Welcome back. The previous lesson found a selective pattern in Nedergaard and Lupyan’s study: people with higher self-reported inner speech performed better on word recall and rhyme judgments, while their scores did not predict a difference on the two other tasks tested. The low-scoring group was not a group verified to have no inner speech.
This final lesson puts that finding beside a methodological critique and evidence from experience sampling. The aim is to assess a public claim without either dismissing the behavioral results or treating a questionnaire score as a diagnosis.
What is the public claim actually claiming?
The title of a public interview, Meet the Neuroscientist Who Discovered That Some People Have No Inner Speech, provides a useful case. Watch selected parts of the interview with study author Johanne Nedergaard. Notice the difference between a headline asserting discovery and a researcher discussing how people were identified.
Meet the Neuroscientist Who Discovered That Some People Have No Inner Speech
In Meet the Neuroscientist Who Discovered That Some People Have No Inner Speech, from Giant's Shoulder, the study author discusses the uncertainty behind the headline.
Watch continuum or category for the question of whether very little inner speech and none are distinct. Then watch recruitment limits for the questionnaire-based selection method, and prevalence uncertainty for why an exact estimate remains difficult.
“There are people who report little inner speech” and “there are people who never experience inner speech” are different claims. So are “a person seldom has spontaneous verbal thoughts” and “a person cannot deliberately produce them.” An interview title can collapse all four into the phrase no inner speech.
Return briefly to the original paper. Its abstract reports lower performance among adults who reported low levels of inner speech, not performance measured in a group independently shown to have none. In its limitations, the authors acknowledge that many members of their low-scoring group reported some inner speech. They also distinguish the prevalence of inner speech in experience from the ability to control its use.
Not Everybody Has an Inner Voice
Revisit Nedergaard and Lupyan's original study to separate its measured results from its proposed name for a possible absence of inner speech.
On the first page, read the Abstract, from its opening through the task findings. In Results → Rhyme judgments → Statistical models (pp. 9–10), scan speed and accuracy; focus on which result was statistically supported. Finally, read Limitations (pp. 15–16), from the first limitation through the section's end. Attend especially to alternative explanations, situational variation, and the authors' description of their low-scoring group.
The paper gives us a meaningful association: questionnaire responses predicted performance on some verbal tasks. It does not establish that complete absence caused poorer performance. Participants were selected by prior questionnaire scores, not assigned an inner-speech condition; the tasks did not directly reveal what each participant consciously experienced while solving them.
Even a small wording difference matters when checking sources. The study found an association with rhyme-judgment accuracy, but not a reliable association with response speed. A retelling that says lower scorers were both less accurate and slower presents the latter as firmer evidence than the original analysis supports.
What does the critique challenge?
A commentary on the study accepts the value of investigating individual differences and the verbal-task findings. Its objection is more specific: low reported prevalence is not evidence that someone has no inner speech at all. The critic points out that the paper does not identify a participant established to lack inner speech completely. That observation challenges what the study demonstrates; it does not demonstrate the opposite proposition that everyone experiences inner speech.
Read the critique as an argument about the strength of an inference, not as a replacement result from another experiment.
Are There Really People With No Inner Voice? Commentary on Nedergaard and Lupyan (2024)
This methodological commentary distinguishes evidence for less reported inner speech from evidence for its complete absence, then examines what other methods can add.
After Research Transparency Statement, read from the central objection to the end of that paragraph. In the following discussion of anecdotes and Descriptive Experience Sampling, start at the paragraph opening “Nedergaard and Lupyan (2024) also back up their claim”; give particular attention to the sampling evidence through that paragraph's end. In the final discussion before Acknowledgments, read proposed validation and the conclusion.
The critique’s argument about the questionnaire is conditional: if it reliably captured the frequency of inner speech, screening over a thousand people might have been expected to identify someone reporting none. That is a reason to examine the instrument and recruitment more closely, not proof that a person with no inner speech cannot exist. A questionnaire item about how one generally thinks may also be answered differently from a question about a rare experience during a particular activity.
The critique likewise tests anecdotes against their details. Someone who says they lack a constant voice but describes occasional verbal “jot notes” is evidence against a constant monologue, not necessarily evidence against all inner speech. This is why the definition matters: a thought can be verbal without sounding like a full, audible conversation.
