Abstract
Behavioral intervention programs for autistic children target a variety of goals including communication, social, academic, and daily living skills. A key component of ethical interventions is social validity, the extent to which interested parties find the program goals, procedures, and outcomes acceptable. However, past research indicates that social validity is rarely assessed from the autistic point of view. Thus, while intervention goals may be acceptable from parent or clinician perspectives, little is known regarding autistic individuals’ views of these goals. In this study, autistic adults were presented with common intervention goals and asked to provide written feedback on them. Though these open-ended questions were presented as an optional survey question, each goal received open-ended feedback from 120 to 189 participants (out of a possible 214 who completed the relevant survey portions). Responses were analyzed using reflexive thematic analysis by a team of nonautistic and autistic researchers. Four themes were identified: “Autistic Way of Being & The Double Empathy Problem”; “Supporting Autonomy, Self-Advocacy, and Interdependence”; “Implementation Matters,", and “Moving Beyond Behavior.” Findings will hopefully aid providers in developing intervention goals that are aligned with the perspectives of the autistic community.
Lay abstract
Intervention programs for autistic children aim to help children with communication, social skills, academic tasks, and everyday behaviors and activities. A major factor in ensuring these programs are ethical is “social validity,” which means ensuring that the program goals, procedures, and outcomes are deemed acceptable by those involved. Historically, the views of autistic individuals on these interventions have been overlooked; instead, researchers have focused on the perspectives of clinicians and parents. This study used a survey with open-ended questions to explore autistic adults’ perspectives on the acceptability of common intervention goals. Our research team, which included both autistic and nonautistic members, analyzed and categorized the survey answers into four themes: “Autistic Way of Being & The Double Empathy Problem”; “Supporting Autonomy, Self-Advocacy, and Interdependence”; “Implementation Matters”; and “Moving Beyond Behavior.” Findings aim to encourage providers to develop intervention goals that align with the perspectives of the autistic community.
Intervention and educational programs for autistic children hinge upon the goals that are created for them. Program goals cover various domains, such as communication, social interaction, academic attainment, daily living, and behavior reduction (Kurth & Mastergeorge, 2010). However, it has been argued that many goals of common autism interventions are ableist (i.e., discriminatory toward those with disabilities) in that they explicitly or implicitly focus on “normalizing” children to appear less autistic (e.g., Roberts, 2020; Robertson, 2010; Shyman, 2016). Such goals may teach autistic children that their natural way of being is considered wrong. Despite researchers and advocates highlighting the potential negative effects of this, there is a scarcity of empirical research into what determines an acceptable intervention goal from the autistic perspective.
Ableist intervention goals are those that encourage autistic individuals to act against what Sinclair (1993) described as their autistic “way of being.” Examples include forcing eye contact (Dalmayne, 2017; Trevisan et al., 2017) and prioritizing spoken language above other communication modes (Schuck et al., 2022a). Focusing on normalization can lead to masking (McGill & Robinson, 2021)—the suppression of autistic traits (Hull et al., 2017)—which has been cross-sectionally linked to poor mental health outcomes (Zhuang et al., 2023). Such goals are likely perpetuated due to the fact that goals are rarely evaluated in behavioral intervention research (Ferguson et al., 2019; Hurley, 2012), even though they are one of the three domains of social validity (i.e., acceptability of an intervention's goals, procedures, and outcomes; Wolf, 1978). Furthermore, despite Wolf (1978) having argued that intervention recipients should be “happy with our efforts and effects” (p. 213), autistic participants are rarely consulted on the development and implementation of interventions (D’Agostino et al., 2019; Hurley, 2012; Monahan et al., 2021). This lack of input has likely contributed to critiques that such interventions are unethical and harmful (Dawson, 2004; Mottron, 2017).
It is thus imperative that autistic individuals are asked for their perspectives on intervention social validity, including intervention goals. While it is clearly necessary to evaluate intervention goals from the perspective of intervention recipients, direct feedback from participants is not always feasible when the target population is young and/or nonspeaking. A viable alternative is to ask autistic adults to provide feedback from their perspective as stakeholders with lived experiences similar to those of the intervention recipients (and potentially also a history of themselves or other autistic family members experiencing similar behavioral interventions). Importantly, eliciting feedback from the autistic community is not only rooted in social validity, but also responsive to neurodiversity advocates, who—in addition to championing the idea that all brains should be accepted—emphasize the importance of the classic disability rights slogan, “nothing about us without us” (Hughes, 2016). When social validity is viewed through the lens of neurodiversity, it is clear that autistic people themselves must be at the forefront of intervention decision-making.
Social validity
Social validity refers to the social significance of intervention goals, the social appropriateness of intervention procedures, and the social importance of the intervention effects (Wolf, 1978). The concept was developed by researchers and practitioners in the field of Applied Behavior Analysis (ABA), which is the scientific discipline in which behavioral interventions originated. Behavioral interventions commonly employ reinforcement techniques to cultivate new skills in autistic children, both at home and in classrooms (Leach, 2010), making them one of the most prevalent forms of intervention (Monz et al., 2019). While teaching via behavioral contingencies may result in behavioral change, early pioneers in the field realized that showing improvement in “objective” outcomes was only meaningful and desirable insofar as stakeholders saw it as such (Wolf, 1978). The title of one of the foundational academic papers about social validity is indicative of this paradigm shift from only focusing on objective outcomes to considering the perspectives of clients, families, and other stakeholders: “Social validity: The case for subjective measurement or how applied behavior analysis is finding its heart” (Wolf, 1978).
