¹² (e.g., FIRST-LIST, SECOND-LIST, THIRD-LIST) on his nondominant hand 26 times in V1 and 22 times in V2. In Table 3, Example 1 illustrates how Brent preserves the list buoy on his nondominant hand across the repeated strings when the phrase includes only one-handed signs (like restaurant on the dominant hand). Example 2 shows how Brent appropriately drops the first-list list buoy on the nondominant hand when he first produces a two-handed sign (yummy) and only re-establishes it when the next listing buoy is called for. Notice here that second-list stays active on the nondominant hand because shrimp eat-f-cl involves only one hand (similar to like restaurant above). Brent does not rearticulate SECOND-LIST with each repetition.

Table 3. Examples of Brent’s use of list buoys in extended discourse

Ex. 1: English translation: “First, (I) like restaurants, like restaurants, like restaurants, like restaurants, like restaurants.”
ASL gloss:
Nondominant Hand	Dominant Hand
first-list
(MAINTAINED)	like restaurant,
(MAINTAINED)	like restaurant,
(MAINTAINED)	like restaurant,
(MAINTAINED)	like restaurant,
(MAINTAINED)	like restaurant
Note: We leave the ASL gloss for restaurant as singular inasmuch as plurality is generally not marked on nouns in ASL
Ex. 2: English Translation: “First, I like to eat BBQ, yummy bone-in ribs (like this); BBQ, yummy bone-in ribs (like this); BBQ, yummy bone-in ribs (like this). Second, I eat shrimp (like this); I eat shrimp (like this); I eat shrimp (like this)”.
Nondominant Hand	Dominant Hand
first-list
(MAINTAINED)	fs-bbq
	yummy fs-ribs bone hold-rib-eat-f-cl,
	fs-bbq YUMMY fs-ribs bone hold-rib-eat-f-cl,
	fs-bbq YUMMY fs-ribs bone hold-rib-eat-f-cl
second-list
(MAINTAINED)	shrimp eat-f-cl,
(MAINTAINED)	shrimp eat-f-cl,
(MAINTAINED)	shrimp eat-f-cl
Note: In both examples, Brent uses the f-cl handling classifier which represents the shape of his hands as he holds the barbeque ribs or the shrimp while eating (“like this”).
The following patterns are not attested (as indicated by the asterisk):
*first-list, fs-bbq yummy fs-ribs bone hold-rib-eat-f-cl
*first-list, fs-bbq yummy fs-ribs bone hold-rib-eat-f-cl
*first-list, fs-bbq yummy fs-ribs bone hold-rib-eat-f-cl
Or
*second-list, shrimp eat-f-cl,
*second-list, shrimp eat-f-cl,
*second-list, shrimp eat-f-cl

These examples illustrate how Brent maintains the larger discourse structure in his narratives. While there is an abundance of repetition, there are still anchors (e.g., list buoys) that he does not repeat, and he preserves the integrity of whole words and signs. In other words, he does not repeat individual letters within fingerspelled words (such as fs-lll-ooo-vvv-eee), nor does he repeat internal movement within a single sign.

3.3.2. Distinctive handshape intrusion: PE and RFR

A second distinctive language pattern is handshape intrusion. In ASL, a sign is made up of four primary components (handshape, location of the sign, movement, and orientation) that are considered contrastive, sublexical linguistic units. In V1, approximately 24% of Brent’s unique utterances include a noncanonical phonological form. The handshape intrusion consists of RFR on signs that do not typically have the ring finger retracted combined with PE on signs that do not typically have the pinky finger extended. For signs that have the pinky finger extended in their canonical form, only the RFR appears in his intrusive forms; similarly, for signs that have the ring finger retracted in their canonical form, only the PE appears in the intrusions. For some signs, both PE and RFR occurred even though their canonical forms contained neither (e.g., signs or fingerspelling that use the handshapes O and W). For examples of handshapes where this intrusion was apparent, see Table 4.

Table 4. Images of Brent’s handshapes with pinky extension (PE) and ring finger retraction (RFR) intrusion paired with illustrations of citation (target) forms in ASL.

Handshapes with intrusive PE; citation form has RFR			Handshapes with intrusive RFR; citation form has PE			Handshapes with intrusive PE + RFR; citation form has neither
Photograph of Brent’s hand in E handshape with pinky extended.	E	Illustration of E handshape in ASL.	Photograph of Brent’s hand in 8 handshape with ring finger retracted.	8	Illustration of 8 handshape in ASL.	Photograph of Brent’s hand in O handshape with ring finger retracted and pinky extended.	O	Illustration of O handshape in ASL.
Photograph of Brent’s hand in D handshape with pinky extended.	D	Illustration of D handshape in ASL.	Photograph of Brent’s hand in Claw-5 handshape with ring finger retracted.	Claw-5	Illustration of Claw-5 handshape in ASL.	Photograph of Brent’s hand in W handshape with ring finger retracted and pinky extended.	W	Illustration of W handshape in ASL.
Photograph of Brent’s hand in G handshape with pinky extended.	G	Illustration of G/1 finger handshape in ASL.	Photograph of Brent’s hand in B handshape with ring finger retracted.	B	Illustration of B handshape in ASL.
Photograph of Brent’s hand in V handshape with pinky extended.	V	Illustration of V handshape in ASL.	Photograph of Brent’s hand in 5 handshape with ring finger retracted.	5	Illustration of 5 handshape in ASL.
Photograph of Brent’s hand in L handshape with pinky extended.	L	Illustration of L handshape in ASL.	Photograph of Brent’s hand in C handshape with ring finger retracted.	C	Illustration of C handshape in ASL.
Photograph of Brent’s hand in R handshape with pinky extended.	R	Illustration of R handshape in ASL.
Photograph of Brent’s hand in P handshape with pinky extended.	P	Illustration of P handshape in ASL.
Photograph of Brent’s hands both in S handshape with pinky fingers extended.	S	Illustration of S handshape in ASL.

The PE/RFR intrusion appeared inconsistently throughout both videos. It appeared to “turn on” and be incorporated into Brent’s signing for an extended period (from 3 to 59 sec), perseverating across signs within and across utterances, then it seemingly “turns off.” There is more of this phonological intrusion in the first half of V1 (28 of 63 utterances; 44.4%) compared to the second half (3 of 63 utterances; 4.8%). In V2, only 17 of 217 utterances (7.8%) incorporated this noncanonical handshape. In both videos, the phonological intrusion is noticeable to the viewer. However, we felt that, in general, the PE and RFR did not disrupt our ability to comprehend Brent’s signing.

We observed PE/RFR intrusion both in one-handed signs (e.g., better, spicy, drink) and two-handed symmetrical signs (e.g., agree, different). Signs produced with PE/RFR also revealed no consistent grammatical category or semantic pattern, for example, agree (v), better (adj), than (conj.), feel (v), and restaurant (n).

PE/RFR intrusion was not a consistent substitution for a specific handshape or more marked handshapes as we found the same signs produced without PE/RFR in other instances. For example, the same two-handed sign story was produced with the PE/RFR intrusion on both hands in one instance, but with a typical Flat-O handshape in two other instances.

On one occasion Brent appeared to show awareness of the PE/RFR intrusion and he effortfully changes his handshape to the expected handshape. When asked about the PE/RFR intrusion pattern, Brent’s mother felt that the onset occurred around the time of a depressive episode he experienced at around age 16. She also expressed that it happens more often when he is anxious. She reported that she can ask him to “sign it again more clearly” and he can sign the phrase again without the phonological intrusion. In Brent’s IEP (age 17;7) observers noted that he periodically uses “incorrect or irregular” handshapes.

3.3.3. Other articulatory issues

Separate from the PE/RFR intrusion, we noted that Brent had some difficulties articulating certain ASL handshapes. As discussed earlier with respect to fingerspelling, Brent displays some difficulty with the letters R, P, K, and Q. When these handshapes appear in lexical signs, for example restaurant (R) and personal (P), the formation was not always consistent with the canonical form: index and middle fingers were not always fully crossed in R (as illustrated in Table 4) and the thumb was not always tucked in between the index and middle fingers in P.

