AI Organisations face a trust crisis. Only 15-35% of Americans trust AI companies to act responsibly.

Companies are losing revenue, customers, and facing legal challenges due to AI bias.

Trust starts at the top. By actively engaging in ethical AI practices, leaders can rebuild trust and benefit both society and their organisations.

Interest is growing in artificial empathy, but so is confusion about what artificial empathy is or needs to be. This confusion makes it challenging to navigate the technical and ethical issues that accompany empathic AI development. Here, we outline a framework for thinking about empathic AI based on the premise that different constellations of capabilities associated with empathy are important for different empathic AI applications. We describe distinctions of capabilities that we argue belong under the empathy umbrella, and show how three medical empathic AI use cases require different sets of these capabilities. We conclude by discussing why appreciation of the diverse capabilities under the empathy umbrella is important for both AI creators and users.

Computational preference elicitation methods are tools used to learn people’s preferences quantitatively in a given context. Recent works on preference elicitation advocate for active learning as an efficient method to iteratively construct queries (framed as comparisons between context-specific cases) that are likely to be most informative about an agent’s underlying preferences. In this work, we argue that the use of active learning for moral preference elicitation relies on certain assumptions about the underlying moral preferences, which can be violated in practice. Specifically, we highlight the following common assumptions (a) preferences are stable over time and not sensitive to the sequence of presented queries, (b) the appropriate hypothesis class is chosen to model moral preferences, and (c) noise in the agent’s responses is limited. While these assumptions can be appropriate for preference elicitation in certain domains, prior research on moral psychology suggests they may not be valid for moral judgments. Through a synthetic simulation of preferences that violate the above assumptions, we observe that active learning can have similar or worse performance than a basic random query selection method in certain settings. Yet, simulation results also demonstrate that active learning can still be viable if the degree of instability or noise is relatively small and when the agent’s preferences can be approximately represented with the hypothesis class used for learning. Our study highlights the nuances associated with effective moral preference elicitation in practice and advocates for the cautious use of active learning as a methodology to learn moral preferences.

Preference elicitation frameworks feature heavily in the research on participatory ethical AI tools and provide a viable mechanism to enquire and incorporate the moral values of various stakeholders. As part of the elicitation process, surveys about moral preferences, opinions, and judgments are typically administered only once to each participant. This methodological practice is reasonable if participants’ responses are stable over time such that, all other relevant factors being held constant, their responses today will be the same as their responses to the same questions at a later time. However, we do not know how often that is the case. It is possible that participants’ true moral preferences change, are subject to temporary moods or whims, or are influenced by environmental factors we don’t track. If participants’ moral responses are unstable in such ways, it would raise important methodological and theoretical issues for how participants’ true moral preferences, opinions, and judgments can be ascertained. We address this possibility here by asking the same survey participants the same moral questions about which patient should receive a kidney when only one is available ten times in ten different sessions over two weeks, varying only presentation order across sessions. We measured how often participants gave different responses to simple (Study One) and more complicated (Study Two) repeated scenarios. On average, the fraction of times participants changed their responses to controversial scenarios was around 10-18% across studies, and this instability is observed to have positive associations with response time and decision-making difficulty. We discuss the implications of these results for the efficacy of moral preference elicitation, highlighting the role of response instability in causing value misalignment between stakeholders and AI tools trained on their moral judgments.

Background: In the early stages of the COVID-19 pandemic, many health systems, including those in the UK, developed triage guidelines to manage severe shortages of ventilators. At present, there is an insufficient understanding of how the public views these guidelines, and little evidence on which features of a patient the public believe should and should not be considered in ventilator triage.

Methods: Two surveys were conducted with representative UK samples. In the first survey, 525 participants were asked in an open-ended format to provide features they thought should and should not be considered in allocating ventilators for COVID-19 patients when not enough ventilators are available. In the second survey, 505 participants were presented with 30 features identified from the first study, and were asked if these features should count in favour of a patient with the feature getting a ventilator, count against the patient, or neither. Statistical tests were conducted to determine if a feature was generally considered by participants as morally relevant and whether its mean was non-neutral.

Results: In Survey 1, the features of a patient most frequently cited as being morally relevant to determining who would receive access to ventilators were age, general health, prospect of recovery, having dependents, and the severity of COVID symptoms. The features most frequently cited as being morally irrelevant to determining who would receive access to ventilators are race, gender, economic status, religion, social status, age, sexual orientation, and career. In Survey 2, the top three features that participants thought should count in favour of receiving a ventilator were pregnancy, having a chance of dying soon, and having waited for a long time. The top three features that participants thought should count against a patient receiving a ventilator were having committed violent crimes in the past, having unnecessarily engaged in activities with a high risk of COVID-19 infection, and a low chance of survival.

