Explainability does not mitigate the negative impact of incorrect AI advice in a personnel selection task.

Despite the rise of decision support systems enabled by artificial intelligence (AI) in personnel selection, their impact on decision-making processes is largely unknown. Consequently, we conducted five experiments (N = 1403 students and Human Resource Management (HRM) employees) investigating how people interact with AI-generated advice in a personnel selection task. In all pre-registered experiments, we presented correct and incorrect advice. In Experiments 1a and 1b, we manipulated the source of the advice (human vs. AI). In Experiments 2a, 2b, and 2c, we further manipulated the type of explainability of AI advice (2a and 2b: heatmaps and 2c: charts). We hypothesized that accurate and explainable advice improves decision-making. The independent variables were regressed on task performance, perceived advice quality and confidence ratings. The results consistently showed that incorrect advice negatively impacted performance, as people failed to dismiss it (i.e., overreliance). Additionally, we found that the effects of source and explainability of advice on the dependent variables were limited. The lack of reduction in participants' overreliance on inaccurate advice when the systems' predictions were made more explainable highlights the complexity of human-AI interaction and the need for regulation and quality standards in HRM.

Comparing preferences for skin cancer screening: AI-enabled app vs dermatologist.

BACKGROUND AND AIM: Skin cancer is a major public health issue. While self-examinations and professional screenings are recommended, they are rarely performed. Mobile health (mHealth) apps utilising artificial intelligence (AI) for skin cancer screening offer a potential solution to aid self-examinations; however, their uptake is low. Therefore, the aim of this research was to examine provider and user characteristics influencing people's decisions to seek skin cancer screening performed by a mHealth app or a dermatologist. METHODS: Two forced-choice conjoint experiments with Nmain = 1591 and Nreplication = 308 participants from the United States were conducted online to investigate preferences for screening providers. In addition to the provider type (mHealth app vs dermatologist), the following provider attributes were manipulated: costs, expertise, privacy policy, and result details. Subsequently, a questionnaire assessed various user characteristics, including demographics, attitudes toward AI technology and medical mistrust. RESULTS: Outcomes were consistent across the two studies. The provider type was the most influential factor, with the dermatologist being selected more often than the mHealth app. Cost, expertise, and privacy policy also significantly impacted decisions. Demographic subgroup analyses showed rather consistent preference trends across various age, gender, and ethnicity groups. Individuals with greater medical mistrust were more inclined to choose the mHealth app. Trust, accuracy, and quality ratings were higher for the dermatologist, whether selected or not. CONCLUSION: Our results offer valuable insights for technology developers, healthcare providers, and policymakers, contributing to unlocking the potential of skin cancer screening apps in bridging healthcare gaps in underserved communities.

Attitudes Toward the Adoption of 2 Artificial Intelligence-Enabled Mental Health Tools Among Prospective Psychotherapists: Cross-sectional Study.

BACKGROUND: Despite growing efforts to develop user-friendly artificial intelligence (AI) applications for clinical care, their adoption remains limited because of the barriers at individual, organizational, and system levels. There is limited research on the intention to use AI systems in mental health care. OBJECTIVE: This study aimed to address this gap by examining the predictors of psychology students' and early practitioners' intention to use 2 specific AI-enabled mental health tools based on the Unified Theory of Acceptance and Use of Technology. METHODS: This cross-sectional study included 206 psychology students and psychotherapists in training to examine the predictors of their intention to use 2 AI-enabled mental health care tools. The first tool provides feedback to the psychotherapist on their adherence to motivational interviewing techniques. The second tool uses patient voice samples to derive mood scores that the therapists may use for treatment decisions. Participants were presented with graphic depictions of the tools' functioning mechanisms before measuring the variables of the extended Unified Theory of Acceptance and Use of Technology. In total, 2 structural equation models (1 for each tool) were specified, which included direct and mediated paths for predicting tool use intentions. RESULTS: Perceived usefulness and social influence had a positive effect on the intention to use the feedback tool (P<.001) and the treatment recommendation tool (perceived usefulness, P=.01 and social influence, P<.001). However, trust was unrelated to use intentions for both the tools. Moreover, perceived ease of use was unrelated (feedback tool) and even negatively related (treatment recommendation tool) to use intentions when considering all predictors (P=.004). In addition, a positive relationship between cognitive technology readiness (P=.02) and the intention to use the feedback tool and a negative relationship between AI anxiety and the intention to use the feedback tool (P=.001) and the treatment recommendation tool (P<.001) were observed. CONCLUSIONS: The results shed light on the general and tool-dependent drivers of AI technology adoption in mental health care. Future research may explore the technological and user group characteristics that influence the adoption of AI-enabled tools in mental health care.