3.3.4. Facial expression: Affect and ASL grammar

An important part of ASL grammar is the use of specific facial expressions as grammatical markers; these expressions are produced simultaneously with manual signs and are distinct from affective expressions (e.g., surprise, happiness). Examples of grammatical facial expressions are raised eyebrows for Yes-No questions or mouth-open to indicate emphasis or intensity. Based on our research team’s many years of experience with nonautistic signers, our overall impression is that Brent shows reduced facial affect, both grammatical and affective.¹³

With respect to affective expressiveness, Brent occasionally smiles (e.g., when he signs embarrass). He also shows a big smile with his greeting at the beginning of the video and with his closing at the end, both of which use the i-love-you sign in a waving motion (this seems to be Brent’s “signature” greeting).

We observed that Brent sometimes uses enactment to convey bodily experiences. For example, he licks his lips while signing about the dessert s’mores, suggesting they are very tasty. He uses a whole-body lean and larger signing movement when talking about his reaction to spicy food. Enactment, or constructed action, is a linguistically acceptable construction in ASL (Quinto-Pozos 2007). In example (e), when talking about the barbeque wings mentioned in the previous utterance, he signs spicy, then stops signing and leans backward and forward with mouth agape as if having a bodily reaction to something spicy. His body movement then stops, and he completes the phrase with the sign drink. The body motion seems to function lexically as it holds a defined slot in the phrase, and it is expressed exactly the same way in the next repetition:

Table

(e)	English translation:
	“Spicy! Whoa! (Gotta take a) drink!”
	ASL Gloss:
	spicy [body-rocking] drink,
	spicy [body rocking] drink

On several occasions, Brent asks his nonpresent social media audience whether they agree, but the expected Yes/No question facial expression (raised eyebrows) was not observed. In one instance (agree), it seems he tucks his chin down, but his eyebrows did not raise up. He also asks his viewers directly “What’s your favorite holiday? Mine is Christmas” without using the canonical Wh-question furrowed eyebrows. We also identified Brent’s use of a rhetorical question, for example, Brent signs what eat, what eat (in context, the English translation was “What does one eat at this restaurant?”), which is followed by him signing i-tell-you (“I’ll tell you”), and he then lists his favorite menu items. In the rhetorical case, the expected nonmanual facial expression would be raised eyebrows with a slight head shake or tilt (Valli et al. 2011), but no clear facial expression is evident. In sum, for each of the linguistic contexts in which the use of a conventionalized facial expression to mark grammatical information is expected for ASL, Brent did not use the form.¹⁴

Interestingly, Brent uses a lexical sign for question (index finger, like the letter X, repeatedly bending) instead of facial expression as a marker of questions. There are languages that use such question-marking particles or lexemes, but English does not (unless one adds “huh?” or “right?” at the end of a sentence). Note the utterances presented in (f); in (f.1) Brent adds the question sign after a repeated series. In (f.2) the question sign appears at the beginning of the utterance and is included in each repetition; and in (f.3) the question sign is positioned at the end of each repeated utterance.

Table

(f)	1. than 2020, than 2020, than 2020 question
	2. question fascinated my favorite story,
	question fascinated my favorite story
	3. agree question,
	agree question

3.4. A comparison of Brent’s fingerspelling and handwriting

As mentioned earlier, Brent produces many fingerspelled words with great accuracy. One way to explore his English word knowledge is to compare his fingerspelling to his writing. Brent’s written and signed “Plan for the day” was collected at age 17;6. Table 5 reproduces, from left to right, the written document, including punctuation and layout; the ASL gloss of Brent’s signing as he reads from the document; his mother’s translation and commentary; and our observational notes regarding the comparison.

Table 5. Comparison of Brent’s handwritten and signed representations of the same semantic content.

Handwriting sample	Brent’s signing of the written text (ASL gloss)	Mother’s interpretation of Brent’s signing	Observational notes
PLAN-FRI	plan friday, plan friday	Plan for Friday	Phrases are written once, but signed twice
1. gV-To-gM A-gPA’S LIVe	fs-glenvale, fs-glenvale	We go for a drive to Glenvale,	City name (Glenvale) is fully fingerspelled, but abbreviated (gV) in writing
	fs-to grandma grandpa home, fs-to grandma grandpa home	which is where Grandma and Grandpa’s house is.	Common ASL signs (grandma, grandpa) are abbreviated gMA gPA in writing; gMA is split at end of line.
	next-down	Next,	Writes LIVe, but signs home (semantically similar)
	two, see, two	Second on the list	Uses discourse organizer for ASL (next-down), signing it only once.
2. BoR-BOL-To MCD.	fs-borge and fs-bolton home, fs-borge and fs-bolton home fs-to mcdonalds, fs-to mcdonalds	We take the route by the homes of Borge and Bolton (these are friends) on our way to McDonald’s from Grandma & Grandpa’s	Fingerspells t-o, reflecting his understanding of English structure of the preposition in the written phrase Friends’ names are fully fingerspelled, but abbreviated in writing
3. CARIBOU —- TO-AAA, BBB, CCC, DDD, EEE’S-LIVE.	next-down caribou, next-down caribou fs-to several home fs-lake fs-to several home fs-lake	Then we go to Caribou. Then we took another drive to see several homes (of friends) by the lake.	CARIBOU is fully written out, not abbreviated. He uses a local sign for caribou (same as deer). Going to Caribou coffee shop is part of Brent’s routine, thus, it may be a well-practiced written form. Five friends’ names are written with only first three letters of their name; the names are not included in Brent’s signed rendition; substitutes fs-to several home (by the) lake

Note. The writing sample includes capitalization and punctuation as represented. Personal names have been anonymized with preservation of the number of letters. The letters “g” and “e” are both lowercase, but written as large as the adjacent upper-case letters. Brent’s mother clarifies that under the 3rd item, he has written the first 3 letters of the names of several friends. We anonymized here by using AAA, BBB, CCC, DDD and EEE.

Brent’s written plan for Friday’s activities is organized as a list with numbered bullets. It includes commas, apostrophes, and periods. Most of the letters are uppercase, with some lowercase letters written quite large, comparable in size to the uppercase letters. Only the words PLAN, LIVE, CARIBOU, and TO are complete; all the remaining written words are abbreviated (e.g., GMA for grandma, MCD for McDonald’s). The middle-justified starting point of the text and the fairly large size of the letters causes him to quickly reach the right edge, sometimes resulting in breaking up the word to the next line.

Brent’s signing of his “Plan for the Day” is similar to his other videos, it includes repeated utterances, lengthy words that are fingerspelled accurately (e.g., fs-glenvale), and some discourse-organizing lexical signs that would be appropriate for describing a list, such as next-down. The dissociation between his signing and writing is evident in that Brent’s writing does not repeat phrases, but he does repeat phrases when he “signs out loud” reading from what he had written. He also uses lexical signs (grandma, mcdonalds) or fingerspells the complete name (fs-borge) of forms that he abbreviated in his writing.

An IEP from age 14;7 states that Brent has dysgraphia with no further elaboration; it also recommends occupational therapy support for handwriting and fine motor skills. The 17;7 observation record states that he does not like writing and that he needs significant encouragement to do it. As Brent enjoys exploring social media, typing and using emojis is another communicative method being encouraged by his parents.

With regard to reading text, we refer to data from an earlier study involving Brent (Bhat et al. 2018, Shield et al. 2022); here his performance on one particular task suggests that Brent, at age 8;5, could go from print to fingerspelling on a letter-by-letter basis. For this task, Brent saw a written English word presented on an iPad and was asked to fingerspell the word. Sometimes he would fingerspell and then produce the lexical ASL sign, indicating that he understood what the English word translates to in ASL. In other instances, he said that he did not know the meaning of the English word (e.g., father). Overall, Brent did well on this task at age 8;5 and generally was able to reproduce isolated printed words using fingerspelling. The IEP (14;7) also states that Brent has good decoding skills for text; even so, his mother reports that currently Brent cannot follow most captions on television and does not “read books for pleasure.”

4. Discussion

In this case study, we have presented the signing patterns of Brent, a 17;7-year-old Deaf autistic native signer of ASL. Brent’s case description makes an important contribution to the autism, deafness, and developmental-differences literatures. Prior studies have focused primarily on DHH autistic individuals who have limited or no expressive language abilities in either speech or sign. Brent has strong ASL abilities and identifies as a proud member of the Deaf Community, but also shows distinctive signing patterns in his ASL usage. We discuss several possible accounts for Brent’s linguistic patterns and explore whether these are ways of signing that can be explained by expected variation already recognized among ASL signers, or whether the patterns are unique to him, possibly influenced by his autism, Temple syndrome, and/or anxiety.