Conclusions: The public generally agreed with existing UK guidelines that allocate ventilators according to medical benefits and that aim to avoid discrimination based on demographic features such as race and gender. However, many participants expressed potentially non-utilitarian concerns, such as inclining to deprioritise ventilator allocation to those who had a criminal history or who contracted the virus by needlessly engaging in high-risk activities.

When two people need a kidney transplant, but only one kidney is available, we need to decide who gets it. If one of the potential recipients needs the kidney because of their own voluntary behavior, but the other is not at all responsible for needing a kidney, then we need to decide whether this fault should be a consideration in favor of the other patient getting the kidney. While there has been considerable philosophical debate on this issue, there is far less research into the views of the public. To explore opinions on these issues, we first asked survey participants to ascribe or deny responsibility for the risky behavior for kidney disease, and for being deprived of a kidney in cases of drinking alcohol, drug abuse, smoking and unhealthy eating when the patient did or did not stop the risky behavior after being diagnosed with kidney disease. Next, we asked participants who should get the kidney when the patient who engaged in risky behavior did or did not know, or have easy access to, the information that the behavior was risky. We found that participants generally ascribed responsibility on the basis of knowledge but allocated the kidney on the basis of access to information. These findings have important implications for moral theories as well as medical policies.

AI systems are often used to make or contribute to important decisions in a growing range of applications, including criminal justice, hiring, and medicine. Since these decisions impact human lives, it is important that the AI systems act in ways which align with human values. Techniques for preference modeling and social choice help researchers learn and aggregate peoples’ preferences, which are used to guide AI behavior; thus, it is imperative that these learned preferences are accurate. These techniques often assume that people are willing to express strict preferences over alternatives; which is not true in practice. People are often indecisive, and especially so when their decision has moral implications. The philosophy and psychology literature shows that indecision is a measurable and nuanced behavior — and that there are several different reasons people are indecisive. This complicates the task of both learning and aggregating preferences, since most of the relevant literature makes restrictive assumptions on the meaning of indecision. We begin to close this gap by formalizing several mathematical \emph{indecision} models based on theories from philosophy, psychology, and economics; these models can be used to describe (indecisive) agent decisions, both when they are allowed to express indecision and when they are not. We test these models using data collected from an online survey where participants choose how to (hypothetically) allocate organs to patients waiting for a transplant.

Given AI’s growing role in modeling and improving decision-making, how and when to present users with feedback is an urgent topic to address. We empirically examined the effect of feedback from false AI on moral decision-making about donor kidney allocation. We found some evidence that judgments about whether a patient should receive a kidney can be influenced by feedback about participants’ own decision-making perceived to be given by AI, even if the feedback is entirely random. We also discovered different effects between assessments presented as being from human experts and assessments presented as being from AI.

The efficient and fair allocation of limited resources is a classical problem in economics and computer science. In kidney exchanges, a central market maker allocates living kidney donors to patients in need of an organ. Patients and donors in kidney exchanges are prioritized using ad-hoc weights decided on by committee and then fed into an allocation algorithm that determines who gets what–and who does not. In this paper, we provide an end-to-end methodology for estimating weights of individual participant profiles in a kidney exchange. We first elicit from human subjects a list of patient attributes they consider acceptable for the purpose of prioritizing patients (e.g., medical characteristics, lifestyle choices, and so on). Then, we ask subjects comparison queries between patient profiles and estimate weights in a principled way from their responses. We show how to use these weights in kidney exchange market clearing algorithms. We then evaluate the impact of the weights in simulations and find that the precise numerical values of the weights we computed matter little, other than the ordering of profiles that they imply. However, compared to not prioritizing patients at all, there is a significant effect, with certain classes of patients being (de)prioritized based on the human-elicited value judgments.

AI systems are now or will soon be sophisticated enough to make consequential decisions. Although this technology has flourished, we also need public appraisals of AI systems playing these more important roles. This article reports surveys of preferences for and against AI systems making decisions in various domains as well as experiments that intervene on these preferences. We find that these preferences are contingent on subjects’ previous exposure to computer systems making these kinds of decisions, and some interventions designed to mimic previous exposure successfully encourage subjects to be more hospitable to computer systems making these weighty decisions.

The efficient and fair allocation of limited resources is a classical problem in economics and computer science. In kidney exchanges, a central market maker allocates living kidney donors to patients in need of an organ. Patients and donors in kidney exchanges are prioritized using ad-hoc weights decided on by committee and then fed into an allocation algorithm that determines who gets what–and who does not. In this paper, we provide an end-to-end methodology for estimating weights of individual participant profiles in a kidney exchange. We first elicit from human subjects a list of patient attributes they consider acceptable for the purpose of prioritizing patients (e.g., medical characteristics, lifestyle choices, and so on). Then, we ask subjects comparison queries between patient profiles and estimate weights in a principled way from their responses. We show how to use these weights in kidney exchange market clearing algorithms. We then evaluate the impact of the weights in simulations and find that the precise numerical values of the weights we computed matter little, other than the ordering of profiles that they imply. However, compared to not prioritizing patients at all, there is a significant effect, with certain classes of patients being (de)prioritized based on the human-elicited value judgments.

