Autistic Adults’ Priorities for Social Media and Mental Health Research: A Coproduced Delphi Consensus Study

Abstract

Background: Social media has been linked to a range of harms and benefits to the mental health and well-being of users, including Autistic social media users, who—some have argued—may be particularly sensitive to both positive and negative effects of the technology. Despite this, there has been relatively little research into the links between social media and mental health in Autistic people, and even less is known about the research priorities of Autistic people themselves.

Methods: We conducted a three-wave Delphi study to identify Autistic adults’ priorities for research into social media and mental health. We also sought to explore community perspectives on ways to undertake research that maximizes impact, relevance, and reach, using thematic analysis of free-text survey responses.

Results: Twenty-one participants completed all 3 stages and identified 29 research priorities (out of 131 initially generated) that reached high levels of endorsement and consensus. While there was some interest in the potential harms of social media, overall, participants prioritized strengths/benefits-focused research. Specifically, there was interest in understanding how social media might be used to connect people—to reliable information, interests, hobbies, peers, and communities; its potential to facilitate self-soothing and emotional regulation as well as if/how social media might be used to reflect and cultivate a positive sense of self and identity in Autistic people. Finally, there was a strong emphasis on undertaking research that can be translated into concrete, practical benefits, including for under-represented, minoritized Autistic populations and those with intersectional identities.

Conclusions: The findings highlight the importance of prioritizing strengths-based research that reflects Autistic adults’ perspectives and investigates social media’s potential to support connection, emotional well-being, and positive identity. Ultimately, it is hoped that a focus on these priorities will translate into practical benefits, particularly for under-represented and intersectional Autistic communities.

Community Brief

Why is this an important issue?

Social media plays a central role in how people communicate, share information, and build communities. For Autistic adults, social media can offer valuable opportunities for connection and self-expression, but also brings challenges, such as the potential for misinformation and cyberbullying. Despite its growing influence, there is little research focused on how social media affects Autistic people’s well-being and even less on what Autistic adults themselves think are the most important questions to explore. Understanding these priorities is essential for guiding research that truly reflects what matters to the Autistic community.

What was the purpose of this study?

The purpose of this study was to find out what Autistic adults see as the most important research topics related to social media and mental health. The researchers also wanted to learn how future research could be designed to have the greatest impact for Autistic people.

What did the researchers do?

The researchers brought together 21 Autistic adults who took part in three rounds of surveys, with opportunities to meet and discuss their ideas with the research team and other participants. This process helped the group to share opinions, reflect on each other’s ideas, and reach agreement on what they thought were the most important research priorities.

What were the results and conclusions of the study?

The group identified 29 top research priorities. While participants mentioned possible harms of social media, they focused mainly on research about strengths and benefits. They wanted to know how social media can help Autistic people connect with reliable information, peers, hobbies, and supportive communities. They also wanted to know how social media might aid emotional regulation and self-soothing and how it can support a positive sense of identity. They particularly wanted to see research leading to practical, real-world benefits, especially for Autistic people from minoritized groups.

What is new or controversial about these findings?

This study differs from much previous research, which has mainly focused on the risks of social media for Autistic people. The findings highlight a clear desire for strengths-based research recognizing social media’s positive potential, as well as Autistic users’ capacities to shape their online experiences. It also stresses the importance of coproducing and collaborating with Autistic people directly in all stages of the research process.

What are potential weaknesses in the study?

The study included a relatively small group of participants, most of whom were White, highly educated women. This means the findings might not represent the full diversity of Autistic perspectives.

How will these findings help Autistic adults now or in the future?

By centering Autistic adults’ voices this study offers valuable guidance for future research and policy. It points toward research that focuses on how Autistic social media users can harness the technology to cultivate connection, community, emotional well-being, and positive identity. Ultimately, we hope that these findings can help shape a future in which social media is a tool for empowerment rather than harm for Autistic adults.

Background

There is a concern that increasing levels of loneliness and mental health difficulties may be linked to the growing use of social media, particularly in young people.¹ [Note: for the purposes of this article, we define social media broadly as “any digital applications or tools that allow users to share content and communicate with others” (p.5)].² While systematic reviews and meta-analyses suggest a small association between higher levels of social media use and poorer mental health,³ there is growing recognition of the potential benefits alongside costs.⁴ Furthermore, there is acknowledgment of a need to understand individual differences and the complex interplay between the individual and the technology in determining outcomes.^5,6 Thus, particular individuals and groups seem to be differentially susceptible to risks and benefits, depending on the nature of engagement and technology used.^7-9

One interindividual difference of interest to social media/mental health links is neurodivergence, particularly autism. Levels of mental health difficulties and suicidality are elevated in the Autistic population¹ and are predicted by loneliness and social disconnection.^10,11 There is also evidence to suggest that many Autistic people turn to the online world as a preferred means of social connection and communication and may use social media in ways that differ from non-Autistic people.¹² For example, one study found that Autistic people spent more time on—and were more positive toward—computer-mediated communication and reported higher levels of online life satisfaction than non-Autistic people.¹³ This is perhaps not surprising given that (some) Autistic people report finding it easier to initiate relationships,¹² and present authentically,¹⁴ that is, without masking, in online spaces than they do in offline spaces.

Against this backdrop, there is relatively little research into the role of social media in the mental health and well-being of Autistic adults—although it is an area of growing interest and research. Furthermore, as per wider autism research,¹⁵ studies of Autistic people’s social media use have often adopted a lens of vulnerability and deficit, focusing on risks and harms. Despite this, existing research includes evidence of potential benefits as well as harms to the Autistic social media user,¹⁶ mirroring general population studies.

While it is beyond the remit of this study to provide a comprehensive overview of research into the proposed harms and benefits of social media use to Autistic people, existing research has commonly referenced the potential for connection,¹⁷ including ways that social media can be used to cultivate relationships,¹⁸ engage with others over shared interests,^12,19 build community over a shared identity,^20,21 and support/be supported by other Autistic people.²² In terms of harms, authors have commonly highlighted the risk of problematic patterns of use,¹⁶ experiences of phishing, grooming, and radicalization^19,23,24 as well as problems related to misinformation and the potential for within-community splits and hostility.²⁰

Aside from a relative lack of research in the field (which is thankfully changing), much existing work has neither been shaped significantly by the Autistic community nor led to outcomes of practical relevance. Thus, despite some high-quality exceptions to this, for example, the work of Van Driel and others,^22,25-27 and a trend toward more participatory, inclusive, and coproduced research with the growth of neuro-affirmative movements,^28,29 community input has often been limited to patient and public involvement panels and research advisory groups,^30,31 which primarily function in a consultancy role, and risk being tokenistic.³² In contrast, researchers have argued for neurodivergent people’s input to be made “across the entire scientific cycle, from shaping theory, proposing hypotheses and contributing to new methods to interpreting and disseminating findings” (p.1).^29,30,33,34 Such an approach is more likely to ensure that research is relevant to people’s everyday lives, tailored to their strengths and needs, and consistent with their values.

For this reason, we sought to identify Autistic adults’ priorities for research into social media and mental health in Autistic people. We used a Delphi method with a panel of Autistic adults to identify community research priorities for Autistic social media research, a method that is recommended when there is not an accepted body of knowledge on a topic and a community consensus perspective is sought.³⁵ In addition, we used reflexive thematic analysis^36,37 of open text responses to a series of questions posed within the Delphi process about how to maximize impact, relevance, and reach of future research.

