Lorie A. Fridell, Ph.D. University of South Florida and Phil Goff, Ph.D. UCLA

Transcription

1 Lorie A. Fridell, Ph.D. University of South Florida and Phil Goff, Ph.D. UCLA

2 My presentation will address: Rethinking Biased Policing : Use the social psychological research on implicit bias to reframe the issue of based policing (BP) Interventions: Discuss the implications of the science for interventions to promote fair and impartial policing Those interventions are encompassed in A Comprehensive Program to Promote Fair and Impartial Policing.

3 I convey the rethinking framework using RACIAL bias.

4 First, what IS biased policing?

5 A conceptual definition Biased policing occurs when the police inappropriately consider race/ethnicity or other characteristics in deciding with whom and/or how to intervene.

6 Decisions that might be inappropriately impacted by e.g., race/ethnicity by a line officer Whom to watch, investigate as suspicious person Whom to stop (pedestrian or driver) Whom to treat with dignity, respect What questions to ask Whom to search From whom to request consent to search Whom to arrest Against whom to use force, excessive force, deadly force etc.

7 Decisions that might be inappropriately impacted by race/ethnicity by a supervisor/commander How much police presence to provide (over-policing, or under-policing) How much input to solicit/use from community members What types of crimes to enforce vigorously; What crimes to ignore In which areas officers will use aggressive tactics How much officer misconduct to tolerate etc.

8

9 This is the key question linked to Rethinking RBP It s linked to assumptions about why/how biased policing occurs. I believe that many police and stakeholders narrowly characterize the nature/causes of BP This has negatively impacted both the discussion and the change efforts. Need to rethink nature, causes, response.

10 Ill-intentioned officers (and ONLY ill-intentioned officers)

11 This characterization is unfortunate, I argue, because: It produces distortions that: Harm relationships between police and racial/ethnic minority communities Lead police to minimize the problem and be very defensive about it. It is wrong; it is too narrow.

12 The characterization produces distortions that harm policeminority community trust As stated: Many stakeholders think illintentioned, bad ( racist ) police are the ones that produce biased policing Many racial/ethnic minorities perceive biased policing to be widespread RESULT: Many racial/ethnic minorities think there are many ill-intentioned, bad ( racist ) police.

13 The characterization produces distortions that lead to police defensiveness, minimizing As stated: Police think ill-intentioned, bad ( racist ) police are the (only) ones that produce biased policing Police believe that bad, ill-intentioned ( racist ) police are few RESULT: Police think the problem isn t that big and that they are being unfairly castigated.

14 Is wrong; it s too narrow

15 The science tells us: Well Meaning Officers Can Produce RBP. Because their human biases impact on their perceptions and decisions.

16 Again, ignoring this manifestation has led to Distortions leading to: Over-estimates on the part of the stakeholders regarding the proportion of officers who are illintentioned, bad, racist Under-estimates on the part of the police regarding the extent of biased policing. Misguided (narrow) reform efforts.

17 In rethinking RBP.. We need to move away from the narrow view that RBP is all about ill-intentioned, bad, racist officers To an expanded view, the implications of which are: Even the best officers, because they are human, might practice RBP Even the best agencies, because they hire humans, will have RBP.

18 We turn to the science of bias. Specifically, the minority-crime implicit association.

19 Social-Psych Research on Unconscious/Implicit Bias Racial stereotyping widespread (e.g., Devine 1989) Encompassing an association between: People of color (Blacks, Hispanics, Arab Descent) and violent, aggressive, criminal, terrorists Per research on implicit bias: This minority/crime association impacts on perceptions and even behavior

20 Unconscious/implicit bias (Cont.) It can operate outside the awareness of the person (Payne, 2001).. even in people who consciously reject racial stereotypes (Devine, 1989).