Surfing in the streets: How problematic smartphone use, fear of missing out, and antisocial personality traits are linked to driving behavior.

Smartphone use while driving (SUWD) is a major cause of accidents and fatal crashes. This serious problem is still too little understood to be solved. Therefore, the current research aimed to contribute to a better understanding of SUWD by examining factors that have received little or no attention in this context: problematic smartphone use (PSU), fear of missing out (FOMO), and Dark Triad. In the first step, we conducted a systematic literature review to map the current state of research on these factors. In the second step, we conducted a cross-sectional study and collected data from 989 German car drivers. A clear majority (61%) admitted to using the smartphone while driving at least occasionally. Further, the results showed that FOMO is positively linked to PSU and that both are positively associated with SUWD. Additionally, we found that Dark Triad traits are relevant predictors of SUWD and other problematic driving behaviors--in particular, psychopathy is associated with committed traffic offenses. Thus, results indicate that PSU, FOMO, and Dark Triad are relevant factors to explain SUWD. We hope to contribute to a more comprehensive understanding of this dangerous phenomenon with these findings.

Troll story: The dark tetrad and online trolling revisited with a glance at humor.

Internet trolling is considered a negative form of online interaction that can have detrimental effects on people's well-being. This pre-registered, experimental study had three aims: first, to replicate the association between internet users' online trolling behavior and the Dark Tetrad of personality (Machiavellianism, narcissism, psychopathy, and sadism) established in prior research; second, to investigate the effect of experiencing social exclusion on people's motivation to engage in trolling behavior; and third, to explore the link between humor styles and trolling behavior. In this online study, participants were initially assessed on their personality, humor styles, and global trolling behavior. Next, respondents were randomly assigned to a social inclusion or exclusion condition. Thereafter, we measured participants' immediate trolling motivation. Results drawn from 1,026 German-speaking participants indicate a clear correlation between global trolling and all facets of the Dark Tetrad as well as with aggressive and self-defeating humor styles. However, no significant relationship between experiencing exclusion/inclusion and trolling motivation emerged. Our quantile regression findings suggest that psychopathy and sadism scores have a significant positive effect on immediate trolling motivation after the experimental manipulation, whereas Machiavellianism and narcissism did not explain variation in trolling motivation. Moreover, being socially excluded had generally no effect on immediate trolling motivation, apart from participants with higher immediate trolling motivation, for whom the experience of social exclusion actually reduced trolling motivation. We show that not all facets of the Dark Tetrad are of equal importance for predicting immediate trolling motivation and that research should perhaps focus more on psychopathy and sadism. Moreover, our results emphasize the relevance of quantile regression in personality research and suggest that even psychopathy and sadism may not be suitable predictors for low levels of trolling behavior.

Non-task expert physicians benefit from correct explainable AI advice when reviewing X-rays.

Artificial intelligence (AI)-generated clinical advice is becoming more prevalent in healthcare. However, the impact of AI-generated advice on physicians' decision-making is underexplored. In this study, physicians received X-rays with correct diagnostic advice and were asked to make a diagnosis, rate the advice's quality, and judge their own confidence. We manipulated whether the advice came with or without a visual annotation on the X-rays, and whether it was labeled as coming from an AI or a human radiologist. Overall, receiving annotated advice from an AI resulted in the highest diagnostic accuracy. Physicians rated the quality of AI advice higher than human advice. We did not find a strong effect of either manipulation on participants' confidence. The magnitude of the effects varied between task experts and non-task experts, with the latter benefiting considerably from correct explainable AI advice. These findings raise important considerations for the deployment of diagnostic advice in healthcare.

Early and Later Perceptions and Reactions to the COVID-19 Pandemic in Germany: On Predictors of Behavioral Responses and Guideline Adherence During the Restrictions.