The generality of decision and game theory has enabled domain-independent progress in AI research. For example, a better algorithm for finding good policies in (PO)MDPs can be instantly used in a variety of applications. But such a general theory is lacking when it comes to moral decision making. For AI applications with a moral component, are we then forced to build systems based on many ad-hoc rules? In this paper we discuss possible ways to avoid this conclusion.

Technological advances are enabling roles for machines that present novel ethical challenges. The study of ‘AI ethics’ has emerged to confront these challenges, and connects perspectives from philosophy, computer science, law, and economics. Less represented in these interdisciplinary efforts is the perspective of cognitive science. We propose a framework – computational ethics – that specifies how the ethical challenges of AI can be partially addressed by incorporating the study of human moral decision-making. The driver of this framework is a computational version of reflective equilibrium (RE), an approach that seeks coherence between considered judgments and governing principles. The framework has two goals: (i) to inform the engineering of ethical AI systems, and (ii) to characterize human moral judgment and decision-making in computational terms. Working jointly towards these two goals will create the opportunity to integrate diverse research questions, bring together multiple academic communities, uncover new interdisciplinary research topics, and shed light on centuries-old philosophical questions.

The elicitors of disgust are heterogeneous, which makes attributing one function to disgust challenging. Theorists have proposed that disgust solves multiple adaptive problems and comprises multiple functional domains. However, theories conflict with regard to what the domains are and how they should be delineated. In this article, we examine clinical evidence of aberrant disgust symptoms in the contamination subtype of obsessive-compulsive disorder, blood-injury-injection phobia, and posttraumatic stress disorder to adjudicate between two prevailing theories of disgust. We argue that the pattern of disgust sensitivities in these psychiatric disorders sheds new light on the domain structure of disgust. Specifically, the supported domain structure of disgust is likely similar to an adaptationist model of disgust, with more subdivisions of the domain of pathogen disgust. We discuss the implications of this approach for the prevention and treatment of psychiatric disorders relevant to disgust.

Psychopathy is a disorder characterized by severe and frequent moral violations in multiple domains of life. Numerous studies have shown psychopathy-related limbic brain abnormalities during moral processing; however, these studies only examined negatively valenced moral stimuli. Here, we aimed to replicate prior psychopathy research on negative moral judgments and to extend this work by examining psychopathy-related abnormalities in the processing of controversial moral stimuli and positive moral processing. Incarcerated adult males (N = 245) completed a functional magnetic resonance imaging protocol on a mobile imaging system stationed at the prison. Psychopathy was assessed using the Hare Psychopathy Checklist-Revised (PCL-R). Participants were then shown words describing three types of moral stimuli: wrong (e.g., stealing), not wrong (e.g., charity), and controversial (e.g., euthanasia). Participants rated each stimulus as either wrong or not wrong. PCL-R total scores were correlated with not wrong behavioral responses to wrong moral stimuli, and were inversely related to hemodynamic activity in the anterior cingulate cortex in the contrast of wrong > not wrong. In the controversial > noncontroversial comparison, psychopathy was inversely associated with activity in the temporal parietal junction and dorsolateral prefrontal cortex. These results indicate that psychopathy-related abnormalities are observed during the processing of complex, negative, and positive moral stimuli.

Rationale: Stimulant use is a significant and prevalent problem, particularly in criminal populations. Previous studies found that cocaine and methamphetamine use is related to impairment in identifying emotions and empathy. Stimulant users also have abnormal neural structure and function of the ventromedial prefrontal cortex (vmPFC), amygdala, and anterior (ACC) and posterior cingulate (PCC), regions implicated in moral decision-making. However, no research has studied the neural correlates of stimulant use and explicit moral processing in an incarcerated population.

Objectives: Here, we examine how stimulant use affects sociomoral processing that might contribute to antisocial behavior. We predicted that vmPFC, amygdala, PCC, and ACC would show abnormal neural response during a moral processing task in incarcerated methamphetamine and cocaine users.

Methods: Incarcerated adult males (N = 211) were scanned with a mobile MRI system while completing a moral decision-making task. Lifetime drug use was assessed. Neural responses during moral processing were compared between users and non-users. The relationship between duration of use and neural function was also examined.

Results: Incarcerated stimulant users showed less amygdala engagement than non-users during moral processing. Duration of stimulant use was negatively associated with activity in ACC and positively associated with vmPFC response during moral processing.

Conclusions: These results suggest a dynamic pattern of fronto-limbic moral processing related to stimulant use with deficits in both central motive and cognitive integration elements of biological moral processes theory. This increases our understanding of how drug use relates to moral processing in the brain in an ultra-high-risk population.

