Evidence status: conditional support A classic 1969 experiment and later cross-cultural research suggest that, under particular conditions, a small request made directly and accepted voluntarily can increase liking for the requester. The evidence base is limited, the mechanism is unsettled, and the result cannot be extended to costly, pressured or power-unequal requests. This is research draft version 0.1.
How Everyday Psychology Takes Shape · Season One: Why Do We Think This Way? · Article 10
Intuition says that we should like people who help us. Gifts and support produce gratitude; this is the familiar direction of interpersonal exchange. Yet there is a seemingly reversed possibility: after doing a small favour for someone, we may sometimes like that person more than we did before.
Popular psychology calls this the Benjamin Franklin effect. A common piece of online advice follows: if you want someone to like you, do not help them—ask them for a small favour. The claim has both an attractive historical story and a classic experiment. It has nevertheless travelled rapidly from “a change of attitude under some conditions” to a supposed button for controlling relationships.
The evidence warrants a careful possibility, not a universal social law.
From borrowing a book to returning prize money
Benjamin Franklin wrote in his autobiography about a member of the Pennsylvania Assembly who was hostile toward him. Instead of flattering the man directly, Franklin asked to borrow a rare book from his collection. The man agreed. After the book was returned, their interactions improved. Franklin concluded that someone who had once done a kindness might become more ready to do another.
The anecdote cannot establish a psychological effect by itself. Courtesy, shared intellectual interest, respect shown for the property or later interactions could all have changed the relationship. In 1969, Jon Jecker and David Landy tried to isolate the role of a request experimentally.
Seventy-four participants took part in a competition administered by an experimenter and won money. Afterwards, some were asked by the experimenter himself to return their winnings because he had used personal funds and was running short. Others received a request through a third party, framed around departmental finances. A control group was not asked to return the money. Participants who directly helped the experimenter later rated him more positively than those who received no request. The third-party request did not produce the same result.
The original paper’s wording already contained the necessary limit: under some circumstances, doing another person a favour can increase liking for that person. The sample was small, the setting unusual, and the request took place in an experiment controlled by the researcher. It justified further investigation, not the claim that any favour request wins affection.
Later research changed the proposed mechanism
The best-known explanation is cognitive dissonance. Someone who initially dislikes a requester but voluntarily helps faces an inconsistency: Why am I helping a person I dislike? Adjusting the attitude to “perhaps this person is not so bad” can reduce the conflict.
The account is plausible but has not been established as the only mechanism. Yu Niiya and colleagues studied American and Japanese participants, comparing direct requests, help without a direct request and changes in attitudes over time. They found that after a direct, small request was accepted, the requester gained liking in some conditions. The same change did not occur when the help was not personally requested. The researchers proposed a different interpretation: a direct request communicates an intention to establish or maintain affiliation. The helper responds to this relational signal rather than merely rationalising their own behaviour.
The two mechanisms make different predictions. Dissonance emphasises inconsistency between voluntary action and prior attitude; affiliation emphasises the request as a signal of trust, dependence or relational openness. Real interaction may also include a sense of competence—another person has indicated that my knowledge is useful—and the effect of subsequent contact.
No large meta-analysis currently establishes the mechanism and boundary conditions with confidence. Stating that the Benjamin Franklin effect “is caused by cognitive dissonance” is stronger than the evidence.
Why a small favour differs from a substantial burden
A request can communicate relational invitation or exploitation. Its meaning depends on cost, freedom, timing and power. Borrowing a publicly available document or asking about familiar expertise may make a person feel that their competence is recognised. Requesting several hours of unpaid weekend work produces a different event.
Genuine choice is crucial. If an employee cannot realistically refuse a manager, apparent helping should not be interpreted as affiliation. Agreement produced by guilt, threat or fear of relational punishment may lead to resentment. Social role and cultural norms also affect whether a request is appropriate and what accepting it means.
Refusal provides limited information. Declining a small favour does not prove dislike; time, ability or boundaries may be the reason. Turning a conditional average effect into a method of reading an individual mind manufactures relational conclusions that are not there.
Repeated requests alter the structure as well. A first small favour may open an interaction. Continued extraction gives the other person new evidence about motive. No psychological effect cancels fairness, reciprocity or the long record of behaviour.
Can this become a practical technique?
If the goal is to manipulate liking, the technique is poorly supported and likely to damage the relationship. Once a request is recognised as a strategy, what might have been an affiliative signal becomes instrumental deception. An effect can exist without being ethically worth deploying.
A healthier practical lesson is to let other people make meaningful contributions of manageable cost. Someone who refuses all help and completes everything alone may unintentionally close opportunities for reciprocal relationship. Asking about genuine expertise, explaining the need clearly and making refusal easy is more respectful than pretending to be helpless.
After help is given, gratitude should identify the contribution rather than imply an ongoing obligation. Reciprocity need not be settled immediately. Relationships are not accounts in which each exchange must be equal, but over time they need to be sustainable and free of one-sided extraction.
The research outcome “liking increased” must also be kept separate from friendship. A change on an experimental rating may be brief. It does not show that trust, intimacy or responsibility has formed. Enduring relationships pass through disagreement, reliability and shared history; they cannot be reproduced by borrowing one book.
How action can help form an attitude
The deeper interest of this small phenomenon is not that it offers a social shortcut. It challenges the intuition that people first possess a complete attitude and then act from it. Action can alter the meaning of a relationship. After helping someone, the helper has a new fact to interpret: I have already invested in this interaction. The requester has also disclosed a need. A subsequent attitude may form partly through understanding what that event meant.
This helps explain why the form of request matters. A task conveyed by a third party contains a behavioural cost but may lack a direct relational signal between requester and helper. A personal request places both people in an encounter that can be accepted or refused. If the effect appears, it belongs to that whole structure, not just to a bias hidden inside the helper’s head.
The conclusion is that helping can sometimes increase liking when the favour is voluntary, modest, personally requested and carries affiliative meaning. The classic experiment and later research offer limited support for this proposition. Its magnitude, duration and mechanism should not be assumed.
The best lesson is not, “Make others work for you and they will like you.” A more defensible one is that relationships sometimes grow through mutual need. An appropriate request gives another person room to contribute and room to decline. That freedom helps determine whether help becomes the beginning of a relationship or evidence of exploitation.
Primary research and further reading
- Jecker, J., & Landy, D. (1969), “Liking a Person as a Function of Doing Him a Favour”.
- Niiya, Y. (2016), “Does a Favor Request Increase Liking Toward the Requester?”.
- Niiya, Y., Ellsworth, P. C., & Yamaguchi, S. (2006), “Amae in Japan and the United States: An Exploration of a ‘Culturally Unique’ Emotion”.
- Niiya, Y., & Ellsworth, P. C. (2012), “Acceptability of Favor Requests in the United States and Japan”.
- Flynn, F. J., & Lake, V. K. B. (2008), “If You Need Help, Just Ask: Underestimating Compliance With Direct Requests for Help”.
Series navigation: Season One overview: How Everyday Psychology Takes Shape
Discover more from Geoffrey Chen
Subscribe to get the latest posts sent to your email.