Methods

Ethical approval was obtained from the UCL Research Ethics Committee (27641/001). The study was designed in collaboration with two neurodivergent coauthors/researchers, one Autistic and one attention-deficit/hyperactivity disorder (ADHD).

Participant recruitment

We recruited Autistic adult participants through our online and offline networks, that is, through advertising on social media (LinkedIn and Twitter/X) and previous research participants who indicated an interest in future research. Participants needed to be 18 years of age or older, with a formal diagnosis of autism, resident in the United Kingdom, and able to understand and respond to written or verbal information in English.

Expertise was conceptualized as lived experience, since we were interested in the research priorities of Autistic adults in general. Social media use was not included as a criterion for participation, since social media use is so common that even nonusers are likely to be affected by virtue of being embedded in a digitally connected culture. Furthermore, we did not want to exclude the views of people who might purposely choose not to engage with the technology.

None of the participants knew the primary researcher (M.S.T.), although several knew one member of the research team (G.P.) through previous research. Participants had the opportunity to meet one another at an introductory and/or closing meeting (details below); to the authors’ knowledge, there was no further interaction between participants throughout the study except for shared (anonymized) survey feedback provided as part of the Delphi method (details below). Some participants may have had preexisting connections, although this was not sought, facilitated, or tracked.

Delphi process

Although there is no standard procedure for a Delphi method, recommendations from several key articles were consulted during study design.^35,38-40 In the Supplementary Material, we include how we addressed key indicators of Delphi method quality,⁴¹ in terms of study design and reporting of findings.

We used a three-round Delphi method design, bookended by an introductory and closing group meeting, since beyond two or three rounds it is reported that participants commonly become fatigued and consensus is usually met.³⁹ We ran an introductory meeting in July 2024, collected data from rounds 1 to 3 between July 2024 and November 2024, and ran a closing group in January 2025.

To accommodate everyone’s needs and maximize engagement, we made the process as flexible as possible. Each round involved a self-completed survey, which we sent as a Word document to each participant by email (returned by the same means); one participant, however, gave responses to round 2 and 3 surveys by phone as they reported struggling to complete them on their own. All correspondence was with a single point of contact (M.S.T.). While we initially asked for each survey to be completed within (approximately) one month, deadlines were extended repeatedly and multiple reminders sent out to maximize response rates.

Introductory meeting

We invited all panel members to an initial online meeting in which we introduced panel members to the study, discussed the methodology, and began to identify research priorities. To help participants prepare for the session and alleviate any anxieties, we sent them an agenda and instructions ahead of the meeting, emphasizing its informal nature and availability of multimedia communication choices via written (Padlet, chat function) and speaking options. For participants who could not attend, or did not want to, they were instead invited to: (1) attend a one-to-one (online) session or (2) be sent written instructions, both of which covered the same core material.

Round 1—generation of research ideas

We sent panel members the round 1 survey, which included questions about basic demographics following the Diversity and Inclusion Survey question guidance,⁴² including age, gender identity, ethnicity, highest educational level, neurodevelopmental, and mental health diagnoses.

We asked panel members to reflect on their own social media use and respond to eight questions about potential harms and benefits of social media to Autistic people and communities (see Supplementary Table S1). The following broad and pragmatic definition of social media² was provided to participants: “any digital platform, application or tool that allows you to share content and communicate with others. So that could include social network sites, like Twitter/X, Instagram or Discord, but could also include messaging apps or gaming where there is a social component.”

We then presented panel members with the following question: “In your opinion what are the most important questions that should be researched when it comes to the role of social media in Autistic people’s mental health?” and asked them to provide up to 10 research ideas in open-text boxes. Research ideas were also generated from the introductory meeting.

To reduce survey length and minimize burden, we integrated highly similar or duplicate research ideas or retained (only) the most clearly worded version. In addition, some responses were reworded slightly to increase comprehensibility and consistency of terminology.

Round 2—rating of research ideas for importance

The round 2 survey included all research ideas derived from round 1 (and the introductory meeting). In addition, we asked panel members to rate each item/research idea (used interchangeably) in terms of importance (“i.e., the extent to which you think it should be prioritised in future research”).

Panel members gave responses on a five-point Likert scale with the following response options: (1) “very unimportant,” (2) “unimportant,” (3) “Neither important nor unimportant,” (4) “important,” and (5) “very important.” In addition, we asked panel members to provide reasons for each of their ratings in free-text response boxes. However, following initial feedback that the process was burdensome, we asked participants to prioritize, first, rating of all items, and second, provision of reasons for their responses solely for items they felt particularly strongly about (either in terms of importance or unimportance).

We then grouped research ideas into the following three categories according to whether items: (A) were highly endorsed (≥70% of panel members rated them as “important” or “very important”) and achieved consensus (the interquartile range of associated importance ratings was ≤1); (B) were “borderline important,” that is, endorsed by ≥50% of panel members (irrespective of level of consensus) and not falling into category (A); and (C) were not important, that is, endorsed by <50% of panel members. Consensus thresholds used in the literature have been found to vary considerably (e.g., 50%–100%)³⁸ and are typically selected arbitrarily^35,38; consequently, we sought precedents and recommendations in the literature, which centered on 70%.^39,40

Round 3—re-rating of “borderline” important research ideas

The round 3 survey included items from category (B), with each item provided with the percentage of panel members who endorsed it (as “important” or “very important”), as well as all reasons provided by panel members for their decisions. We then asked participants to re-rate any items included on the basis of the information provided.

Items retained for inclusion in the (final) list of research priorities included all items assigned to category (A) from round 2, as well as any items from category (B) that subsequently met category (A) criteria following re-rating by the panel at round 3. Thus, we retained items that were heavily endorsed and reached consensus, as well as any that met these criteria after consideration of other panel members’ ratings and reasons for their ratings.

The round 3 survey also included six questions, with free-text response boxes, which were included to generate ideas about how to maximize impact, relevance, and reach of future research: “Having thought about research priorities in this area”: (i) “…do you think there are any important factors that researchers should consider when undertaking research into the role of social media use in the mental health of the Autistic community?”; (ii) “…can you think of any ways that researchers can increase the impact and practical relevance of their research to the Autistic community?”; (iii) “…do you think that there are particular groups or identities within the Autistic community that would be at risk of being under-represented in this research?”; and (iv) “What do you think would be the main barriers to recruitment and engagement in the research?”; (v) “Do you have any ideas how research might overcome these barriers?”; (vi) “Do you have any other thoughts or reflections that you would like to share about the role of social media in the mental health of the Autistic community, or research related to this?”

We analyzed written responses using inductive, reflexive thematic analysis, following the guidelines of Braun and Clarke^36,37 and Campbell and colleagues.⁴³ We used a critical realist epistemology, recognizing “individuals’ perspectives and their situations as real phenomena that causally interact with one another” (p.20).⁴⁴

Specifically, two authors (M.S.T. and G.P.) independently familiarized themselves with the data. M.S.T. generated initial codes to categorize data with similar meanings, and these were reviewed by G.P. Both authors then independently defined and named themes on the basis of identified codes and associated quotes and came together to review and agree upon final themes. M.S.T. and G.P. conversed throughout the process, reflecting on our positionality, neurodiverse experiences, and academic roles, and how these may impact on how we frame themes.