This chapter proposes that a remaining frontier of morality research is to understand how and why humans perform moral and immoral actions, particularly those related to unjustified violence. It is argued that the best way to do this would be to incorporate neuroscience research using rodent models of negative intersubjectivity into empirical investigations of morality. The relationship between negative intersubjectivity and moral action is reviewed, and rodent models of negative intersubjectivity are evaluated. The chapter concludes by discussing how these rodent models can be used and expanded to develop treatments for clinically violent behavior, as well as to deepen our understanding of other types of morally relevant action.

A large hunk of research in moral psychology is devoted to self-reports, which represent the end product of a complex and diverse bundle of underlying cognitive processes.1 There is more to the moral processing, however, than what can be discerned from introspection or straightfor-ward paper-and-pencil methodologies. A complete account must include all of the processes — explicit or implicit, articulated or unspoken — that go into everyday moral responses.

Psychopathy research is plagued by an enigma: Psychopaths reliably act immorally, but they also accurately report whether an action is morally wrong. The current study revealed that cooperative suppressor effects and conflicting subsets of personality traits within the construct of psychopathy might help explain this conundrum. Among a sample of adult male offenders (n = 100) who ranked deserved punishment of crimes, Psychopathy Checklist-Revised (PCL-R) total scores were not linearly correlated with deserved punishment task performance. However, these null results masked significant opposing associations between task performance and factors of psychopathy: the PCL-R Interpersonal/Affective (i.e. manipulative and callous) factor was positively associated with task performance, while the PCL-R Social Deviance (i.e. impulsive and antisocial) factor was simultaneously negatively associated with task performance. Importantly, these relationships were qualified by a significant interaction where the Interpersonal/Affective traits were positively associated with task performance when Social Deviance traits were high, but Social Deviance traits were negatively associated with task performance when Interpersonal/Affective traits were low. This interaction helped reveal a significant non-linear relationship between PCL-R total scores and task performance such that individuals with very low or very high PCL-R total scores performed better than those with middle-range PCL-R total scores. These results may explain the enigma of why individuals with very high psychopathic traits, but not other groups of anti-social individuals, usually have normal moral judgment in laboratory settings, but still behave immorally, especially in contexts where Social Deviance traits have strong influence.

The goal of this chapter is to provide a comprehensive review of the scientific evidence for and against the hypothesis that psychopaths are impaired in their moral judgments. In our view, contrary to popular opinion, the current literature does not provide evidence suggesting psychopaths have severe moral cognition deficits. However, as will become clear, few firm conclusions can be reached about moral cognition in psychopaths without further research. So far there are very little data examining moral judgment or decision making in psychopaths, partially because historically psychopathy research has been practicably and financially difficult to implement. Another reason is that psychopaths are often pathological liars, so it is hard to determine what they really believe. Additional obstacles arise because different researchers have used inconsistent criteria for diagnosing psychopathy and because few scientific tests of moral judgment or belief are established and/or standardized. To interpret this limited evidence, it is critical that both psychopathy and moral judgments be defined carefully. We therefore begin by discussing both of these definitions in turn.

How people judge something to be morally right or wrong is a fundamental question of both the sciences and the humanities. Here we aim to identify the neural processes that underlie the specific conclusion that something is morally wrong. To do this, we introduce a novel distinction between “moral deliberation,” or the weighing of moral considerations, and the formation of a “moral verdict,” or the commitment to one moral conclusion. We predict and identify hemodynamic activity in the bilateral anterior insula and basal ganglia that correlates with committing to the moral verdict “this is morally wrong” as opposed to “this is morally not-wrong,” a finding that is consistent with research from economic decision-making. Using comparisons of deliberation-locked vs. verdict-locked analyses, we also demonstrate that hemodynamic activity in high-level cortical regions previously implicated in morality—including the ventromedial prefrontal cortex, posterior cingulate cortex, and temporoparietal junction—correlates primarily with moral deliberation as opposed to moral verdicts. These findings provide new insights into what types of processes comprise the enterprise of moral judgment, and in doing so point to a framework for resolving why some clinical patients, including psychopaths, may have intact moral judgment but impaired moral behavior.

Evolutionary approaches to dissecting our psychological architecture underscore the importance of both function and structure. Here we focus on both the function and structure of our neural circuitry and report a functional bilateral asymmetry associated with the processing of immoral stimuli. Many processes in the human brain are associated with functional specialization unique to one hemisphere. With respect to emotions, most research points to right-hemispheric lateralization. Here we provide evidence that not all emotional stimuli share right-hemispheric lateralization. Across three studies employing different paradigms, the processing of negative morally laden stimuli was found to be highly left-lateralized. Regions of engagement common to the three studies include the left medial prefrontal cortex, left temporoparietal junction, and left posterior cingulate. These data support the hypothesis that processing of immoral stimuli preferentially engages left hemispheric processes and sheds light on our evolved neural architecture.