21 Blink

22 Why/how does the race-crime blink response manifest?

23 Stereotypes sometimes fill in when we are facing ambiguous stimuli Study: Subjects experienced an ambiguous bump on the part of a stranger The subjects interpreted the bump by African Americans differently.than bumps by whites. Bump was perceived as more threatening when performed by an African American. Replicated to show this phenomenon true for both white and non-white subjects.

24 Subway Study When asked to reflect back with their imperfect memories to identify who was holding the deadly razor in a subway scene. Subjects were more likely to (erroneously) identify a person of color.

25 Research The Black-Crime Association Influences Visual Processing

26 Visual Perception Study: Methods Eberhardt, Goff, Purdie, & Davies (2004). Journal of Personality and Social Psychology. White male undergraduates Primed with Black male faces, White male faces, or no faces Completed object recognition task

27 Levels of Degradation Frame

28 Crime Object

29 Crime Object

30 Crime Object

31 Also: Objects that were not related to crime.

32 Visual Perception Study: Hypotheses If there is a Black-crime implicit association Participants primed with Black male faces should be faster to identify crime-relevant objects than those primed with White male faces. [The crime objects would be more accessible. ] There should be no effect of prime for crimeirrelevant objects.

33 Frame number Object Identification Crime-Relevant Crime-Irrelevant White Prime No Prime Black Prime

34 Visual Perception Study: Conclusions Exposure to Black male faces facilitated the identification of crime-relevant objects. Exposure to White male faces inhibited the identification of crime-relevant objects.

35 Is there some fact at the root of stereotypes including the Black-Crime association? Absolutely. But the issue for policing is in the application of this stereotype or call it group characteristic to all members of a group. Officers should not police based on group characteristics (whether one thinks they are stereotypes, biases, whatever) Can produce unsafe, ineffective, unjust policing.

36 Research The Black-Crime Association Influences Behavior

37 Correll, Park, Judd, & Wittenbrink (2002). Journal of Personality and Social Psychology, 83(6).

38

39

40

41

42

43

44 Sum of Results Reaction Times: Subjects are slower to see Gun with a white target. slower to see No Gun when it is a black target rather than a white target. Errors -- These are the most likely errors: Incorrectly shoot the Black with NO GUN Incorrectly NOT shoot the White WITH gun

45 but also recently researchers found a Turban Effect.

46 The Turban Effect Unkelbach, Forgas and Denson, JESP 44, 2008 Computer shoot, don t shoot scenario Stimuli varied by appearance of Islamic head dress We predicted and found a significant bias for participants to shoot more at Muslim targets (p. 1409).

47 Peruche and Plant (2005, 2006) e.g., Peruche/Plant, 2006, Basic and Applied Social Psych 28(2)

48 Background Lab studies on shoot-don t-shoot decisions using POLICE subjects. One agency, 50 officers Methods: Computer simulation Faces with objects Quickly: Shoot or don t shoot.

49

50

51

52 Components of Research (cont.) Measured the errors made and whether they varied as a function of the race of the person s picture. Officers completed surveys: INCLUDING: Asked about quality of the officers contact with Black people at work and in their personal lives. Amount and nature of contacts (as positive or not).

53 Key findings Found Black-crime blink response in these subject police officers Found a weaker blink response in officers who reported greater positive personal contacts with Blacks This reflects contact hypothesis

54

55 Potential Remedies: Extinguish (or weaken) the blink responses. Reduce implicit bias by harnessing power of the contact hypothesis Raise consciousness so that the person can implement controlled (unbiased) behavior.

56 Generally, pretty tough, BUT..

57 How would you extinguish? To extinguish association between race and threat You would expose person to repeated RANDOM pairings of race and threat. Arguably, high quality firearms training does just that. e.g., computerized role plays With random pairings of threat and demographics

58 Correll et al., 2007 [JPSP, 92(6)] Seemed to demonstrate the efficacy of high quality firearms training Shoot, non-shoot methodology Looked at speed and accuracy Used both police and non-police subjects.