In March 2020, the German government enacted measures on movement restrictions and social distancing due to the COVID-19 pandemic. As this situation was previously unknown, it raised numerous questions about people's perceptions of and behavioral responses to these new policies. In this context, we were specifically interested in people's trust in official information, predictors for self-prepping behavior and health behavior to protect oneself and others, and determinants for adherence to social distancing guidelines. To explore these questions, we conducted three studies in which a total of 1,368 participants were surveyed (Study 1 N=377, March 2020; Study 2 N=461, April 2020; Study 3 N=530, April 2021) across Germany between March 2020 and April 2021. Results showed striking differences in the level of trust in official statistics (depending on the source). Furthermore, all three studies showed congruent findings regarding the influence of different factors on the respective behavioral responses. Trust in official statistics predicted behavioral responses in all three studies. However, it did not influence adherence to social distancing guidelines in 2020, but in 2021. Furthermore, adherence to social distancing guidelines was associated with higher acceptance rates of the measures and being older. Being female and less right-wing orientated were positively associated with guidelines adherence only in the studies from 2020. This year, political orientation moderated the association between acceptance of the measures and guideline adherence. This investigation is one of the first to examine perceptions and reactions during the COVID-19 pandemic in Germany across 1year and provides insights into important dimensions that need to be considered when communicating with the public.

The relationship between health IT characteristics and organizational variables among German healthcare workers.

Health information technologies (HITs) are widely employed in healthcare and are supposed to improve quality of care and patient safety. However, so far, their implementation has shown mixed results, which might be explainable by understudied psychological factors of human-HIT interaction. Therefore, the present study investigates the association between the perception of HIT characteristics and psychological and organizational variables among 445 healthcare workers via a cross-sectional online survey in Germany. The proposed hypotheses were tested using structural equation modeling. The results showed that good HIT usability was associated with lower levels of techno-overload and lower IT-related strain. In turn, experiencing techno-overload and IT-related strain was associated with lower job satisfaction. An effective error management culture at the workplace was linked to higher job satisfaction and a slightly lower frequency of self-reported medical errors. About 69% of surveyed healthcare workers reported making errors less frequently than their colleagues, suggesting a bias in either the perception or reporting of errors. In conclusion, the study's findings indicate that ensuring high perceived usability when implementing HITs is crucial to avoiding frustration among healthcare workers and keeping them satisfied. Additionally healthcare facilities should invest in error management programs since error management culture is linked to other important organizational variables.

Do as AI say: susceptibility in deployment of clinical decision-aids.

Artificial intelligence (AI) models for decision support have been developed for clinical settings such as radiology, but little work evaluates the potential impact of such systems. In this study, physicians received chest X-rays and diagnostic advice, some of which was inaccurate, and were asked to evaluate advice quality and make diagnoses. All advice was generated by human experts, but some was labeled as coming from an AI system. As a group, radiologists rated advice as lower quality when it appeared to come from an AI system; physicians with less task-expertise did not. Diagnostic accuracy was significantly worse when participants received inaccurate advice, regardless of the purported source. This work raises important considerations for how advice, AI and non-AI, should be deployed in clinical environments.

Hand(y) hygiene insights: Applying three theoretical models to investigate hospital patients' and visitors' hand hygiene behavior.

BACKGROUND: Improving hand hygiene in hospitals is the most efficient method to prevent healthcare-associated infections. The hand hygiene behavior of hospital patients and visitors is not well-researched, although they pose a risk for the transmission of pathogens. Therefore, the present study had three aims: (1) Finding a suitable theoretical model to explain patients' and visitors' hand hygiene practice; (2) Identifying important predictors for their hand hygiene behavior; and (3) Comparing the essential determinants of hand hygiene behavior between healthcare professionals from the literature to our non-professional sample. METHODS: In total N = 1,605 patients and visitors were surveyed on their hand hygiene practice in hospitals. The employed questionnaires were based on three theoretical models: a) the Theory of Planned Behavior (TPB); b) the Health Action Process Approach (HAPA); and c) the Theoretical Domains Framework (TDF). Structural equation modeling was used to analyze the data. To compare our results to the determinants of healthcare workers' hand hygiene behavior, we searched for studies that used one of the three theoretical models. RESULTS: Among patients, 52% of the variance in the hand hygiene behavior was accounted for by the TDF domains, 44% by a modified HAPA model, and 40% by the TPB factors. Among visitors, these figures were 59%, 37%, and 55%, respectively. Two clusters of variables surfaced as being essential determinants of behavior: self-regulatory processes and social influence processes. The critical determinants for healthcare professionals' hand hygiene reported in the literature were similar to the findings from our non-professional sample. CONCLUSIONS: The TDF was identified as the most suitable model to explain patients' and visitors' hand hygiene practices. Patients and visitors should be included in existing behavior change intervention strategies. Newly planned interventions should focus on targeting self-regulatory and social influence processes to improve effectiveness.