We first want to ensure that our readers understand how we frame Brent’s ASL distinctiveness and his overall communication abilities. We do not subscribe to a position that there is a “right way to be deaf” or a singular way to sign ASL (see Friedner & Block 2017 and Kusters et al. 2017, for a critique of the rigid framing of “normative” ASL and the “capital D” Deaf Culture). We recognize that the intersectionality of autism and deafness likely contributes significantly to Brent’s unique way of signing and we endorse Friedner & Block’s (2017) position that the “umbrella of the Deaf Community” should accept signing variations from individuals like Brent. In this discussion, we follow Prizant’s (2015) work which emphasizes the importance of understanding the root causes of what might be considered distinctive behaviors that are often an autistic individual’s strategies for coping with anxiety and emotional dysregulation. While we end this paper’s Discussion with consideration of strategies that may help to reduce Brent’s repetition and phonological intrusion, and promote Brent’s innovated strategies for supporting his communication (e.g., lexical substitutions for facial expressions), we would only advocate for such approaches if Brent, who is now 19 years old, expressed agency in desiring an intervention as it relates to his quality of life. In the absence of such agency, Brent’s community of supporters may start from the position of considering how environmental contexts may be contributing to these patterns.

To summarize, our analysis shows that Brent’s ASL narratives demonstrated strong ASL fluency and understanding of the world. His desire for social engagement and community connection is promoted by his family and the professionals in his sphere of support. This social connection is mostly realized with signing adults rather than peers. Brent’s focused interest in restaurants, hotels, and home improvement stores brings him joy and his parents create stimulating experiences that connect with these interests and that also introduce him to new topics (e.g., recent ones include elections and highway systems). Brent’s ASL strengths include sophisticated semantic content and vocabulary, metalinguistic awareness, and fingerspelling ability.

4.1. Repetition: What might underlie it?

In both videos we analyzed, Brent shows abundant repetition of his signed phrases (nearly every utterance is repeated three to five times in V1, and two repetitions are consistently documented in V2). His use of repetition occurs even when signing from written phrases that are only written once on the page. For some interlocutors, this dominant signing pattern may affect their comprehension of Brent’s message; it could also affect his social interactions, especially with peers. A signing pattern with four or five repetitions of each phrase could also reduce Brent’s opportunities for canonical turn-taking in ASL conversation. What might underlie his use of repetition?

4.1.1. Echolalia and other repetitive behaviors

The first repetition explanation we investigated is a known trait among some autistic individuals termed echolalia. Sign echolalia has been documented in two other studies of deaf autistic signers (Poizner et al. 1990, Shield et al. 2017a). In these reports, signers typically repeated the signed utterances of others, whereas Brent’s utterances are his own self-generated expressions. Scahill & Challa (2016) describe repetitive phrasing, but only cite fixed phrases (e.g., repeating a specific phrase spoken in a favorite movie). In Brent’s signing there are no such stereotyped phrases (apart from his signature “greeting” of hello i-love-you and the oft-repeated sign yummy, which was expected inasmuch as both narratives reflect his interest in favorite restaurant menu items).

4.1.2. Developmental language error

Next, we considered whether Brent’s repetition pattern could reflect a kind of child language error that has persisted into adolescence. Sign-internal repetition has been documented in the signing of young children and in child-directed signing (Holzrichter & Meier 2000, Meier et al. 2008, Pizer et al. 2011). Based on our analysis, Brent does not show a pattern whereby the number of movement cycles internal to a sign is increased.

4.1.3. Stuttering

Tetnowski & Donaher (2022) report that stuttering and autism can co-occur. However, the patterns of autistic stutterers do not always look like the stuttering of children who stutter but have no other diagnosis. For example, word-medial and word-final disfluencies are observed in autism, but not often in stutterers. Brent’s repetition does not appear to mirror any of these stuttering patterns (word initial, medial, or final) as he mostly produces complete repetition of full phrases. Moreover, in his fingerspelling, he exhibits no word-initial letter repetition, nor any other word-internal letter repetition. Based on our review of the literature, Brent’s full phrasal repetition is not widely attested as a speech disfluency pattern. One other speech disfluency disorder, termed “cluttering,” (Ward 2011) is related to rapid speech that impacts intelligibility. While Brent occasionally signs at a faster rate, we felt that his phrasal repetitions were intact and intelligible.

4.1.4. Stereotypy

We considered whether Brent’s phrasal repetition could be considered a kind of stereotypy that has been associated with autism, such as “stimming.”¹⁵ For autistic individuals, stimming (or motor stereotypy) can be a self-soothing behavior that provides relief from excessive sensory stimulation and/or anxiety (Charlton et al. 2021). In their study of motor stereotypies, Goldman et al. (2008) found that arm and hand/finger movements were recorded in 58% of 129 children with autism in their study. Thus, it is possible that repeatedly signing an utterance, with its large limb movements, could be functioning as a stim for Brent. Autistic adults report that they unconsciously engage in repetitive, rhythmic motor behavior (such as hand flapping) as a response to an overwhelming environmental situation, noisy thoughts, or uncontainable emotion (Kapp et al. 2019). In Kapp’s study, the autistic individuals interviewed were hearing; they reported instances of repetitive vocalizations (“vocal stimming”) such as muttering, grunting, or whistling. Still, we are unable to find any report identifying complete repetition of self-generated phrases as a vocal stim.

Brent’s parents report that he does have some stimming motor actions (eye/face rubbing, handwringing, and pacing) and vocalizations (e.g., grunting) that he seems to use to self-soothe when anxious or frustrated; this pattern is also documented in the IEPs. This shows that, even when the use of the manual modality is dominated by signed communication, physical or motor stimming is not precluded in an autistic individual. In one instance, Brent paused his signing to rub his eyes; he did not continue to sign while stimming. According to his mother, Brent does not typically sign while pacing, handwringing, or face rubbing. The IEP (17;7) and parent report suggest that Brent may be counting some of his repetitive behaviors. For example, he seems to be counting his laps while pacing, the number of fist-bump greetings he does, the number of pictures he takes with a cell-phone camera, or the number of times his mother says a good morning greeting.

In summary, known linguistic and repetitive behavior patterns related to autism do not seem to provide an adequate account of Brent’s pattern of phrasal repetition.

4.1.5. Perseverative patterns: Anxiety and obsessive-compulsive disorder

We now explore an explanation for Brent’s repetitive language pattern that is motivated by his psychological state. Kerns et al. (2014) report that anxiety is observed in approximately 40% of autistic individuals (though rates vary widely, from 11-84% of samples; White et al. 2009); research also suggests a strong association between emotion regulation difficulties and anxiety disorders in autistic youth (Conner et al. 2020). A second cluster of autism traits (sensory sensitivities, insistence on sameness, restricted/focused interests and repetitive behaviors) has also been associated with anxiety symptoms (Rodgers et al. 2012).

Brent receives treatment for generalized anxiety disorder. Parent interviews and his IEPs presented abundant information about his anxiety and how it impacts his daily life. According to his parents, Brent’s phrasal repetitions increase when his anxiety increases, until it becomes so overwhelming that he effectively shuts down, stops signing, and resorts to head shakes (refusals), grunts, and points. Across our two language samples, there is considerable variation in the number of times a phrase is repeated. The first video was variable in frequency number, but with a higher average number of repetitions; in the second video, Brent’s phrases were consistently repeated twice.

The elevated prevalence of anxiety co-occurring with autism has been attributed by some researchers to neurobiological differences that affect how some autistic individuals experience unexpected situations (intolerance of uncertainty) or other stressors (Herrington et al. 2017, Rodgers & Ofield 2018). Brent prefers sameness for morning and bedtime routines and can become upset if the routine is not adhered to. He tolerates changes if those around him give him clear notification of the alteration in advance. Brent has areas of intense focus, a trait often observed in autistic individuals (DSM-5-TR, Restrictive and Repetitive Behaviors, or RRBs). As noted, his favorite topics are restaurants (and favorite menu items), hotels, and home improvement stores. This kind of topic focus can be a repetitive pattern that serves to avoid areas of uncertainty, thereby indirectly reducing anxiety.