59 RESULTS: Both samples exhibited robust racial bias in response speed. Quicker to make decision when armed black or unarmed white (slower when unarmed black or armed white) But (1) police, overall, were more accurate and fast (PHEW!) And, KEY: (2) Bias was less likely to manifest in errors in police subjects.

60

61 Harness contact hypothesis Again, blink response is weaker in people who have had positive, personal contact with people who are different from them Intervention: Positive personal contact.

62

63 Raising consciousness to promote controlled (unbiased) response Research has shown: Increasing blink consciousness can, in a motivated person, lead to controlled (unbiased) behavior. Per Dovidio (2000): Making a non-prejudiced person aware of his/her automatic biased responses produces the motivation and ability to consciously override the automatic response with a controlled response that reflects one s egalitarian beliefs. See also Devine, 1989.

64 This science, in sum Indicates that it is not merely a small cadre of illintentioned police who might be producing biased policing Even the best police officers, because they are human, might practice biased policing Even the best agencies will have biased policing because they are hire humans to do the job.

65 This broader conceptualization (this rethinking )of biased policing can: Produce a more constructive discussion of the issue between police and community Provide motivation and direction/guidance for intervention.

66

67 The old equation Many ill-intentioned ( racist ) police Racially Biased Policing

68 As shared: This has led to great police defensiveness and inaction Mid-sized east coast agency.. Fridell: (without science) These are ways your agency can respond to RBP and perceptions Command Staff Member: If we implement these recommendations, we ll be admitting we are racist.

69 The old equation Many ill-intentioned ( racist ) police Racially Biased Policing

70 Last year, I heard the converse Resident stakeholder in one of my 2-day (command-level) trainings Heard about the science He said, I came in here thinking all police were racist. Now I don t.

71 The old equation Many ill-intentioned ( racist ) police Racially Biased Policing

72 The common stakeholder/police perspective can be: BP isn t produced solely by ill-intentioned people. Even the best officers might practice RBP Even the best agencies, because they hire humans, will have biased decisions. Allows Stakeholders to hear from police: Yes, this is a issue. Reduces police defensiveness and can motivate change.

73

74 EVERY agency, because they hire humans, must be proactive in producing fair and impartial policing.

75 Also: That we must attend in our change efforts to both. Ill-intentioned officers Well meaning humans who aspire to fair and impartial policing.

76 Areas of intervention associated with a Comprehensive Program To Produce FIP Recruitment and Hiring Policies prohibiting biased policing Training Leadership/supervision and accountability Outreach to diverse communities Assessing institutional practices/priorities Measurement.

77 Interventions x Ways RBP Manifests Recruitment and Hiring Policy Well-Meaning XX XX Ill-intentioned XX Training Supervision and Accountability Assessing Policies, Practices XX XX XX XX XX Measurement XX XX

78 Not just race/ethnicity We ve focused so far on racially biased policing Has been the bias getting the most attention There is a body of scientific research Other potential biases, based on: Income Age Ability to speak English Sexual orientation etc. The comprehensive program can be applied to all.

79 Per Comprehensive Program to Produce FIP Every agency should attempt to hire diverse personnel and people who can police in an unbiased fashion Every agency should have a meaningful policy that tells officers when they can and cannot use race and ethnicity (and other demographics) to make law enforcement decisions. Every agency needs to have effective academy and inservice training directed to the ways that racial and other biases might manifest in even the best departments

80 Every agency should have strong accountability mechanisms in place to promote professional behavior and those mechanisms ought to be directed, as well, to promoting fair and impartial policing. Every agency should have ongoing efforts to strengthen the relationships between the department and the diverse communities that it serves Every agency should assess the costs/benefits of collecting (vehicle/pedestrian) stop data And make an INFORMED decision regarding whether to include data collection as part of its response.

81 Turn now to coverage of one element Recruitment and Hiring Policies prohibiting racially biased policing Training Leadership/supervision and accountability Outreach to diverse communities Assessing institutional practices/priorities Measurement.