Social validity is often listed as a criterion for determining whether an intervention meets the standards for an evidence-based practice (Horner, 2005; Reichow et al., 2011). Social validity consists of objective and subjective evaluation of intervention acceptability and feasibility. Assessing social validity can take many forms, such as questionnaires (see Carter & Wheeler, 2019 for a list of instruments), direct observation, normative comparison (Kazdin & Matson, 1981), data on maintenance of intervention procedures and outcomes (Kennedy, 2002), and qualitative interviews (Leko, 2014). Assessments can be made by both direct consumers (those who directly receive the intervention, such as autistic children and, if parent training is involved, their parents) and indirect consumers (e.g., other family members or peers; Schwartz & Baer, 1991). However, there is no standardization of social validity evaluation in the field, and researchers disagree about who the ideal candidates are for such assessments (Snodgrass et al., 2022).
Despite the concept of social validity being around for over half a century and the numerous methods for assessing the social validity of intervention goals, procedures, and outcomes, many articles do not report on it (Callahan et al., 2017; D’Agostino et al., 2019; Ledford et al., 2016, Snodgrass et al., 2018), especially with regard to assessing social validity of intervention goals (e.g., Ferguson et al., 2019; Hurley, 2012). In the case of early autism intervention, social validity assessment is further complicated by the fact that many participants are very young children and/or individuals with limited cognitive/communicative capacities who would therefore be unable to complete many of the procedures designed to assess the social validity of behavioral interventions. Thus, even when social validity is assessed, it is often measured by asking parents, teachers, or clinicians for their opinions as opposed to directly asking the participants (D’Agostino et al., 2019; Hurley, 2012). While these perspectives are informative and can provide useful information, it is nevertheless imperative to ascertain how the children themselves feel about interventions, as they are the stakeholder whom the interventions are most impacting.
Neurodiversity
The neurodiversity movement acknowledges the inherent brain differences within the human species and opposes the medical model which dominates in many Western healthcare and education systems. The medical model stresses disease/disability symptoms, centers perceived deficits, and sees disability as being internal to an individual and as something to be treated or corrected when possible (Marks, 1997). While suitable for conditions requiring a cure like cancer or epilepsy, this approach can be problematic when applied to other groups, such as autistic people.
Stemming from the autistic rights movement, neurodiversity spans all neurological variations and is not exclusive to autism (Walker, 2014). While there is no unified definition of neurodiversity among advocates, several key ideas are often recognized: (1) no brain type is “standard” or “right”; (2) neurodiversity should be recognized and celebrated; and (3) in addition to individual impairments, society plays a role in disabling neurodivergent individuals (den Houting, 2018; Dwyer, 2019; Walker, 2014). Advocates call for acceptance of neurodivergent individuals while also acknowledging the importance of offering necessary supports and services (den Houting, 2018; Dwyer, 2022; Robertson, 2010).
Though they recognize the importance of support, many neurodiversity advocates are fierce critics of behavioral intervention (e.g., Des Roches Rosa, 2020; Michael, 2018; Milton, 2020; Sequenzia, 2016). Two of the major criticisms of behavioral intervention put forth by autistic and nonautistic advocates are (1) the historical use of aversive stimuli, such as slaps and electric shocks, as punishment (e.g., Simmons & Lovaas, 1969; though it should be noted that such aversives are not entirely a thing of the past, as the Judge Rotenberg Center still uses electric shocks with their clients to enforce compliance and treat refractory “problem behaviors;” Blenkush & Cunningham, 2023; Neumeier & Brown, 2020; Perone et al., 2022), and (2) intervention goals that attempt to reduce autistic characteristics and pursue “normality.” Research has shown that self-stimulatory behavior, which is often a target of behavior reduction (Leekam et al., 2011), is recognized by autistic individuals as frequently involuntary, soothing for some, and unnecessarily stigmatized (Jaffey & Ashwin, 2022; McCarty & Brumback, 2021; Kapp et al., 2019). Emphasizing normalization has even led some advocates to label the intervention “autistic conversion therapy” (Sequenzia, 2016).
It has been argued that behavioral intervention can align with the neurodiversity perspective in theory (Schuck et al., 2022b). For example, Naturalistic Developmental Behavioral Interventions (Schreibman et al., 2015) prioritize child-led activities, play-based learning, and emphasize effort over perfection, apparently consistent with neurodiversity values. While support for some of these principles has been found amongst autistic adults (Schuck et al., 2022a), criticism of even these more naturalistic approaches remains (Autistic Self Advocacy Network, n.d.; Des Roches Rosa, 2020; Schuck et al., 2022a), with some autistic individuals voicing concerns that behavioral interventions may promote masking or lead to poor mental health outcomes (Cumming et al., 2020; McGill & Robinson, 2021). Thus, research on all aspects of social validity is imperative. An intervention might have acceptable procedures but if used for unapproved goals, its validity is undermined. A recent study found that autistic adults viewed certain intervention targets, like play skills and autism traits, as less appropriate for autistic children than did professionals and parents (Waddington et al., 2024). Similarly, while Chazin and colleagues (2024) found that ranking of goals by importance was similar across stakeholder groups, there were slight differences in the median ratings of such goals (e.g., the median rating by caregivers and practitioners regarding “Tolerating loud sounds” was medium priority, whereas autistic individuals generally found it to be a somewhat low priority). Such discrepancies underscore the importance of including autistic insights when designing and implementing interventions, necessitating further exploration into why there are differing views.