Obsessive-Compulsive Disorder (OCD) has been reported as co-occurring in up to 37% of autistic individuals (Leyfer et al. 2006). However, Kerns et al. (2014) argue that there is considerable overlap in the characteristics of OCD and the RRBs associated with autism, making it difficult to determine whether a dual diagnosis of OCD and autism is warranted. They suggest that there may be an atypical anxiety pattern that includes characteristics of OCD that is unique to autistic individuals. They explain that, for autistic individuals, compulsive rigidity makes the world feel “just right” and their repetitive behaviors create a positive valence, reducing anxiety around uncertainty or social interactions. By contrast, OCD often produces feelings with negative valence, including increased worry or catastrophizing if repetitions or behaviors are not followed. In sum, autistic individuals may have a unique anxiety pattern that should not be confused with OCD.

In considering whether Brent’s phrasal repetition is related to anxiety, we noted earlier that he also has perseverative or counting behaviors outside of linguistic contexts. And, as described above, he has some stimming motor actions (face/eye rubbing, handwringing, and pacing) that his parents and school staff report are ways that he soothes himself when anxious. Taken together, we find it plausible that Brent’s phrasal repetition and his counting pattern is consistent with the atypical anxiety pattern that Kerns et al. (2014) describe for autistic individuals. What needs further exploration is the novel idea that language in the manual modality may itself be a candidate for physical stimulation that is self-soothing (i.e., stimming) for an autistic individual. Repeating signs or phrases may be a kind of manual stim that provides relief during anxiety-elevated situations. Goldman et al. (2008) found that hand/finger stereotypies were more associated with autism, and less so for nonautistic individuals who have developmental language disorders.

4.1.6. An important linguistic clue: Brent’s repetitions are conditioned by discourse structure

In reviewing these accounts for Brent’s phrasal repetition, we must highlight an important linguistic finding. His repetitions are not “across the board” as his repetitions are conditioned by discourse structure. That is, certain linguistic structures are not repeated: list buoys (FIRST-LIST, SECOND-LIST, THIRD-LIST …), discourse connectors (next, and), and side-comments (yummy!). This finding suggests that Brent’s phrasal repetitions are not a pervasive language disfluency (e.g., stuttering). We also conclude that Brent’s repetition pattern does not align with specific unconventional language patterns that are associated with autism such as echolalia (Luyster et al. 2022; Shield et al. 2017).

4.2. Handshape intrusion: What might account for the intrusion of PE and RFR?

Brent’s second area of linguistic distinctiveness is the intrusion of unexpected handshapes into his signing. We identified two primary articulatory patterns: PE and RFR. How do we explain them?

4.2.1. Typical phonological variation seen in ASL signers?

Sign linguists have shown that natively-exposed signers will occasionally add an extended pinky finger to their signs, beyond those ASL handshapes that already have PE in their canonical form (5, B, F, I, Y). Hoopes (1998) found that PE predominantly occurred with signs that had certain handshapes (A, G/1, X, Bent-B, Bent-1, Flat-O) and maintained that PE does not occur in signs in which the middle finger is extended (V: see), unless all the fingers are extended, or in signs in which the fingers are all in a closed position (S: save). Hoopes also suggested that repetition (especially if a sign was associated with the topic of the conversation) could trigger PE. Brent’s PE handshape patterns do not align with Hoopes’ proposed phonotactic constraint; although he adds PE to signs that would be predicted by Hoopes to allow it, e.g., more (Flat-O) and different (G), he also produced many signs with PE that contradict the predictions (e.g., V: look-at; S: year).

Keane (2014) makes a different argument regarding PE in fluent signers. He proposes that this phonological variation is due to co-articulation with a preceding or following form that is inherently +PE. We identified multiple instances of +PE following a sign that did not have PE (e.g., think, in which only the index finger is extended, and other, in which only the thumb is extended), suggesting that co-articulation of preceding PE is unlikely the driver of Brent’s added PE.

With regard to the RFR that was prevalent in Brent’s signing, we are not aware of any RFR variation attested among native signers of ASL. Indeed, the RFR intrusion results in a number of handshapes that are not attested in the phonological inventory of ASL, such as the Claw-5+RFR, 8+RFR¹⁶, B+RFR, 5+RFR, and C+RFR handshapes shown in Table 4. In other words, these handshapes are not typically produced by signers of ASL, either in their canonical forms or as products of coarticulation or sociolinguistic variation. Indeed, these distinctive handshapes are, in our view, difficult to produce, suggesting that they are not the result of a lax style of signing or a failure to achieve a phonological target.

To determine if these handshapes are readily attested in other signed languages, we examined the handshapes included in the handshape font created by the Centre for Sign Linguistics and Deaf Studies at the Chinese University of Hong Kong. This handshape font includes all attested handshapes in a set of Asian signed languages (G. Tang, personal communication, January 6, 2023). We found that all of the handshapes included in the handshape font portfolio that include RFR also include the retraction of another finger, as illustrated in Figure 1.

Figure 1. Handshapes from the Chinese University Hong Kong font that include retracted ring finger (RFR). The first nine handshapes from left to right are attested in American Sign Language (ASL); while the last handshape on the right one is not productive in ASL, it does have specific usage in the southwest region of the United States related to cactus plants.

10 illustrated handshapes, from left to right: 3, Index+Pinky “Horns”, I-Love-You, U, V, R, L, Bent-3, U+Thumb, Cactus.

None of Brent’s handshapes with intrusive RFR shown in Table 4 are included in the handshape font, and all of the handshapes in the font that have RFR include retraction of at least one other finger (usually the pinky, but sometimes the middle finger, index finger, or thumb). All handshapes that include RFR in ASL also have retraction of the thumb (as in the number seven), middle finger (as in i-love-you), or the pinky (as in the number three).

In Brent’s case, it seems that he often adds either PE or RFR, or both, to arrive at an outcome handshape that has both PE and RFR. This is best illustrated by his fingerspelling of fs-wing (which canonically would have the W handshape that is -PE and -RFR). In this case, Brent adds PE and RFR with the resulting W substitution looking like the number seven. Brent’s mother also reported that when he signs the numbers two, four, and seven, they often look identical, thereby causing frequent confusion. This could be because he adds PE to the number two, and adds RFR to the number four, resulting in perceptual similarity to the number seven which already has PE and RFR (Figure 2). The uniqueness and pervasiveness of Brent’s intrusion does not align with either Hoopes’ or Keane’s phonological variation accounts of PE.

Figure 2. Four American sign Language (ASL) handshapes that are often rendered identical in Brent’s signing (all are realized as a 7 handshape) due to the intrusion of pinky extension (PE) and retracted ring finger (RFR). At times, this ‘7’ handshape is produced with index and middle finger close together.

Illustration of V handshape in ASL. Illustration of W handshape in ASL. Illustration of 4 handshape in ASL. Illustration of 7 handshape in ASL. Photograph of Brent’s hand in 7 handshape.

4.2.2. Related to a general language impairment?

Most cases of phonological intrusion come from the adult aphasia and dementia literatures, but there are some cases of children with articulation disorders that involve phonological substitution for gaps in their phonological inventories. Forrest et al. (2000) evaluated hearing children with articulation disorders who generated substitutions for sounds that were absent from their inventories. The children with consistent substitutions showed a stronger response to therapeutic intervention. In Brent’s case, his PE or RFR intrusion is not in response to a gap in his handshape inventory as we found examples of the same signs (e.g., better) with and without the intrusion. Leonard & Leonard (1985) discuss a case study of a child (age 4;8) who substitutes a “favorite sound” (a velar) only in phonological contexts where the child experienced articulation difficulty. In considering this possibility with Brent’s intrusion pattern, we could argue that PE and RFR are “favorite phonological features,” still we must contend with the fact that articulation difficulty was not identified with the specific phonological contexts where the substitutions occurred. That is, we observed PE intrusion with R handshape contexts (where he does have articulation disfluency) and with S handshape contexts where he does not. We conclude that Brent’s pattern is best explained by intrusion and does not align with substitution accounts provided by studies of hearing children with articulation disorders.

