Why Talking to Yourself in the Third Person Changes How You Feel

Talking to yourself in the third person changes how you feel because using your own name instead of "I" creates psychological distance from the experience, and distance reduces emotional reactivity. "Why is Anthony upset about this?" gets processed more like a question about another person than a question about you, which quiets the self-referential machinery that makes an emotion feel overwhelming. The unusual part, and the reason this line of research stands out, is that the shift appears to cost nothing: brain measures suggest it does not draw on the effortful control that most emotion-regulation strategies require.

The behavioral evidence

Ethan Kross and colleagues at the University of Michigan published "Self-Talk as a Regulatory Mechanism: How You Do It Matters" in the Journal of Personality and Social Psychology in 2014. It reported seven studies with 585 participants total.

The manipulation was small to the point of seeming trivial. Before a stressful task, participants reflected on their feelings using either first-person pronouns or their own name and second-person pronouns. Everything else was held constant.

Participants using non-first-person language appraised upcoming stressors as challenges rather than threats. They performed better on tasks such as giving a speech to evaluators, as rated by observers who did not know the condition. They ruminated less about the experience afterward. And the effects held for socially anxious participants, which matters because many regulation strategies work well for people who need them least.

Kross's group has continued this line of work, and the lab's public page on distanced self-talk collects the papers.

The neural evidence, and what "without effort" actually means

Jason Moser led the study that gave this research its neuroscience, published in Scientific Reports in 2017, with Kross among the co-authors. It ran two experiments using different methods.

In the first, 29 participants viewed aversive images while EEG recorded event-related potentials. The measure of interest was the late positive potential, a well-validated marker of emotional reactivity. Under first-person reflection, negative images produced a larger late positive potential than neutral ones, as expected. Under third-person reflection, that difference was not significant in the late window from roughly one to six seconds. The emotional amplification was reduced.

The second experiment used fMRI with 50 participants recalling negative autobiographical memories. Third-person reflection produced less activity in the left medial prefrontal cortex and anterior cingulate, regions associated with self-referential processing. The emotional content was still there. The degree to which it was being processed as being about the self was lower.

The claim about effort rests on a specific comparison. The researchers also measured markers associated with cognitive control, including the stimulus preceding negativity in the EEG study, and found no significant increase under third-person reflection. This is what separates the finding from ordinary reappraisal, which typically shows up as increased recruitment of prefrontal control regions and feels effortful to do.

It is worth being precise about the logic here. "No extra effort" is a null result on a particular measure. A null result is weaker evidence than a positive one, especially at these sample sizes, and it does not prove the absence of an effect. It is consistent with the effortlessness claim rather than proof of it.

The replication

This is where the story is better than usual for a headline-friendly psychology finding. A direct replication published in Biological Psychology in 2022 ran 67 participants through an emotion regulation picture task, comparing distanced self-talk against detached reappraisal while recording ERPs. It reproduced the core result: distanced self-talk reduced the late positive potential without increasing the pre-processing demands that reappraisal shows.

Independent replication with a larger sample is not something most widely publicized effects can claim. It does not make the finding certain, but it puts distanced self-talk on considerably firmer ground than most of what gets written about under the heading of self-talk.

4 forms of distanced self-talk, ranked by evidence

1. Use your own name. "Why is this bothering Anthony?" This is the exact manipulation used in the Kross and Moser studies, so it has the most direct support. It also feels the strangest to do, which is worth knowing in advance so you do not conclude the awkwardness means it is not working.
2. Use second person. "You have handled worse than this." Grouped with name use as non-first-person language in this research and tested alongside it. Slightly less awkward, and widely used spontaneously, which is part of what prompted the research in the first place.
3. Ask rather than assert. The studies framed reflection as a question about the self rather than a declaration about the self. Questions invite an answer instead of a rebuttal, which matters given that self-referential assertions can trigger the counterargument problem described in why some affirmations make people feel worse.
4. Distance in time. Asking how much this will matter in a year is a related technique from the broader self-distancing literature. It has its own supporting studies but was not the manipulation in the two papers above, so it belongs in a separate evidential bucket rather than borrowing their credibility.

What this research does not establish

The samples are small. Twenty-nine participants for an EEG study and 50 for an fMRI study were normal for the field in 2017 and are modest by current standards. Neuroimaging with small samples has a documented history of producing effects that shrink on closer examination.

The authors themselves report a result that did not come out as predicted. They expected third-person reflection to modulate amygdala activity and it did not, which they describe as surprising and attribute to the memory-based paradigm. A paper that reports its own inconvenient null is more trustworthy than one that does not, and it is also a reminder that the mechanistic picture is incomplete.

The outcomes are immediate. These studies measured reactivity during and shortly after a laboratory task. There is no evidence here about what happens if you do this daily for a year, and no evidence that it treats an anxiety disorder or depression. It is a moment-to-moment regulation tool with laboratory support, not a clinical intervention.

Finally, this is a different thing from the affirmation practices it often gets bundled with. Distanced self-talk changes the grammar of how you reflect. It does not assert anything flattering about you, which may be exactly why it avoids the failure mode that trips up positive self-statements.

Using it

The practical version is unfussy. When something lands badly, restate the situation using your name. Ask what the person you just named should do next. That is the whole intervention, and the evidence suggests the benefit comes from the grammatical shift rather than from anything you conclude.

It pairs naturally with the instructional approach that dominates the performance literature, where naming the next concrete step outperforms generating enthusiasm. The comparison between instructional and motivational self-talk covers which one to reach for and when. Timing may also matter more than it appears, given what is known about how the morning brain handles self-directed thought, though that connection has not been tested directly with distanced self-talk.