Current study
The current study aimed to directly address the lack of autistic perspectives on intervention and educational goals. Autistic adults were asked to provide written feedback on common intervention goals via an online survey. Participant responses were analyzed using thematic analysis (Braun & Clark, 2006, 2019) to answer the overarching research question: What do autistic adults think about common intervention goals for autistic children and why? While we initially considered analyzing and presenting themes at the level of individual goals, our initial analyses indicated many sentiments were repeated across goals, so we decided to look for themes more broadly. After analysis commenced, an additional research question was added based on participants’ responses: What improvements can interventionists and educators make to ensure goals are in line with the needs of the autistic community?
Method
This study was approved by the University of California, Santa Barbara Institutional Review Board (IRB #26-23-0047).
Participants
Participants were required to be ≥18 years old and identify as autistic (with or without a clinical diagnosis). Participants also needed to be able to read English and type responses. A convenience sample was recruited online via social networking sites (e.g., Facebook, Twitter, Reddit) and by reaching out to autism organizations to send the advertisement to their listservs. A variety of organizations were approached (e.g., autistic behavior therapist groups, groups with explicit anti-ABA stances) for a wide range of perspectives.
A total of 235 individuals participated (see Baiden et al., 2024), though this report focuses on 214 participants who answered at least one open-ended question about intervention goals. Participants were on average 34.86 years (SD = 11.27); 108 identified as female, 51 as nonbinary/genderqueer, 37 as male, and 18 identified as other or did not state gender. Most (n = 161) reported a clinical diagnosis of autism, while the remaining (n = 53) identified as autistic and did not have a formal diagnosis. The majority identified as White (n = 180). The remaining participants identified as mixed race (n = 6), Hispanic (n = 6), Asian (n = 4), Native American (n = 3), and Black (n = 2). Thirteen participants did not answer the ethnicity/race question or indicated that they preferred not to state their ethnicity/race. Of the 193 participants who answered the question about where they lived, most reported living in the United States (n = 135; 69.95%). The majority of non-US participants reported living in Canada, with others living in the UK, Australia, and Scandinavian countries (other countries were represented as well but are not listed due to potentially being able to identify those participants). Participants’ average score on the RAADS-14 (Eriksson et al., 2013), a measure of autistic traits, was 32.53 (SD = 6.61), well above the autism cut-off of 14. Two participants (both of whom reported clinical diagnoses) scored less than 14 but were not excluded (see below for more detail about the RAADS-14). Just over a quarter of participants (n = 62; 29.0%) reported having received behavioral intervention as a child; 105 reported not receiving any and 41 were unsure if they had or not.
Procedure
After providing consent, participants were directed to study questionnaires. Consent and questionnaires were all available via Qualtrics. Participants first filled out the RAADS-14 (Eriksson et al., 2013), a screening instrument designed to assess autistic traits. The RAADS-14 was developed to be a shorter version (14 versus 80 items) of the Ritvo Autism and Asperger Diagnostic Scale-Revised (RAADS-R; Ritvo et al., 2011). Items are rated on a scale of 0–3, with a total possible score of 42. In a validation study conducted with 1,233 participants with and without psychiatric diagnoses (including autism), it was determined that a cut-off score of ≥14 had 97% specificity in terms of correctly identifying autistic individuals (Eriksson et al., 2013). The RAADS-14 was included in the current study in order to characterize our sample; however, we did not consider participants’ scores when determining eligibility. We chose to use the RAADS-14 due to its brevity, fairly strong psychometric properties compared to other measures (Baghdadli et al., 2017; Sturm et al., 2024), and anecdotal evidence from two autistic authors that many in the autistic community prefer the phrasing of items on the RAADS to those on other readily available self-report questionnaires such as the Autism Spectrum Quotient (AQ; Baron-Cohen et al., 2001). After completing the RAADS-14, participants were directed to the study survey and asked to provide demographics at the end.
Survey
The data analyzed for this report focuses on feedback about common intervention goals (i.e., frequently identified intervention targets including both skill building and behavior reduction) for autistic children. All survey questions were developed by nonautistic and autistic individuals (authors RKS, KMBP, PD, and ZJW) based on personal experience with intervention services (e.g., KMBP has extensive experience as a Board Certified Behavior Analyst and is familiar with many intervention goals that frequently show up in clients’ service plans) and literature review of both academic and nonacademic sources. The authors strove to include questions about goals that were likely to be both acceptable (e.g., focusing on quality of life, Robertson, 2010) and unacceptable (e.g., reducing nondangerous stimming; Bascom, 2011) based on prior writings by autistic individuals.
