Guesses about strangers' tastes were more accurate when the strangers differed more from the guesser
Researchers testing 118 people found guesses about strangers' tastes were more accurate when the strangers differed more from the guesser. Participants still started from their own ratings, and Phys.org calls the result surprising.
The Scientist · Science desk
What happened
- Each participant first rated how likely they were to do 105 everyday activities on a numeric scale, which fixed their own anchor on record.
- They then guessed how likely two fictional people and two stereotyped groups, city and village dwellers, were to do 75 of those activities, after 30 were excluded.
- In experiment 1 they placed a stranger relative to a familiar individual, then chose whether the stranger was closer to a neutral group fixed at 5 or to a rated group.
Why it matters
- What doesn't add up contradiction Phys.org's source describes anchoring as a cognitive bias, yet accuracy was higher for targets further from the self, so the bias did not simply cost accuracy here.
- What it limits constraint The targets were fictional people with brief descriptions and stereotyped groups, so the result does not show that people read real strangers better when those strangers differ.
- What it limits constraint With no accuracy scores or effect sizes in the available account, the size of the distance effect cannot be judged from it.
The authors take their frame from navigation. In the spatial domain, they write, "egocentric reference frames encode locations relative to the observer, whereas allocentric frames represent spatial relations relative to external landmarks." [1] Applied to people, the self is the observer and other people's ratings are landmarks. Distance from the self tracks similarity, so more similar preferences count as closer [12]. The authors say research has shown that "specific subregions of the hippocampal formation integrate egocentric and allocentric information during such computations." [2] For social reference frames the evidence is thinner: how they combine beyond simple self-other comparisons is, in the authors' word, "less explored." [3]
The design tests that directly. Experiment 2 swapped the roles of individuals and groups from experiment 1 [11]. The swap lets the authors check whether the pattern depends on the kind of landmark a person starts from. According to the Phys.org report, anchoring persisted with both individuals and groups as comparison points [5].
The thing this doesn't tell you is how large the effect is. The Phys.org account does not report accuracy scores or effect sizes, or how the 118 participants divided between the two experiments [7]. It also describes the strangers as constructed: two fictional individuals with brief descriptions and two stereotyped groups [9]. These were judgments about described characters, so the experiments cannot say whether the pattern holds for real strangers.
We would commit to a narrow reading. In a lab task with constructed targets, guesses were more accurate for targets who sat further from the participant [6], even though participants kept using their own preferences as the starting point [5]. Phys.org's word for that is "surprisingly." [13]
One question we would put to the full paper is whether the distance effect survives when the gap between the two alternatives in each forced choice is held constant.
What to watch
- Accuracy scores and effect sizes in the full Royal Society Open Science paper, including how the 118 participants split across the two experiments.
- Whether the distance effect holds when the gap between the two forced-choice alternatives is held constant.
- Any replication that uses real acquaintances or real strangers' stated ratings as targets.
Clarity's read
What the record supports and how the coverage leans. The claims behind it follow.
Reality
- Evidence40
- Adoption
- Not scored
- Hype gap−5
- Incentives
- Not scored
- Confidence40
Claim ledger
Ranked by verification strength, evidence, and original report placement.
- [1]
The authors write: "In the spatial domain, egocentric reference frames encode locations relative to the observer, whereas allocentric frames represent spatial relations relative to external landmarks."
- [2]
The authors write that extensive research has shown "specific subregions of the hippocampal formation integrate egocentric and allocentric information during such computations."
- [3]
The authors write that whether and how multiple social reference frames are integrated, beyond simple self-other comparisons, is "less explored."
- [4]
Egocentric anchoring is a cognitive bias in which people use their own preferences, beliefs and perspectives as a mental starting point to understand or predict what another person thinks, feels or prefers.
- [5]
A study in Royal Society Open Science suggests the tendency to start from one's own preferences persists when both individuals and groups act as comparison points.
- [6]
Participants were more accurate in their judgments when a stranger's inferred preferences differed more from their own.
- [7]
The researchers conducted two behavioral experiments with 118 participants.
- [8]
In the anchoring phase, participants chose how likely they were to do 105 different activities on a number-based scale.
- [9]
Participants were introduced to two fictitious individuals with brief descriptions and two stereotypical groups ('city dwellers' and 'village dwellers'), and guessed how likely each would be to engage in 75 of the activities; 30 were excluded.
- [10]
In experiment 1, participants inferred a stranger's preference relative to one of the familiar individuals using their previous ratings, then in a two-alternative forced choice task decided whether that stranger's preference was more similar to a neutral group (always a 5) or one of the two groups they had previously rated.
- [11]
In experiment 2 the procedure was reversed: participants inferred a stranger's preference relative to a familiar group, then decided whether it was more similar to a neutral individual or one of two individuals they had rated.
- [12]
In the social-map analogy, people use their own preferences as a starting point and adjust away from it; distance from the self relates to similarity, with more similar preferences ranking as closer.
- [13]
Phys.org's report describes the finding that participants were more accurate for stranger preferences differing more from their own as surprising.
Get this ledger inside ChatGPT. Ask about any of our stories and the answer brings its figure, each claim’s verdict and the sources. Free, no account.
Sources
1 independent publisher whose own reporting we read for this story.
- phys.orgJudging strangers' preferences may actually be easier when they're less like our own
1 article · October 16, 2026
Topics and entities
Follow any of these and your For You feed starts watching them — no settings page required.
Entities
- Royal Society Open ScienceFollow
- Phys.orgFollow
- Egocentric anchoringFollow