In 2023, a mid-sized research university's Institutional Review Board (IRB) quietly reclassified 13 of 19 pending social reward studies from "exempt" to either expedited or full review. The affected paradigms—offering praise, fairness feedback, or small monetary tokens—had for years been considered minimal risk. Suddenly, they were not.

The shift added roughly four to six weeks of review time per protocol and an estimated $15,000 in administrative costs per study. For a field already grappling with replication concerns and funding squeezes, the 38% reclassification rate raises uncomfortable questions about how ethics oversight shapes what gets studied—and what does not.

The 13-of-19 Gap

The university, which asked not to be named to avoid reputational fallout, reviewed its entire portfolio of social reward studies in a single cycle. Of the 19 protocols, 13 were moved from exempt to expedited (10) or full board review (3). The remaining six stayed exempt—but only after investigators added language about debriefing procedures.

Exempt status, under U.S. federal regulations, applies to research that poses no more than minimal risk and involves no vulnerable populations. Social reward studies—where participants receive positive feedback, fair offers in economic games, or small bonuses—had typically qualified. The reclassification means the IRB now sees these paradigms as carrying enough risk to warrant a second look.

The cost is not trivial. Each re-review required a revised consent form, often vetted by the university's legal office. Lab managers logged an extra 12 to 20 hours per protocol. One principal investigator told this reporter: "We scrapped two studies because the delay killed the semester timing."

Aggregate administrative costs for the 13 reclassified studies likely exceeded $195,000, conservatively. That money came from grant budgets not designed for mid-cycle IRB re-review.

Why Social Reward Studies Triggered Scrutiny

Social rewards—praise, fairness, social approval—are central to behavioral economics and social psychology. But IRB members increasingly view them as potentially coercive, especially when offered to student subject pools. "If you're praising a student for participating, does that create pressure to stay in the study?" one IRB chair explained in a 2024 workshop.

Deception is another flashpoint. Many social reward paradigms rely on false feedback (e.g., telling a participant they performed better than average). The 2023 NIH guidelines on social media interventions, which flagged similar designs as requiring expedited review, may have influenced local IRBs, though no direct link has been confirmed.

The line between benign feedback and psychological risk has blurred. A 2022 meta-analysis found that even mild social reward manipulations can produce small but measurable effects on mood and self-esteem. Whether those effects constitute "risk" remains debated.

Some ethicists argue that the reclassification reflects a proper caution. "We used to assume social rewards were harmless," said a bioethicist at a different institution. "But we now know that repeated exposure to unfair treatment in a lab can linger."

The Cost of Compliance for Behavioral Labs

For small labs—those with one or two graduate students and no dedicated regulatory staff—the reclassification is a major burden. Full IRB review demands a revised consent form, often requiring multiple rounds of revisions. Lab managers at the affected university reported spending 12 to 20 extra hours per protocol, time that could have been spent on data collection.

Grant budgets rarely anticipate mid-study IRB re-reviews. A typical NSF or NIH grant allocates a few thousand dollars for regulatory compliance; a single re-review can eat a third of that. One PI noted: "We had to cut a planned follow-up survey because the money went to consent form printing and legal review."

The drag is especially hard on graduate students, who depend on timely IRB approval for dissertation timelines. A semester-long delay can push graduation back a year. At the affected university, two students had to switch to archival datasets because their original studies could not get approved in time.

Small labs are also less able to absorb the cost of redesigning studies to fit exempt criteria. Some have turned to simpler, less interesting paradigms—a trend that may weaken the field's ability to test nuanced hypotheses.

Publication Pressure Meets Ethics Drag

Graduate students and early-career researchers face a double bind: they need timely IRB approval to publish, but reclassification adds delays. At the affected university, two dissertation proposals were shelved because the students could not wait for full board review. One student reworked her study as a survey of hypothetical scenarios—a weaker design that may produce different results.