The survey included 19 questions about intervention goals. Questions were phrased “[Goal] is a good intervention goal.” Participants were first asked to rate the degree to which they agreed with this statement using a Likert scale (this data is presented in Baiden et al., 2024). Participants were then asked, “OPTIONAL: Is there anything you’d like to add regarding your feelings toward this statement?” Participants could write as much as they wanted in the text-box or they could skip it entirely. This report focuses on participants’ typed responses to these open-ended questions. Exploratory graph analysis (EGA; Golino et al., 2020) of the participants’ agreement ratings generated a structure of three latent factors (termed “communities” in EGA) of goals (see Baiden et al., 2024 for more details regarding this analysis). The three communities were named Uncontroversial, Controversial, and Social based on the fact that the first community was composed of goals that were rated highly, the second was composed of goals that received low ratings, and the third had varying endorsement ratings, yet were all related to social skills (see Table 1 for which goals were included in each community). Though the survey requires further validation, the qualitative research questions for the current study are answerable regardless of the quantitative instrument's validity.
Intervention goals included in the survey organized by community structure based on Baiden et al. (2024).
Note. All statements were worded, “[Goal] is a good intervention goal.” *Indicates open-ended responses were analyzed in the current study. aEndorsement was defined as a participant indicating they somewhat agreed, agreed, or strongly agreed that it was a good goal. Adapted with permission from The Social Validity of Behavioral Interventions: Seeking Input from Autistic Adults by Baiden, K. M. P., Williams, Z. J., Schuck, R. K., Dwyer, P., and Wang, M. (2024). Journal of Autism and Developmental Disorders. https://doi.org/10.1007/s10803-024-06297-3. Table adapated and reprinted under CC BY 4.0 license (http://creativecommons. org/licenses/by/4.0/).
Data analysis
Participant responses were analyzed using Braun and Clark's (2006, 2019) approach to reflexive thematic analysis: (1) getting familiar with the data; (2) data coding; (3) generating initial themes from data; (4) developing themes; (5) refining and naming themes; (6) writing up the results in a report. This process is iterative, as work in one step may necessitate going back to a prior step.
Due to the large amount of data (over 100 participants responded to each question), the research team decided to only code and analyze responses to the two most and two least endorsed goals from each community (Uncontroversial, Controversial, Social; see Table 1). This data condensation (Miles et al., 2018) was warranted given that our research question was not goal-specific and instead aimed to understand autistic adults’ perspectives on intervention goals in general. Nonetheless, by choosing goals in each of the three areas that varied in endorsement, we felt that a wide range of participant responses would be represented.
All coding took place via Microsoft Excel. Coders included the first author (a fourth-year graduate student in education at the time) and undergraduate, postgraduate, and graduate research assistants (RAs). To start, coders independently read responses to the 19 statements, each taking notes in an analytic memo that included a summary, initial patterns, potential codes, and/or questions about the data (Miles et al., 2018). Coders met multiple times to brainstorm codes and practice applying them. Codes were applied at the individual response level. Multiple codes could be applied to different aspects of a single response, and some responses could have no codes. Codes were added, removed, and refined based on coder discussion. This process was reiterated until the coders believed codes adequately fit the data.
Next, the first author coded all responses. RA coders each reviewed the first author's coding of responses to 2–3 goal statements, highlighting areas of disagreement. The first author then reviewed the RAs’ double coding and resolved any discrepancies. The purpose of double coding was not to become “reliable”; it was to ensure that “the coding approach [wa]s collaborative and reflexive, designed to develop a richer, more nuanced reading of the data, rather than seeking a consensus on meaning” (Braun & Clarke, 2019; p. 594).
The first author then generated initial themes based on the codes. Themes were discussed and debated among the research team. Such peer debrief among people from different backgrounds and perspectives help strengthen the trustworthiness and credibility of our findings.
Author positionality
Six authors are or were doctoral students at the time of the study—three in education (including a Board Certified Behavior Analyst and a former special education teacher), one in developmental psychology, one in an MD/PhD program (with a clinical background in psychiatry and working towards a PhD in neuroscience), and one in school psychology. Two authors were psychology undergraduates. An additional psychology undergraduate helped with coding but is not an author. Authors’ research interests cover social validity of interventions for autistic people, understanding of neurodiversity, sensory processing in autistic adults, psychiatric care for autistic and neurodivergent individuals, educational experiences of autistic students, and mental health screening in education.
Four authors identify as autistic. Three authors have experience with naturalistic behavioral interventions for autistic children, two of the undergraduates provided peer support at a university autism center, and one author sees autistic patients in a medical setting. All authors support the neurodiversity movement and aim to reduce iatrogenic intervention effects.
Community involvement
Autistic collaborators were part of the entire research process. Two autistic authors were involved in survey development. The other two autistic authors helped develop the coding scheme and coded data. All four of them reviewed and revised the manuscript.
Findings
Though the open-ended question presented after each goal statement was labeled as optional, the majority of participants provided a response. The number of responses per statement ranged from 120 responses to Toilet training is a good intervention goal to 189 in response to Reducing motor stimming is a good intervention goal. Analysis of participant responses generated four themes: (1) Autistic Identity & The Double Empathy Problem; (2) Supporting Autonomy, Self-Advocacy, and Interdependence; (3) Implementation Matters; and (4) Moving Beyond Behavior.
