Shared novelty, self-expansion, and how the 36 questions were misreported
One experiment assigned couples to a novel joint activity and measured reported relationship quality immediately afterwards. A separate procedure, designed to generate closeness between strangers in a laboratory hour, was later rewritten in the press as a method for falling in love.
The model
Arthur Aron and Elaine Aron proposed in 1986 that people are motivated to expand their sense of self — their capabilities, perspectives, and resources — and that close relationships are one of the principal ways this happens, as a partner's characteristics come to be experienced as partly one's own. The model predicts that the sense of expansion is strongest early in a relationship, when there is most to take in, and that it declines as a partner becomes familiar.
From that prediction follows a testable one: joint activity that is novel and stimulating should restore something of the earlier state, while activity that is merely agreeable should not.
The 2000 experiment
Arthur Aron, Christina Norman, Elaine Aron, Colin McKenna, and Richard Heyman reported a set of studies in 2000 in the Journal of Personality and Social Psychology. In the laboratory portion, married couples were randomly assigned to a physically awkward joint task — moving together across a gymnasium while bound at the wrist and ankle, carrying a cushion without using hands or arms — or to a comparison task that was pleasant but undemanding. Relationship quality was measured by questionnaire before and after. Couples in the novel-task condition reported higher relationship quality afterwards than couples in the pleasant-task condition.
The design is a genuine experiment: assignment was random, so the comparison is not vulnerable to the objection that couples who already did stimulating things together were simply happier to begin with. What it delivers is a short-run, laboratory-measured, self-reported difference between two conditions.
The 36 questions
The other widely circulated piece of this work is a 1997 paper by Arthur Aron, Edward Melinat, Elaine Aron, Robert Vallone, and Renee Bator in Personality and Social Psychology Bulletin. Pairs of participants, mostly strangers, worked through escalating reciprocal questions for around forty-five minutes; a comparison group made small talk. The reciprocal-disclosure group reported greater closeness at the end of the session.
What the paper describes is a laboratory procedure for generating temporary closeness between strangers under controlled conditions, built to give researchers a way of studying closeness experimentally. It was not designed as a couples exercise, it was not tested as one, and the paper makes no claim about love.
The popular version dates from 2015, when Mandy Len Catron published an essay in The New York Times describing her own use of the questions. The essay is clear about being a personal account. The coverage that followed was less careful, and the procedure entered general circulation as a technique for producing love on demand — a description that the 1997 paper does not support and does not attempt.
How much weight the findings carry
Both studies are small by present standards, with participant numbers in the dozens per condition. Outcome measures were taken immediately after the manipulation, so what was captured is a short-run shift in reported feeling rather than a durable change; follow-up, where it existed, was brief. Effect sizes were moderate. Neither study has accumulated a body of well-powered, preregistered direct replications, which is the standard the field moved toward after the replication debates that began around 2011. The Open Science Collaboration reported in 2015 that a large proportion of published psychology findings produced weaker effects on direct replication than in their original reports; nothing exempts this literature from that general finding.
None of this makes the studies uninformative. Random assignment is worth a great deal, and the self-expansion model has generated a coherent and testable research programme. It does mean the honest description is narrow: under laboratory conditions, a novel shared task produced a measurable short-run difference in self-reported relationship quality relative to a pleasant one.
What the evidence does not establish
That the laboratory effect persists beyond the session, since follow-up was short or absent. That novelty in a laboratory corresponds to novelty in ordinary life, where activities are chosen rather than assigned. That the 1997 procedure produces attachment, love, or any lasting bond, none of which it measured. That effects of this size, obtained from small samples without direct replication, would hold at the magnitude first reported.
Sources
- Aron, A., & Aron, E. N. (1986). Love and the Expansion of Self: Understanding Attraction and Satisfaction. Hemisphere.
- Aron, A., Melinat, E., Aron, E. N., Vallone, R. D., & Bator, R. J. (1997). The experimental generation of interpersonal closeness: A procedure and some preliminary findings. Personality and Social Psychology Bulletin, 23(4), 363–377.
- Aron, A., Norman, C. C., Aron, E. N., McKenna, C., & Heyman, R. E. (2000). Couples' shared participation in novel and arousing activities and experienced relationship quality. Journal of Personality and Social Psychology, 78(2), 273–284.
- Catron, M. L. (2015). To fall in love with anyone, do this. The New York Times, 9 January 2015.
- Open Science Collaboration (2015). Estimating the reproducibility of psychological science. Science, 349(6251), aac4716.