82

83 What agencies generally do as part of Racial Profiling training Cultural Diversity Training 4 th Amendment Professional traffic stops Necessary, but not sufficient.

84 I ll report on fair and impartial policing curriculums for Academy/recruit, in-service patrol First-line supervisors Leadership/command NOT just racial bias.

85 Recruit/Patrol Training Under Development Funded by COPS Office Curriculum Development Team (CDT): Police (sergeants, chiefs, CPD trainer) Academic experts on RBP and cops Social psychologists who study implicit/unconscious bias (Eberhardt, Plant, Dovidio)

86 Training under development (Cont.) CDT identified the messages for training; we want the recruit, patrol officer: To Understand: That even good people have biases and that might well include him/her That biases/stereotypes impact on what we perceive/see and can (unless prevented) impact on what we do Policing based on biases/stereotypes is unjust, unsafe, ineffective. To have tools that help him/her (1) recognize the blink response, (2) implement controlled (unbiased) responses.

87 Training under development (cont.) Piloted twice (Idaho, Massachusetts) Interactive, exercises, videos, etc. Again, not just RACE bias We don t get to it for 45 minutes Lead off with Susan Boyle Movie clips, discussions Use exercises to show how policing based on stereotypes/biases is unsafe/ineffective e.g., Woman with a gun

88 Skills Module Beware your implicit biases; challenge what you think you see Pantomime DV role play Sub theme: Focus on individual/situation factors (versus drawing conclusions based on demographics).

89 Skills (cont) Avoid profiling by proxy Video: Shows officer being called because gays are making out in the park Analyze response options with an FIP lens Discussion: Black man sitting in a car in a white neighborhood

90 Skills (Cont.) Reduce ambiguity Slow it down when feasible Recall: Stereotypes fill in when stimuli are ambiguous Make those stimuli UN-ambiguous; find out more info Engage with people Harness the power of the contact theory Make the community you serve unambiguous.

91 Including officer safety and effective policing

92 Some student comments I will go into each situation with this training in the back of my head. This is about the best training on this subject I have had. I and several others thought that (due to the class title) it would be a boring class taught by liberals jamming info on how we were racist down our throats, rather than the good class it turned out to be. Returning Cop: I had [been impacted by stereotypes/biases] as an officer but did not know it until now. I could have used you guys back in 1995! Piloting one more time, then two train-the-trainer sessions.

93 First Line Supervisor Training: Topics How bias manifests in even well-intentioned people Science of bias How to identify officers who may be manifesting bias (What are they looking for? What are potential info sources?) We discuss difficulty of ID g biased behavior. How to intervene with officers when they have concerns about bias. How might bias manifest in their own work/decisions. How to talk about bias with individuals (officers, community members), community groups, media.

94 Two pilots (Des Moines and Prince William County) Overwhelmingly reported that they WILL use the training in their daily jobs as supervisors. Comments: I will better recognize bias and be able to address it with officers. Examples presented made me reflect on my personal behavior as well as behavior of my subordinates. (I learned) how to deal with biases that we all have. (Some of the most useful info included) the fact that bias can occur and the officer doesn t realize it.

95 Command Level Training Small classes (n=30) Multiple agencies (great to have stakeholders, too) TOPICS: Re-thinking biased policing The social science of human bias and the implications for policing The benefits and elements of a comprehensive program to produce fair and impartial policing Coverage of each element How to implement a comprehensive program.

96 Summary and Conclusions Old thinking : Biased policing is produced only by ill-intentioned officers. Has negatively impacted on discussion and change efforts Science helps us rethink biased policing Even well-intentioned officers. Even the best officers might practice biased policing because they are human Even the best agencies, because they hire humans, must be proactive in producing fair and impartial policing..

97 Elements of that proactive effort: Recruitment and Hiring Policies prohibiting racially biased policing Training Leadership/supervision and accountability Outreach to diverse communities Assessing institutional practices/priorities Measurement.

98