4.2.3. Related to autism?

There are several autism-related phenomena attested in the literature which could be at play here; we address each in turn.

4.2.3.1. Phonological development

A significant minority of hearing autistic children struggle with phonological development in spoken languages. For example, Rapin et al. (2009:75) found that 24% of 62 hearing autistic children ages 7-9 had “persistently and severely impaired” phonology. In general, studies have conflicted about whether such differences are a matter of a delayed, but typical, developmental trajectory (McCleery et al. 2006, Wu et al. 2020) or of idiosyncratic phonological development. For example, Cleland et al. (2010) studied a group of 69 hearing autistic children ages 5-13. Of these, 41% produced phonological errors, which included both typical developmental phonological processes (e.g., gliding, cluster reduction, and final consonant deletion) as well as idiosyncratic errors (e.g., phoneme-specific omissions of nasal consonants as well as initial consonant deletions).

Brent’s articulatory challenges with particular ASL handshapes (R, P, K, and Q) could be viewed as developmental inasmuch as these may be relatively complex handshapes, but his inconsistent PE/RFR intrusions and his ability to produce canonical form handshapes of the same (or similar) signs at other times, suggests this pattern is not developmental in nature. As mentioned, the intrusion pattern seems to have developed during adolescence. Interestingly, Brent does not show any evidence of palm orientation reversal, which is the primary form of phonological difference that has been documented thus far in signing autistic children (Shield & Meier 2012, Shield et al. 2020).

4.2.3.2. Neologisms

While the invention of novel linguistic forms (neologisms) is sometimes attested in autistic children (Kanner 1943, Rutter 1965, Volden & Lord 1991), and Brent’s innovation of abbreviations like “ffm” (Favor For Me) fits this pattern, his inconsistent usage of idiosyncratic handshapes does not seem to be a kind of neologism.

4.2.3.3. Stimming

As previously discussed with phrasal repetitions, we considered whether the phonological intrusions of PE and RFR might reflect a kind of involuntary body movement related to Brent’s autism (e.g., stimming). In these narrative samples, we only observed one instance of Brent pausing his story and engaging in face rubbing. However, on this occasion (V2) he is also signing off-the-point (as a way to say his eye rubbing isn’t related to his narrative); in this instance, PE is added to the sign (which uses a G handshape) and this intrusion perseverates while he rubs his eye. Thus, we are unable to determine whether PE and/or RFR would intrude if he just rubbed his eyes or face without signing. Additional language samples with concurrent stimming will help us to answer this question.

4.2.4. Related to a movement disorder?

We also considered nonautism related movement disorders, such as akathisia, that involve restlessness, compulsive or uncontrollable body movements (Lohr et al. 2015). This condition is not typically related to anxiety but coping with the persistent bodily movement can cause anxiety. Akathisia is associated with antipsychotic medications or drug withdrawal and mainly occurs with lower limbs; it is also associated with health conditions like Parkinson’s disease or Traumatic Brain Injury. Brent’s IEP does mention some difficulties with eye-hand coordination, but no specific examples were provided. In consideration of the above, we conclude that Brent’s persistent handshape intrusion, and repetition of phrases for that matter, do not seem to align with a movement disorder or medication-related explanation.

4.2.5. Related to Temple syndrome?

Brent does have some of the traits associated with Temple syndrome that are reported in the literature (Juriaans et al. 2022). For example, he has a shorter stature and small-for-age hands. It is possible that Brent’s articulation difficulty with R, K, P, and Q handshapes could be related to his smaller hands. One feature associated with Temple syndrome, clinodactyly, is characterized by an abnormally bent or curved finger, most typically the pinky finger, as a persistent configuration from birth.¹⁷ Brent’s PE appears intermittently enough that it seems less plausibly explained by clinodactyly, which is a permanent condition, and not observed in Brent’s case. The retraction of Brent’s ring finger, a highly marked configuration, could possibly be facilitated by joint hypermobility (documented in 67% of Juriaans et al.’s cases), but the evaluation records and Brent’s mother did not endorse this explanation. Even with smaller hands and joint hypermobility as possible contributors to some articulation dysfluency, we were not compelled by a physical/muscular account of the PE/RFR pattern of the handshape intrusion that we documented in Brent’s signing. That the intrusion “turns on and off” seems to be motivated by some factor other than purely muscular.

4.2.6. Related to anxiety?

We considered whether the pervasive PE/RFR phonological intrusions could be related to Brent’s anxiety. A connection between anxiety and stuttering is attested in the spoken language literature, although the relationship is characterized as complex (Ezrati-Vinacour & Levin 2004); nonetheless, we have been unable to find a source documenting phonological intrusion of the type we observe with Brent that is explained by anxiety. It seems that Brent has some awareness of the intrusion based on his ability to self-monitor and self-correct; the intrusion seems somewhat conscious because he selectively adds either PE or RFR, or both, depending on the target sign so that his final hand configuration is +PE and +RFR when the intrusion occurs. While PE/RFR could be something like a “favorite configuration” for Brent (cf. Leonard & Leonard’s 1985 description of “favorite sound”), it does not function as a substitution for a phonological gap.

4.2.7. An important clue: The onset of Brent’s intrusion co-occurred with a major depressive episode

After considering possible accounts related to language variation, disfluency, systematic substitution, autism, or movement disorders, none seems to provide a compelling explanation for Brent’s frequent and idiosyncratic PE and RFR intrusion while signing. It is interesting that the onset of Brent’s pattern of handshape intrusions may be contemporaneous with Brent’s major depressive episode at age 16. His mother firmly expressed the belief that somehow Brent continues the handshape substitution as a remaining element of this episode.

Because handshape intrusion does not happen all the time, and most people can still understand Brent’s message even with PE/RFR, it is not something that is being directly addressed in his current IEP. When there is a communication breakdown related to his idiosyncratic signing, people in Brent’s sphere of support may ask him to “sign more clearly” and, his mother reports, he usually can sign again without the PE/RFR. While there is some connection in the literature between anxiety and stuttering, we are unaware of an association between depression and phonological intrusion. Even so, Brent seems aware that he signs this way and the intrusion (either PE or RFR or both) may be a “favored” or “comforting” configuration.

4.3. Brent’s reduced facial expression

A third area of distinctiveness in Brent’s signing lies in his use of facial expressions. Signed languages use facial expressions for both affective and grammatical purposes, and it is our impression that both types of facial expressions were less evident in Brent’s signing. Many studies of hearing autistic individuals have found that they produce fewer facial expressions, and expressions of shorter duration, than nonautistic individuals; see Trevisan et al. (2018) for a meta-analysis of 37 published articles. However, very few studies have investigated how deaf autistic signers produce facial expressions. One early case study of an adult deaf autistic signer described a notable lack of facial expressions, despite native exposure to ASL from birth by her Deaf parents (Poizner et al. 1990).

More recently, Denmark et al. (2014) investigated how 13 deaf autistic signers of BSL (ages 9;0-17;0) comprehended facial expressions as compared to a group of 12 deaf nonautistic children and adolescents who were matched for chronological age and nonverbal intelligence. The deaf autistic children recognized significantly fewer emotional facial expressions produced during signing than did the nonautistic group. This work established that the face- and emotion-processing challenges documented in hearing, speaking autistic children extend to deaf, signing children. Later, Denmark et al. (2019) compared similar groups of deaf autistic and nonautistic British children and adolescents on their ability to produce emotional facial expressions while retelling a narrative based on their viewing of a video story that contained no dialogue. Adult native signers then rated the facial expressions produced by the children during their retelling as identical, similar, or missing compared to those of an adult signer model. Overall, autistic and nonautistic signers produced a similar number of facial expressions during their narratives; however, the autistic group produced significantly fewer expressions that were rated as “identical” to the adult model’s than did the nonautistic group. This work suggests that the facial expressions produced by deaf autistic signers are qualitatively different from those produced by typical signers.

