Your Favorite Website Has Been Studied. Your Interview & Interrogation Training Hasn't Changed Since 1955.

Part 2 of a series on Science-Based Interviewing and grocery store psychology.

There is a map of you.

Not a topographical—a heat map. A colored image showing where human eyes go, how long they stay, what they skip, and what they never see at all. Maps like it exist for many things you looked at today, and they exist because entire professions decided that guessing was unacceptable and going with the gut was less profitable.

Grocery store item placement is studied. Demand interview and interrogation training that has been tested for effectiveness as well.

Start where we started: the grocery store. We already know retail professionals tuned the lighting and vented the bakery toward the door. But it goes further than atmosphere. Retailers map foot traffic through the aisles. They map which shelf your eye hits first and which one your hand actually reaches for. Shelf position and the number of facings measurably change what people notice and buy, and that has been tested and published (Chandon et al., 2009).

The grocery store aisle you walk down to shop for your favorite products was drawn by data.

Then the advertisers. They put people in front of ads with calibrated equipment and recorded eye movement down to the fixation—where the gaze landed, how long it held, and what order it moved in. They found that even the viewer's goal reshapes the entire scan path across an ad (Rayner, Miller, & Rotello, 2008). There is enough of this work now to fill review articles summarizing decades of it (Higgins, Leinenger, & Rayner, 2014). Peer-reviewed. Replicated. Maybe even boring to most.

Then the web. Many serious sites you visit have been click-mapped, scroll-mapped, and heat-mapped. Designers can see the exact line where readers stop scrolling. They can see the button nobody presses. They test in the lab, then validate in the field—and every A/B test running right now is a live field experiment on real human beings, run by ordinary customers, continuously, forever. Not in a lab. In the wild. With real people making real decisions. The results come back within hours, and the less effective techniques get deleted (seems like common sense even).

Even your résumé has been through the machine. When the career-search firm Ladders ran an eye-tracking study on professional recruiters, it clocked the average first-pass résumé screening at 7.4 seconds—up from six when they ran it in 2012 (Ladders, 2018).

Someone measured, to the tenth of a second, how long a human gaze rests on a page before deciding your future—so that a person applying for a job could move their name and titles into the exact path the eye actually travels. Getting better, using better techniques, and methods is not reserved for Fortune 500 boardrooms. The job seeker gets it. The small-business owner running a landing page gets it. The person stocking an endcap gets it. They went to the data to decide what are best practices.

The thing they all have in common

None of those people are outliers that have top-tier business acumen. They are not braver than you. They share this one thing in common: They asked for evidence about their own actions, and they wanted feedback on their own professions. They simply wanted to get better at what they did. There are no tricks. They tested their methods. They tested in the field. They read results that sometimes told them they were wrong. And then—this is where public and private sector investigators need to pay attention—they changed what they did.

Nobody in those professions treats an outdated practice as a personal afront. Nobody defends a layout or vendor out of loyalty. When the heat map says the headline is dead, the headline moves. When the A/B test says version B wins, version A dies without a ceremonious funeral. Being proven wrong is not a humiliation in these professions. It is the product. It is Tuesday.

You are surrounded by this. It seems like ordinary behavior and business as usual in many industries.

Other areas of law enforcement already do this

Researchers put police officers in front of staged encounters with eye trackers running and compared rookies to veterans—and found the experienced officers fixated differently, locking onto hands and likely threat areas faster and scanning more efficiently than novices (Huhta et al., 2022). Law enforcement was willing to wire up its own people and study their eye movements to sharpen threat detection and situational awareness. It embraced the research which identified potential training gaps for officer safety.

Sample eye-tracking data. In both images, each individual circle represents a participant’s point of fixation. The size of each circle represents visit duration, with larger circles representing longer durations. The number within each circle represents the order of fixation following stimulus onset (Huhta et al., 2022).

Until relatively recently in the U.S., we had just never turned it towards the interview room. Towards how we conduct interviews and interrogations.

Law enforcement isn't a field that couldn't understand the data. It is a field that chose where to look when it came to one portion of the profession. In this case we stood staring back at the mid-twentieth century for what to do.

The one room nobody mapped

Walk into an interview room.

There is no heat map. There is no A/B test. There is no field data flowing back from ten thousand contacts telling you which opening produces more accurate information. For the conversation that decides whether a fraudster gets caught, whether a witness is believed, and whether a person's life comes apart—there is a chair, a habit, and a mostly outdated accusatory curriculum.

And that curriculum, that interviewing playbook, has not been upgraded in a long time.

What still dominates the interview room and the field is a family of outdated, accusatorial legacy practices—methods that guess at guilt off of posture and nerves, an idea researchers have tried and failed to validate for decades and set a single goal: a confession. The most widely used of them is, by its own company's description, the most common approach to questioning subjects in the world. A nationwide survey found two-thirds of police departments reported most or some of their officers were trained in that model. And it is not only police—its clients include private security, insurance and credit-card fraud investigators, and corporate loss-prevention teams (Harder, 2026). It is running inside businesses right now. Possibly yours.

