Skip navigation
Disciplinary Self-Help Litigation Manual - Header
× You have 2 more free articles available this month. Subscribe today.

Study Shows Innocent People Choose False Guilty Pleas and False Testimony to Gain Benefits

The study included college students from the United States, South Korea, and Japan. It found that 30-50% of all innocent study participants were willing to plead guilty and up to 56% of those participants falsely implicated another participant.

The criminal justice structure imagined by the founders of the U.S. Constitution envisioned a system of trials to resolve criminal cases. This was in accord with the common law. In 1783, an English court wrote: “[A] confession from the mind by the flattery of hope, or by torture or fear, comes in so questionable a shape . . . that no credit ought to be given to it.” Rex v. Warickshall, 168 Eng. Rep. 234 (1783).

Plea bargaining rose to the fore after the American Civil War. Appellate courts, however, summarily rejected them. In Wilson v. United States, 162 U.S. 613 (1896), the U.S. Supreme Court said that the “true test of admissibility is that the confession is made freely, voluntarily and without compulsion or inducement of any sort.”

That ruling, however, did not put an end to plea bargains. It merely pushed the practice into the shadows. In the early 1900s, New York defense attorneys stood outside night court with monetary based plea offers. For $300, a case could be resolved with 10 days in jail, $200 for 20 days, and $300 for 30 days in jail. In Chicago, “fixers” negotiated between the prosecution and the defense.

It was in that era that criminal statutes and criminal codes increased dramatically. The phenomenon of overcriminalization was at its height in 1919 with the start of Prohibition. By 1925, about 90% of criminal cases were resolved via plea bargains. Still, the Supreme Court rejected them. In Walker v. Johnston, 312 U.S. 275 (1941), the Court found a guilty plea unconstitutional because of the inducements made to obtain it.

The original federal code listed about 30 crimes. The exact number currently in that code is unknown, but studies have found there are about 5,000 crimes in the code and another 300,000 contained in federal regulations. With overcriminalization, courts became burdened with resolving the cases brought before them by prosecutors. Trials are costly, and they require large amounts of time and personnel.

The Supreme Court finally pulled plea bargaining from out of the shadows with its ruling in Brady v. United States, 397 U.S. 742 (1970) (“Although Brady’s plea of guilty may well have been motivated in part by a desire to avoid a possible death penalty, we are convinced that his plea was voluntarily and intelligently made and we have no reason to doubt that his solemn admission of guilt was truthful.”). For the last decade, about 97% of all criminal cases are resolved via a plea bargain.

Japan had an informal plea bargaining system until it amended its Code of Criminal Procedure effective June 1, 2018. The Code recognizes pleas only in cases where the defendant agrees to provide information about certain enumerated crimes, such as drug and weapons crimes and economic crimes that affect the public fiscally or execution of public duties like bribery.

South Korea’s criminal justice system is based upon that imposed when Japan colonized South Korea from 1905 to 1945. Korean prosecutors are said to wield “absolute power and authority over defendants.” Without a law or procedure to regulate plea bargains, prosecutors “conduct negotiations informally in criminal proceedings,” said a former economic crime investigator with the Korea’s national police.

This study was spearheaded by Andrew M. Pardiek, Associate Professor of Law at Southern Illinois University of Law; Vanessa A. Edkins, Ph.D, Professor of Psychology and Head of the School of Psychology at Florida Institute of Technology; and Lucian E. Dervan, Professor of Law and Director of Criminal Justice Studies at Belmont University College of Law (collectively, “Authors”).

With the push to introduce formal plea bargaining in Japan and South Korea and the concern over the innocence problem in the U.S., the Authors of the study under review here “engaged in a multi-year international collaboration with law and psychology researchers to administer psychological studies to test the prevalence of false pleas by the innocent in each country.”

“The goal was to mirror real-world plea bargaining conditions as closely as possible in a laboratory, by accusing students of academic dishonesty and offering them a plea,” wrote the Authors. “While academic discipline is not precisely equivalent to traditional criminal penalties, the anxiety experienced by students accused of cheating and anticipating punishment is similar in form to the anxiety experienced by one charged with a criminal offense.”

The volunteer students were told they were signing up to participle in a psychological inquiry into individual versus group problem solving. When they arrived for the experiment, they were met by a second participant who was actually a confederate working with the researchers. They were then met by a research assistant who lead them into a private room.

