Vitalii Vodolazskyi/ShutterstockLabels like autism, attention deficit hyperactivity disorder (ADHD) and dyslexia are not new. But the way we understand them is changing.In recent years, researchers have increasingly worked with neurodivergent people rather than simply studying them from the outside. That change has brought better access to diagnosis, more inclusive approaches in schools and workplaces and a growing challenge to the idea that neurological difference is something to be fixed.Language sits at the heart of that change. But getting it right can feel daunting. Should we say “a person with autism” or “an autistic person”? Are medical terms respectful, or do they quietly reinforce stigma? And who gets to decide these things anyway?For years, professionals were encouraged to use person-first language – phrases such as “person with autism” – to emphasise humanity over diagnosis. But research published in 2016 upended that assumption. Autistic people themselves, it turned out, largely preferred identity-first language: “autistic person”.That finding has been repeated many times since. Until our recent study, however, very little was known about whether the same preferences applied across the wider neurodivergent community. So, our research team – all neurodivergent – set out to discover just that. Read more: What autistic people – and those with ADHD and dyslexia – really think about the word ‘neurodiversity’ In our new study, we surveyed more than 900 neurodivergent adults across the UK about their terminology preferences. Participants identified with a range of diagnoses, including autism, ADHD, dyslexia, dyspraxia, dyscalculia, Tourette syndrome and stuttering. For each, we presented a list of commonly used terms.Some were identity first, such as “dyslexic”. Others were person first, such as “person with dyslexia”. We asked people to rate how likeable and how offensive they found each term. Crucially, we also asked why. Those open-text responses revealed far more than a simple preference list.What we foundOverall, most groups preferred identity-first language. Terms like “autistic people” or “dyslexic people” were seen as more likeable and less offensive. There were important exceptions. People with Tourette syndrome and people who stutter tended to prefer person-first terms.And when we looked more closely, the picture became more complicated still. Some groups – particularly people with ADHD – felt that none of the available terms really fit. Many said existing labels were vague or failed to capture the full reality of their lives. “Attention deficit”, for example, was seen as too narrow. People described ADHD as affecting far more than focus, shaping energy, emotions, creativity and daily functioning in ways the term barely hints at. shutterstock. Vitalii Vodolazskyi/Shutterstock In other words, the issue was not just how language was structured, but whether it worked at all.Across diagnoses, people spoke powerfully about how certain words made them feel. Words such as “disorder” were widely disliked. Many felt they implied something broken or defective, rather than acknowledging that difficulties often arise because society is not designed with neurodivergent people in mind. Several participants said these terms reinforced stereotypes and shaped how others treated them.Respect, identity and disagreementParticipants were also clear about one thing: people should be allowed to describe themselves in the way that feels right to them. Even among autistic participants – a group with a well-established preference for identity-first language – many stressed that others should be free to choose person-first terms if that reflected their own identity.Community infighting over “correct” language was seen as unhelpful. Several people pointed out that neurodivergent communities face far bigger challenges than internal policing of words, including discrimination, exclusion and lack of support.At the same time, participants drew a clear line between self-description and professional language. They felt that teachers, doctors, researchers and journalists should follow group-level community preferences when speaking in general terms – and be open to correction when they get it wrong. Who is using the language, and in what context, mattered enormously. Read more: Why it’s time to rethink the notion of an autism ‘spectrum’ What emerged most clearly from our study was that debates about language are rarely just about words. They are about power. About who gets to define whom. And about whether neurodivergent people are seen as fully human, with authority over their own lives and identities. Participants were often less concerned with perfect terminology than with intent, respect and action. Terminology discussions are not just about language, but about the dehumanisation and associated stigma of people considered “disordered” or “abnormal”. Language shapes action. How we treat people is shaped by whether we see them as being worthy of the same dignity and respect that we afford to those we see as fully human. As such, self-determination, autonomy and respect sit at the centre of such language debates. We recommend listening to neurodivergent people to find out about their preferences and using the words that they prefer, instead of solely being led by traditions which have developed without the input of the communities we are referring to. When it comes to dignity and respect, actions speak louder than words. People want to feel respected and accepted for who they are, regardless of the labels people use to describe their differences.Aimee Grant receives funding from the Wellcome Trust and UKRI.Monique Botha receives funding from the Leverhulme trust, Royal Society of Edinburgh. The Carnegie trust, and the Economic and social research council. I am also an advisory member of the Donaldson's Trust. Amy Pearson does not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and has disclosed no relevant affiliations beyond their academic appointment.