Disgust, an emotion related to avoiding harmful substances, has been linked to moral judgments in many behavioral studies. However, the fact that participants report feelings of disgust when thinking about feces and a heinous crime does not necessarily indicate that the same mechanisms mediate these reactions. Humans might instead have separate neural and physiological systems guiding aversive behaviors and judgments across different domains. The present interdisciplinary study used functional magnetic resonance imaging (n = 50) and behavioral assessment to investigate the biological homology of pathogen-related and moral disgust. We provide evidence that pathogen-related and sociomoral acts entrain many common as well as unique brain networks. We also investigated whether morality itself is composed of distinct neural and behavioral subdomains. We provide evidence that, despite their tendency to elicit similar ratings of moral wrongness, incestuous and nonsexual immoral acts entrain dramatically separate, while still overlapping, brain networks. These results (i) provide support for the view that the biological response of disgust is intimately tied to immorality, (ii) demonstrate that there are at least three separate domains of disgust, and (iii) suggest strongly that morality, like disgust, is not a unified psychological or neurological phenomenon.

Calls for Ethical AI have become urgent and pervasive, especially as ethical issues surrounding AI products at tech companies are increasingly scrutinized by the public. Yet even after a first wave of responses to these calls coalesced around Ethical AI principles to guide decision-making and a second wave generated technical tools to mitigate specific ethical issues, multiple lines of evidence indicate that these Ethical AI principles and technical tools have only a limited impact on the daily practices of AI users and producers. In other words, there is a big gap between what we publish in academic papers and what AI creators need to generate AI products that reflect society’s values. Ethical AI is by no means the only field to have this problem. However, when medical and ecology fields documented similar gaps between their fields’ scientific discoveries and the practices and products that people actually use, they invested tremendous resources into subfields that developed evidence about how to translate what was done in the lab to adopted solutions. I argue in this commentary that it is our research community’s moral duty to invest in our own subfield of “Translational Ethical AI” that will determine how best to ensure AI practitioners can implement the Ethical AI technical tools we publish in academic venues in production settings. Further, I offer concrete steps for doing that, drawing on insights gleaned from other translational fields. Closing the “Ethical AI Publication-to-Practice gap” will be a considerable transdisciplinary challenge, but one of the AI research community has the unique expertise, political leverage, and moral responsibility to tackle.

Big Data and Artificial Intelligence have a symbiotic relationship. Artificial Intelligence needs to be trained on Big Data to be accurate, and Big Data’s value is largely realized through its use by Artificial Intelligence. As a result, Big Data and Artificial Intelligence practices are tightly intertwined in real life settings, as are their impacts on society. Unethical uses of Artificial Intelligence are therefore a Big Data problem, at least to some degree. Efforts to address this problem have been dominated by the documentation of Ethical Artificial Intelligence principles and the creation of technical tools that address specific aspects of those principles. However, there is mounting evidence that Ethical Artificial Intelligence principles and technical tools have little impact on the Artificial Intelligence that is created in practice, sometimes in very public ways. The goal of this commentary is to highlight four interconnected areas society can invest in to close this Ethical Artificial Intelligence publication-to-practice gap, maximizing the positive impact Artificial Intelligence and Big Data have on society. For Ethical Artificial Intelligence to become a reality, I argue that these areas need to be addressed holistically in a way that acknowledges their interdependencies. Progress will require iteration, compromise, and transdisciplinary collaboration, but the result of our investments will be the realization of Artificial Intelligence’s and Big Data’s tremendous potential for social good, in practice rather than in just our hopes and aspirations.

Diverse disciplines are interested in how the coordination of interacting agents’ movements, emotions, and physiology over time impacts social behavior. Here, we describe a new multivariate procedure for automating the investigation of this kind of behaviorally-relevant “interactional synchrony”, and introduce a novel interactional synchrony measure based on features of dynamic time warping (DTW) paths. We demonstrate that our DTW path-based measure of interactional synchrony between facial action units of two people interacting freely in a natural social interaction can be used to predict how much trust they will display in a subsequent Trust Game. We also show that our approach outperforms univariate head movement models, models that consider participants’ facial action units independently, and models that use previously proposed synchrony or similarity measures. The insights of this work can be applied to any research question that aims to quantify the temporal coordination of multiple signals over time, but has immediate applications in psychology, medicine, and robotics.