Autistic way of being & the double empathy problem
A common thread across many responses was the idea that autistic people are inherently different—but not less—than neurotypical people. Participants referred to concepts such as quality of life or interpersonal skills in relation to being autistic (for example, “quality of AUTISTIC life—not NT [neurotypical] life”; “Autistic interpersonal skills will always be different than neurotypical interpersonal skills - and that's ok.”). Another example are intense/special interests 1 : a participant that disagreed with reducing fixations highlighted the importance of understanding that “the ASD kid likes deep and narrow and that's ok. And that neurotypical kids like general and nonspecific and that's ok.”
A common concern was that interventions could promote masking in autistic children. One participant explained, “It's difficult and depressing to be forced to emulate skills that are not typical for my neurotype.” Many were particularly concerned about improving sensory tolerance. Participants compared this to situations like: “Would you put a neurotypical child in a room with speakers blaring at full volume and force them to learn to tolerate it?” Another wrote: “This is physically painful, and we end up just having to pretend we’re doing ok so we don’t get in trouble.” A participant suggested interventions should help children “understand what helps increase their capacities and…reduce overstimulation.” Others believed such goals were inappropriate as they would lead autistic individuals to simply hide distress.
Participants emphasized that goal interpretation can differ, with acceptability related to alignment with an autistic identity, or way of being. They asked: “Who is deciding what interpersonal skills are, and how will that be measured? Is it just ‘to taste’, judged by a Neurotypical person?” and cautioned: “Be mindful of what you label as ‘inattention’ and ‘hyperactivity’—often natural autistic behavior (stimming, lack of eye contact) get labeled as that.” Amid concerns about professional misinterpretations, participants struggled to label goals definitively as “good.” They suggested that professionals might harbor biases due to misunderstandings about autism. One remarked: “I don't trust NTs to teach conversational skills to neurodivergent people in a respectful way.”
Across multiple goals, participants’ responses evoked the double empathy problem (Milton, 2012). One participant described how neurotypical people often make less effort to understand how to best interact with autistic people: “NTs need to learn the interpersonal skills to interact with autistics and treat us with respect, but they almost never do.” Observing that neurotypical people lacked respect toward autistic people led many to suggest that neurotypical people should learn more about autism and neurodiversity (e.g., “there should be just as much focus on the allistic (non-autistic) learning as there is on the autistic learning”). One participant drew a parallel between neurotypicals’ lack of understanding and masking: “Making autistics more palatable to neurotypicals by masking our autism doesn't do anything to help spread awareness and de-stigmatize autism…We should focus our attention on making neurotypicals less judgmental.” This kind of cross-neurotype education was also seen as a solution to improve breakdowns communication: Why is the autistic person the only one that needs to be flexible? Helping both autistic and non autistic people understand how others experience empathy and listening differently and using that understanding as a lens to be more accepting and supportive improves communication skills for everyone.
Some participants added that providers can explain neurotypical/societal norms and expectations to autistic children as long as they make it clear that they were not required/expected to conform. One participant suggested presenting such expectations as “some tools you'll find useful,” identifying education of neurotypical societal norms as “getting an extra set of stuff to use,” highlighting it as a choice. Another similarly suggested “fram[ing] these skills as ‘the things you need to learn to deal with neurotypicals’ and not ‘the things that every person is supposed to know to be normal.’”
Supporting autonomy, self-advocacy, and interdependence
Some participants questioned whether complete independence—accomplishing things on their own—was a reasonable goal for which autistic people should strive. Participants raised two related ideas: first, that autonomy—the idea that people should be able to have agency and choice in life—and self-advocacy—being able to ask for what you need—were more important than independence; second, that interdependence—relying on others—is an undervalued concept in Western society.
Some participants opposed viewing independence as a success metric for interventions: “The idea that disabled people are only successful when fully independent physically and that full independence means financially being able to support themselves is outdated and gross.” They noted that Western-defined independence is not feasible for all autistic individuals. Participants believed in promoting autonomy and self-advocacy for autistic people to access necessary support. This idea was elaborated upon by multiple participants: Allowing the autistic person to drive that [increasing independence] while being there to offer support in helping them to make those decisions (including helping them to make better informed decisions and helping them to have more tools/options to choose from). Attempts to obtain someone else's idea of independence often leads to poor mental health and burnout as we hit adulthood… Independence comes from autonomy- the ability to think and act for one's own self.
Some emphasized that nobody is truly independent and that interdependence/support from others is an important part of the human experience: “We are all interdependent and expecting people with disabilities to go it alone is silly…I would prioritize choice and autonomy”; “Even the neurotypicals I know aren't completely independent…Independence is a spectrum.” Many respondents believed that practitioners should not deny autistic individuals support when requested, even if they have previously not required support. One said, “Be careful about the ‘you can do this, so you should’ trap—something that's exhausting without support shouldn’t be done without support all the time.” Another participant echoed that, “Codependence is a reality for many autistic people and should be accepted and embraced. Those with more needs are right to accept support where they need it and don’t deserve to be stigmatised for doing so.” Another further reflected on how focusing on independence as the ultimate goal could inadvertently promote stigma: It's okay to rely on others and need support—even very high levels of support—it doesn't make anyone less deserving of life or respect or fair treatment. Saying promoting independence is a good goal is inadvertently saying those without independence are somehow defective; this is clearly not the case.
