Case File 311 · Investigation Psychology

Why Is It Hard to Abandon an Investigation Theory?

It is hard because of how human cognition works, not because investigators are careless or dishonest. Once a theory forms, several well-documented psychological processes make it resistant to change:

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

It is hard because of how human cognition works, not because investigators are careless or dishonest. Once a theory forms, several well-documented psychological processes make it resistant to change: confirmation bias filters new evidence to support the existing view, belief perseverance keeps the theory alive even when its original basis is undermined, sunk-cost thinking makes the invested effort feel wasted if the theory is dropped, and the emotional and professional stakes of admitting a wrong direction create additional resistance. These are not weaknesses unique to investigators. They are general features of human reasoning that become especially consequential in criminal cases, where the cost of holding onto a wrong theory can be a wrongful conviction.

Key takeaways

Multiple cognitive biases work together to make abandoning an investigation theory psychologically difficult, even for experienced professionals.
Confirmation bias, belief perseverance and hindsight bias reinforce each other, creating a self-sustaining cycle.
Sunk-cost thinking and professional reputation concerns add non-cognitive pressures that make changing course feel costly.
Findley and Scott's analysis of wrongful convictions found that tunnel vision runs through nearly every documented case.
Structured debiasing techniques, such as generating alternative hypotheses and assigning a "devil's advocate," are more effective than simply telling investigators to stay open-minded.

What cognitive biases keep a theory in place?

Several biases reinforce one another once a working theory has formed.

Confirmation bias is the tendency to seek, interpret and remember information in ways that support an existing belief. Nickerson's 1998 review in the Review of General Psychology catalogued the phenomenon across more than two hundred studies and called it "perhaps the best known and most widely accepted notion of inferential error." In an investigation, confirmation bias means that evidence supporting the current theory is noticed and weighted heavily, while evidence pointing elsewhere is overlooked, dismissed or explained away.

O'Brien (2009) demonstrated this in an experiment where participants read a mock police file. When asked to name a suspect midway through, nearly everyone identified the obvious candidate. When a stronger suspect emerged in the second half of the file, participants who had already committed to the first suspect continued to favour that person. Those who were asked to generate reasons why their suspect might be innocent showed less bias.

Belief perseverance keeps a belief in place even after its foundation has been removed. Ross, Lepper and Hubbard's 1975 experiment showed that participants continued to hold beliefs about their own abilities after being told the supporting feedback was fabricated. In investigations, this means that a theory can survive the discrediting of the evidence that created it, because the investigator has already built a web of supporting explanations around it.

Hindsight bias makes the chosen theory feel inevitable in retrospect. Once investigators have settled on a direction, looking back makes it seem as though the evidence always pointed that way. This creates a false sense of certainty that makes it harder to imagine the case going differently.

Anchoring bias gives the first piece of information disproportionate influence over subsequent judgements. An early tip, an initial witness statement or a first forensic result can set the direction for the entire investigation. Later evidence is interpreted relative to this anchor rather than on its own merits.

What non-cognitive factors make it harder?

Beyond pure cognition, practical and social pressures compound the problem.

Sunk-cost thinking. Investigations consume time, staffing, budget and political capital. The more resources invested in a theory, the harder it becomes to discard it. Staw's 1981 research on escalation of commitment, published in Organizational Behavior and Human Decision Processes, showed that people consistently invest more resources in failing courses of action precisely because they have already invested so much. Investigators who have spent months building a case face the same pull.

Professional and reputational stakes. Changing theories can feel like an admission of failure, particularly when cases have attracted public attention. Findley and Scott's 2006 analysis noted that tunnel vision affects not only investigators but prosecutors, judges and defence lawyers, all of whom may have professional reasons to resist revisiting a settled theory.

Organisational culture. In hierarchical organisations, challenging a senior investigator's theory can carry professional risk. Rossmo's research on criminal investigative failures identified groupthink and organisational dynamics as contributors to tunnel vision alongside individual cognitive biases. When the institutional culture discourages dissent, the cognitive barriers to abandoning a theory are reinforced by social ones.

What does the wrongful conviction research show?

Findley and Scott's 2006 article in the Wisconsin Law Review, "The multiple dimensions of tunnel vision in criminal cases," analysed the role of cognitive biases in wrongful convictions revealed by DNA exonerations. They found that tunnel vision was a theme running through nearly every case, and that it was not a single error but a cluster of reinforcing tendencies.

Their analysis identified several patterns:

Investigators focused on a suspect early and filtered subsequent evidence through that lens.
Evidence inconsistent with the chosen theory was dismissed as unreliable, irrelevant or explainable.
The case file grew to support one theory while alternative explanations were not pursued.
By the time the case reached trial, the apparent weight of evidence felt overwhelming, even though much of it had been shaped by the initial assumption.

Kassin, Bogart and Kerner's 2012 study reinforced this picture. In DNA exoneration cases involving false confessions, the confession typically came first, and other evidence errors followed. The confession created a framework that biased subsequent evidence evaluation, making the case appear stronger than it was.

How can investigators counteract these pressures?

Research points to several strategies that are more effective than general advice to "keep an open mind."

Generate alternative hypotheses. Lord, Lepper and Preston's 1984 study found that asking people to consider why the opposite conclusion might be true was significantly more effective at reducing bias than asking them to be fair and objective. In practice, this means investigators should regularly ask: if this suspect did not do it, who else could have, and what evidence would support that?

Assign a devil's advocate. Designating a team member to challenge the prevailing theory provides a structured check against groupthink and confirmation bias. The challenge must be genuine, not performative: the advocate needs access to all the evidence and the authority to raise concerns.

Test competing hypotheses. Rather than asking whether the evidence supports the current theory, investigators can ask whether the evidence distinguishes between competing theories. This shifts the question from "does the evidence fit?" to "could the evidence fit more than one explanation?"

Document reasoning in real time. Recording the information available and the reasoning applied at each decision point creates a record that can be reviewed without the distortions of hindsight bias.

Seek disconfirming evidence. Actively looking for evidence that contradicts the current theory is the most direct counter to confirmation bias, though it goes against the natural tendency to seek confirmation.

How does Murder Book explore the difficulty of abandoning a theory?

Murder Book: The Art of Interrogation is an upcoming narrative interrogation game from Gungrounds, an independent studio in Zagreb, Croatia. Its development screenshots show a branching case map with multiple achievement endings, meaning the player's interpretation of the evidence determines the outcome. A player who commits to one theory early may miss contradictions that would point in a different direction.

The game's in-game casebook lists behaviours such as Stonewalling, Avoidance and Fabrication, and its evidence board links contradictory statements. But these cues are prompts for further investigation, not verdicts. A player who treats an early suspicion as settled may find that the case leads to a "not guilty" result, shown among the game's case outcomes. The structure reflects the real difficulty this article describes: it is far easier to commit to a theory than to let it go when the evidence shifts. Murder Book is in development for PC (Steam) and mobile, with release currently planned for the end of Q2 2027.