The research was conducted by Ge Wang, Haixin Teng, and Zengyang Xu of the School of Public Administration at Central China Normal University in Wuhan, and it addresses one of the most persistent blind spots in the study of algorithmic governance. Scholars have long argued that automation transforms street-level bureaucracy, echoing Michael Lipsky’s classic insight that frontline workers are the ultimate policymakers because they exercise discretion in every individual encounter. Earlier work by Bovens and Zouridis described a drift from street-level to system-level bureaucracy, in which decision-making migrates from human judgment to automated systems. But the new study asks a sharper question: what happens to how bureaucrats enforce — not just what they decide — when AI enters the loop, and how does felt accountability reshape that relationship?

To answer it, the team designed a preregistered survey experiment embedded in the real-world context of honking violations, a common and familiar enforcement scenario for traffic police. Participants were 356 street-level police officers drawn from a representative sample, making this one of the more methodologically robust attempts to measure enforcement style directly among practitioners rather than students or hypothetical respondents. Invalid questionnaires were excluded before hypothesis testing, based on substantial missing data, incomplete experimental responses, or failure to pass embedded attention-check items, and the experimental design was preregistered to guard against selective reporting. Data supporting the findings are available from the corresponding author upon reasonable request, and the preregistration is publicly archived on the Open Science Framework.

The experiment crossed two manipulations: whether officers received AI support in the enforcement scenario, and which type of accountability they experienced. In public administration, accountability is conventionally divided into process accountability, where officials must justify the procedures and reasoning behind their decisions, and outcome accountability, where they are judged by the results those decisions produce. Decades of psychological research, from Tetlock onward, have shown that these two forms of felt accountability trigger distinct information-processing strategies, often in opposite directions. The Chinese enforcement context makes the setting particularly instructive, given its documented history of campaign-style enforcement and shifting regulatory styles among frontline officials.

The core findings are cleanly delineated. AI-supported enforcement, on its own, pushed officers toward a more formal and coercive style of enforcement — rule-invoking, sanction-first, procedural — while weakening the educational component, the practice of explaining, persuading, and teaching violators why compliance matters. This matters because enforcement style is not cosmetic. A substantial body of regulatory scholarship, including work by May and Wood on inspectors at the regulatory front lines and by Braithwaite and colleagues on enforcement pyramids, links style directly to compliance outcomes, citizen trust, and the legitimacy of the state. A coercive turn among officers algorithmically nudged into formalism could therefore ripple outward into how millions of daily encounters between citizens and the state actually feel.

Accountability, meanwhile, produced effects of its own that align closely with prior experimental literature. Officers who felt process accountability gravitated toward a more educational, prioritization-focused, and accommodative enforcement style — they reasoned more about how and why they enforced, reserved harsh measures for the cases that mattered most, and left more room for discretion and dialogue. Officers under outcome accountability moved in the opposite direction, reinforcing a more formal and coercive posture, consistent with the idea that being judged purely on results encourages defensive, box-ticking enforcement designed to be blame-proof.

The most consequential and arguably most surprising result is the interaction. Process accountability did not buffer officers against the algorithmic push toward formalism and coercion; it intensified it. Under process accountability, the presence of AI support further reinforced formal and coercive enforcement while further weakening the educational approach. The authors’ interpretation, grounded in the combined-effects framework they develop, is that when officers must justify their processes, the documented, auditable recommendation of an algorithm becomes an attractive anchor — a defensible procedural basis for action. Following the machine feels procedurally safe, and in doing so officers may shed the relational, educative practices that algorithmic systems cannot capture or document. The accountability mechanism designed to make officers more thoughtful may, in the presence of AI, make them more mechanical.

These findings carry immediate implications for governments racing to deploy AI in policing, from automated violation detection to algorithmic risk assessment and case triage. The study suggests that technology procurement and accountability reform cannot be designed in isolation. An agency that pairs algorithmic enforcement tools with process-oriented oversight — often considered the ethically preferable arrangement — may inadvertently deepen the very dehumanization of frontline encounters that critics of algorithmic governance fear. Conversely, outcome accountability’s independent drift toward coercion compounds the problem. If policymakers want AI to augment rather than hollow out street-level discretion, the results imply that accountability frameworks must be engineered to explicitly protect and incentivize the educational, judgment-rich dimensions of enforcement that machines neither practice nor document.