Implementation matters
While some goals were clearly controversial (see Table 1 for the quantitative ratings), participants often brought up the notion that a goal might be acceptable dependent on the definition of the goal (as discussed above)—but also on the specific intervention techniques employed: Improving conversational skills as a goal is not exactly the problem. The approach to improving conversational skills is the problem. Does that mean forcing a child to constantly verbally orally confirm what it is they want when their body language and other communication is obvious?
Participants provided several suggestions for providers to address intervention goals without causing harm. First, alter the environment and provide accommodations (“allowing the child to move instead of staying seated, for example”) instead of focusing on “unwanted” behavior. A participant explained the importance of considering toileting accommodations: Traumatizing a child into using a toilet when they have unknown co-occurring issues that make it difficult (for example, proprioception difficulties making it impossible for the person to sense the need to eliminate) when accommodations are readily and easily available is needless traumatization.
Some participants endorsed taking a strengths-based approach with autistic children. They emphasized allowing children to pursue their passions (“If they can do things they like to do, they will feel joy and life will feel worth living”) and utilizing these interests in achieving their goals (“Do we want to teach people compliance…or do we want to empower people to work with their neurology to their maximum potential?”). One participant encouraged, “If they like trains, try to creatively incorporate trains into the lesson.” Some participants described special interests as assets to enhance social, executive functioning, and coping skills amongst autistic people. Others pointed out that they were able to make a career out of their “fixations.”
Moving beyond behavior
Participants were clear that providers need to look past observable behavior to understand how to best support autistic people. Many mentioned trying to understand the root cause of the target behavior when implementing an intervention. This was especially true for reducing dangerous behavior and toileting goals. For example, in response to the statement about reducing dangerous behavior, a participant stated: “Yes [it's a good goal], but the underlying reason also needs to be addressed. No one runs into the street for the sheer fun of it, there's usually an underlying sensory or communication issue.” Another responded about toileting: “Difficulty with the toilet is 99% of the time sensory, not behavioral. Don’t approach a sensory problem with behavior therapy.” Participants encouraged providers to consider “What is happening at the school that has this child in fight-flight-or-freeze so often that they have to run into the street to get away? Maybe work on that and the kid will stop running.” The consideration of root causes is based not only in respect for people's inner thoughts and feelings but also in practicality—the interventions simply would not be as effective if only behavior was considered. A participant explained how, without identification of root causes, developing positive coping strategies would be difficult: I have unintentionally put myself in dangerous situations because I was overstimulated, in meltdown, and physically incapable of knowing where I was and making rational decisions…Telling me in the moment…isn't going to reduce that behavior because in the moment, it isn't a rational choice. Help me identify and reduce the impact of stressors that would cause me to do such a thing.
In response to other goals, participants pointed out that outward behavior can often be misinterpreted. Such misinterpretation can lead providers to target behaviors that are functional and/or nonharmful. Many participants saw stimming as a coping/self-regulation mechanism and/or as communication. One participant stated: “Stimming helps me regulate, if you want me to stop you'd better give me a different way to do that…I have tried not stimming and it just means that I'm anxious and unproductive for the next hour.” Another echoed: “Removing an autistic person's coping mechanism is just cruel, and will likely result in increased anxiety and/or a meltdown.” Some identified stimming as an expression of joy, stating that “If I'm waving my arms this is an expression of emotion; it is a smile I make with my hands.” Reduction of stimming could have detrimental effects that are likely antithetical to most intervention goals; a participant likened it to “trying to eliminate the very thing you would be claiming to want to improve simply because it doesn't fit someone else's preconceived ideas of what ‘happy’ looks like.” The same participant also shared that her “daughter is very clear and tells me verbally that when she flaps her hands it means she is happy.”
As with stimming, many participants also viewed “fixations” as a source of joy, anxiety management, or communication. They viewed “reducing fixations” as a harmful goal, as it takes away the mental and emotional safe space that special interests or special objects create. One response expressed that having the ability to share special interests with others “provides connection, communication opportunities, self-esteem, self-worth, joy (and likely some happy flappies), peace, feeling seen/heard/loved, and so much more.”
The other behavior that appeared to often get misinterpreted by providers was the need to move around (both in terms of reducing hyperactivity/inattention and increasing school readiness). Participants felt that motion was often misinterpreted as inattention, when movement could actually facilitate learning for autistic people. A participant explained: “When I'm the most engaged, I often look the most tuned out. I get concerned when ‘paying attention’ turns into a set of expected behaviors that might be more for show.” Another felt similarly: “A lot of what was labeled inattentiveness were actually things I was doing to be able to pay attention.” Instead of discouraging movement, participants suggested educators allow students to learn in ways suited to them: “Children need to be allowed to stim (including non-disruptive walking and vocal stimming) in stressful situations, like classrooms.” One respondent pondered: “Why can’t they have ‘being able to move about’ as part of their IEP [Individualized Education Program] when they go to school?”