There is a useful balance here. The original authors do not claim that their low-scoring participants were all completely without inner speech; they explicitly flag that limitation. The critic argues that stronger language used elsewhere about a complete lack nevertheless outruns the data. The disagreement concerns the step from measured variation to categorical absence, not whether the reported word-task associations should be ignored.
What can complementary methods establish?
Descriptive Experience Sampling (DES) investigates particular moments rather than asking for a general self-description. Participants note what was in awareness just before an unpredictable signal, then discuss sampled moments in detailed interviews. This can clarify whether a report of “talking to myself” refers to experienced words, a nonverbal thought later put into words, or something less easily classified.
The following interview explains the procedure and why an initial self-description may change during careful discussion.
No. 1 Inner Experience Expert: Inner Monologue, Aphantasia and How Your Mind Lies to You
In No. 1 Inner Experience Expert: Inner Monologue, Aphantasia and How Your Mind Lies to You, from Giant's Shoulder, DES researcher Russell Hurlburt describes how interviews refine reports about inner experience. Watch for the method, while treating his interpretations as interpretations rather than a direct readout of anyone's mind.
Watch description problems for examples of how someone may use “self-talk” more broadly than the interviewer does. Then watch sampling procedure for the unpredictable signals, immediate notes, and repeated interviews.
DES answers a more precise question than a general questionnaire: What did this person report experiencing at these sampled moments, under the distinctions developed in the interviews? It still cannot inspect every moment. In the study discussed by the critique, two participants reported no inner speech in their everyday-life samples but reported it frequently while golfing. That example shows why none observed in one set of contexts cannot be expanded to none possible in any context. The study involved ten participants, so its percentages should not be treated as population prevalence estimates.
Careful interviewing also brings its own judgment calls. Asking for the exact sequence of words can help distinguish experienced speech from a verbal description supplied afterward. But an especially narrow criterion might miss condensed or fragmentary forms that another researcher would count as inner speech. Repeated interviewing may improve clarity while also changing how a participant attends to or describes experience. Neither a questionnaire nor DES is an infallible test.
The methods are most useful when their different targets stay visible:
| Evidence | What it can support | What it cannot establish on its own |
|---|---|---|
| General questionnaire | A person’s reported tendency toward verbal inner experience | The contents of every moment, or complete absence |
| DES with interviews | Carefully described experience at sampled moments | That unsampled contexts contain no inner speech |
| Verbal-task performance | An association with, or under an intervention a contribution from, speech-related processing | Direct observation of a conscious inner voice |
Earlier lessons also considered interference from repeated speech during a task. Such experiments can help test whether speech-related processing contributes to performance under specified conditions. They do not, by themselves, tell us whether a particular participant usually experiences an inner voice—or whether someone who reports little of it can never use verbal processing.
A defensible appraisal
A fair assessment of the interview title is partly supported, but stronger than this study establishes. Research gives good reason not to assume that everyone has a continuous inner monologue. Nedergaard and Lupyan add evidence that differences in self-reported inner speech have selective behavioral correlates: in their sample, higher verbal scores predicted better word recall and more accurate rhyme judgments, but not a detected difference on their switching or visual-discrimination measures.
The unsupported leap is from those findings to a confirmed group of people who never experience inner speech. The low-scoring group included reports of some inner speech. The overt-speech comparisons provide clues about possible strategies, but speaking aloud was not randomly assigned. The questionnaire does not settle what happened at every moment or whether a person could produce inner speech deliberately.
A stronger investigation of complete absence would need to specify what counts—including condensed, non-auditory verbal thought—and seek converging evidence across questionnaires, carefully interviewed samples in varied contexts, and attempts to elicit inner speech deliberately. Behavioral tasks could help investigate consequences, provided their results were not mistaken for direct reports of experience. Even then, a failure to observe inner speech would require a carefully bounded conclusion rather than certainty about every possible circumstance.
Conclusion
The central distinction of this course is between what someone reports experiencing, what they can do when prompted, and which processes help on a task. The original anendophasia study finds selective associations worth explaining; the critique shows why they do not confirm complete absence; experience sampling offers a closer view of particular moments while leaving other moments and contexts open.
The most defensible position is neither “everyone silently thinks in words” nor “a low questionnaire score diagnoses anendophasia.” People’s inner experiences vary, and whether some people entirely lack inner speech—as opposed to experiencing it rarely or in particular contexts—remains an open empirical question.