The study also speaks to a growing theoretical conversation about artificial discretion — the hybrid of human and machine judgment described by Young, Bullock, and Lecy — and about why bureaucrats trust AI recommendations, with recent experimental work suggesting that confirmation of professional judgment drives algorithmic acceptance. By demonstrating that enforcement style is a joint product of technology and institutional design, the Chinese policing experiment moves the field beyond the simple question of whether bureaucrats accept algorithmic advice toward the richer question of what they become when they do. As cities worldwide wire AI into everything from traffic cameras to welfare fraud detection, the transformation of the frontline worker may prove to be the most important, and least visible, policy outcome of the algorithmic state.

Subject of Research: Experimental study of how AI support and accountability mechanisms transform street-level police enforcement styles

Article Title: Artificial intelligence, accountability mechanisms, and the transformation of street-level enforcement style: experimental evidence from frontline policing

Article References: Wang, G., Teng, H., & Xu, Z. (2026). Artificial intelligence, accountability mechanisms, and the transformation of street-level enforcement style: experimental evidence from frontline policing. Global Public Policy and Governance, 6(2), 176-198. https://doi.org/10.1007/s43508-026-00146-y

Image Credits: AI Generated

DOI: 10.1007/s43508-026-00146-y

Keywords: artificial intelligence, street-level bureaucracy, policing, enforcement style, accountability, process accountability, outcome accountability, algorithmic governance, police discretion, survey experiment, public administration, coercion

The research team, led by Lina Harvold Ellingsen-Dalskau of the Norwegian Institute of Public Health, set out to address a persistent blind spot in health services research. Immigrants are known to carry an elevated risk of mental health problems, yet they use specialist psychiatric services less than expected, and there are reasons to suspect that when they do reach hospital care, the quality of their experience may differ from that of the majority population. Previous evidence on psychiatric inpatient experiences among immigrant groups has been scarce, partly because measuring patient experience in a way that is fair and comparable across cultural and linguistic groups is itself a formidable methodological challenge. The new study confronts that challenge head-on, using a national survey infrastructure and openly acknowledging the limits of cross-group comparison.

The quantitative backbone of the study is the Psychiatric Inpatient Patient Experience Questionnaire with Continuous Electronic Measurement, known as PIPEQ-CEM. This instrument was deployed continuously in psychiatric inpatient institutions across Norway from January 2020 to June 2022, producing an unusually rich dataset of 8,184 patients, of whom 89.5 percent were Norwegian-born. Respondents completed the questionnaire alongside items covering demographic background, experiences of coercion, and indicators of their mental health. The researchers then used linear regression to test whether a patient’s region of birth was associated with the key outcomes: the overall experience score, ratings of specific dimensions of care, reported coercion, and the perceived change in mental health over the course of the admission.

The headline quantitative result is one of broad similarity. On overall patient experience scores, the study found few statistically meaningful differences between immigrant groups and the majority population. That null finding matters, because it pushes back against any assumption that immigrant patients necessarily rate psychiatric hospitals more harshly across the board. Yet beneath the aggregate similarity, the analysis uncovered patterns that cannot be ignored. Some immigrant groups rated specific aspects of care, including the initial welcome they received on admission, the privacy they were afforded, and the perceived benefit of their treatment, lower than the Norwegian-born majority did. These are not peripheral details; the first hours of an admission shape trust, and privacy is a core dignity concern in any locked ward environment.

More striking still were the differences in coercion and clinical trajectory. Some immigrant groups reported higher levels of coercive measures during their inpatient stay, and also reported less improvement in their mental health between admission and discharge. Coercion, which in psychiatric settings can include involuntary medication, restraint, seclusion, or being admitted against one’s will, is one of the most ethically charged dimensions of mental health care, and international policy has long aimed to minimize it. Its uneven distribution across population groups raises questions that the survey data alone cannot answer, about how risk is assessed, how communication difficulties influence clinical decisions, and whether patients from some backgrounds are more likely to experience care as something done to them rather than with them.

