Something Always Sticks – Strategic Bullshit and the Reliance on Uninformed Sources

Last registered on May 27, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Something Always Sticks – Strategic Bullshit and the Reliance on Uninformed Sources
RCT ID
AEARCTR-0018403
Initial registration date
May 20, 2026

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
May 27, 2026, 10:49 AM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Primary Investigator

Name
Mathilde Lea Editha Bechdolf
Affiliation
Otto-von-Guericke-Universität (OVGU) Magdeburg
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Abdolkarim Sadrieh
PI Affiliation
Otto von Guericke University Magdeburg
Contact Investigator
PI Name
Jannik Thomas Greif
PI Affiliation
Otto von Guericke University Magdeburg
Contact Investigator

Additional Trial Information

Status
In development
Start date
2026-05-06
End date
2027-05-06
Keywords
Behavior, Environment & Energy, Governance, Lab
Additional Keywords
Communication, Common-Pool Resource, Behavioral Bias, Environmental Tipping Point, Voice Messages
JEL code(s)
C92, D62, D63, H40, Q20, Q22, Q50
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
Strategic bullshit, defined as uninformed, but goal-oriented communication, is assumed to affect the behavior of the recipients (Frankfurt 2005). We conduct lab experiments to assess, whether the strategic bullshit broadcasted by a sender can sway the participants in a common-pool resource (CPR) game to adapt their extraction levels. The CPR has an unknown tipping-point. If total extraction surpasses the tipping-point, the resource’s regeneration rate falls to a low level. The senders either receive a bonus, if the tipping-point is surpassed (“pro-depletion”), or if it is not surpassed (“pro-sustainability”). Hence, even if they are not informed on the tipping-point, the senders may attempt to communicate strategically to achieve their pro-depletion or pro-sustainability goal. The CPR participants are neither informed on the type of sender they are matched to, nor on the exact level of the tipping-point. However, in some treatments, the CPR participants know that the sender is not informed of the tipping-point (“uninformed sender”), while in others they know that the sender is informed (“informed sender”). Comparing the effect of the messages sent by the uninformed versus the informed senders, we can estimate the power of strategic bullshit.
External Link(s)

Registration Citation

Citation
Bechdolf, Mathilde Lea Editha, Jannik Thomas Greif and Abdolkarim Sadrieh. 2026. "Something Always Sticks – Strategic Bullshit and the Reliance on Uninformed Sources." AEA RCT Registry. May 27. https://doi.org/10.1257/rct.18403-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Participants take part in a common-pool resource (CPR) game in groups of three, in which they repeatedly decide how many units to extract from a shared resource with an unknown tipping point. Prior to the CPR game, groups listen to an audio message recorded by a sender. Participants are told that senders are financially incentivized to either promote depletion or sustainability of the resource, but are not informed about which incentive their specific sender faces. Participants are however told whether their sender was informed or uninformed about the tipping point. Groups play in total 4 CPR games with 15 periods each. Some groups receive no messages and serve as a baseline.
Intervention Start Date
2026-05-06
Intervention End Date
2027-05-06

Primary Outcomes

Primary Outcomes (end points)
Individual extraction level and implemented group extraction level in the common-pool resource game.
Primary Outcomes (explanation)
Extraction is measured at two levels. First, the individual extraction choice, defined as the number of units each participant submits in each of the 15 periods per game. Second, the implemented group extraction level, defined as the median of the three group members' individual choices, which determines the actual extraction outcome for all group members in that period. Both outcomes are directly observed in the experiment.

Secondary Outcomes

Secondary Outcomes (end points)
Socio-demographic characteristics (e.g. age, gender, field of study), social preferences (e.g. risk preferences, trust, altruism), and cultural variables (e.g. power distance, collectivism).
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The experiment consists of three phases, each conducted in the laboratory with separate subject pools. In Phase 1, 40 speakers are recruited and each records messages for 6 scenarios, varying in the information available to the sender and the sender's financial incentives. In Phase 2, 36 raters evaluate 20 audio messages each, with each message rated by at least 3 raters. Raters are financially incentivized based on agreement with other raters. In Phase 3, 100 groups of 3 players (300 players total) participate in common-pool resource games. 20 groups serve as a control treatment and receive no audio messages.
Experimental Design Details
Not available
Randomization Method
Subjects are recruited electronically and randomly assigned to treatments. Scenario assignment within sessions is conducted automatically by the experimental software.
Randomization Unit
Groups of three subjects are randomly assigned to treatments and scenarios.
Was the treatment clustered?
Yes

Experiment Characteristics

Sample size: planned number of clusters
Phase 3: 100 groups of 3 players
Sample size: planned number of observations
Phase 3: 80 treatment groups and 20 control groups play the 15 period CPR game 4 times, resulting in 6,000 median extractions (100 groups × 4 games × 15 periods) and 18,000 individual extraction decisions (300 players × 4 games × 15 periods).
Sample size (or number of clusters) by treatment arms
Phase 3: 100 groups of 3 players each will be recruited, resulting in 300 players. 20 groups are assigned to the control treatment. The remaining 80 treatment groups are distributed across the four treatment conditions (informed pro-sustainability, informed pro-depletion, uninformed pro-sustainability, uninformed pro-depletion), with 20 groups per condition.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Phase 3 recruits N = 100 groups (300 participants), allocated as 20 groups to each of 5 conditions: control plus four treatments in a 2×2 design (Information × Incentive). A priori power analysis (G*Power 3.1, two-sample t-test, two-sided, α = 0.05, 1−β = 0.80) shows that the primary contrast H2 (Informed vs. Uninformed, pooled across incentive: 40 vs. 40 groups) detects d = 0.65, and H1 (Uninformed vs. Control: 40 vs. 20 groups) detects d = 0.76. These effect sizes represent conservative estimates, as the regression analyses exploit the full panel structure of the data (15 periods × 4 games per group), which substantially increases statistical power relative to the group-level t-tests.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Institutional Review Board Certificate of the German Association for Experimental Economic Research e.V,
IRB Approval Date
2026-04-28
IRB Approval Number
gDRa6AIP
Analysis Plan