In addition to our observation that Brent produces fewer facial expressions than would be expected, we also find that he seems to rely on alternate strategies to convey meanings that would typically be carried by facial expression. For example, he relies on a manual sign he produces at the beginning or end of an utterance (glossed as question) in order to signal that he is asking a question (as in example (f)), but does not raise his eyebrows, as would be expected for yes/no questions in ASL. This lexical strategy could reflect his understanding that he needs to signal that he is asking a question rather than making a statement, and also his difficulty with conveying that information through facial expression alone. We note that, in the conversational dyad video we have recently reviewed, Brent uses a full range of question signs (e.g., why, where, how), but our preliminary analysis finds that these are produced without accompanying grammatical facial expression; again, he uses the question lexical strategy frequently when asking questions of his conversational partner. Similarly, he sometimes uses enactment to convey bodily experiences (lip-licking to indicate tastiness; body leaning and larger signing to express intensity); these, too, may serve as alternatives to facial expressions. Brent may have difficulty interpreting the facial expressions of others (as suggested by Denmark et al. 2014) and may best understand those questions in which adults themselves use the question sign.

4.4. Dysgraphia can co-exist with fluent fingerspelling

Stone et al. (2015:8) argue that there is a predictive relationship between “the rapid and accurate decoding of fingerspelling and accurate decoding of printed words, suggesting a common underlying decoding, or fluency, skill that develops in different modalities.” In their study, fingerspelling decoding skill predicted reading fluency over and beyond ASL proficiency alone. Our comparison of Brent’s use of fingerspelling to his writing showed that this kind of predictive relationship may not be established in his productions.

Brent fingerspells frequently (22.6% of his sign tokens compared to the 10-15% average reported by Padden & Gunsauls 2003), can produce lengthy fingerspelled words, and is highly accurate with respect to what the target letter is (but is not always fully accurate in the articulation of certain letters from the manual alphabet). On several occasions, Brent fingerspells a concept when a known, common sign exists. However, Brent’s writing abilities are not commensurate with his fingerspelling abilities (see Table 5) suggesting a dissociation between these skills. For example, Brent shortens words that he can fingerspell (e.g., fs-glenvale becomes gV in his writing) or sign (e.g., grandma is written as gMA). He substitutes a different English word as a translation equivalent (e.g., he writes live when he signs home). We speculate that maybe, with his excellent sequential and visual memory skills, Brent has memorized many strings of “through the air” fingerspelled letters without necessarily establishing a connection with printed text.

It may be that Brent’s approach to fingerspelling learning has not followed Padden’s (2006) characterization of “learning to fingerspell twice” in which she documents that, for native ASL signing children, they first learn fingerspelled patterns as larger sign-like gestalts (#bus), not as individual hand configurations representing letters of the alphabet (B-U-S). Once they make connections between the fingerspelled letters and print, they develop the alphabetic principle and other concepts of spelling. Padden (1991) suggests that as early as age four deaf children of deaf parents explore the connections between fingerspelling and print, including fingerspelling to themselves while attempting to write and even attempts to create written characters that mimic the hand configuration of the fingerspelled letter. While we see many elements signaling Brent’s understanding of a relationship between printed words, fingerspelling, and sign equivalents, without more systematic testing of Brent’s production and comprehension of print, fingerspelling and sign vocabulary, we struggle to develop a comprehensive account of the evident dissociation between his written and signed forms. We see that, even at age six, Brent was able to do letter-matching from print to fingerspelled form and could correctly fingerspell words on a screen up to nine letters long, even when he did not know the meaning. But we also see that Brent relies on abbreviations in both fingerspelling (c-c-f for “Cheesecake Factory”) and writing (gV); perhaps this is a strategy he uses to ease the writing or fingerspelling load. Yet this is countered by the fact that Brent effectively increases his “spelling load” by choosing to fingerspell often (even when there is a common sign available), uses chaining sequences frequently, and even creates fingerspelled shortcuts for signed phrases (ffm for “favor for me”). Whether or not Brent considers audience design in choosing abbreviations is not known.

We also note that differences in handwriting quality and speed have been documented in some autistic individuals (see Handle et al. 2022, for a review). Some studies (e.g., Hellinckx et al. 2013; Johnson et al. 2013; Rosenblum et al. 2016) suggest that handwriting differences may be associated with autism-related challenges with manual dexterity, visual-motor integration, and other graphomotor skills. In Brent’s case, where his fingerspelling clearly eclipses his handwriting, further research is needed to understand the nature of the manual skills required for handwriting versus typing versus fingerspelling and how autism may interact with these specific abilities. In addition, Brent may be able to leverage pictures, including emojis or sign illustrations, to aid his comprehension and production of written systems.

We refer to another case study of “Adam,” a deaf natively-exposed ASL signer, where the association between fingerspelling and reading/writing was evident. Quinto-Pozos et al. (2017) found that Adam had a significant sequential memory issue which appeared to affect his decoding of any fingerspelling beyond 3 letters. This processing difficulty had a considerable impact on his ability to produce fingerspelled strings. Adam’s co-occurring dysgraphia (difficulty with reading and writing) was argued to be related to his difficulty with establishing a connection between fingerspelling and print. Interestingly, with his strong visual memory skills, Brent seems to be able to decode and memorize lengthy fingerspelled forms, but seemingly has not established a complete system of English spelling knowledge that connects to the print mode of English.

4.5. Considerations for supporting Brent

Rutherford & Johnston (2022:3) explain that the clinical application of the neurodiversity paradigm “is still in its infancy.” As researchers who are not autistic, but who hold a neurodiversity-affirming position, we have tried to provide a linguistic description of Brent’s distinctive ASL patterns. We are not advocating a specific clinical approach to reduce or change his patterns. We used strengths-focused framing before characterizing Brent’s linguistic distinctiveness. Like Prizant (2015), we emphasize the importance of understanding the root causes of repetitive behaviors such as anxiety and emotional dysregulation. Autistic adults who were told to stop stimming or who underwent unwanted interventions to reduce stimming in childhood report a strong sense of resentment (Kapp et al. 2019), reinforcing the idea that seeking to extinguish or reduce the behavior may be a misguided and potentially harmful goal.

Given Brent’s metalinguistic awareness and motivation for positive social outcomes, individuals in Brent’s support sphere may be able to glean his own position on his ASL distinctiveness and support his choices on whether and how his distinctive signing might be addressed, or not. More importantly, his support sphere can attend to the environmental contributors to Brent’s responses and collaborate with him on how to navigate those stressors, depending on situational needs, risks, and complexity (Rutherford & Johnston 2022).

Recognizing that common measures of anxiety were developed on neurotypical children and thus may not “entirely capture the essence of anxiety in ASD,” Rodgers & Ofield (2018:60) have led efforts to develop autism-informed scales and treatment approaches. For example, they suggest a modified Cognitive Behavioral Therapy (CBT) approach for autistic individuals that gives them autonomy and confidence in targeting their anxiety reduction. An important consideration is that Brent’s phrasal repetitions may be a form of self-regulation and he may need it to navigate daily situations. Rutherford & Johnston (2022) suggest using neurodiversity-affirming tools to gather information about how an individual experiences aspects of themselves and what makes them happy. If Brent’s phonological intrusions are directly related to his experience with depression, then those intrusions may continue to appear according to the status of his mental health; the intrusions may also be serving a kind of self-regulating function. As such, the intrusions may be helpful indicators to those around him to know when he is experiencing depression or anxiety.

Finally, we see that Brent has developed an adaptive linguistic strategy that works for him: a lexical sign as a substitute to mark questions (in light of his reduced facial expression). Furthermore, this adaptive strategy may indicate that Brent understands others’ needs and mental states. As such, strategies like this should be recognized, and even expanded on, as they are intrinsically developed (or selectively deployed from his input experience), and thus clearly meet Brent’s needs.

5. Conclusions

We have presented a case study of the language patterns of a signer with a constellation of attributes that have not previously been described in the literature: a natively-exposed, fluent ASL-signing, Deaf autistic adolescent. Our case-study approach with its linguistic focus illuminates one possible trajectory of linguistic development in autistic individuals. In particular, Brent’s strong ASL fluency is important since most studies of deaf autistic signers to-date have focused on children whose language level is well below what is expected for their age. We have highlighted how Brent has many strengths in his ability to sign ASL: he is able to produce lengthy and mostly well-paced signed utterances with advanced vocabulary; he shows evidence of metalinguistic awareness; he is a skilled fingerspeller; he uses advanced discourse structures such as list buoys to organize his utterances; and he has favorite topics about which he likes to express his opinions and convey his experiences. We also offered an interdisciplinary lens to our analysis of Brent’s distinctive patterns of repetition and phonological intrusion into his ASL.