Many modern data sets require inference methods that can estimate the shared and individual-specific components of variability in collections of matrices that change over time. Promising methods have been developed to analyze these types of data in static cases, but only a few approaches are available for dynamic settings. To address this gap, we consider novel models and inference methods for pairs of matrices in which the columns correspond to multivariate observations at different time points. In order to characterize common and individual features, we propose a Bayesian dynamic factor modeling framework called Time Aligned Common and Individual Factor Analysis (TACIFA) that includes uncertainty in time alignment through an unknown warping function. We provide theoretical support for the proposed model, showing identifiability and posterior concentration. The structure enables efficient computation through a Hamiltonian Monte Carlo (HMC) algorithm. We show excellent performance in simulations, and illustrate the method through application to a social mimicry experiment.

To demonstrate the capability of computer vision analysis to detect atypical orienting and attention behaviors in toddlers with autism spectrum disorder. One hundered and four toddlers of 16-31 months old (mean = 22) participated in this study. Twenty-two of the toddlers had autism spectrum disorder and 82 had typical development or developmental delay. Toddlers watched video stimuli on a tablet while the built-in camera recorded their head movement. Computer vision analysis measured participants’ attention and orienting in response to name calls. Reliability of the computer vision analysis algorithm was tested against a human rater. Differences in behavior were analyzed between the autism spectrum disorder group and the comparison group. Reliability between computer vision analysis and human coding for orienting to name was excellent (intra-class coefficient 0.84, 95% confidence interval 0.67-0.91). Only 8% of toddlers with autism spectrum disorder oriented to name calling on >1 trial, compared to 63% of toddlers in the comparison group (p = 0.002). Mean latency to orient was significantly longer for toddlers with autism spectrum disorder (2.02 vs 1.06 s, p = 0.04). Sensitivity for autism spectrum disorder of atypical orienting was 96% and specificity was 38%. Older toddlers with autism spectrum disorder showed less attention to the videos overall (p = 0.03). Automated coding offers a reliable, quantitative method for detecting atypical social orienting and reduced sustained attention in toddlers with autism spectrum disorder.

Current tools for objectively measuring young children’s observed behaviors are expensive, time-consuming, and require extensive training and professional administration. The lack of scalable, reliable, and validated tools impacts access to evidence-based knowledge and limits our capacity to collect population-level data in non-clinical settings. To address this gap, we developed mobile technology to collect videos of young children while they watched movies designed to elicit autism-related behaviors and then used automatic behavioral coding of these videos to quantify children’s emotions and behaviors. We present results from our iPhone study Autism & Beyond, built on ResearchKit’s open-source platform. The entire study—from an e-Consent process to stimuli presentation and data collection—was conducted within an iPhone-based app available in the Apple Store. Over 1 year, 1756 families with children aged 12–72 months old participated in the study, completing 5618 caregiver-reported surveys and uploading 4441 videos recorded in the child’s natural settings. Usable data were collected on 87.6% of the uploaded videos. Automatic coding identified significant differences in emotion and attention by age, sex, and autism risk status. This study demonstrates the acceptability of an app-based tool to caregivers, their willingness to upload videos of their children, the feasibility of caregiver-collected data in the home, and the application of automatic behavioral encoding to quantify emotions and attention variables that are clinically meaningful and may be refined to screen children for autism and developmental disorders outside of clinical settings. This technology has the potential to transform how we screen and monitor children’s development.

In spite of recent advances in the genetics and neuroscience of early childhood mental health, behavioral observation is still the gold standard in screening, diagnosis, and outcome assessment. Unfortunately, clinical observation is often subjective, needs significant rater training, does not capture data from participants in their natural environment, and is not scalable for use in large populations or for longitudinal monitoring. To address these challenges, we developed and tested a self-contained app designed to measure toddlers’ social communication behaviors in a primary care, school, or home setting. Twenty 16-30 month old children with and without autism participated in this study. Toddlers watched the developmentally-appropriate visual stimuli on an iPad in a pediatric clinic and in our lab while the iPad camera simultaneously recorded video of the child’s behaviors. Automated computer vision algorithms coded emotions and social referencing to quantify autism risk behaviors. We validated our automatic computer coding by comparing the computer-generated analysis of facial expression and social referencing to human coding of these behaviors. We report our method and propose the development and testing of measures of young children’s behaviors as the first step toward development of a novel, fully integrated, low-cost, scalable screening tool for autism and other neurodevelopmental disorders of early childhood.

Introduction: It is unknown how the brain coordinates decisions to withstand personal costs in order to prevent other individuals’ distress. Here we test whether local field potential (LFP) oscillations between brain regions create “neural contexts” that select specific brain functions and encode the outcomes of these types of intersubjective decisions.

Methods: Rats participated in an “Intersubjective Avoidance Test” (IAT) that tested rats’ willingness to enter an innately aversive chamber to prevent another rat from getting shocked. c-Fos immunoreactivity was used to screen for brain regions involved in IAT performance. Multi-site local field potential (LFP) recordings were collected simultaneously and bilaterally from five brain regions implicated in the c-Fos studies while rats made decisions in the IAT. Local field potential recordings were analyzed using an elastic net penalized regression framework.

