When faced with disagreement on a topic that matters, many people instinctively reach for reassurance: they tell the other person their heart is in the right place. A new study suggests this familiar move may be less effective than an alternative approach. Researchers at Johns Hopkins and partner institutions tested two strategies against each other in a controlled setting and found that naming the other side's likely arguments and validating them produced roughly twice the relational benefit of praising their underlying intentions.

The experiment, published on 12 August in Communications Psychology, involved 579 US adults recruited through Prolific on 17 May 2024, with 558 remaining after exclusions. Participants spent about ten minutes on the study, for which they were paid $2.41. The researchers—Li Li Hwangpo, Lindi R. Shepard, Lisa Nehring, Nan Mu, Katherine J. Cornwall and Hunter Gehlbach, based at Johns Hopkins School of Education, Indiana University Bloomington, and the University of Wisconsin–Madison School of Medicine and Public Health—registered their hypotheses on the same date as recruitment.

How the experiment was structured

Participants first read background material on climate change education policy, learning that New Jersey was the first state to require it across grades and content areas. They then stated their position on a national mandate for climate education in K–12 public schools and rated the strength of their view. Only after this initial response were they randomly assigned to one of three conditions within their opinion group.

All participants then encountered a mock Facebook post from a fictional public-school teacher named Taylor Harris—a gender-neutral name with gender-neutral pronouns and minimal biographical detail to prevent participants from identifying preexisting commonalities. The post argued against the participant's stated position. In the control condition, that counter-argument stood alone.

The two treatment conditions added framing before the counter-argument. In one, the teacher outlined arguments that someone holding the participant's view might make, called them sound, and proceeded with phrases like "despite these very valid reasons…" or "despite these very real concerns…" depending on which side the reader supported. In the other treatment, the teacher named the participant's likely motives and endorsed them, saying "while I appreciate these intentions and our shared motivations…" before moving into the counter-argument. Critically, the teacher never asked the participant anything; the arguments being affirmed were guesses.

Where the two approaches diverged

Both treatments improved relational outcomes compared to the bare counter-argument, but the argument-validation approach outperformed the intention-praising approach by a substantial margin. On whether the teacher appeared to have taken the reader's perspective, the argument version produced a covariate-adjusted Cohen's d of 1.12 (95% CI 0.88 to 1.34) versus 0.45 (0.26 to 0.67) for the intention version. Perceived similarity showed a gap of 0.66 against 0.33. Expected relationship with the teacher came in at 0.69 against 0.36. On fairness—whether the teacher and information seemed balanced—the effect reached 1.37 against 0.56, the largest effect in the entire study.

The authors note that "across these four outcomes, the argument-affirming SPT treatment produced effects that were approximately twice those of the intention-affirming SPT treatment," though their discussion later narrows this to "for most outcomes" and their abstract describes it as what "estimated effect sizes further suggested." The finding surprised the research team, who "had tentatively expected the opposite," reasoning that validating someone's deeper intentions would penetrate more deeply into their identity.

The limitations and caveats

Two results did not withstand scrutiny. A fifth relational measure—whether readers reciprocated by working harder to understand the teacher—showed significance for the argument condition at p = .017, but its confidence interval included zero once a Holm-Bonferroni correction for multiple comparisons was applied. The intention condition was not significant to begin with (p = .220). The authors report three deviations from their preregistered analysis: a Holm-Bonferroni correction as an additional robustness check, an exploratory opinion-change model without the initial-opinion covariate as a sensitivity check, and exploratory comparisons of reading time as an attention check.

Most importantly, neither treatment moved opinions measurably further than the control. Across the entire sample, opinions shifted 0.80 points toward the opposing view on a ten-point scale, but the gap between treatment and control was small and non-significant: p = .195 for the argument condition and p = .602 for intentions. Effect sizes were d = 0.13 (95% CI -0.05 to 0.33) for arguments and d = 0.05 (-0.15 to 0.25) for intentions.

The fairness effect, the study's largest finding, contains a built-in confound. Argument-condition readers who initially supported the mandate estimated that 26.99% of the post's information favored their side, compared to 8.40% among control-group supporters. By design, the treatment posts did carry more of the participant's position because the affirmation was attached to the front. Calling a two-sided message fairer than a one-sided message describes what readers were actually shown.

The four relational measures also correlate strongly with one another, with intercorrelations ranging from r = .71 to .85. The authors acknowledge this themselves, noting that "although these constructs are theoretically distinct and have reasonable internal consistency (Cronbach's ɑ = .86 to .91), a broader range of distinct outcomes might benefit future studies." The two-to-one comparison between treatments was never formally tested; it is a comparison of point estimates, with individual ratios ranging from about 1.9 to about 2.5. On two of the four outcomes, the confidence intervals between the two treatment arms overlap.

The perceived-perspective-taking scale also conflates two distinct concepts. It averages four items—effort, motivation, clarity and accuracy—into a single score, preventing researchers from separating the first from the last. The authors state directly that the scale "did not distinguish perceived SPT accuracy from perceived SPT effort," and individuals "might feel that a perceiver is trying hard to take their perspective but doing so inaccurately." The paper's opening anecdote illustrates the same gap: environmental campaigners who opened town halls in coal country by praising coal "appeared to have been appreciated," yet "merely trying to take the coal miners' perspectives did not ensure accurate SPT."

The scope of the findings

This is a single experiment on a single issue with a US sample. The encounter was "text-based, impersonal, and unidirectional, potentially making the stakes for taking the perspective of the teacher fairly low," lacking the facial expression, gesture and tone of a real argument. The "U.S. political context is unique, potentially affecting the international and cultural generalizability of our findings," the authors write. The paper also arrived with an early-access banner warning that it is an unedited manuscript in which "there may be errors present which affect the content."

The sample itself leaned toward support for the policy. On a ten-point scale from strong opposition to strong support for the mandate, the average starting position was 7.54, and participants rated themselves slightly liberal on average. None of the participants were arguing with someone they would have to face at Thanksgiving.

That distinction carries weight. Nobody in this experiment was arguing with someone they love. This reading comes from a ten-minute study conducted by people without clinical training, and it is not guidance for anyone's relationship. A disagreement corroding a marriage or family belongs with a couples or family therapist, and no choice of opening sentence replaces that work.

What the authors recommend

The paper's practical-implications section addresses advocacy groups, teachers, and specifically "bloggers, op-ed writers, journalists, and others addressing broad, ideologically diverse audiences," suggesting they "might begin by acknowledging the arguments or intentions of those who disagree." This section uses less cautious language than the rest of the article—the gesture "can produce a powerful bridge," such efforts "are likely to enhance audience perceptions of the communicator"—and it extends beyond a design in which nobody spoke to anybody, nothing was measured after ten minutes, and the bridge in question is a self-reported score about a person who does not exist.

The narrower claim holds up. If the goal is for the person you are about to contradict to think better of you afterward, the evidence here supports making a visible attempt to state their case before you dismantle it. Persuading them any faster is not on offer: the paper reports "small, non-significant differences" between treatment and control on opinion change. The familiar reassurance that you know they mean well, the polite version most people reach for, came second in the one setting this experiment tested.

Source: Silicon Canals