Closing meeting

All panel members were invited to an online closing group meeting, where we presented the findings, solicited feedback and reflections on the process and experience of participating, discussed future steps, and celebrated the work undertaken. All those not attending the closing meeting were sent a summary.

Study evaluation

Finally, a focus group of three panel members who took part in the study were recruited to evaluate the study and comment on, and suggest edits to, a late draft of this article (see Supplementary Material).

Reimbursement

We remunerated participants at a rate of £25/hour for their time (through online gift vouchers), in accordance with guidelines by the National Institute for Health and Care Research.⁴⁵ Participants could therefore receive a maximum of £95 for their time, reflecting attendance of initial and final group meetings (each one-hour long), and completion of three surveys (estimated completion time: 35 mins each). Panel members who evaluated the study were given additional vouchers worth £45.

Results

We initially recruited 26 participants. Of these, we excluded four (15%) (three at stage 1 and one at stage 2), due to strong indicators that they were “scammers,” that is, not meeting study inclusion criteria but presumably engaging for financial purposes.⁴⁶ Several participants responded very quickly to the study advert and requests for information, from email addresses with multiple random numbers embedded, at times of day that suggested a non-U.K. time zone; correspondence was also uncharacteristically curt and formal, and while scheduling was very straightforward, all participants retained their cameras off during meetings and engaged either minimally or else offered information that was highly impersonal and medicalized.

The resulting 22 participants completed survey 1 (100% retention) (Table 1). One (4%) dropped out after stage 1, leaving 21 who completed all three surveys (95% retention). Of the initial 22 participants, 11 identified as women (n = 11; 50%), 5 as men (23%), and 6 as non-binary (27%). The mean age was 36.42 years (SD = 13.94; range = 18–64). Participants were highly educated and most were White (n = 18; 82%). Half (n = 11; 50%) reported being formally diagnosed or self-identified as ADHD. Mental health difficulties were common, with almost half (n = 10; 45%) reporting anxiety and 36% (n = 8) reporting depression. Participants reported typically using social media 2.92 hours and 3.64 hours per day on a weekday and weekend day, respectively. This is greater than commonly reported, up-to-date, global average figures for general population samples, for example, 2.35 hours.⁴⁷

Table 1.

Panel Member Characteristics

Variable	Level	N (%) /M (SD)
Gender	Man	5 (22.73)
Gender	Woman	11 (50)
Non-binary	6 (27.27)
Age	—	36.42 (13.94)
Ethnicity	White	18 (81.81)
	Black British	1 (4.55)
	Multiple Ethnicities	2 (9.09)
Prefer not to say	1 (4.55)
Highest Educational level	^aO levels, GCSE or equivalent	1 (4.55)
	A levels, or equivalent	4 (18.18)
	Undergraduate degree	10 (45.45)
Postgraduate degree	7 (31.81)
^bOther neurodevelopmental difference	ADHD	11 (50)
	Dyspraxia	5 (22.73)
	Dyslexia	2 (9.09)
	Dysgraphia	1 (4.55)
Tourette syndrome	1 (4.55)
^cMental health difficulties	Anxiety	10 (45.45)
	Depression	8 (36.36)
	PTSD / complex PTSD	5 (22.73)
	Disordered eating	3 (13.64)
	OCD	1 (4.55)
Bipolar disorder	1 (4.55)
Typical daily social media use (mins)	Weekday	175.18 (84.27)
Weekend day	218.63 (117.77)

Group level characteristics are presented for the 22 panel members who completed survey 1.

O Levels, GCSEs, and A Levels are U.K. educational qualifications, with GCSEs taken at age 16 covering a broad curriculum, A Levels taken at age 18 offering specialized study, and O Levels being an older system still used in some countries.

Other neurodevelopmental differences and mental health difficulties include formally diagnosed, self-identified, or self-suspected.

Mental health difficulties reflect current difficulties.

ADHD, attention-deficit/hyperactivity disorder; GCSE, General Certificate of Secondary Education; OCD, Obsessive-Compulsive Disorder; PTSD, Post-Traumatic Stress Disorder.

Delphi stages—descriptives

Stage 1 generated 125 research ideas. We added another six items based on research ideas generated by participants in the introductory meeting. We removed 33 that were highly similar to other items (and integrated them into other items where relevant), resulting in 98 distinct research items rated in subsequent stages.

For ease of interpretation, we clustered items into foci and subfoci for subsequent surveys (Table 2). This resulted in four foci: (i) the user (n = 35), (ii) uses, costs, and benefits (n = 40), (iii) the technology (n = 10), and (iv) solutions and responses (n = 13). As is evident from the relative frequency of items in each focus, participants were more interested in research on the user and their use of social media as well as resulting costs and benefits, rather than the technology itself or potential solutions and responses to the technology.

Table 2.

Foci and Subfoci of Research Ideas Generated, with Number of Research Ideas Falling into Each at Stage 1 (First Number in Bracket) and Retained at the End of the Study, That is, Research Ideas That Were Rated as High Priority and High Consensus (Second Number in Bracket)*

Foci	Subfoci
1. Uses, costs, and benefits (40/17)	(i) Connection and disconnection (11/4)
	(ii) Pursuing interests (2/1)
	(iii) Knowledge, learning, and mis/information (6/4)
	(iv) Impacts and general well-being (12/6)
(v) Risks, safety, and safety management (9/2)
2. The user (35/9)	(i) User characteristics, interindividual differences, diversity, and inclusion (9/1)
2. The user (35/9)	(ii) Identity, representation, and autism culture (17/5)
(iii) Autonomy, advocacy, and power (9/3)
3. The technology (10/0)	(i) Platforms, features, and affordances (10/0)
4. Solutions and responses (13/3)	(ii) Strategies (4/2)
(iii) Creation of Autistic spaces (9/1)

*For example, 35 research ideas focused on the user at stage 1, but only 9 of these were retained in the final list (stage 3).

Following initial rating of items in survey 2, 22 items fell into category A (highly endorsed/high consensus), 52 into category B (borderline important), and 24 into category C (not important). For survey 2 items, the average percentage of panel members who provided written justification for their rating was 24.88% per item (SD = 7.16).

After participants re-rated category B items in survey 3 and had the opportunity to review other panel members’ free text responses for each item alongside its overall percent endorsement, a further seven items made it into category A. This resulted in a (final) total of 29 research ideas that participants endorsed highly and reached consensus (see Table 3). For survey 3, the average percentage of panel members who changed their ratings from survey 2 was 9.82% per item (SD = 10.72).

Table 3.

