The events of this past summer and fall, starting with the killing of George Floyd, followed by discussions associated with the election, have brought a great deal of attention to implicit-bias training. The media reported on the varying views of this training—some good and some critical; some with facts, some with falsehoods, some with both.

Fair and Impartial Policing (FIP) has trained thousands of law enforcement professionals over more than a decade, using a curriculum that is constantly evolving to capture the most up-to-date and reliable research available. We have prepared this document to respond to the media reports and, in so doing, we share FACT and FICTION about the science of implicit bias and the effectiveness of implicit-bias training for police. We include results from the first comprehensive evaluation of implicit-bias training for police, which occurred during the course of our training the 36,000 sworn personnel of NYPD. Throughout, we refute the common misunderstandings about the training, the science behind it, or both.

FICTION: The science of implicit bias is not strong.

The FIP training program is based on decades of research conducted by social psychologists, which confirm:

  • The existence of implicit bias;
  • That implicit biases manifest in a variety of ways, including implicit associations, attentional bias, confirmation bias, and outgroup bias; and
  • That skills may be successfully implemented by individuals to reduce and manage these biases.1

Those who suggest that the science of implicit bias is weak are simply wrong. They typically base their claims on the fact that the reliability and validity of the Harvard produced Implicit Association Test (IAT) have been challenged.2 It is a FACT that the reliability and validity of the IAT as a measure of implicit bias is debated in academic research circles, but the IAT is only one (the most well-known) of a number of measures of implicit bias.

There is virtually no debate in the scientific community about the existence of implicit bias.


FICTION: An objective of implicit-bias training is to eliminate biases.

Some individuals criticize implicit-bias training by referring to the research showing that it is challenging to reduce biases and near impossible to eliminate them. But eliminating biases is NOT the objective of implicit-bias training. Some FACTS:

  • Humans develop implicit associations (stereotypes) across their lifetimes, beginning at a young age.
  • Historical and cultural influences that reinforce those associations remain ongoing and virtually omnipresent in our environment.
  • Therefore, it would be an unrealistic goal—and contrary to the science—to think that an 8-hour implicit-bias training session would reduce, much less eliminate, trainees’ implicit biases.
  • While FIP does include coverage of the mechanisms that individuals can use over the long term to weaken/reduce implicit associations …..

The key goal of FIP training is to make attendees aware of their biases, help them focus on critical decision points where those biases may have negative consequences, and give them strategies and skills to reduce the likelihood that those biases will impact their behaviors.


FICTION: Implicit-bias training produces a “backlash” in trainees.

You may have heard claims that implicit-bias training produces a “backlash” that makes racist attitudes and/or biased behavior worse.3 This claim is based on a FACT that several studies have shown that, when researchers direct subjects to repress their biases, racist attitudes and discriminatory behavior increase.4

However, science-based implicit-bias training programs do not direct people to repress their biases. FIP promotes bias awareness, not suppression.


FICTION: There is no evidence that implicit-bias training is effective

Critics claim there are no evaluations to show that implicit-bias training works. In FACT, well-constructed implicit-bias training can be very effective, and we know this from (a) the reports of training participants and (b) evaluation research.

With regard to training participants, the overwhelming majority of police professionals, of all ranks, who have been through the FIP training program rate our training as a “5” on a 5-point scale and similarly rate our trainers as a “5.” Typical comments from trainees
are reflected here:

  • In my 12 years of being a police officer in the XXPD I can honestly say that this was by far the best training I’ve received to help me do my job better, safer and serve my community better. … I believe that because of today’s class I will have more tools to become a better police officer.
  • I came to this class against my will. My mind was quickly changed from “this class is a waste of time” to “why have I not had this class sooner in my 20 years of service.”
  • I loved that this class is not about how cops are bad. It was about recognizing problems and how to fix them. I’m sold.
  • Absolutely relevant and essential subject matter for the police world.
  • Powerful material and applications for every-day use on and off duty.

In addition to trainees’ own reports, there is a significant and growing body of evaluation research showing that implicit-bias training impacts attitudes, intentions, and behavior in various trainee audiences (e.g., police, educators, medical personnel).


  • Compared to members of control groups, implicit-bias trainees are more aware of bias and concerned about discrimination,5 have increased motivation to behave in a bias-free manner,6 and intend to use bias-reducing and bias-managing techniques.7
  • Studies document reductions in biased behavior as a result of this training8 (e.g., after implicit-bias training, university science departments hired more women and minorities compared to control departments that did not receive the training). 9

The first comprehensive evaluation of implicit-bias training for police was conducted during the course of FIP’s training of NYPD. The independent outcome evaluation assessed the training’s impact on (1) the attitudes, knowledge, skills, and behaviors of line-level officers; and (2) supervisors’ amenability to and implementation of the FIP content.10 The findings are consistent with those reported above.


  • FIP-trained individuals (compared to a control group) are more concerned about discrimination as a social problem and more motivated to act in an unbiased way. They are more likely to recognize that there is bias in policing and that it is a legitimate concern of the public as opposed to merely a “fiction” produced by the  media.11
  • FIP-trained officers are more likely to recognize how implicit biases might affect police professionals, such as understanding that even well-intentioned officers could be impacted by unconscious biases.
  • We believe that our training reduces biased behavior, however it was not a surprise to us that the evaluation team did not successfully detect behavioral changes. The researchers wrote, “Estimating the effect of a single training curriculum on officers’ decisions…. may well be akin to finding the proverbial needle in a haystack.” They recognized that the search for this “needle in the haystack” is particularly complicated when an agency is undergoing not just one change (such as implicit-bias training), but many at the same time, which was the case in New York City.
  • Implicit-bias training has a strong, positive impact on first-line supervisors. The evaluation researchers characterized their results as “remarkable” in terms of the extent to which the FIP-trained supervisors embraced and implemented their important roles to (1) monitor their subordinates for signs of biased behavior, and (2) intervene with officers in an effort to correct biased behavior.