Results: Rats voluntarily entered an innately aversive chamber to prevent another rat from getting shocked, and c-Fos immunoreactivity in brain regions known to be involved in human empathy—including the anterior cingulate, insula, orbital frontal cortex, and amygdala—correlated with the magnitude of “intersubjective avoidance” each rat displayed. Local field potential recordings revealed that optimal accounts of rats’ performance in the task require specific frequencies of LFP oscillations between brain regions in addition to specific frequencies of LFP oscillations within brain regions. Alpha and low gamma coherence between spatially distributed brain regions predicts more intersubjective avoidance, while theta and high gamma coherence between a separate subset of brain regions predicts less intersubjective avoidance. Phase relationship analyses indicated that choice-relevant coherence in the alpha range reflects information passed from the amygdala to cortical structures, while coherence in the theta range reflects information passed in the reverse direction.

Conclusion: These results indicate that the frequency-specific “neural context” surrounding brain regions involved in social cognition encodes outcomes of decisions that affect others, above and beyond signals from any set of brain regions in isolation.

Simultaneous neural recordings taken from multiple areas of the rodent brain are garnering growing interest due to the insight they can provide about spatially distributed neural circuitry. The promise of such recordings has inspired great progress in methods for surgically implanting large numbers of metal electrodes into intact rodent brains. However, methods for localizing the precise location of these electrodes have remained severely lacking. Traditional histological techniques that require slicing and staining of physical brain tissue are cumbersome, and become increasingly impractical as the number of implanted electrodes increases. Here we solve these problems by describing a method that registers 3-D computerized tomography (CT) images of intact rat brains implanted with metal electrode bundles to a Magnetic Resonance Imaging Histology (MRH) Atlas. Our method allows accurate visualization of each electrode bundle’s trajectory and location without removing the electrodes from the brain or surgically implanting external markers. In addition, unlike physical brain slices, once the 3D images of the electrode bundles and the MRH atlas are registered, it is possible to verify electrode placements from many angles by “re-slicing” the images along different planes of view. Further, our method can be fully automated and easily scaled to applications with large numbers of specimens. Our digital imaging approach to efficiently localizing metal electrodes offers a substantial addition to currently available methods, which, in turn, may help accelerate the rate at which insights are gleaned from rodent network neuroscience.

One of the goals of neuroscience is to identify neural networks that correlate with important behaviors, environments, or genotypes. This work proposes a strategy for identifying neural networks characterized by time- and frequency-dependent connectivity patterns, using convolutional dictionary learning that links spike-train data to local field potentials (LFPs) across multiple areas of the brain. Analytical contributions are: (i) modeling dynamic relationships between LFPs and spikes; (ii) describing the relationships between spikes and LFPs, by analyzing the ability to predict LFP data from one region based on spiking information from across the brain; and (iii) development of a clustering methodology that allows inference of similarities in neurons from multiple regions. Results are based on data sets in which spike and LFP data are recorded simultaneously from up to 16 brain regions in a mouse.

The local field potential (LFP) is a source of information about the broad patterns of brain activity, and the frequencies present in these time-series measurements are often highly correlated between regions. It is believed that these regions may jointly constitute a “brain state,” relating to cognition and behavior. An infinite hidden Markov model (iHMM) is proposed to model the evolution of brain states, based on electrophysiological LFP data measured at multiple brain regions. A brain state influences the spectral content of each region in the measured LFP. A new state-dependent tensor factorization is employed across brain regions, and the spectral properties of the LFPs are characterized in terms of Gaussian processes (GPs). The LFPs are modeled as a mixture of GPs, with state- and region-dependent mixture weights, and with the spectral content of the data encoded in GP spectral mixture covariance kernels. The model is able to estimate the number of brain states and the number of mixture components in the mixture of GPs. A new variational Bayesian split-merge algorithm is employed for inference. The model infers state changes as a function of external covariates in two novel electrophysiological datasets, using LFP data recorded simultaneously from multiple brain regions in mice; the results are validated and interpreted by subject-matter experts.

Sleep and metabolism are intertwined physiologically and behaviorally, but the neural systems underlying their coordination are still poorly understood. The hypothalamus is likely to play a major role in the regulation sleep, metabolism, and their interaction. And increasing evidence suggests that hypocretin cells in the lateral hypothalamus may provide particularly important contributions. Here we review: 1) direct interactions between biological arousal and metabolic systems in the hypothalamus, and 2) indirect interactions between these two systems mediated by stress or reward, emphasizing the role of hypocretins. An increased understanding of the mechanisms underlying these interactions may provide novel approaches for the treatment of patients with sleep disorders and obesity, as well as suggest new therapeutic strategies for symptoms of aging, stress, or addiction.