Final List of Research Ideas That Were Deemed to be High Priority and Heavily Endorsed

Priority ranking	Theme No.	% endorsement	Research idea
1st	1(iv)	95.24	Do Autistic people use social media to self-regulate, self-distract and/or self-soothe when they are feeling dysregulated or overstimulated? If so, how? And is it effective?
2nd	1(i)	90.48	Is social media an effective means of peer support and community for Autistic individuals, and what is the impact of this on self-esteem, well-being, belonging, and identity?
2nd	1(i)	90.48	What roles do online communities play in providing social support for Autistic people, and how does this impact on self-esteem, well-being, belonging, and identity?
2nd	2(ii)	90.48	How does the portrayal of Autism and Neurodivergence in social media content affect the self-image and mental health of Autistic social media users?
3rd	1(iv)	85.71	How does social media positively and/or negatively impact on the mental health of Autistic people?
3rd	4(ii)	85.71	What does a safe and supportive social media space for Autistic people look like?
4th	1(i)	80.95	Is social media use driven by choice or necessity for Autistic people? E.g., does the availability versus absence of local, in-person social opportunities influence social media use?
4th	1(i)	80.95	Does social media use promote inclusivity for vulnerable intersections of the Autistic community?
4th	1(iii)	80.95	What impact does misinformation and trauma/deficit narratives have on community identity in Autism?
4th	1(iv)	80.95	What types of social media platforms, platform features, patterns of engagement, social media content, online communities, and individuals they connect with do Autistic people find are linked to more positive experiences and feelings? And, what makes it a positive experience?
4th	1(iv)	80.95	What types of social media platforms, patterns of engagement, social media content, online communities, and individuals they connect with do Autistic people find are linked to more negative experiences and feelings? And, what makes it a negative experience?
4th	2(i)	80.95	How can we ensure that the online Autism community/communities are more inclusive and diverse?
4th	2(ii)	80.95	In what way does community knowledge and shared experiences on social media help Autistic people define and understand autism and make sense of themselves?
4th	2(ii)	80.95	Is social media an effective way to promote positive identity for Autistic people and Autistic young people (18–25-year olds) within isolated communities, e.g., isolated by language, literacy barriers, financial restrictions, etc.?
4th	2(iii)	80.95	Exploring how to increase allistic (non-Autistic) carers’ and support workers’ willingness to help Autistic individuals with medium and high support needs to access social media and increase their autonomy in online spaces.
5th	1(ii)	76.19	What role does social media play in helping Autistic people discover and connect with special interests, and how does this influence their mental health and sense of purpose?
5th	1(iii)	76.19	How does social media help bridge a resource/information gap for Autistic individuals?
5th	1(iii)	76.19	Exploring ways Autistic people can use social media to engage with new research, and examining the effects of this on misinformation and concerns around being Autistic and the risk of poor personal outcomes.
5th	1(iii)	76.19	How does social media aid older individuals with diagnosis and understanding a diagnosis?
5th	1(iv)	76.19	Do Autistic people experience sensory overload when they use social media?
5th	1(v)	76.19	How can social media companies, and society more generally, support Autistic users and help protect them against hate crime, online abuse and online sexual grooming, etc.?
5th	2(ii)	76.19	How do Autistic and non-Autistic people communicate online/on social media? (Including similarities and differences).
5th	2(ii)	76.19	Does social media have a positive influence on identity formation and/or function as a positive extension of personhood for self-diagnosed and newly diagnosed Autistic individuals?
5th	2(iii)	76.19	What can we learn from online Autistic communities when working to spread awareness and acceptance of autism more broadly?
6th	1(iv)	71.43	What are the benefits for Autistic people of social media use and online Autistic communities?
6th	1(v)	71.43	How do Autistic people perceive and manage online bullying and harassment, and what strategies do they use to cope with such experiences?
6th	2(iii)	71.43	In what ways does social media increase Autistic people’s autonomy?
6th	4(i)	71.43	What sort of online content has a significant emotional impact on Autistic people, and how do Autistic people manage this/what strategies do they use?
6th	4(i)	71.43	How do Autistic people disconnect from, or take a break from, social media if/when they feel they need to? What strategies do they use? And are they effective?

Priority ranking is based on % endorsement only.

Research priorities

Mirroring the findings from survey 1, the vast majority of the 29 items retained related to the user (n = 9 items; 31.03%) and uses and costs and benefits of engagement (n = 17; 58.62%), with little interest in researching the social media technology itself (n = 0; 0%) or potential solutions and responses thereto (n = 3; 10.34%).

Several items highlighted an interest in how social media can “positively and/or negatively impact on the mental health of Autistic people” (item 5) as well as exploration of “what types of social media platforms, platform features, patterns of engagement, social media content, online communities and individuals they connect with” (items 10 and 11) are connected with positive and negative online experiences. Nevertheless, there was a clear emphasis on research that explores the “benefits for Autistic people of social media use” (item 25) over harms.

In terms of benefits to explore, these included the potential for social media to connect users with “special interests” (item 16), “bridg[e] a resource/information gap” (item 17), and engage with “new research” (item 18) as well as to “self-regulate, self-distract and/or self-soothe” when “dysregulated or over-stimulated” (item 1). Panel members were also interested in researching the potential for social media to build “peer support and community” (item 2) and “promote inclusivity” (item 8) as well as the potential impacts of this on “self-esteem, well-being, belonging and identity” (item 3), particularly within particular subgroups and/or “vulnerable intersections of the autistic community” (item 8).

The panel was interested in the “portrayal of autism and neurodivergence” on social media, as well as the potential impacts of this on “self-image and mental health” (item 4). Specifically, participants were interested in how social media might be used to “spread awareness and acceptance of autism” (item 24). This included the potential for social media to help Autistic people to “define and understand autism and make sense of themselves” (item 13; added underlining), and “promote positive identity for autistic people” (item 14). Thus, the panel was interested in whether social media has “a positive influence on identity formation and/or function as a positive extension of personhood” (item 23), including for particular (potentially more vulnerable) subgroups, for example, “young people” and “isolated communities” (item 14), “older individuals” (item 19) as well as the “self-diagnosed and newly diagnosed” (item 23).

Relatedly, the panel was interested in how social media might be used to increase “autonomy” (item 27) for Autistic people, including for “individuals with medium and high support needs” (item 15), and how to “ensure that the online autism community/communities are more inclusive and diverse” (item 12), including “for vulnerable intersections” thereof (item 8).

With respect to potential harms of social media engagement, participants prioritized research into the role of social media in “hate crime, online abuse and online sexual grooming” (item 21), “online bullying and harassment” (item 26), and “sensory overload” (item 20) as well as the potential role of “misinformation and trauma/deficit narratives […] on community identity in autism” (item 9).

In terms of managing the online world, participants were keen to see research focusing on individual responses as well as more societal/systemic responses. With respect to the former, participants prioritized research exploring “strategies” that Autistic people use to manage and cope with “online bullying and harassment” (item 26) and online content that has a “significant emotional impact” (item 28), as well as how “autistic people disconnect from, or take a break from, social media if/when they feel they need to” (item 29). With respect to more systemic responses, participants wanted research to examine what a “safe and supportive social media space for autistic people [might] look like” (item 6), and to explore how “social media companies, and society more generally, [can] support autistic users and help protect them against hate crime, online abuse and online sexual grooming” (item 21).

Ideas for maximizing impact, relevance, and reach

As noted, we analyzed free text responses, provided in response to six questions presented in survey 3, to identify how the research prioritized by participants should be conducted. This generated a wealth of rich data, which we summarize (in brief) below and in Table 4, although we present a table of full codes identified, with associated quotes, in Supplementary Table S2.

Table 4.