To get closer to those questions, the team turned to the free-text comments that patients attached to their questionnaires. In total, 390 comments from immigrant patients and 125 from Norwegian-born patients were subjected to thematic analysis, a qualitative method that identifies recurring patterns of meaning across large bodies of unstructured text. The analysis revealed four interlocking themes that were specific to immigrant patients’ experiences: their expectations of and prior knowledge about psychiatric care before admission; their sense of belonging and trust within the ward environment; communication with staff; and access to information about their treatment and rights.

Each of these themes points to a concrete mechanism through which care can go wrong for patients unfamiliar with the system. A patient who has never encountered a psychiatric hospital before, or whose expectations are shaped by stigma or frightening stories from their country of origin, may arrive with fear that staff do not recognize or address. A patient who cannot follow rapid conversations in Norwegian, or who receives written information only in a language they do not read, may miss crucial explanations about medication, legal safeguards, or discharge planning. A patient who feels culturally invisible on the ward may withhold trust precisely at the moment when trust is the foundation of therapeutic benefit. The qualitative findings thus give psychological texture to the numbers, suggesting how small differences in welcome, privacy, and information flow could accumulate into lower perceived treatment benefit and a slower recovery trajectory.

The study’s authors are careful about what their measurement can and cannot claim, and that caution is itself an important finding. Patient-reported experience measures, or PREMs, assume that patients share a common framework for judging care: what counts as a good welcome, what counts as acceptable privacy, what a helpful treatment feels like. When those frameworks differ across cultural groups, identical scores may not mean identical experiences, and different scores may reflect different expectations rather than different treatment. The thematic analysis helps here, because it shows that contextual factors, such as prior knowledge of psychiatric care and the experience of belonging, actively shape how patients interpret and report their care. Any fair comparison of patient experience across diverse populations must therefore be accompanied by culturally appropriate measurement instruments, not simply translated versions of majority-population questionnaires.

The practical implications extend well beyond Norway’s borders. Health systems across Europe and North America serve increasingly diverse populations, and mental health services sit at the intersection of several well-documented inequities: barriers to access, communication gaps, stigma that varies across communities, and clinical decision-making that may be more coercive for some groups than others. This study suggests a template for detecting such inequities early, through continuous electronic measurement rather than one-off surveys, combined with systematic analysis of patient comments that can reveal why the numbers look the way they do. For ward managers, the actionable levers are tangible: invest in the admission experience, protect privacy deliberately, ensure interpreters and translated materials are genuinely available, and build staff competence in cross-cultural communication so that trust can form quickly even in crisis admissions.

For researchers, the message is that methodology and equity are inseparable. A PREM that works for the majority population is not automatically a valid instrument for everyone, and treating aggregate similarity as evidence of equity risks obscuring the disparities, in coercion and in recovery, that surface when the data are disaggregated by region of birth. The Norwegian study demonstrates that it is possible to run such disaggregated analysis at national scale while still listening carefully to patients in their own words. What emerges is a dual mandate for psychiatric services: deliver care that is culturally sensitive in practice, and measure patient experience in ways that are culturally appropriate in method. Only when both halves of that mandate are met can health systems honestly claim to know how well they are serving all of the patients in their care, and only then can the higher coercion and flatter recovery curves reported by some immigrant groups be understood, and ultimately eliminated.

Subject of Research: Comparison of patient-reported experiences, coercion, and mental health outcomes between immigrant and majority populations in Norwegian psychiatric inpatient care

Article Title: Patient-reported experience measures (PREMs), coercion, and mental health outcomes in psychiatric inpatient care: a national mixed-methods study comparing immigrant and majority populations

Article References: Ellingsen-Dalskau, L. H., Bjertnæs, Ø., Kjøllesdal, M. K. R., & Iversen, H. H. (2026). Patient-reported experience measures (PREMs), coercion, and mental health outcomes in psychiatric inpatient care: a national mixed-methods study comparing immigrant and majority populations. BMC Health Services Research. https://doi.org/10.1186/s12913-026-15511-0

Image Credits: AI Generated

DOI: 10.1186/s12913-026-15511-0

Keywords: patient-reported experience measures, PREMs, psychiatric inpatient care, coercion, mental health outcomes, immigrant health, health services research, Norway, cross-cultural psychiatry, mixed-methods study, PIPEQ-CEM, health equity