What do these practices teach? Among other things, a suspect who avoids eye contact, slouches, shifts in his chair, or touches his nose is lying. Behavior provoking questions will tell you who is lying.

The advertising industry spent decades building instruments to learn how eyes actually work, then published it, then revised it, then kept testing it—all to move a conversion number two percent. Accusatorial interviewing made a claim about what behavioral lie detection, taught it to two-thirds of American police, sold it worldwide, and never checked. Criminologist Richard Leo's assessment, after nearly a half-century of research: "folk psychology … complete bunk." Investigators cannot reliably beat chance at detecting deception, and one 2020 analysis found that even the method's own founder, using his own technique, could not accurately tell whether a suspect was lying (as quoted in Harder, 2026).

Same organ. Same seventy years. Some professions bought eye trackers and read data to improve. The other bought a story.

Except somebody did map interviewing and interrogation training

Interview and interrogation training is a choice.

The interview room has been tested—just not by the people defending the old way. The data is inconvenient for anyone selling a proprietary model that has never been put through this kind of testing.

Researchers went and gathered the evidence nobody selling the old method ever bothered to gather. When Meissner and colleagues pooled the evidence comparing two approaches, legacy accusatorial methods and science-based interviewing methods (information-gathering methods), the pattern held across both: information-gathering methods increased true confessions and reduced false ones, while accusatory methods did the reverse (Meissner et al., 2014).

Then Russano and colleagues took it into live casework testing science-based interviewing curricula. Real investigators—federal, state, and local—recorded their suspect interviews before and after training in science-based methods and handed the tapes over to be coded. When investigators used the science-based tactics, suspects cooperated more, disclosed more information, and—through that cooperation and disclosure—made more admissions. When accusatorial tactics were used, suspects pushed back: more counter-interrogation behavior, less cooperation, and less information. Same investigators. Same real cases. The two approaches moved the room in opposite directions (Russano et al., 2024, 2026).

That is the A/B test. It has been run. In the field. On real cases. The results are published, and they, along with other interview and interrogation research, point one direction—science-based interviewing is better for business.

So now it is a business decision, not a philosophical one.

Whether you are a corporate investigations manager, a chief, a sheriff, a training director, or the investigator who has to sit in the chair—you are the one holding the A/B results. So what is your call?

More information or less information? More reliable information or less reliable information? More cooperation or more resistance? A statement that holds up in a courtroom, or a confession you spend the next two years defending with less information to test it?

The grocer does not agonize over this. When the test comes back, version A dies without a funeral. Nobody calls it disloyal. They move the shelf, they bank the two percent, and it is Tuesday.

The accusatory interview and interrogation training bill

Nearly 30% of DNA exonerees in the United States since 1989 falsely confessed, (Innocence Project, n.d.,) many of them interrogated with the confrontational methods these courses teach. Nebraska produced the Beatrice Six—what the New Yorker called the largest DNA exoneration involving false confessions in American history. And every wrongful conviction is two failures, not one, because there is a guilty party still walking free, unpunished.

These cases are not an abstraction, and they are not rare enough to wave off. IXI keeps a searchable database of false confession exonerations—more than 480 of them, drawn from the National Registry of Exonerations, filterable by state and by crime, each one linked back to the record. Take a look at the cases in your state and review the circumstances. How was the interview run? What did the investigator believe walking in? And what did it cost to undo?

Read them the way a retailer reads a failed test—not as an indictment of the people in the room, but as information. Most of those investigators were doing exactly what they had been trained to do. Even with good intentions, the method failed them as surely as it failed the person across the table. Training problems are a fixable problem. Professions get better this way. So do organizations, and so do the investigators inside them. Nobody has to keep running version A, nor should they if it is less effective.

There is a heat map for the cereal aisle. There is one for the ad, the landing page, and the résumé. Millions of dollars and decades of peer-reviewed attention in science all pointed at helping people be a little less wrong about a two-percent margin. Law enforcement itself owns the instruments and has used them on its own officers to improve practices. And the upgraded interview has been field-tested on real cases, with the results published.

And the room where we decide what is true still runs on 1950s ideas—because, well, I don't know, that is a question for the reader.

You are surrounded by people who demand evidence about their actions and feedback on their craft and who are better at their jobs for it. We are still running techniques that produced a false confession in its first famous case. We are still running techniques that have been shown to be less effective in gathering information...the point of the whole endeavor.

We can't upgrade anything?

One Last Science-Based Interviewing Distinction

First, my stake in this. I am a practitioner co-author of the 2026 study, and I teach science-based interviewing for a living. That is why the standard below is the one I want applied, including on me. If you take one thing from this article into your next training decision, make it this: when you are choosing interview and interrogation training, the arbiter is not the instructor's résumé, the agency down the road, the certification on the wall, how entertaining the instructor was, or how good the two days felt in the room. Every vendor in this space will tell you their method works. Testimonials are not evidence. Confidence is not evidence. Longevity is not evidence—the accusatory model has been around since the fifties, and so has its failure rate.