Themes Identified from Free Text Responses Given to Questions 1 to 6 Presented in Survey 3 on How to Maximize Impact, Relevance, and Reach of Future Research in the Field

Theme	Example quotes
Many different voices	- “Make no assumptions. If you do feel the necessity to refer to a stereotypical view of autistic people, make this very clear. There really is a vast array of autistic people” (P12).
Many different voices	- “I think the nuance is important, therefore and the devil is probably in the detail of investigating individual differences both with autistics, non-autistics and then different populations within each group” (P7).
- “Our community has many sub-communities and there’s many varieties of autism, for lack of a better framing, this should be considered in this work” (P10).
Making research accessible right from the start	- “Executive function challenges — difficulty getting started or completing the process of registering interest and participating” (P2).
Making research accessible right from the start	- “There is often a stigma associated with mental health issues, which can be compounded for autistic individuals. This stigma, along with a general distrust of research and institutions, can be a significant barrier to participation” (P11).
- “Reaching different cultures” (P4).
Undertaking research that makes a positive difference to autistic people’s lives	- “…any work done on this subject needs to be done with autistic people and the findings disseminated to our community” (P10).
	- “The potential for the research to have applications which deliver immediate practical benefits for Autistic people, rather than just generating interesting findings” (P2).
- “Ensure you are working with a diverse group of experts by experience and/or participants” (P13).

Also, see Supplementary Table S2.

We identified three themes. Theme 1, many different voices, emphasized a need for nuanced research that does not fall into simplistic, reductive narratives, but recognizes and captures the rich array of Autistic experiences. This was deemed particularly important in the context of research that has traditionally marginalized specific groups and intersectionalities. Panel members spoke of a need “to ensure that [we are] hearing from as many autistic voices as possible because we all vary” (P9), highlighting a number of “sub-communities” (P10), as well as a range of interindividual differences to be considered when researching Autistic people’s experiences. This included references to age, for example, “older late dx [diagnosed] or not dx people” (P17); ethnicity, for example, “Black and Brown Autistic people […] Gypsy, Roma and Traveller people” (P2); gender identity and sexuality [e.g., “the LGBTQ+ community” (P17)]; those with additional support needs, for example, Autistic people with “physical disability” (P17), “multiple health conditions” (P2), and/or “learning disability” (P8); or communication differences, for example, those who are “illiterate or dyslexic, or have sensory impairments” (P3) or experience “situational mutism” (P7) as well as those whose movement is restricted, for example, those in “residential care settings” (P2) or “in-patient services” (P2).

Theme 2, making research accessible right from the start, reflected on a range of factors, many relating to differences identified in Theme 1, which panel members felt might prevent Autistic people from participating in research if not carefully and thoughtfully designed to maximize accessibility. These included cognitive/neuropsychological factors, for example, “executive function challenges” (P2) and “cognitive load/demand” (P7); lack of trust in the context of discrimination and distrust, for example, “systemic racism can deter black and brown individuals from participating in research” (P11), and “stigma associated with mental health issues, […] can be compounded for autistic individuals” (P11); technological barriers [“not all potential participants may have access to the necessary technology” (P12)]; and time constraints, for example, “potential participants may be time-poor due to work, caregiving responsibilities, or other commitments” (P11).

Theme 3, undertaking research that makes a positive difference to Autistic people’s lives, reflected calls for transparent, accessible, coproduced research that is nonstigmatizing and empowering, with the potential to make genuine, concrete differences to Autistic people’s lives. This was highlighted in the context of research that has often adopted a pathologizing and stigmatizing lens, which risks doing harm to Autistic people. As part of this, diversity and representation were again central, with calls for the “diversity of the autistic community [to be] reflected” (P2), in all aspects of the research, that is, not only in the participants recruited but also in the “recruitment strategies” used (P11), “materials that are produced” (P2), and within the “research team and advisory group” (P2).

From a pragmatic perspective, participants also emphasized a number of concrete ways to maximize accessibility, which revolved primarily around “simplifying processes as much as possible” (P2), for example, providing “very clear instructions” (P7 and P11), and “get[ting] Autistic people to test them before they are finalised” (P2). Furthermore, participants emphasized a need to offer lots of different ways of engaging participants to meet individuals’ preferences, for example, offering “a multitude of ways to respond to questions” (P10), and “sharing [of] findings using a range of evidence-based resources” (P2).

Discussion

This study identified community priorities for research into the role of social media in the mental health of Autistic people. We used a rigorous, three-wave Delphi method,³⁵ which offered 22 Autistic panel members opportunities to rate, and subsequently re-rate, identified priorities after having considered other panel members’ responses and overall levels of endorsement for each item. We made great effort to maximize flexibility of engagement, which helped to deliver a high level of participant retention and was reflected in positive feedback in the study’s evaluation (Supplementary Material). Ultimately, we identified 29 research priorities, which were highly endorsed and reached high levels of consensus.

Our Autistic panel members agreed that future research should prioritize potential pros and cons of social media to Autistic people, as well as potential ways to harness benefits and ameliorate harms.¹⁶ However, they emphasized the importance of not focusing solely on harms. This is pertinent given previous criticisms of social media research for adopting a primarily “concern-centric” approach to understanding its associations with well-being⁴⁸ as well as common misinformed representations of Autistic people that center on deficit and vulnerability,⁴⁹ creating the potential for the intersection of two catastrophizing/negatively focused narratives.

In terms of potential responses to issues relating to social media, there was a call for research that focuses more on the user and user’s behavior, as opposed to the technology or technological solutions. This potentially reflects an emphasis on individual agency and choice, intersecting with the (aforementioned) desire to move away from narratives of deficit and vulnerability.

With respect to potential benefits, the study foregrounded an interest in the use of social media to access peer support and connect with a community, that is, build social capital and connectedness.¹⁸ This is unsurprising given existing research and theory, in both Autistic and general population samples, which suggests that when benefits of social media are accrued, these are typically underpinned by experiences that relate to satisfaction of core needs such as acceptance and belonging.^5,16,50

Autistic panel members also recommended research into the potential for social media to facilitate access to reliable information, including about autism itself postdiagnosis. This is topical given debates around the potential impact of online mis-information, including in relation to neurodivergence.⁵¹ It also highlights a potential gap between what is wanted and what is available online with respect to information about Autistic experiences and neurodivergence more generally,²⁰ a theme that was also reflected in discussions in the introductory meeting.

The study also highlighted interest in the potential for social media to aid self-soothing and emotion regulation during periods of dysregulation and overstimulation. This is interesting given common concerns around the potential for social media to over-stimulate and dys-regulate neurodivergent social media users.⁵² It also speaks to concerns raised by panel members that Autistic people with high support needs may have their access to supportive technologies blocked by well-meaning supporters. Nonetheless, the potential for social media to dys-regulate was also identified as a research priority,¹⁶ highlighting a desire for nuance.

The research also emphasized interest in how Autistic people and experiences are portrayed online, and how these shape understanding and acceptance (or rejection) of Autistic people.²⁰ The primary focus was on the impact of narratives around autism (and neurodivergence more generally) on Autistic people themselves as well as the Autistic community (i.e., individual and shared identity/ies).⁵³ This makes sense in the context of literature on the role of Autistic experiences of discrimination, stigma, and masking⁵⁴; thus, negative, deficit-based narratives of autism (and neurodivergence more generally) are known to drive internalized stigma and self-criticism, with very real and harmful consequences for mental health.^32,55 It is not surprising, therefore, that there is considerable interest in how social media, with its huge potential to connect and spread information,² might be harnessed for the purposes of Autistic advocacy and the construction of a strong sense of shared identity, as well as a desire not to focus solely on issues of vulnerability and harm.^21,56

It is noteworthy that fewer priorities identified related to the potential harms of social media. Nonetheless, there was an interest in research into risks, including hate crime, online abuse, and sexual grooming, reflecting some of the existing research in the field^19,23,24 as well as broader research into the general population.⁵⁷ In addition, as noted, there was interest in the potential for sensory overload/overwhelm as a result of online engagement⁵² as well as the impact of misinformation and spread of deficit-based narratives of autism. Once again, this emphasizes a central concern around the portrayal of autism and Autistic people online and its role in identity formation.

