Starmans, C. & Bloom, P. (2012). Windows to the soul: Children and adults see the eyes as the location of the self. Cognition 123(2), 313–318. DOI 10.1016/j.cognition.2012.02.002.
Key takeaways
From the reading, NOTE-995.
- An indirect question. Instead of asking where the self is, the study asks which of two pictures shows an object “closer to Mary”. If the self were spread evenly or nowhere, all positions should be equal.
- Head over body, eyes within the head. On a girl, both groups chose the fly on the head over the lower body (children 70%, adults 76% of such trials). Children chose eyes over mouth (p = .05); adults chose them equally.
- The eyes, not the head. On an alien with its face on its chest, adults chose the fly at the chest-eyes over every other position (77% chose it most often). Children were split between the eyes (49%) and the top of the head (39%), not significantly different.
- Not what the character can see. On a profile, both groups chose a snowflake in front of the eyes most often, and chose one behind the head, out of sight, over one in front of the feet, in sight (adults p = .05, children p = .01).
- What it shows is a judgement about others. The conclusion that people “intuitively think of the self as occupying a physical location within the body, close to the eyes” assumes that “closer to Mary” means “closer to Mary’s self”. The proposal that this rests on “a more intuitive or phenomenological sense of where in our bodies we reside” is not tested.
Standing in the record
Filed on 2026-10-20 at the owner’s request, in the round of experimental work near the rubber hand illusion. It is a folk-judgement study in the strict sense, of children and adults judging pictures, and it sits near the rubber hand topic rather than building on it. Read in full: Active.
It bears on none of the rubber hand THEORYs. THEORY-766, THEORY-769 and THEORY-767 concern what makes a body part felt as one’s own; THEORY-458 concerns ownership against agency. Nothing here manipulates or measures either. It bears on THEORY-768 only by analogy: Blanke & Metzinger place the origin of the weak first-person perspective inside self-location, and these participants place a drawn person at the eyes. That is a judgement about where someone else is, not a measure of anyone’s felt self-location, so it neither meets nor bears against that THEORY’s promotion test.
On the record’s self holdings it is about where “the self” is located in the eyes of ordinary judges, and the paper slides between person, self and soul, which ADR-024 and TERM-045 keep apart. Under QUESTION-048 it establishes something about the attribution, not about the thing attributed. It does not bear on THEORY-417’s questions of personal identity: nothing here is about persistence or what a person is. It carries no instruction for machine-learning practice.
Filed alongside: Alsmith & Longo 2014 (LIT-1266), which criticises it as “projected” self-location and finds the torso as well as the head when people point at their own bodies.