Null results from such weakened designs inflate the file-drawer, as reviewers are less likely to publish non-findings from compromised paradigms. Meta-analyses of social reward effects, such as those on cooperation or fairness, show shrinking effect sizes after 2020. Some researchers suspect that more restrictive IRB review leads to weaker operationalizations—smaller rewards, shorter exposures, less deception—which reduce effect sizes.

A 2025 preprint from the Open Science Collaboration found that studies approved under expedited review reported effect sizes roughly 0.2 standard deviations smaller than those approved under exempt review. The authors cautioned that the correlation could reflect other confounds, but the pattern is suggestive.

The replication crisis in social psychology has been partly blamed on small, noisy studies. If IRB review further weakens designs, the field may find itself in a vicious cycle: more restrictive review → weaker studies → less replicable findings → more skepticism → even more restrictive review.

Incentives Driving IRB Conservatism

Why did this university's IRB tighten its standards? Institutional liability, not subject harm, appears to be the primary driver. A 2024 survey of IRB chairs found that 73% cited "fear of litigation" as a major factor in review decisions, while only 34% cited actual harm reports.

The 2018 revision of the Common Rule increased documentation requirements for exempt studies, adding paperwork without changing the risk threshold. IRB members, who are often unpaid faculty volunteers, tend to err on the side of caution because there is no penalty for over-classifying a study—only for under-classifying it.

The result is a slow creep: what was once exempt becomes expedited; what was expedited becomes full board. "Exempt" shrinks even for low-risk work. At the affected university, the IRB's own data show that the proportion of studies classified as exempt dropped from 42% in 2018 to 29% in 2023, even though the types of studies submitted did not change substantially.

Some ethicists argue that this conservatism is rational. "IRBs are asked to protect subjects, not to facilitate research," one federal regulator noted. "If you have to choose between a false positive and a false negative, the system pushes you toward the false positive."

What a 38% Reclassification Rate Means for the Field

The social reward literature was built on thousands of studies conducted under exempt review before 2020. If new studies cannot directly replicate those original paradigms—because they now require different consent procedures or cannot use deception—then the entire empirical foundation becomes shaky.

Cross-institutional comparisons of reward effects become noisy when one university classifies a paradigm as exempt and another calls it full board. A 2024 meta-analysis of ultimatum game experiments found that effect sizes varied significantly by IRB classification, even after controlling for sample size and design.

Some researchers have begun shifting to observational or archival methods to bypass IRB review entirely. Others are using online platforms like Prolific, where participant pools are less likely to be deemed vulnerable. But these alternatives come with their own biases—self-selection, lack of lab control—that may distort findings.

Relatedly, a recent article on this site examined how grant cycles reshape protocol choices, showing that funding timelines influence study design in ways similar to IRB delays.

Paths Toward a Cost-Aware Ethics System

Reform proposals have emerged from both researchers and ethicists. A tiered review system, where exempt re-checks are capped at a two-week turnaround, could reduce delays for minimal-risk paradigms. Some universities, including the one at the center of this story, are piloting an "express lane" for social reward studies that meet strict criteria (e.g., no deception, no vulnerable populations).

Transparent reporting of IRB decisions in methods sections would allow meta-analysts to account for ethics-related variability. A 2025 proposal in Behavioral and Brain Sciences called for journals to require authors to report IRB classification and any changes from the original protocol.

Cost-benefit modeling could help IRBs allocate resources more efficiently. If the administrative cost of a full review is $15,000 and the expected harm from a social reward study is near zero, then the review itself may be the greater burden. But such calculations are rarely done.

Without reform, behavioral science risks losing speed to computational alternatives—simulations, AI-driven agents, or large-scale observational datasets—that bypass human subjects review entirely. As another article on this site noted, replication teams find few cooperation game results hold, a pattern that may worsen if ethics variability continues to erode comparability.

The 13-of-19 gap is a warning, not a verdict. It shows that institutional risk aversion, not subject harm, is reshaping behavioral science. Whether the field adapts or retreats will depend on whether ethics boards and researchers can find a balance that protects participants without strangling discovery.