Central to issues of identity, there was a heavy focus in the research priorities identified as well as in the ideas for maximizing impact, relevance, and reach, on extending research to include and understand the online lives of minoritized groups and intersectional identities. This mirrors findings of a systematic review of information technology use by Autistic people, which highlighted underrepresentation of women, transgender, non-White, low income, and minimally and nonspeaking Autistic people as well as those with co-occurring intellectual disability in the research.¹⁶ The issues raised by our panel go further, however, in suggesting additional groups that may be overlooked as well as associated barriers to engagement, including consideration of non-native English speakers, those from gypsy, Roma, and traveller groups, late diagnosed and older adults, individuals with high support needs, and people in inpatient services, residential care settings, and with multiple physical and/or mental health conditions (Supplementary Table S2).

The findings also emphasized a desire for genuinely co-delivered and participatory research methods that will deliver concrete benefits to Autistic communities^30,58 and highlighted a number of ideas about what this might look like. This included the importance of transparent processes that build trust, clear and flexible engagement methods, the use of adapted materials tailored to the needs and preferences of the communities they hope to engage, and greater inclusivity, not only with respect to participants and communities targeted but also with respect to researchers undertaking the research. While many of these ideas have been highlighted in previous writings and work on participatory methodologies,³² they reinforce these as central concerns of the Autistic community, including with respect to social media and mental health research.

Implications

Many of the research priorities highlighted by the panel align with existing research,⁵² including research into the potential for social media to connect Autistic people to information,²⁰ support,⁵⁹ interests,^12,19 and peers,¹⁸ as well as explore issues of identity, belonging, and community.⁶⁰ Many of the research priorities linked to potential risks of engagement are also well represented in the literature, for example, the potential for social media to feed misinformation,⁵¹ amplify reductive, deficit-focused narratives,⁶¹ and expose the user to online abuse, bullying, and sexual grooming.²³ Such concordance between prioritized and existing research is reassuring, however, and suggests that further work in these areas is warranted. Participants’ ideas for ways to maximize impact, relevance, and reach also highlight the need to extend this research to explore how such processes operate within under-represented groups.

In addition, however, a number of novel—or under-researched—areas were identified, most notably—as described above—the potential for social media to support Autistic people “to self-regulate, self-distract and/or self-soothe.” Another promising avenue for research highlighted involves the adoption of a longitudinal, developmental perspective to explore how social media interacts with identity development across the lifespan as well as through the journey of identification (whether formally diagnosed or not) as an Autistic person. Thus, a number of research priorities identified related to the role of “identity formation,” including for “young people (18–25-year olds)” and “newly diagnosed Autistic individuals,” alongside interest in the role of social media use for “older Autistic people” and “older late dx [diagnosed] or not dx people.” Such a developmental, life-course perspective has been identified previously as lacking but much needed in autism research.⁶²

Finally, while the research priorities identified focused more on the user and user behavior rather than the technology or technological design (see Table 2), any future research exploring the roles of social media in mental health and well-being must move beyond reductive, deterministic assumptions that treat all technologies as the same. Instead, they must take into consideration the complex and nuanced ways that the features and affordances of the technology interact with interindividual differences to either support or undermine mental health and well-being.^5,63

Although the practical, real-world implications of the study will come to fruition when the research priorities identified are explored through meaningful, coproduced research, our findings nevertheless highlight a number of practical implications beyond research. First, the diverse, complex, and nuanced array of issues raised by the participants in relation to their social media use and mental health reinforce calls for people’s online lives to be given due attention in mental health services, both in terms of benefits to harness and risks to ameliorate.⁵ Second, the prioritization of research into understanding the user and their experiences online over the technology and its features and affordances, coupled with the focus on issues of empowerment and representation, speak to the importance of centering Autistic people’s autonomy and agency when addressing online well-being, whether that be through intervention development, public health policy, or online regulation.

Limitations

There are several limitations to our work. First, while there is no standard recommended size for a Delphi study, our panel was modest in size, potentially limiting generalizability. However, previous reviews and commentaries have highlighted that published studies typically include panels of 15–20 participants (although panels considerably smaller and larger are not uncommon), with potential trade-offs with larger panels, for example, >30, including reduced response rates and saturation of accuracy, reliability, and data quality.^35,38,64 Furthermore, our sample size was in line with proposed indicators of Delphi method quality.⁴¹

Second, generalizability may also have been compromised by biases in recruitment, including recruitment through existing contacts, which likely contributed to an over-representation of White, highly educated women, echoing concerns raised by the group (and in the focus group’s evaluation of the study) about the under-representation of minoritized and intersectional identities. This is potentially problematic, since by centering the research priorities of such a group, we risk reinforcing patterns of inequality that we would hope to address, for example, by driving research that is not of broad relevance. However, we note that the sample was relatively diverse with respect to age and gender identity, and further, that far from overlooking issues of minoritization and under-representation, this was in fact at the core of many participants’ responses.

Relatedly, our decision to include formally diagnosed participants only further risked privileging certain Autistic people’s experiences. This is particularly problematic given existing barriers to obtaining a formal diagnosis and risks compounding many of the issues described. Furthermore, engagement with the study may also have been limited by some of the systemic and structural impediments raised by the participants, for example, challenges to engagement due to technological barriers, time constraints, carer responsibilities, and so on.

Third, we did not orient participants toward existing research in the field at the start of the study (e.g., through an initial scoping review of extant literature) as is sometimes undertaken in Delphi groups.³⁵ Consequently, research priorities identified mirrored some of the extant literature. However, we made this decision a priori in order not to bias panel members with ideas about what has already been done, and hence (potentially) what is/is not possible—a recognized risk of such an approach.⁶⁵

Finally, members of the focus group who evaluated the study noted that the surveys were long and time-consuming; nonetheless, this feedback was balanced with a sense that they were not difficult to complete and that the research focus was an important one. Relatedly, one participant said that they would have dropped out if they had not been given one-to-one support to complete the surveys, reinforcing the importance of being as flexible as possible in supporting engagement.

Conclusions

In summary, we identified 29 research priorities for social media and mental health research, rated as of high importance and high consensus by a sample of 21 Autistic adults as well as ideas for how to maximize impact, relevance, and reach in relation to this. The findings suggest a drive for research into the potential benefits and harms of social media use, but with a focus on the former. Specifically, there was an interest in understanding how social media might be used to connect people: to information, to interests and hobbies, and to other people as well as to communities. There was also interest in how social media might be harnessed to increase agency and autonomy in shaping representations of, and narratives around, autism in the media, to cultivate and reflect a positive and healthy sense of self and identity, collective and individual. Indeed, identity was central to many priorities and themes identified, with a call for greater understanding of the roles of social media in autism to include its relevance to marginalized, disempowered, and minoritized groups and intersectional identities.

Footnotes

Acknowledgments

The authors would like to thank all panel members who took part in this study.

Authorship Confirmation Statement

M.S.T.: Conceptualization, methodology, formal analysis, investigation, data curation, writing—original draft, project administration, and funding acquisition; K.R.: Conceptualization and writing—review and editing; W.M.: Conceptualization and writing—review and editing; E.P.: Conceptualization and writing—review and editing; and G.P.: Conceptualization, methodology, formal analysis, writing—review and editing, and funding acquisition. The article has been submitted solely to this journal and is not published.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was funded by UCL Grand Challenges (Grant No: 550216/156425).