The arbiter is the science. And by science I do not mean a footnote in a sales deck. I mean the three places a method has to survive before anyone should teach it to investigators. The lab, where you can know the ground truth and count what a technique actually produces—and where the accusatory approach reliably produces confessions from the guilty and the innocent, while information-gathering methods produce true admissions without dragging the false ones along with them (Meissner et al., 2014). The field, where real investigators work real cases and someone measures what changed—more information, more cooperation, and no loss of the confessions everyone is afraid to give up (Russano et al., 2024, 2026;). And the practitioners and prosecutors themselves, surveyed on the job, who tell researchers which techniques hold up and are most useful (Brimbal et al., 2026).

Why the convergence is the point

Experiments can isolate cause-and-effect relationships. Also in the laboratory, researchers have accurate knowledge about what has happened but cannot reproduce the stakes of a real interrogation room. Field data have all the stakes; however, in contrast to the laboratory, there is limited ground truth. In field studies one can observe differences, but also lack of ground truth cannot measure things such as accuracy in memory recall. Surveys capture what practitioners believe but not what they do. Each method's shortcoming is supported where the others are strong—taken together they all arrive at the same conclusions. That is what a strong result in applied science looks like, and it is a far stronger basis for a curriculum decision than any single study.

That is a bar the accusatory model cannot clear. It has one number to offer—confession count—and confession count cannot tell you which of those confessions was true.

So ask the question the next time someone pitches you a course. Where was this tested, who tested it, and what happened to the people who went back to work with it? If the answer is a story, keep shopping. If the answer is a citation, a field study, and the investigators who kept using it, you are looking at the upgrade.

Better interviews. Better investigations.

Interview and Interrogation References

Aviv, R. (2017, June 19). Remembering the murder you didn't commit. The New Yorker. https://www.newyorker.com/magazine/2017/06/19/remembering-the-murder-you-didnt-commit

Brimbal, L., Hill, S. L., Maguire, E. R., & Soderstrom, M. F. P. (2026). Prosecutors' perceptions of the effectiveness of police interview practices. Psychology, Crime & Law. Advance online publication. https://doi.org/10.1080/1068316X.2026.2700431

Chandon, P., Hutchinson, J. W., Bradlow, E. T., & Young, S. H. (2009). Does in-store marketing work? Effects of the number and position of shelf facings on brand attention and evaluation at the point of purchase. Journal of Marketing, 73(6), 1–17. https://doi.org/10.1509/jmkg.73.6.1

Harder, J. J. (2026, July 9). A famed police interrogation technique led to infamous false confessions in Nebraska. We still use it. Flatwater Free Press. https://flatwaterfreepress.org/a-famed-police-interrogation-technique-led-to-infamous-false-confessions-in-nebraska-we-still-use-it/

Higgins, E., Leinenger, M., & Rayner, K. (2014). Eye movements when viewing advertisements. Frontiers in Psychology, 5, 210. https://doi.org/10.3389/fpsyg.2014.00210

Huhta, J.-M., Di Nota, P. M., Surakka, V., Isokoski, P., & Ropo, E. (2022). Experience-dependent effects to situational awareness in police officers: An eye tracking study. International Journal of Environmental Research and Public Health, 19(9), 5047. https://doi.org/10.3390/ijerph19095047

Ladders, Inc. (2018). Eye-tracking study. The Ladders.

Meissner, C. A., Redlich, A. D., Michael, S. W., Evans, J. R., Camilletti, C. R., Bhatt, S., & Brandon, S. (2014). Accusatorial and information-gathering interrogation methods and their effects on true and false confessions: A meta-analytic review. Journal of Experimental Criminology, 10(4), 459–486. https://doi.org/10.1007/s11292-014-9207-6

Rayner, K., Miller, B., & Rotello, C. M. (2008). Eye movements when looking at print advertisements: The goal of the viewer matters. Applied Cognitive Psychology, 22(5), 697–707. https://doi.org/10.1002/acp.1389

Russano, M. B., Kelly, C. E., & Meissner, C. A. (2021). From the ivory tower to the interrogation room: Training and field evaluation research on suspect interviewing. In R. Bull & I. Blandón-Gitlin (Eds.), The Routledge international handbook of legal and investigative psychology (pp. 287–310). Routledge.

Russano, M. B., Meissner, C. A., Atkinson, D. J., Brandon, S. E., Wells, S., Kleinman, S. M., Ray, D. G., & Jones, M. S. (2024). Evaluating the effectiveness of a 5-day training on science-based methods of interrogation with U.S. federal, state, and local law enforcement investigators. Psychology, Public Policy, and Law, 30(2), 105–120. https://doi.org/10.1037/law0000422

Russano, M. B., Meissner, C. A., Jones, M. S., Rothweiler, J. N., Taylor, P. J., Cory, C., & Brandon, S. E. (2026). Evaluating the effectiveness of a practitioner-designed science-based interviewing and interrogation course: A collaborative training and research effort. Legal and Criminological Psychology, 31, 273–300. https://doi.org/10.1111/lcrp.70021