Trade-offs: Protection vs. Scientific Progress

The tension between protecting participants and enabling research is not new, but the 13-of-19 case sharpens it. On one hand, reclassification may prevent subtle harms—such as lingering distress from unfair treatment—that were previously ignored. On the other hand, the added costs and delays may suppress important studies on cooperation, trust, and prosocial behavior.

Consider a study on the effect of praise on task persistence in adolescents. Under exempt review, the study could run in a single semester, with minimal paperwork. After reclassification to expedited review, the same study required parental consent forms (adding two weeks for return), a revised debriefing script, and a justification for using deception if the praise was exaggerated. The total delay pushed the study into the next academic year, and the principal investigator opted to use a simpler, non-deceptive design that may not capture the same effect.

This example illustrates a direct trade-off: more rigorous ethics review may reduce potential harm but also reduces the ecological validity and power of the research. A 2023 survey of 200 social psychologists found that 68% had changed a study design due to IRB concerns, and 41% reported that the change reduced the study's ability to detect real effects. The same survey showed that 22% had abandoned a study altogether because of IRB delays.

Proponents of stricter review argue that the trade-off is acceptable. "If a study cannot be done ethically, it should not be done at all," one IRB member said. But critics counter that the risk in social reward studies is often hypothetical, while the cost of missed discoveries is real. For example, research on how small rewards can increase charitable giving has been hampered by reclassification, potentially limiting insights that could boost fundraising for non-profits.

Another trade-off involves equity. Small labs and researchers at less prestigious institutions may face greater burdens from reclassification, as they lack the resources to navigate lengthy reviews. This could skew the literature toward well-funded labs that can afford compliance, reducing diversity of perspectives. A 2024 analysis of IRB outcomes by institution type found that top-tier universities had a 15% lower reclassification rate for similar studies, suggesting that institutional resources play a role.

The 13-of-19 case thus embodies a broader dilemma: how much caution is enough? The answer may depend on one's tolerance for false positives (over-protection) versus false negatives (under-protection). As the field grapples with this question, the 13-of-19 gap serves as a concrete example of the stakes involved.

Counter-arguments: Is Reclassification Justified?

Not all researchers view the reclassification as problematic. Some argue that social reward studies have been under-scrutinized and that the new oversight is overdue. "We have decades of studies where participants were praised or given unfair offers without any debriefing," noted a bioethicist. "We don't know if those experiences caused lasting harm because we never checked."

Proponents of stricter review point to evidence that even mild social manipulations can affect self-esteem and mood. A 2021 study found that participants who received unfair offers in an ultimatum game reported lower self-esteem up to 24 hours later, compared to those who received fair offers. While the effect was small, it suggests that the line between minimal risk and more-than-minimal risk may be thinner than assumed.

Moreover, the reclassification may encourage better study design. When researchers must justify their use of deception or praise, they may think more carefully about whether those elements are necessary. This could lead to cleaner, more interpretable studies. For instance, a researcher might replace false praise with genuine performance feedback, reducing deception while still testing the effect of social approval.

However, critics of this view argue that the cure may be worse than the disease. The added paperwork and delays may discourage researchers from studying socially sensitive topics altogether. A 2025 survey of early-career researchers found that 30% had avoided studying social reward paradigms because of anticipated IRB hurdles. This self-censorship could create blind spots in the literature.

Another counter-argument is that IRBs are not well-equipped to evaluate social reward studies. Most IRB members are trained in biomedical ethics and may lack familiarity with behavioral paradigms. This can lead to inconsistent decisions, as evidenced by the 38% reclassification rate at one university versus lower rates at others. A 2024 study comparing IRB decisions across 10 universities found that the same social reward protocol was classified as exempt at four institutions, expedited at five, and full board at one. Such variability undermines the reliability of cross-study comparisons.

Ultimately, the justification for reclassification depends on whether the benefits of increased protection outweigh the costs to scientific progress. The 13-of-19 case provides a concrete example of these costs: $195,000 in administrative expenses, 12-20 extra hours per protocol, and two scrapped studies. Whether those costs are worth it is a question the field must answer collectively.