Supplementary Material

References

1. Twenge

, Haidt

, Blake

, McAllister

, Lemon

, Le Roy

. Worldwide increases in adolescent loneliness. J Adolesc. 2021;93:257–269; doi: 10.1016/j.adolescence.2021.06.006

2. Nesi

, Choukas-Bradley

, Prinstein

. Transformation of adolescent peer relations in the social media context: Part 1—A theoretical framework and application to dyadic peer relationships. Clin Child Fam Psychol Rev. 2018;21(3):267–294; doi: 10.1007/s10567-018-0261-x

3. Appel

, Marker

, Gnambs

. Are social media ruining our lives? A review of meta-analytic evidence. Review of General Psychology. 2020;24(1):60–74; doi: 10.1177/1089268019880891

4. Marciano

, Lin

, Sato

, Saboor

, Viswanath

. Does social media use make us happy? A meta-analysis on social media and positive well-being outcomes. SSM - Mental Health. 2024;6:100331.

5. Tibber

, Silver

. A trans-diagnostic cognitive-behavioural conceptualisation of the positive and negative roles of social media use in adolescents’ mental health and wellbeing. tCBT. 2022;15(e7).

6. Ward

, Dill-Shackleford

, Mazurek

. Social media use and happiness in adults with autism spectrum disorder. Cyberpsychol Behav Soc Netw. 2018;21(3):205–209; doi: 10.1089/cyber.2017.0331

7. Orben

, Przybylski

, Blakemore

, Kievit

. Windows of developmental sensitivity to social media. Nat Commun. 2022;13(1):1649; doi: 10.1038/s41467-022-29296-3

8. Beyens

, Pouwels

, van Driel

, Keijsers

, Valkenburg

. The effect of social media on well-being differs from adolescent to adolescent. Sci Rep. 2020;10(1):10763; doi: 10.1038/s41598-020-67727-7

9. van der Wal

, Beyens

, Janssen

LHC

, Valkenburg

. Social media use leads to negative mental health outcomes for most adolescents. Preprint. 2025:1–37.

10.

10. Hymas

, Badcock

, Milne

. Loneliness in autism and its association with anxiety and depression: A systematic review with meta-analyses. Rev J Autism Dev Disord. 2024;11(1):121–156; doi: 10.1007/s40489-022-00330-w

11.

11. van Bentum

, Sijbrandij

, Huibers

, Begeer

. Occurrence and predictors of lifetime suicidality and suicidal ideation in autistic adults. Autism. 2024;28(9):2282–2294; doi: 10.1177/13623613231225901

12.

12. Gillespie-Lynch

, Kapp

, Shane-Simpson

, Smith

, Hutman

. Intersections between the autism spectrum and the internet: Perceived benefits and preferred functions of computer-mediated communication. Intellect Dev Disabil. 2014;52(6):456–469; doi: 10.1352/1934-9556-52.6.456

13.

13. Van Der Aa

, Pollmann

MMH

, Plaat

, Van Der Gaag

. Computer-mediated communication in adults with high-functioning autism spectrum disorders and controls. Res Autism Spectr Disord. 2016;23:15–27; doi: 10.1016/j.rasd.2015.11.007

14.

14. Jedrzejewska

, Dewey

. Camouflaging in autistic and non-autistic adolescents in the modern context of social media. J Autism Dev Disord. 2022;52(2):630–646; doi: 10.1007/s10803-021-04953-6

15.

15. Burnham Riosa

, Chan

, Maughan

, Stables

, Albaum

, Weiss

. Remediating deficits or increasing strengths in autism spectrum disorder research: A content analysis. Adv Neurodev Disord. 2017;1(3):113–121; doi: 10.1007/s41252-017-0027-3

16.

16. Hassrick

EMG

, Holmes

, Sosnowy

, Walton

, Carley

. Benefits and risks: A systematic review of information and communication technology use by autistic people. Autism Adulthood. 2021;3(1):72–84; doi: 10.1089/aut.2020.0048

17.

17. Davidson

. Autistic culture online: Virtual communication and cultural expression on the spectrum. Soc Cult Geogr. 2008;9(7):791–806; doi: 10.1080/14649360802382586

18.

18. Mazurek

. Social media use among adults with autism spectrum disorders. Comput Human Behav. 2013;29(4):1709–1714; doi: 10.1016/J.CHB.2013.02.004

19.

19. Barber

. Social media and autism spectrum conditions. Practice Nursing. 2017;28(7):292–298; doi: 10.12968/pnur.2017.28.7.292

20.

20. Skafle

, Gabarron

, Nordahl-Hansen

. Social media shaping autism perception and identity. Autism. 2024;28(10):2489–2502; doi: 10.1177/13623613241230454

21.

21. Parsloe

. Discourses of disability, narratives of community: Reclaiming an autistic identity online. Journal of Applied Communication Research. 2015;43(3):336–356; doi: 10.1080/00909882.2015.1052829

22.

22. Koteyko

, van Driel

, Vines

. Autistic sociality on Twitter: Enacted affordances and affiliation strategies. Discourse and Communication. 2022;16(4):385–402; doi: 10.1177/17504813211070655

23.

23. Macmillan

, Berg

, Just

, Stewart

. Online safety experiences of autistic young people: An Interpretative Phenomenological Analysis. Res Autism Spectr Disord. 2022;96:101995; doi: 10.1016/j.rasd.2022.101995

24.

24. Faccini

, Allely

. Rare instances of individuals with autism supporting or engaging in terrorism. JIDOB. 2017;8(2):70–82; doi: 10.1108/JIDOB-11-2016-0022

25.

25. Koteyko

, Van Driel

, Billan

, Pena

, Vines

. Stigma management strategies of autistic social media users. Autism Adulthood. 2025;7(3):273–282; doi: 10.1089/AUT.2023.0095

26.

26. Barros Pena

, Koteyko

, Van Driel

, Delgado

, Vines

. “My Perfect Platform Would Be Telepathy” - Reimagining the Design of Social Media with Autistic Adults. In: Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. ACM; 2023:pp. 1–16; doi:10.1145/3544548.3580673

27.

27. Barile

, Edwards

, Dhingra

, Thompson

. Associations Among County-Level Social Determinants of Health, Child Maltreatment, and Emotional Support on Health-Related Quality of Life in Adulthood. Psychol Violence. 2015;5(2):183–191; doi: 10.1037/a0038202

28.

28. Kapp

, Pripas-Kapit

, Dekker

, et al. Autistic Community and the Neurodiversity Movement—Stories from the Frontline. ( Kapp

, ed.). Springer Nature; 2020.

29.

29. Pellicano

, den Houting

. Annual Research Review: Shifting from ‘normal science’ to neurodiversity in autism science. J Child Psychol Psychiatry. 2022;63(4):381–396; doi: 10.1111/jcpp.13534

30.

30. Sonuga-Barke

EJS

, Chandler

, Lukito

, et al.; RE-STAR team*. Participatory translational science of neurodivergence: Model for attention-deficit/hyperactivity disorder and autism research. Br J Psychiatry. 2024;224(4):127–131; doi: 10.1192/bjp.2023.151

31.