As always, however, case studies are case studies. One must be careful not to generalize. The language samples we utilized were samples of convenience and were monologues; we did not conduct any language testing or elicitation. More case studies like this one are needed, as well as systematic studies of fluent deaf autistic signers.

5.1. Implications

We now discuss a few implications of this work for various stakeholders, including the research community investigating autism, clinicians, teachers, and the deaf and autistic communities themselves.

5.1.1. For linguists, psychologists, and other researchers

The distinctive patterns documented in Brent’s signing contribute to the research and clinical literatures on language in autism. Some of Brent’s patterns, such as phrasal repetition and intermittently phonologically intrusive handshapes, do not appear to have clear analogues in autistic individuals who use a spoken language; this finding prompts us to refine our theories of the different ways that language can interact with autism (or other contributing factors such as anxiety). Specifically, we introduced the heretofore unexplored notion that language in the visual-gestural modality may interact with autism-related motor patterns (stimming) in unique ways.

A recent UK study by Hodkinson et al. (2023), found that the frequency and pattern of autism-related traits were not significantly different between 65 deaf and 41 hearing autistic individuals (ages 2-18). Based on parent/caregiver report, ratings in certain domains (social interaction, verbal/nonverbal communication, RRBs) were found to be very similar across the groups. Hodkinson et al. (2023:10) suggest that given these similarities, “the language-related features of autism may be linguistic (i.e., can be signed or spoken) rather than specifically auditory.” The main area of difference they did find between deaf and hearing participants was in the quality of their peer relationships; deaf autistic individuals showed more difficulties in this area, presumably due to the greater communication barriers they face in social contexts. While this is a large study of deaf autistic individuals, it appears that only 24 of the 65 deaf participants used BSL. As described, the Autism Diagnostic Interview-Revised instrument (adapted by the authors for deaf populations) does not go into detail regarding aspects of sign language fluency (e.g., use of facial expression for grammar, rather than affect or social connection). It is not evident whether any of the deaf autistic signers in their study were as fluent in BSL as Brent is in ASL.

Finally, we add to the understanding of the relationship between fingerspelling and spelling knowledge with respect to print. That a signing individual can memorize and produce long strings of fingerspelled words, but still not be able to write them, suggests that dysgraphia can co-exist with fingerspelling accuracy.

5.1.2. For diagnosticians

On average DHH children are diagnosed significantly later than hearing children (Meinzen-Derr et al. 2014). Autism may be underdiagnosed in the deaf population because deaf children are a heterogeneous group who come from a wide array of language backgrounds, so clinicians may assume that delays in language could stem from the late language exposure experienced by many deaf children of hearing parents. Obtaining an autism diagnosis for DHH children is made more difficult than for hearing children because of the paucity of instruments that have been adapted for the signing modality, but there are recent efforts to adapt gold-standard instruments such as the Autism Diagnostic Observation Schedule-2 (ADOS-2) and Autism Diagnostic Interview-Revised (ADI-R) for use with DHH populations (Allgar et al. 2021, Phillips et al. 2022, Wright et al. 2022). Even so, challenges remain as few clinicians who are both qualified to administer these instruments and capable of communicating in a signed language (McCann et al. 2022, McFayden et al. 2023). Brent’s mother conveyed to us that, while she and his father were grateful to have received a fairly early diagnosis from autism experts, a remaining challenge throughout Brent’s education has been the need for more “Deaf-Plus” expertise and for appropriate educational and social support within the context of deaf education.

Describing how autism may present in a fluent signer, such as Brent, can help inform how diagnosticians approach the screening and identification of autism, and of other intersectionalities such as anxiety, in deaf children. While Temple syndrome is extremely rare, Brent’s collection of traits aligns with Juriaans et al.’s (2022) findings, based on their clinical study of 15 individuals with Temple syndrome, that 67% had joint hypermobility, 40% had autism-like behaviors, and 33% had OCD. Thus, it may be that Brent’s Temple syndrome contributes to and overlaps with OCD-like traits that may be characteristic of anxiety patterns associated with autism (Kerns et al. 2014), and could be the primary source of some of the phenomena documented here, such as the intrusive RFR handshape, which may be facilitated by joint hypermobility.

5.1.3. For teachers and interventionists

The finding of Brent’s alternate strategies for facial expression and using list buoys to organize his discourse is instructive for adaptive approaches to teaching these structures to autistic students. If facial expressions are challenging, alternative strategies exist, and these should be used to encourage language development and to scaffold the development of other skills.

Similarly, we wonder if the modality of sign affords possibilities for self-regulation through the body and limb movements of sign itself. Brent’s repetition of phrases may serve a regulatory function; he may derive a calming effect through the repeated production of signed phrases (and the limb movements needed to execute them) that is unavailable to autistic speakers. As such, we encourage teachers and interventionists to seek to identify potential causes of dysregulation and anxiety and, possibly, based on the individual, to view repetitive signing as a solution rather than a problem. For some individuals, as they feel more regulated, the repetitive behavior that serves the purpose of emotional regulation may also decrease naturally. Importantly, Prizant (2015:27) suggests that “seeking to eliminate behavior without fully understanding its purpose is not only unhelpful; it also shows a lack of respect for the individual. Worse, it can make life more difficult for the person with autism.” Joyce et al. (2017) illustrate how young autistic people can self-report and provide insight into their repetitive behaviors with thoughtful interview approaches.

5.1.4. For deaf autistic people, their parents, and their communities

Friedner & Block (2017:288) propose that “in combination, deaf studies and autistic studies together can result in a new way of understanding non-normative communication and embodiment.” They maintain that the continued negative framing of stimming and the perpetuation of language ideologies of normative ASL makes it difficult to “recognize and authorize signing that may be distinctive” (Friedner & Block 2017:293). In this case study, we identified distinctive strategies and solutions that were created by Brent, such as lexical question markers and embodiment forms, in place of “normative ASL” features. By highlighting these linguistic strategies, we validate “ways of signing” that are just as effective from a linguistic lens. In discussing Brent’s patterns of phrasal repetition and phonological intrusion, we attempt to de-stigmatize the outcomes to foster more acceptance of “non-normative ASL.” By exploring different accounts for these patterns, we shed light on possible approaches that could reduce the patterns, but only if Brent, or any such deaf autistic individual, seeks to embrace that outcome.

As a research team of nonautistic individuals, we are committed to a neurodiversity-affirming approach and are open to conversations with our research partners and community allies. While Brent may not be able to read our final manuscript, we have worked closely with his parents to ensure the accuracy of our interpretations. His mother shared with our research team the following comment: “I so appreciate that you see in Brent, what we see in him … .and are willing to do the work to show how his language is both unique and extraordinary.” Brent’s powerful use of ASL, including his distinctive signing patterns and creative solutions, is itself an important contribution to this research that we hope many will learn from. We have tried to pay close attention to our data and to our writing choices to promote neurodiversity-affirming interpretations and avoid ableist language. We encourage researchers working in this space to review Nicolaidis et al.’s (2019), AASPIRE Practice-based Guidelines for the Inclusion of Autistic Adults in Research as Co-Researchers and Study Participants; and Singleton et al.’s (2014, 2015) work on ethical conduct in research involving Deaf participants.

5.2. Future directions

Our case study provides an in-depth examination of Brent’s two ASL narratives. For future work, we plan to examine additional videos of Brent engaged in natural dyadic conversation. As our two language samples are both social media monologues, it will be important to document Brent’s language use across different communicative contexts; we expect the patterns documented here to be fairly robust as Brent’s parents have validated their presence across many settings they have observed.

We also have the impression that Brent’s usage of ASL classifiers and spatial verb agreement is reduced, possibly related to perspective-taking differences, or general difficulties with visual-spatial reference (as reflected in lower-than-average WISC score and parent report of challenges with map-like representations). We plan to assess Brent’s production and comprehension of classifiers and spatially-modified signs with more structured instruments as a complement to our analysis of his natural language samples.

