Translated from the Portuguese original. Read the original (Português)
Echolalia is the repetition, in echo, of speech heard before: from someone else, a video, a song, or the person's own earlier speech. It can be immediate (seconds later) or delayed (days, weeks, months later). For decades it was treated as "meaningless" repetition, but research has shown the opposite: most of the time echolalia communicates, organizes, and rehearses language. It is also considered a common phase of typical development, before two and a half years of age. The target of clinical work today is not to erase the repetition — it is to expand the person's functional communication.
Echolalia is the repetition of previously heard utterances, with full or partial correspondence in form to the model. In the DSM-5-TR, it appears as an example of "stereotyped or repetitive speech," one of the items in the domain of restricted, repetitive patterns of behavior, interests, or activities of Autism Spectrum Disorder; the ICD-11 (WHO, code 6A02) describes it among the features of atypical language development in autism, alongside pronoun reversal. More recent literature argues that classifying it only as a stereotypy is incomplete: it behaves like an atypical communication pattern, with identifiable functions.
Three ways of repeating
Echolalia is not a single thing. The difference between the types lies in the time interval and in how much the speech was modified — and that distinction matters, because the mitigated form tends to be read as a step toward the person's own speech.
- Immediate: the person repeats right away, or a few seconds after hearing it.
- Delayed: the person repeats it minutes, days, weeks, or even months later — snippets from cartoons, songs, commercials, or things said by someone at home. This is what many people call "speaking in scripts."
- Mitigated: the repetition comes altered — a pronoun changes, a word is added or removed, the intonation shifts from a question to a request.
Repeating is not speaking without meaning
The turning point in understanding echolalia came from patient observational work. In a study published in 1981, Prizant and Duchan followed four autistic boys (ages 4 to 9) at home and at school over eight months and analyzed, on video, 1,009 utterances one by one — the context of the communication, the structural features of what was said, the response latency, and the behavioral and linguistic aspects of each occurrence. The result went against the prevailing idea: the repetitions were distributed across identifiable functional categories, not random noise. In a later re-reading of that data, a 2021 narrative review calculated that only about 4% of the observed repetitions had no apparent purpose.
The study described seven functional categories of immediate echolalia. Three years later, Prizant and Rydell did the same for delayed echolalia and arrived at fourteen categories, which varied along three dimensions: how much the speech was directed at someone, how much the person appeared to understand what they were saying, and how much it related to the situation. Here are some of those functions:
Turn-taking
The person repeats to take their turn in the conversation. They have understood that this is the moment to say something, but don't yet have their own sentence available — so they return what they just heard. It is a legitimate form of participation in the interaction, not a lapse of attention. In Prizant and Rydell's analysis, delayed echolalia also appears serving to call someone, protest, direct the other person's action, and sustain social contact.
Requesting, affirming, and naming
Repeating "Want juice?" can be the way the person found to ask for juice — and it is precisely in this function that mitigated echolalia tends to appear, with the sentence adjusted. Repeating the whole question can also function as a "yes": it is the available way to agree. And there is the declarative function, in which the person repeats to name or point at something, often together with a gesture.
Rehearsal and self-regulation
Sometimes the repetition is a rehearsal: the person repeats — often whispering, at low volume — like someone using speech to process what they heard. In other cases, it is self-regulatory: the repetition accompanies the person's own action, produced together with the movement, in a function that is more cognitive than communicative — here echolalia comes close to other stereotypies. There remains the non-focused category: the repetition comes out with no apparent intention, directed at no one, and can occur at moments of strong excitement or discomfort — the most automatic of the seven varieties.
It also happens in typical development
Echoed repetition is not, by itself, a sign of autism: it is widely considered a common phase of typical language development. Echoed repetition appears frequently in neurotypical children under 30 months and gradually gives way to the child's own speech as vocabulary and grammar become organized. Repeating whole chunks of speech seems to be, for some children, a pathway into speaking in more than one word.
In autism, echolalia is considerably more frequent and can persist: estimates in the literature indicate that echolalia appears in about 75% to 80% of autistic people who develop speech, and it can remain present well beyond early childhood. When the person does not develop enough speech to communicate, augmentative and alternative communication resources come into play — and these are also verbal behavior.
Echolalia and "echoic" are not the same thing
Anyone reading about verbal behavior will come across both terms, and they are easily confused. Worth separating:
The distinction has a practical consequence: the echoic prompt is a widely used and very useful teaching tool — and the fact that the person repeats well is, in this case, a resource, not a problem to eliminate.
What helps — and what the evidence actually shows
Care is needed here, because the literature has changed direction in recent years. A 2023 systematic review (Blackburn, Tueres, and colleagues) gathered 15 studies published between 1980 and 2023. The interventions, in general, reduced echolalia — but the review concluded that the quality of the studies was very low, with substantial inconsistency in definitions, implementation, and measures, which prevents evidence-based recommendations. Two procedures showed up with more consistent results:
- Cues-Pause-Point (McMorrow and Foxx, 1986): the adult structures the question, makes a deliberate pause, and points to a cue, so that the correct answer — rather than the echo — is what the person produces. Four studies in the review tested it, all showing a drop in echolalia.