The hypocretins (abbreviated ‘Hcrts’ – also called ‘orexins’) are two neuropeptides secreted exclusively by a small population of neurons in the lateral hypothalamus. These peptides bind to two receptors located throughout the brain in nuclei associated with diverse cognitive and physiological functions. Initially, the brain Hcrt system was found to have a major role in the regulation of sleep/wake transitions. More recent studies indicate Hcrts may play a role in other physiological functions, including food intake, addiction, and stress. Taken together, these studies suggest a general role for Hcrts in mediating arousal, especially when an organism must respond to unexpected stressors and challenges in the environment.

Although stress is an extensively investigated phenomenon, the effects of specific stressors on the pharmacologic activity of routinely administered drugs are less well characterized. We designed the present study to investigate the effect of handling stress on catecholaminergic responsivity following an acute methylphenidate (MP, Ritalin) challenge in the medial prefrontal cortex (mPFC). Norepinephrine (NE) and dopamine (DA) levels were simultaneously measured in 15-min samples of PFC dialysate using HPLC coupled with electrochemical detection. Sprague-Dawley rats were handled for 15 min, which produced an increase from basal extracellular DA and NE levels. Handling stress attenuates the DA response when administered 2 h prior to IP MP, whereas handling stress enhances the DA response when administered simultaneously with IG MP. These findings suggest that persistent alterations in mesocorticolimbic DA-ergic activity are induced by a short exposure to restraint stress as evidenced by the altered response to MP challenge.

Motor-related regions of parietal and prefrontal cortices have been shown to selectively activate when observers passively view objects that afford manual grasping. Yet, it remains unknown whether these cortical responses depend on prior motor-related experience with the object being observed. To address this question, we asked participants to undergo fMRI scanning while viewing exemplars of two different categories of graspable objects: one associated with extensive motor experience (door knobs) and one associated with no self-reported motor experience (artificial rock climbing holds). Despite participants’ lack of experience grasping climbing holds, these objects were found to generate a systematic response in several visuomotor-related regions of cortex-including left PMv and left AIP. Interestingly, however, the response to door knobs did not include activity in any motor-related regions, being limited instead to a comparatively small bilateral area of lateral occipital cortex, relative to the more spatially extensive response in occipital and temporal cortex that was observed for climbing holds. This result suggested that object-specific responses in both visual- and motor-related cortex may in fact negatively correlate with object-specific motor experience. To test this possibility, we repeated the experiment using participants having extensive self-reported experience grasping climbing holds (i.e., veteran indoor rock climbers). Consistent with our hypothesis, both climbing holds and door knobs generated activity limited to lateral occipital cortex. Taken together, these data support the proposal that repeated real-world motor experience with an object category may lead to reduced implicit analysis in both motor- and visual-related regions of cortex.

Visual spatial attention has long been associated with facilitatory effects on visual perception. Here, we report that spatial attention can also modulate implicit visuomotor processing in dorsal regions of human cortex. Participants underwent fMRI scanning while performing a voluntary attentional orienting task that varied the category of a task-irrelevant object in the attended location (tool vs. non-tool). Data were then analyzed as a function of the attended location (left vs. right visual field) and the object category in that location. We found that the fMRI BOLD response in two visuomotor-related regions–the supplementary motor area (SMA) and the left inferior parietal lobule (IPL)–showed an interaction between the location of attention and the location of the tool in the bilateral display. Further, these responses were statistically distinct from those regions in dorsal cortex showing activity modulated only by the tool location or only by the attended location. While the effects of attending non-foveally within the visual field have been well documented in relation to visual perception, our findings support the proposal that voluntary visuospatial attention may also have consequences for the implicit planning of object-directed actions.

Although the need for data science methodological training is widely recognized across many disciplines, data science training is often absent from PhD programs. At the same time, master’s-level data science educational programs have seen incredible growth and investment. In 2018, Duke University initiated a National Science Foundation (NSF)-funded program to determine whether master’s-level data science programs that universities have already invested in could be leveraged to reduce data science education barriers doctoral students face. Doctoral fellows from diverse fields worked with teams of master’s students from Duke’s Master in Interdisciplinary Data Science program on applied capstone projects focused on the doctoral fellows’ own disciplines and dissertation research. Fellows also gained access to the master’s program’s courses and professional development resources. We examined the implementation, experience, and effect of this integration into Master in Data Science program infrastructure using qualitative data collection with doctoral fellows, master’s students, and fellows’ doctoral advisors. Master’s students participating in doctoral-led capstones benefited from their doctoral fellows’ mentorship, project management, and content knowledge. Participating doctoral students showed increased learning of data science techniques and professional skills development. While some fellows’ research was advanced through the capstones, data also showed mismatches between selected master’s program goals and doctoral students’ needs. Overall, this pilot indicated potential promise in harnessing existing Master in Data Science programs to bolster doctoral students’ data science learning and professional readiness while also identifying areas for improving future such efforts.