They were given three logic problems and instructed to work together in solving them without anyone else present. Once those problems were completed, the assistant returned to present three more logic problems with instructions to work alone on the problems.

After the assistant left the room, half of the confederates encouraged the participant to cheat and work together. The confederate would repeatedly ask the participant what they got for each question. The study participants who acquiesced and offered assistance were placed in the “guilty conduction.”

In the other half of cases, the confederate worked out the problems. Absent an unprompted attempt to cheat, the participant was placed in the “innocent condition.” In all cases, the participant was accused of cheating.

Once the second set of questions were completed, the research assistant returned to collect the answers and left the room. About five minutes later, the assistant returned and said there was a problem and asked to speak to the students individually. The confederate’s name was called, and the two left the room.

Five minutes later, the research assistant returned and sat near the participant, telling them, “You and the other student had the wrong answer on the second and third individual questions. The chances of you both getting the exact same wrong answer are really small – in fact they are like less than 4%. When this occurs, we are required report it to the professor in charge and she may consider this a form of academic dishonesty.”

The participant was then given two options. The first was much like probation – in return for an admission of academic dishonesty, the participant was offered the loss of any promised compensation and the student’s academic advisor would be informed of the dishonesty.

The second option mimicked a decision to go to trial. Students were told that if they contested the accusation a charge would be brought before the academic review board. The procedures of such a review were explained. If found innocent, they would receive the compensation owed them for the study. When explaining penalties if found guilty, the researchers created a “harsh” condition for half the students and a “lenient” condition for the other half.

The lenient condition would require the students to attend an eight-hour ethics training seminar with a pass/fail exam at the end. The harsh condition included the same punishment as well as 10 hours of community service. The disparity was meant to test the effect on sentence disparity on decisions to plead. The conditions may not have included jail or prison time, but they still constituted threats to the participant’s time, representing a deprivation of liberty interests.

To further mirror the criminal plea-bargaining process, participants were offered representation to serve the function of counsel. The student advocate then offered a one-page document that reiterated the right to defend oneself in front of the review board. While the representation was cursory, it was similar to that which most misdemeanor defendants in the U.S. receive, and the Authors said counsel access for the accused may be similarly limited in Japan and South Korea.

The Authors used the options to create two studies. If the participant in Study 1 chose to accept the plea deal, they were presented with and encouraged to complete and sign a sheet of paper that asked them to indicate who initiated the cheating. The researchers said those who chose to take the deal “should be naturally primed by accepting the plea to move to the next logical step of implicating their ‘co-defendant.’”

Study 2 “depended on implicating another, rightly or wrongly. It tested the willingness of those in the non-cheat condition to falsely accuse a third party in exchange for a lighter punishment.” It also tested their willingness “to testify against an innocent third party.”

Differing, but similar, results were found for each country. Of the 181 U.S. participants aged 18-25 who made it to the study’s end stage, 51.9% pleaded guilty, and 48.1% rejected the plea offer. For those in the guilty condition, the rate of pleas when faced with the harsh punishment was 73.1% and 72.4% when faced with the lenient punishment. For the “innocent” participants, 45.9% pleaded guilty to the harsh punishment and 40% to the lenient punishment.

The data demonstrated the “participants’ willingness to implicate others.” Of the 94 who chose to plead guilty, 73 agreed to identify the instigator of the cheating. Most of those in the guilty condition identified the confederate as the instigator, but 18.8% of them said they initiated the cheating.

Amongst those in the innocent condition, 58.5% falsely implicated the confederate as the instigator. “In other words, it was more likely than not that the plea bargains generated misinformation and implicated innocent people when individuals were willing to discuss instigation,” the Authors wrote.

The second study tested plea bargains more directly by requiring cooperation through the implication of another. There were 118 participants in this study. For the guilty participants, 71.4% pleaded guilty, and 65.8% of the innocent participants accepted the plea deal. “This means that guilt was actually not a significant predictor of plea acceptance – the rates were high across the board, regardless of guilt or innocence,” wrote the Authors. “Adding in the extra requirement of implicating a fellow student in order to receive a deal did not lead to fewer innocents pleading, but instead we saw even more accepting the deal than in Study 1.”