More than 90% of the FIP-trained supervisors reported that they would actively monitor their subordinates’ behavior for bias and they would follow up with subordinates if they detected signs of biased behavior.


FACT: Implicit-bias training is not the silver bullet.

Critics who claim that, if implicit-bias training worked as designed, biased/abusive behavior on the part of police would disappear12 neither understand the science of bias  education nor the frailties of humans.

Well-constructed, data-driven implicit-bias training:

  • Helps police and other criminal justice officials recognize biases they may not know they have—biases that may be inconsistent with their consciously held beliefs.
  • Leads attendees to acknowledge that, even if they have been acting in good faith to be fair, they can be unaware of their own blind spots, unaware of the times when they fail.
  • Gives participants techniques to be alert to the critical decision points— particularly those where discretion is involved—at which they are most likely to behave in a biased manner.
  • Gives participants the motivation and strategies to reduce and manage their biases so this does not occur.

Implicit-bias training will not cure a racist cop of animus towards minority groups, but will inform the police officer of good intentions that every officer is a part of the problem of biased policing and also a significant part of the solution.


Implicit-bias training for police is not “the answer” to the problem, but it is a necessary component of multidimensional efforts.


Download the PDF article here.

1 See Fridell, L. (2017). Producing bias-free policing: A science-based approach. Springer Publishers.
2 Goldhill, O. (2017). The world is relying on a flawed psychological test to fight racism. Quartz, 12/3/2017 at MacDonald, H. (2017). Are we all unconscious racists? City Journal, 10/18/2017 at
3 Journalists misapplying the backlash research to implicit bias training include Hobbes, M. (2020). ‘Implicit bias’ trainings don’t actually change police behavior. The Huffington Post, 6/12/2020 at
_sig=AQAAADQerzJEnrJcPjIgUMoryB_pVGAMnI9GgK7Xy8GtfxWuPZklSaRWM3eGzkdkrX1TlkH6m6VQN7ejhg9zohPShGYyUOyFBLQet1DFlXUHnWHstNSz4DHHqM6DqMcO1eNcrYgCPXQZ7UVyP9oqu1hQP8pBPP5sh-xl0SxoHAArri and Mahbubani, R. (2020). Officers already get training to deal with biases they may not know they have, but there’s no evidence it actually works. Insider, 6/16/2020 at
4 Gawronski, B., Deutsch, R., Mbirkou, S., Seibt, B., & Strack, F. (2008). When ‘just say no’ is not enough: Affirmation versus negation training and the reduction of automatic stereotype activation. Journal of Experimental Social Psychology, 44(2008): 370 – 377. Macrae, C.N., Bodenhausen, G.V., Milne, A.B., & Jetten, J. (1994). Out of mind but back in sight: Stereotypes on the rebound. Journal of Personality and Social Psychology, 67(5): 808 – 817.
5 Devine, P.G., Forscher, P.S., Austin, A.J. & Cox, W.T.L. (2012). Long-term reduction in implicit bias: A prejudice habit-breaking intervention, 48 J. Experimental Soc. Psychol, 1267. Carnes, M., Devine, P.G., Isaac, C. Manwell, L.B., Ford, C.E., ByarsWinston, A., Fine, E. & Sheridan, J.T. (2012). Promoting institutional change through bias literacy, 5 J. Diversity Higher Educ. 63, 63-64. Carnes, M., Devine, P.G., Manwell, L.B., Byars-Winston, A., Fine, E., Ford, C.E., Forscher, P., Isaac, C., Kaatz, A., Magua, W., Palta, M. & Sheridan, J. (2015). The effect of an intervention to break the gender bias habit for faculty at one institution: A cluster randomized, controlled trial, 90 Academic Med. 221, 224.
6 Carnes et al., 2015. Sekaquaptewa, D., Takahashi, K., Malley, J., Herzog, K. & Bliss, S. (2019). An evidence-based faculty recruitment workshop influences departmental hiring practice perceptions among university faculty, 38 Equality, Diversity, and Inclusion: An International Journal 188, 208.
7 Devine et al., 2012. Carnes et al., 2012.
8 Devine et al., 2012. Forscher, P.S., Mitamura, C., Dix, C., Cox, E.L., Cox, W.T.L. & Devine, P.G. (2017). Breaking the prejudice habit: Mechanisms, time course, and longevity, 72 J. Experimental Soc. Psychol. 133, 133, 143. Sekaquaptewa, et al. (2019). Smith, J.L., Handley, I.M., Zale, A.V., Rushing, S. & Potvin, M.A. (2015). Now hiring! Empirically testing a three-step intervention to increase faculty gender diversity in STEM, 65 BIOSCIENCE 1084. Devine, P.G., Forscher, P.S. Cox, W.T.L., Kaatz, A., Sheridan, J. & Carnes, M. (2017). A gender bias-breaking intervention led to increased hiring of female faculty in STEM departments, 73 J. Experimental Soc. Psychol. 211, 213-4. Burroughs, E.A. et al., (2017). Reducing bias in faculty searches, 64 Notices of the AMS 1304. Russell, J. & Summers, A. (2013). Reflective decision-making, and foster care placements, 19
9 Smith et al., 2015. Devine et al., 2017. Burroughs et al., 2017.
10 The Impacts of Implicit Bias Awareness Training in the NYPD found at
11 The report authors provide the statistical significance for indexes comprised of multiple items, such as those listed above. All of the differences between the FIP-trained and not-FIP trained groups were statistically significant at least at the .05 level; in fact, all but one were significant at the .01 level.
12 Hobbes, M. (2020).