Utilizing behavioral theories to explain hospital visitors' observed hand hygiene behavior.

BACKGROUND: Hand hygiene is essential for infection prevention. This study aimed to find a suitable theoretical model and identify critical facilitators and barriers to explain hospital visitors' hand hygiene practice. METHODS: Visitors in 4 hospitals were observed and asked to give explanations for using or not using the hand rub dispenser. The written explanations of N = 838 participants were coded according to three theoretical models: Theory of Planned Behavior, Health Action Process Approach (HAPA), and Theoretical Domains Framework (TDF). RESULTS: Self-reported hand hygiene behavior differed from observed behavior, with 15.75% wrongly claiming to have cleaned their hands. Critical facilitators for hand hygiene were attitude toward the behavior,subjective norm, outcome expectancies, risk perception, planning, action control, knowledge and skills, motivation and goals, and social influences. Key barriers included perceived behavioral control; barriers and resources; memory, attention, and decision processes; and environmental context and resources. CONCLUSIONS: Visitors' self-reported hand hygiene behavior is over-reported. Both HAPA and TDF were identified as suitable theoretical models for explaining visitor's hand hygiene practice. Future behavior change interventions should focus on (1) visibility and accessibility of cleaning products; (2) informing laypeople about their role regarding infection prevention; and (3) leveraging social influence processes.

The effect of persuasive messages on hospital visitors' hand hygiene behavior.

OBJECTIVE: Hospital visitors pose a risk for transmitting pathogens that can cause health care-associated infections. The present study aimed to test an evidence-based intervention to improve visitors' hand hygiene behavior through persuasive messages. METHOD: For the 14-week-long field experiment, 7 signs were designed according to the principles of persuasion proposed by Cialdini: reciprocity, consistency, social-proof, unity, liking, authority, and scarcity. Each sign was displayed on a screen for 1 week directly above the hand-rub dispenser in a hospital lobby. After each 1-week posting, the screen was blank for 1 week. RESULTS: An electronic monitoring system counted 246,098 people entering and leaving the hospital's lobby and 17,308 dispenser usages. The signs based on the authority and the social-proof principles significantly increased the hand-rub dispenser usage rate in comparison to the average baseline usage rate. CONCLUSIONS: These results indicate that simple and cost-efficient interventions can initiate expedient behavior change in hospitals. However, the findings also highlight the importance of careful planning and rigorous pretesting of material for an intervention to be effective. Theoretical and practical implications of these findings are discussed. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

How a smiley protects health: A pilot intervention to improve hand hygiene in hospitals by activating injunctive norms through emoticons.

Hand hygiene practice in hospitals is unfortunately still widely insufficient, even though it is known that transmitting pathogens via hands is the leading cause of healthcare-associated infections. Previous research has shown that improving knowledge, providing feedback on past behaviour and targeting social norms are promising approaches to improve hand hygiene practices. The present field experiment was designed to direct people on when to perform hand hygiene and prevent forgetfulness. This intervention is the first to examine the effect of inducing injunctive social norms via an emoticon-based feedback system on hand hygiene behaviour. Electronic monitoring and feedback devices were installed in hospital patient rooms on top of hand-rub dispensers, next to the doorway, for a period of 17 weeks. In the emoticon condition, screens at the devices activated whenever a person entered or exited the room. Before using the alcohol-based hand-rub dispenser, a frowny face was displayed, indicating that hand hygiene should be performed. If the dispenser was subsequently used, this picture changed to a smiley face to positively reinforce the correct behaviour. Hand hygiene behaviour in the emoticon rooms significantly outperformed the behaviour in three other tested conditions. The strong effect in this field experiment indicates that activating injunctive norms may be a promising approach to improve hand hygiene behaviour. Theoretical and practical implications of these findings are discussed.