Discussion
This study aimed to answer the following research questions: What do autistic adults think about common intervention goals for autistic children, and why? What improvements can interventionists and educators make to ensure future goals are in line with the needs of the autistic community? Intervention goals were clearly an important topic for the autistic adults in this study, as evidenced by the number of participants who elected to respond to the optional open-ended questions. Thematic analysis of responses generated four themes related to intervention goals for young autistic children: (1) Autistic Way of Being & The Double Empathy Problem; (2) Supporting Autonomy, Self-Advocacy, and Interdependence; (3) Implementation Matters; and (4) Moving Beyond Behavior. These themes are discussed below with particular attention to recommendations to improve intervention social validity.
Autistic way of being, autonomy, interdependence as guiding concepts
Autistic Way of Being & The Double Empathy Problem and Supporting Autonomy, Self-Advocacy, and Interdependence highlight concepts that must be central to any intervention or education program. Participants expressed concerns about the costs of masking when forced to pursue goals that were antithetical to their neurotype, a theme echoed in previous research (e.g., Cumming et al., 2020; Mcgill & Roberts, 2021; Schuck et al., 2022a). Aligned with the findings from a recent study on community perspectives in Australia and New Zealand on intervention goals (Waddington et al., 2023), this study found that autistic adults do not believe goals that reduce autism characteristics such as stimming or focused interests are appropriate or should be prioritized. Interventionists would benefit from seeking out information about autistic identities (e.g., Davies et al., 2023) to broaden their view of what it means to be authentically (and successfully) autistic.
Participants highlighted the double standard facing autistic individuals, whereby they are expected to align with neurotypical norms, a reciprocity seldom expected of neurotypicals (Milton, 2012). To remediate this lack of reciprocity, participants advocated for neurotypical professionals to deepen their understanding of neurodiversity and autism. Such understanding impacts the social validity of intervention goals, as what is acceptable to autistic individuals is often misconstrued by professionals. The autistic participants in this study endorsed goals that specifically enhanced autistic quality of life (see Milton, 2020 for a similar point). Some proposed that professionals should teach neurotypicals that conformity by autistic people is not obligatory or necessary, even if they do sometimes learn about neurotypical norms. Echoing this, a participant in Cumming et al.'s (2020) study valued learning about “ways to respond normally” but also “would have liked to have someone explain to me the reason why these interactions were necessary” (p. 78). This strategy illustrates the benefit of recognizing neurotypical-autistic differences respectfully, without imposing potentially detrimental expectations.
Finally, pushing for clients to become autonomous self-advocates was stressed in many responses. While autonomy and self-advocacy are generally seen as important skills for all children to develop, participants were clear to differentiate autonomy from independence. According to our participants, autonomy and self-advocacy should coexist harmoniously with whatever level of support is needed. This is related to the idea of relational autonomy (Davy, 2015) and must be applied when working with autistic children, especially given that autistic individuals are likely to display a spiky developmental profile (Doyle, 2020) that can fluctuate day by day (den Houting, 2018). Ultimately, it may be socially valid to work on a specific goal with a child, but if a provider insists that the child display a skill independently all the time, the social validity may decrease.
Implementation matters and moving beyond behavior: practical take-aways
Participants felt conflicted about certain implementation methods potentially undermining acceptable goals, a sentiment reflected in the Implementation Matters theme (see Waddington et al., 2023 for a similar finding). They expressed concerns about negative side effects from interventions, reflecting others’ descriptions of potential harms (e.g., Des Roches Rosa, 2020; Michael, 2018; Milton, 2020; Sequenzia, 2016; Williams, 2018). This is particularly worrisome given that autism behavioral intervention research has done a poor job of assessing adverse events during clinical studies (Bottema-Beutel et al., 2021; Dawson & Fletcher-Watson, 2022). Participants were also concerned that, while some goals might be socially valid, they are only acceptable insofar as the professionals actually address the root problem, in line with the Moving Beyond Behavior theme. Functional analysis procedures (Hanley et al., 2003) used to understand environmental triggers and reinforcers of behavior can be helpful, but it is crucial to interrogate why something is triggering or reinforcing; that is, take the individuals’ emotions and motivations into consideration (Bearded Behaviorist & Naturally Effective Behavior LLC, 2023).
Participants suggested improved intervention implementation by providing appropriate accommodations and environmental changes, offering alternatives, and adopting a strengths-based approach. Despite widespread agreement on the importance of these principles, these approaches are not consistently put into practice. For example, a recent study revealed that accommodations typically miss physical environment and quality of life concerns for autistic populations (Leifler et al., 2021). Additionally, providers may have a tendency to excessively modify behaviors that are not harmful to the individual or others around them. Last, while strengths-based programs are beneficial (e.g., Lee et al., 2020, 2024), lack of environmental adjustments as well as lack of acceptance have been identified as barriers to autistic children being able to capitalize on their strengths (Clark & Adams, 2020).