31. Tan

, Crane

, Haar

, Heyworth

, Poulsen

, Pellicano

. Reporting community involvement in autism research: Findings from the journal Autism. Autism. 2024; doi: 10.1177/13623613241275263/SUPPL_FILE/SJ-DOCX-1-AUT-10.1177_13623613241275263.DOCX

32.

32. Nicolaidis

, Raymaker

, Kapp

, et al. The AASPIRE practice-based guidelines for the inclusion of autistic adults in research as co-researchers and study participants. Autism. 2019;23(8):2007–2019; doi: 10.1177/1362361319830523

33.

33. Fletcher-Watson

, Adams

, Brook

, et al. Making the future together: Shaping autism research through meaningful participation. Autism. 2019;23(4):943–953; doi: 10.1177/1362361318786721

34.

34. Pellicano

, Stears

. Bridging autism, science and society: Moving toward an ethically informed approach to autism research. Autism Res. 2011;4(4):271–282; doi: 10.1002/aur.201

35.

35. Taylor

. We Agree, Don’t We? The delphi method for health environments research. Health Environments Research and Design Journal. 2020;13(1):11–23; doi: 10.1177/1937586719887709

36.

36. Braun

, Clarke

. Thematic analysis. In: APA handbook of research methods in psychology, Vol. 2: Research designs: quantitative, qualitative, neuropsychological, and biological. ( Cooper

, Camic

, Long

, Panter

, Rindskopf

, Sher

., eds.) American Psychological Association; 2012: pp. 55–71.

37.

37. Braun

, Clarke

. Thematic analysis: a practical guide. Sage; 2021.

38.

38. Nasa

, Jain

, Juneja

. Delphi methodology in healthcare research: How to decide its appropriateness. World J Methodol. 2021;11(4):116–129; doi: 10.5662/wjm.v11.i4.116

39.

39. Sumsion

. The Delphi technique: An adaptive research tool. Br J Occup Ther. 1998;61(4):153–156; doi: 10.1177/030802269806100403

40.

40. Iqbal

, Pipon-Young

. The Delphi method. Psychologist. 2009;22(7):598–600.

41.

41. Landeta

, Lertxundi

. Quality indicators for Delphi studies. Futures & Foresight Science. 2024;6(1); doi: 10.1002/ffo2.172

42.

42. Molyneaux

, Wellcome Trust D&I Team, Hunt

. Diversity and Inclusion Survey (DAISY) Question Guidance – Working Draft (V2). 2022.

43.

43. Campbell

, Orr

, Durepos

, et al. Reflexive thematic analysis for applied qualitative health research. TQR. 2021;26(6); doi: 10.46743/2160-3715/2021.5010

44.

44. Maxwell

. Critical realism as a stance for designing qualitative research. In: The SAGE handbook of qualitative research design. Sage; 2022. doi:10.4135/9781529770278.n10

45.

45. NIHR. Payment Guidance for Researchers and Professionals. NIHR; 2022.

46.

46. Pellicano

, Adams

, Crane

, et al. Letter to the Editor: A possible threat to data integrity for online qualitative autism research. Autism. 2024;28(3):786–792; doi: 10.1177/13623613231174543

47.

47. We Are Social, Meltwater. Digital 2025 - Global Overview Report; 2025.

48.

48. Orben

, Weinstein

, Przybylski

. Only holistic and iterative change will fix digital technology research. Psychol Inq. 2020;31(3):235–241; doi: 10.1080/1047840X.2020.1820221

49.

49. Bottema-Beutel

, Kapp

, Sasson

, Gernsbacher

, Natri

, Botha

. Anti-ableism and scientific accuracy in autism research: A false dichotomy. Front Psychiatry. 2023;14:1244451; doi: 10.3389/fpsyt.2023.1244451

50.

50. Clark

, Algoe

, Green

. Social network sites and well-being: The role of social connection. Curr Dir Psychol Sci. 2018;27(1):32–37; doi: 10.1177/0963721417730833

51.

51. Aragon-Guevara

, Castle

, Sheridan

, Vivanti

. The reach and accuracy of information on autism on TikTok. J Autism Dev Disord. 2025;55(6):1953–1958; doi: 10.1007/s10803-023-06084-6

52.

52. Van Driel

, Vines

, Barros Pena

, Koteyko

. Understanding autistic adults’ use of social media. Proc ACM Hum-Comput Interact. 2023;7(CSCW2):1–23; doi: 10.1145/3610048

53.

53. Karaminis

, Botha

, Longley

, et al. Language matters in british newspapers: A participatory analysis of the autism UK Press Corpus. Autism in Adulthood. 2024; doi: 10.1089/aut.2023.0105

54.

54. Pearson

, Rose

. A conceptual analysis of autistic masking: Understanding the narrative of stigma and the illusion of choice. Autism Adulthood. 2021;3(1):52–60; doi: 10.1089/aut.2020.0043

55.

55. Turnock

, Langley

, Jones

CRG

. Understanding stigma in autism: A narrative review and theoretical model. Autism in Adulthood. 2022;4(1):76–91; doi: 10.1089/aut.2021.0005

56.

56. Edwards

, Love

AMA

, Jones

, Cai

, Nguyen

BTH

, Gibbs

. ‘Most people have no idea what autism is’: Unpacking autism disclosure using social media analysis. Autism. 2024;28(5):1107–1119; doi: 10.1177/13623613231192133

57.

57. Subrahmanyam

, Šmahel

. The Darker Sides of the Internet: Violence, Cyber Bullying, and Victimization. In: Digital youth-the role of media in development, 1st ed. Springer; 2011, pp. 179–200.

58.

58. Pellicano

, Dinsmore

, Charman

. What should autism research focus upon? Community views and priorities from the United Kingdom. Autism. 2014;18(7):756–770; doi: 10.1177/1362361314529627

59.

59. Saha

, Agarwal

. Modeling social support in autism community on social media. Netw Model Anal Health Inform Bioinforma. 2016;5(1):1–14; doi: 10.1007/S13721-016-0115-8

60.

60. Bellon-Harn

, Saar

, Santhanam

, Heydari

. The diagnostic journey of autistic women as shared on TikTok. Research in Autism. 2025;120:202529; doi: 10.1016/J.REIA.2025.202529

61.

61. Bakombo

, Ewalefo

, Konkle

ATM

. The influence of social media on the perception of autism spectrum disorders: Content analysis of public discourse on YouTube Videos. Int J Environ Res Public Health. 2023;20(4):3246; doi: 10.3390/IJERPH20043246

62.

62. Amaral

, Anderson

, Bailey

, et al. Gaps in current autism research: The thoughts of the autism research editorial board and associate editors. Autism Res. 2019;12(5):700–714; doi: 10.1002/AUR.2101

63.

63. Tibber

, Wang

, Zhang

. The role of platform brand in the association between social media use, stress and educational attainment. Int J Hum Comput Interact. 2023;39(8):1594–1605; doi: 10.1080/10447318.2022.2064035

64.

64. Boulkedid

, Abdoul

, Loustau

, Sibony

, Alberti

. Using and reporting the Delphi method for selecting healthcare quality indicators: A systematic review. PLoS One. 2011;6(6):e20476; doi: 10.1371/journal.pone.0020476

65.

65. Keeney

, Hasson

, McKenna

. A critical review of the Delphi technique as a research methodology for nursing. Int J Nurs Stud. 2001;38(2):195–200; doi: 10.1016/S0020-7489(00)00044-4

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.03 MB

0.01 MB

0.00 MB

0.02 MB