Of those who succumbed to the confederate’s requests to cheat, about two-thirds indicated they were the instigator of the cheating. Of the innocent participants, “52% stated that the confederate was the one who started cheating – again this is when no cheating actually occurred. The most concerning part was that of our 50 innocent individuals, 88% were willing to testify that cheating had, in fact, occurred and that the confederate was involved in the event,” said the Authors.

This data “suggest the burden of contesting innocence and the risks of increased punishment influence behavior,” wrote the Authors. It also “shows this influence precludes reliance on the veracity of a bargained statement or cooperation.”

In South Korea, 69 students completed the study. Those faced with the harsher punishment chose to plead guilty in 68.4% of the cases, and 62.5% of those facing the lenient punishment pleaded guilty. For those in the non-cheat condition, only 27.3% of those facing the harsher punishment pleaded guilty while 72.7% chose the academic review board. Meanwhile, 29.4% pleaded guilty when facing the lenient punishment.

Of the 63 students who accepted the plea offer, 59% gave evidence about the instigator, with 39% indicating they initiated the cheating. “With the innocent participants, among those participants who were asked to cheat but refused and still pled guilty, all nine students indicated the other person instigated the cheating,” the researchers found. “Among those who were not asked to cheat but pled guilty, three out of five falsely stated that they instigated the cheating while the other two falsely implicated the other student.”

As to Study 2, 78 South Korean students completed the study. Only 12.2% of the innocent participants chose to plead guilty, and 72.9% of the guilty participants pleaded guilty. The results show that 88.9% of the guilty participants correctly implicated the confederate as the instigator. “For the five innocent students who pleaded guilty, two stated they themselves instigated the cheating while the other three implicated the confederate.”

In Japan, the researchers “encountered a number of difficulties administering the experiment.” Only 27 students aged 18-27 completed Study 1. Of them, “35.7% of the innocent participants and 61.5% of the guilty participants accept[ed] the plea bargain offer.”

The data showed 57.1% of those in the cheat condition who faced the harsher punishment pleaded guilty while 66.7% of those facing the lenient punishment accepted the plea. Amongst the non-cheat condition, those facing harsh punishment accepted the deal in 28.6% of the cases, and 42.9% of those facing lenient punishment accepted the plea offer.

The Authors made some conclusions based on the data. With forty to fifty percent of the American study participants “willing to falsely condemn themselves in return for a perceived benefit,” it suggests the 1970 Brady decision was wrong “to place confidence in the ability of individuals to assert their right to trial in the face of uncertainty and risk.”

It also suggests “it is wrong to place confidence in the information provided as a result of the plea bargain, including information against co-defendants provided in formal proceedings by those who have pleaded guilty.”

“We now have laboratory evidence that plea bargaining’s innocence problem extends beyond individuals offering false confessions that implicate themselves for leniency,” the Authors wrote. “This new data demonstrates that individuals are willing to provide false testimony against others in return for benefits.”

The Authors said the study did not advocate eliminating plea bargaining as the answer. They recognized this would simply overburden the courts in the U.S. They acknowledged that “at its core, plea bargaining holds the potential of being beneficial to all parties and, therefore, eliminating plea bargaining may be as detrimental to the parties as to the system itself.” The Authors hope to “find a path forward that better protects the rights of those within the system, while simultaneously valuing accuracy more an efficiency.”

Informal plea bargaining was not seen as an answer. Pretrial detention has been found to make it more likely that false guilty pleas will occur. The current study also found that people are highly risk averse. People sometimes have an overwhelming need to “just be done with it” even if it means accepting a penalty wrongly imposed.

One reform suggested by the Authors is a limit on pretrial detention. The trial penalty must also be limited, so people who reject a plea and choose a trial are placed in a less coercive situation. Next, plea bargaining must be more transparent. Finally, defendants should be given more time to consider plea offers.

In closing, the Authors said, “We would do well to recognize that basic human nature will lead the innocent to confess and those accused to falsely implicate others. The time has come to recognize these failures in our criminal justice systems and begin the process of building a new and better plea bargaining system.” 


As a digital subscriber to Criminal Legal News, you can access full text and downloads for this and other premium content.

Subscribe today

Already a subscriber? Login



Stop Prison Profiteering Campaign Ad 2
Advertise here
The Habeas Citebook: Prosecutorial Misconduct Side