To combat these issues and ensure interventions are socially valid, professionals may benefit from taking an approach that emphasizes strengths-based learning with listening to the child's needs and desires. One such approach is to utilize Universal Design for Learning (UDL; Meyer et al., 2014) along with design thinking (DT; Stanford d. School, n.d). UDL provides a way for educators and interventionists to design learning opportunities accessible to all children, while DT provides an explicit process for ensuring whatever is designed actually meets the user's needs (Lambert et al., 2021). By starting from a point of empathy (a core component of DT), social validity is likely to increase.
Limitations and future directions
Though this study sheds light on how autistic adults feel about common intervention goals for autistic children, several limitations require discussion. First, while large for a qualitative study, we utilized convenience sampling to obtain our data. This method of sampling may result in a sample that does not accurately reflect the diversity of experiences within the autistic community. It also means that the study may be biased towards individuals who have a particular interest in our research topic or who share certain characteristics that are not generalizable to all autistic people. The participants were also predominantly female-identifying and White. Thus, some of the findings may not apply as well to males or other racial/ethnic groups and may also fail to capture the intersectional experiences of autistic adults who do not identify as female or White. It has been argued that, currently, truly “un-masking” is only a realistic, viable option for some White autistic people, as displays of autistic traits in Black and brown people may lead to harm due to the color of their skin (Cerda, 2023; Hammond, n.d.). Hammond (n.d.) suggests that Black people are always enacting ABA as they navigate a world built on white supremacy. In this view, ABA is simply a “symptom” of a broken system that simultaneously serves Black people—since altering their behavior can save their lives—and oppresses them. While the current findings regarding acceptance would likely still hold for minority groups, it is possible that some findings, such as concerns about masking, may present differently in other groups. Similarly, masking may have been particularly important to the participants in this study since most identified as female or nonbinary/genderqueer. Since studies have shown that masking is more common in autistic women and nonbinary individuals as opposed to men (Cook et al., 2021), it is possible autistic men might focus on other aspects of intervention goals. Also crucial in future research is the recruitment of more participants who received behavioral intervention as a child, as well as the elicitation of perspectives of those currently engaged in such programs. Additionally, although many advocates argue that perspectives of speaking autistic adults can still shed light on issues that are relevant to nonspeaking individuals (Des Roches Rosa, 2019; Thornton, 2021), our sample only consists of individuals who were able to read and type, and we assume that few of them are nonspeaking (especially given findings that most minimally verbal autistic children/adolescents also have significant challenges with receptive language, including reading; Chen et al., 2024). Thus, more work must be done to assess the perspectives of nonspeaking/minimally verbal autistic people.
Notably, 11 participants indicated in their responses that they disliked the word “intervention.” While the survey questions were designed to apply to services beyond behavioral intervention, it is possible that “intervention” may have made them feel defensive. The word “intervention” could have deficit-focused, negative connotations, as it could imply that there is a need to intervene upon (i.e., remove or remediate) autistic traits or fundamental parts of an autistic identity, which goes against the principle of neurodiversity that there is no “correct” way of being. The term may also elicit memories of past ethical breaches of trust within the field of autism interventions. It therefore may be preferable in future research on this topic to use other language, such as “supports” or “educational approaches.”
Last, before any analysis took place, we condensed our data such that only responses to 12 out of 19 goal statements were analyzed and reported. While likely that we have reached/are near saturation given the vast amount of data that was reviewed, it is possible that new perspectives/themes would be uncovered in the unanalyzed portion of the data. However, the generation of any new themes would not invalidate the ones generated in the current analysis.
Conclusion
In this study, autistic adults provided feedback on common intervention goals for young autistic children. They highlighted several areas where professionals could help autistic children obtain a higher quality of life, namely: understanding and emphasizing an autistic way of being, encouraging autonomy and self-advocacy, using appropriate intervention procedures that do not cause disproportionate psychological or physical harm, and looking beyond outward behavior to help solve root causes. Findings suggest that some intervention goals are likely to be unacceptable to a large proportion of autistic people (e.g., reducing stimming). On the other hand, identifying acceptable goals requires far more nuance and careful consideration. Nevertheless, two keys to ensuring goals are delivered in a socially valid manner are to adopt the neurodiversity perspective and embrace autism acceptance.
Footnotes
Acknowledgements
We would like to thank all of the participants who took the time to participate in this study.
Author note
Rachel K. Schuck is now affiliated with the San Diego State University Department of Administration, Rehabilitation, and Post-Secondary Education. Alicia Geng is now affiliated with the University of California, Santa Barbara Department of Counseling, Clinical, and School Psychology. Patrick Dwyer is now affiliated with the La Trobe University Olga Tennison Autism Research Centre (OTARC). An earlier version of this work was presented in the first author's dissertation.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: While none of the authors expects to directly profit from publication of this paper, we aim to be maximally transparent in our potential conflicts of interest as suggested by 
. RKS and KMPB are both certified in Pivotal Response Treatment, and KMPB is also a board certified behavior analyst who provides PRT at a community agency. PD and ZJW both serve on the autistic researcher review board of the Autism Intervention Research Network on Physical Health (AIR-P), and ZJW is a member of the family advisory committee of the Autism Speaks Autism Learning Health Network Vanderbilt site. ZJW also serves as a consultant for Roche on multiple projects related to autism intervention and clinical trials.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.