- Tact training (teaching to label): four studies; three recorded a drop in echolalia and all four recorded an increase in labeling and in correct responses without a prompt. In other words, the gain came from teaching what to say, not from blocking the repetition.
The shift in targetA 2025 scoping review (Dinello and Gladfelter), covering 74 articles, found an uncomfortable pattern: almost all the studies reported reduced echolalia, and very few reported increased speech or communication attempts. Since echolalia is a functional form of communication — and autistic self-advocates oppose interventions aimed at eliminating autistic traits — the review recommends that the target stop being "reduce echolalia" and become expanding functional communication. The 2023 review points in the same direction, calling for alignment with neuroaffirming principles.
In practice, this means: finding out what that repetition serves at that moment (requesting? participating? calming down?), teaching a form of communication that serves the same function, allowing response time, using the person's repetition repertoire as a starting point — and measuring what matters, which is whether the person manages to say what they want to say. See ABA and functional analysis.
About the "gestalt language processor"It circulates widely online that children who use a lot of delayed echolalia would be "gestalt language processors," with their own protocol (Natural Language Acquisition). The intuition behind this — that echolalia is meaningful and should not simply be suppressed — is well supported. The label and the protocol, however, still lack empirical support: critical analyses published in 2024 and 2025 point out that the concept is conceptually problematic and that treatment studies evaluating it are lacking. Treat it as a hypothesis under discussion, not as an established finding.
When it is worth seeking an evaluation
Echolalia, by itself, does not confirm any diagnosis. It makes sense to seek a professional evaluation when:
- the repetition is the main means of communication for a child well past two and a half years of age, or language development is otherwise different from what is expected;
- the person seems to want to say something and cannot — there is frustration or impairment in daily life, at school, or at home;
- echolalia appears newly in someone who used to speak differently, especially in teenagers and adults. Echolalia is not exclusive to autism: it is also described in catatonia (where it is listed among the DSM-5-TR criteria), in Tourette syndrome, in some aphasias, in dementias such as the behavioral variant of frontotemporal dementia, and in cases of intellectual disability. In this scenario, a medical evaluation is necessary.
Language and communication are evaluated by a qualified team — typically speech-language pathology, together with psychology and medicine as the case requires. See also early signs and autism diagnosis.
Test your understanding
Quick quiz
Three questions to reinforce the concept. You see the explanation right away.
Further reading
- American Psychiatric Association. DSM-5-TR: Diagnostic and Statistical Manual of Mental Disorders. 5th ed., text revision, 2022.
- World Health Organization (WHO). ICD-11 — 6A02 Autism spectrum disorder.
- Prizant, B. M. & Duchan, J. F. "The functions of immediate echolalia in autistic children". Journal of Speech and Hearing Disorders, 46(3), 241–249, 1981. And: Prizant, B. M. & Rydell, P. J. "Analysis of functions of delayed echolalia in autistic children". Journal of Speech and Hearing Research, 27(2), 183–192, 1984.
- Pruccoli, J., Spadoni, C., Orsenigo, A. & Parmeggiani, A. "Should Echolalia Be Considered a Phonic Stereotypy? A Narrative Review". Brain Sciences, 11(7), 862, 2021.
- Blackburn, C., Tueres, M. et al. "A systematic review of interventions for echolalia in autistic children". International Journal of Language & Communication Disorders, 58(6), 1977–1993, 2023.
- Dinello, A. & Gladfelter, A. "Intervention Techniques Targeting Echolalia: A Scoping Review". American Journal of Speech-Language Pathology, 34(3), 1528–1543, 2025.
- Hutchins, T. L., Knox, S. E. & Fletcher, E. C. "Natural language acquisition and gestalt language processing: A critical analysis of their application to autism and speech language therapy". Autism & Developmental Language Impairments, 9, 2024.
- Xie, F., Pascual, E. & Oakley, T. "Functional echolalia in autism speech: Verbal formulae and repeated prior utterances as communicative and cognitive strategies". Frontiers in Psychology, 14, 1010615, 2023.
- McMorrow, M. J. & Foxx, R. M. "Some direct and generalized effects of replacing an autistic man's echolalia with correct responses to questions". Journal of Applied Behavior Analysis, 19, 289–297, 1986.
- Skinner, B. F. Verbal Behavior, 1957; and Cooper, J. O., Heron, T. E. & Heward, W. L. Applied Behavior Analysis, 3rd ed., 2020.
This entry is educational and does not replace assessment, diagnosis or treatment by a qualified health professional. Every person is unique — for a specific case, seek individual guidance.