Case File 307 · Investigation Psychology

What Is Anchoring Bias in an Investigation?

By Murder Book Editorial TeamPublished by Gungrounds, the studio making Murder BookPublished 3 October 2026Updated 3 October 2026

Anchoring bias is the tendency to rely too heavily on the first piece of information encountered when making subsequent judgments. In a criminal investigation, the anchor is often the earliest significant evidence: a witness identification, a fingerprint database hit, a confession or a tip. Once that anchor is set, later evidence tends to be evaluated in relation to it rather than on its own merits. Research in cognitive psychology and forensic science shows that anchoring can affect decisions made by investigators, forensic examiners, prosecutors and judges, and that even experienced professionals are not immune. Understanding anchoring is essential to understanding how tunnel vision takes hold.

Key takeaways

Anchoring bias causes people to give disproportionate weight to the first significant piece of information they encounter.
In investigations, early evidence such as a witness identification or database hit can anchor the entire case to one suspect.
Research shows that even experienced judges and forensic examiners are affected by irrelevant anchors.
Anchoring is closely related to confirmation bias and tunnel vision but describes a specific mechanism: the outsized influence of initial information.
Safeguards include delaying hypothesis formation, blind forensic analysis and structured consideration of alternatives.

Where does the concept of anchoring come from?

Amos Tversky and Daniel Kahneman introduced the anchoring and adjustment heuristic in their landmark 1974 paper in Science, "Judgment under Uncertainty: Heuristics and Biases." They demonstrated that when people make estimates under uncertainty, they tend to start from an initial value, the anchor, and adjust from it. The adjustments are typically insufficient, so the final estimate remains closer to the anchor than the evidence warrants.

In one well-known experiment, participants were asked to estimate a numerical answer after first seeing an arbitrary number produced by a spinning wheel. Those who saw a higher number consistently gave higher estimates, even though the wheel had nothing to do with the question. Tversky and Kahneman concluded that these heuristics are "highly economical and usually effective, but they lead to systematic and predictable errors."

How does anchoring affect criminal investigations?

In an investigation, the anchor is not a random number; it is often the first compelling lead. A fingerprint match, a witness pointing to someone, or an early confession can become the starting point around which all subsequent reasoning is organised.

The Barrister Magazine described the process this way: a critical event at the start of a case anchors the investigation down a certain path, and what follows are confirmation and expectation effects that keep the case anchored to the same initial assumption. Examples include someone flagged on CCTV, a fingerprint or DNA database hit, or a convincing eyewitness account.

Once the anchor is set, confirmation bias reinforces it. Investigators tend to seek evidence that supports the anchored theory and to interpret ambiguous evidence as consistent with it. The combination of anchoring and confirmation bias is a primary driver of investigative tunnel vision.

Can anchoring bias affect forensic experts?

Yes. Dror, Charlton and Péron (2006) showed that fingerprint examiners were vulnerable to contextual anchoring. When examiners were given the same fingerprints they had previously analysed, but with misleading background information suggesting the prints were not a match, four out of five changed their original conclusions. The physical evidence was identical; only the contextual anchor had shifted.

Kassin, Dror and Kukucka (2013) extended this finding across multiple forensic domains. They described how a confession can serve as an anchor that corrupts subsequent forensic analysis, witness memories and investigative decisions. They termed this forensic confirmation bias and proposed that forensic examiners should evaluate evidence without knowledge of confessions or suspect information to reduce its influence.

The 2009 National Academy of Sciences report, Strengthening Forensic Science in the United States, recognised cognitive bias, including anchoring, as a concern across all forensic disciplines and called for further research into its effects on expert judgment.

Does anchoring affect judges and sentencing?

Englich and Mussweiler demonstrated in a series of studies that even experienced criminal judges are susceptible to anchoring effects. In one experiment, judges were given identical case materials but different sentencing demands from the prosecutor: some received a demand for two months' imprisonment, others for 34 months. Despite having the same facts, judges who received the higher demand imposed significantly longer sentences on average (28.7 months versus 18.78 months).

The researchers found that judges appeared to test whether the prosecutor's demand was appropriate by selectively retrieving information consistent with it, a process that anchored their final decision to the initial number. This finding is striking because judges are trained to evaluate evidence independently and are aware of the need for impartiality.

How is anchoring different from confirmation bias and tunnel vision?

These three concepts are related but describe different things:

Anchoring bias is about the outsized influence of initial information. It explains why the first lead matters so much.
Confirmation bias is about the ongoing tendency to favour hypothesis-consistent information. It explains how the anchor gets reinforced over time.
Tunnel vision is the broader investigative pattern that results when anchoring, confirmation bias, belief perseverance and institutional pressures combine.

Anchoring sets the starting point. Confirmation bias keeps the investigation on the same track. Tunnel vision is the outcome. Understanding all three helps explain why investigations can go wrong even when individual participants are acting in good faith.

What can reduce the influence of anchoring?

Researchers have suggested several approaches:

Delay hypothesis formation. Gathering more evidence before committing to a theory reduces the chance that one early finding will dominate the case.
Blind forensic procedures. Presenting evidence to forensic examiners without case context, as Kassin, Dror and Kukucka (2013) recommended, limits the opportunity for an anchor to bias their analysis.
Consider alternative explanations. Actively generating competing hypotheses forces engagement with possibilities beyond the anchored theory.
Use structured decision-making. Formal review processes and checklists can prompt investigators to ask whether their reasoning has been anchored by early information.
Awareness training with limits. Tversky and Kahneman (1974) noted that people remain susceptible to anchoring even when warned about it. Awareness is a starting point, not a solution on its own.

How does Murder Book relate to anchoring?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Players enter each case with initial information, statements from suspects and an in-game casebook listing behavioural cues such as Stonewalling, Stress Response and Fabrication. The first impression a suspect makes can function as an anchor, but development screenshots show that cases have branching paths and multiple endings, including a "not guilty" result. A player who stays anchored to their initial reading of the evidence risks reaching the wrong conclusion.

The game's design encourages players to keep testing their assumptions against new information rather than committing to the first theory that comes to mind. This reflects the core lesson of anchoring research: the first piece of information you encounter shapes your thinking more than you expect, and resisting that pull requires deliberate effort.

Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.