We aim to use a more detailed analysis to investigate the solutions that Brent has perhaps innovated for the marking of phrase endings, such as body leaning or eye gaze shifting. Another signing pattern we observed that will require further analysis is Brent’s frequent activation (e.g., mirroring) of his nondominant hand during one-handed signs. And, finally, we may have an opportunity to examine historical video data (recordings of Brent signing as a child) to consider the precursors to, or trajectories of, the linguistic patterns documented in the present sample. We anticipate that this further work will broaden, and deepen, our understanding of Brent’s unique linguistic abilities.

Declaration of Interest Statement

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Acknowledgments

First, we thank Brent and his parents for allowing us to learn from Brent and hopefully making a meaningful contribution to him, to Brent’s sphere of support, and to the research and clinical communities working at the intersection of autism and deafness.

We are grateful for the expert feedback from our consultant, Pamela Wright, a Deaf ASL-fluent sign language researcher. She reviewed the video recordings used in this dataset and validated the patterns we documented. She also endorsed our interpretations and possible accounts of Brent’s patterns. She suggested future analyses to explore Brent’s use of pauses, body positioning, and gaze shifts as potentially important markers in his ASL signing– all of which we intend to pursue.

The handshape fonts were created by Centre for Sign Linguistics and Deaf Studies, Chinese University Hong Kong.

We thank the anonymous reviewers who offered us excellent commentary and suggestions on earlier versions of the manuscript. We also appreciate the thoughtful feedback we received from several generous colleagues: Juliet Goldbart, Ava Gurba, Jon Henner, and Justin Power.

Data Availability Statement

As the video recordings in this dataset are identifiable, they are not available for open access. Please contact the corresponding author for questions.

Disclosure Statement

The authors declare that they have no conflict of interest.

Additional information

Funding

Jenny Singleton received support from her Stony Brook University research funds for compensating the Deaf consultant on the project. Aaron Shield received support from NIDCD grant 1F31-DC0011219 and Autism Science Foundation Research Enhancement Grant 14-04. Kristin Walker received support from the National Science Foundation Graduate Research Fellowship Program under Grant No. (2234683). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, National Institutes of Health, or the Autism Science Foundation.

Notes

¹ We use the convention of capitalizing the term “Deaf” when referring to individuals, like Brent, who identify as members of the ASL-using Deaf community or when describing the Deaf community as a whole. We use lowercase “deaf” when referring to the diagnosis of hearing loss. We also use the abbreviation DHH to refer to a population of Deaf and/or Hard of Hearing individuals.

² Early language deprivation for DHH children has been associated with myriad health, mental health, and cognitive difficulties, in particular with theory of mind (Baron-Cohen et al. 1985, Hall 2017, Hall et al. 2017, Glickman & Hall 2018, Schick et al. 2007).

³ We recognize that the term “native signer” is contested in the literature (see Cheng et al. 2021 for this important discussion); in our case study, it is important that Brent has been exposed to fluent ASL signing language models from birth so that we can eliminate delayed or deprivation of language input as a potential contributing factor to differences in Brent’s ASL signing. Both of Brent’s parents are hearing, but acquired ASL from their deaf parents (Brent’s four deaf grandparents) from birth. In this paper, we use the terminology “natively-exposed ASL signer” rather than “native signer” to focus on the linguistic context beginning at birth rather than a label for the signer’s linguistic identity or fluency.

⁴ The evaluations use different terminology with respect to echolalia, and so we cannot be certain what type is evident. At age 17;7 the evaluation reports “patterns of echolalia that impacted the examiner’s ability to understand his messages”; at age 9, Brent’s ADOS report included code A3, immediate echolalia, noting “occasional echoing.”

⁵ We are fully aware of the challenges of establishing reliability and validity of psycho-educational and linguistic assessments used with DHH children (Haug & Mann 2008, Henner et al. 2017), DHH children with language disorders (Henner et al. 2018, Quinto-Pozos et al. 2014), and DHH autistic children (Allgar et al. 2021, Mood & Shield 2014, Phillips et al. 2022, Szymanski & Brice 2008, Wright et al. 2022). Many standardized instruments such as the WISC-V and ADOS-2 are normed on hearing children; thus, deaf norms are not available (but see Phillips et al. 2022 for progress on this front). Sometimes evaluators do not possess adequate ASL fluency to detect disfluencies present in child signers. We are unable to verify the integrity of the many assessments reported in Brent’s IEPs and related observations/reports.

⁶ We have followed Chen Pichler et al. (2010) in terms of the conventions we use to transcribe ASL signs, such as using English words in small capital letters to refer to ASL signs, using the “fs” prefix with italics for fingerspelled items (fs-hardrockcafe), using the # prefix (#PIZZA) when the fingerspelled word is lexicalized, and using the plus sign (DIFFERENT++) when the movement of a sign is repeated. Glossing does not include much information about form or grammatical markers.

⁷ There is difficulty establishing age-appropriate norms for ASL vocabulary knowledge as most instruments are designed for ASL acquisition in early childhood (Hoffmeister et al. 2022, Novogrodsky et al. 2014). For example, ASL-LEX 2.0 (asl-lex.org) provides acquisition norms for certain ASL vocabulary up to age 57 months. Sign equivalents for WRECKED, FASCINATED, EMBARRASS, OPINION do not appear in the ASL-LEX 2.0 corpus.

⁸ We clarify here that there are several ASL signs used in the community for the translation-equivalent of autism or autistic; in recent years, a new ASL sign representing the term neurodivergent has emerged. We do not know if Brent knows the English translation-equivalents of these signs; we do know that he corrected his mother’s ASL sign for the new one.

⁹ Fingerspelling generally does not indicate capitalized letters. While many of Brent’s fingerspelled items might be capitalized in print (e.g., Hard Rock Café), we keep all the letters lowercase in our examples to avoid an implication that he knows, or omits, capitalization.

¹⁰ This lexicalized fingerspelling of #pizza in ASL is illustrated as Variant 3 on this site: https://www.lifeprint.com/asl101/pages-signs/p/pizza.htm

¹¹ When Brent signed question he was using a lexical sign wherein the extended index finger bends at the second knuckle several times (similar to the fingerspelled letter ‘X’); this has the function of making the adjacent statement into a question. In the English translation, we used the word “right?” as the closest equivalent of a lexicalized question marker, but we do not assume Brent intends to elicit audience agreement as this translation might suggest.

¹² Here, we use the convention of writing FIRST-LIST, SECOND-LIST, etc. to distinguish these list buoys (where the signer uses their dominant hand index finger to point to the tip of fingers on the non-dominant hand) from lexical sign for FIRST, SECOND, etc.

¹³ Connected eye gaze with one’s interlocutor is also expected during signed conversations. Brent’s IEP evaluation at age 17;7 indicates that he struggles with maintaining eye contact and tends to look away from conversations at unexpected times. Our source video examples are not recordings of naturalistic, dyadic ASL conversation so we cannot directly evaluate Brent’s eye gaze patterns; however, we have observed a recent video of Brent in conversation with an adult family friend. While we have not yet completed a full analysis of this video, we observed that Brent maintains excellent conversational eye gaze behavior throughout.

¹⁴ While not yet fully analyzed, we can report that in Brent’s third video (where he “interviews” his adult family friend) he uses the following ASL question signs (which, why, what, where, when, how, how-old); there is no observable facial expression accompanying any of these question signs in the context where an interlocutor is physically present with Brent.

¹⁵ We use the terms “stimming” and “stim” without stigma and to promote greater academic acceptance of this term preferred by many in the autism advocacy movement. Kapp et al. (2019) and Charlton et al. (2021) discuss the movement’s efforts to reclaim the term “stimming” or “loud hands” (Bascom 2012) to counter social judgment of autistic behaviors that serve a soothing, regulating function.

¹⁶ To clarify, the 8+RFR handshape is distinct from what might be called a “Horns” or “Evil-Eye” handshape (index and pinky finger extended straight and middle and ring fingers are fully retracted and held in by thumb). In Brent’s case, the ring finger is fully retracted without being held in by the thumb; the thumb is still in contact with the middle finger.

¹⁷ His mother reports that when Brent was around age 5, he had surgery for a trigger thumb. This was before his Temple syndrome diagnosis. It is unclear whether the two are related; she reports that he can straighten all fingers when asked.