The relational bottleneck as an inductive bias for efficient abstraction.

A central challenge for cognitive science is to explain how abstract concepts are acquired from limited experience. This has often been framed in terms of a dichotomy between connectionist and symbolic cognitive models. Here, we highlight a recently emerging line of work that suggests a novel reconciliation of these approaches, by exploiting an inductive bias that we term the relational bottleneck. In that approach, neural networks are constrained via their architecture to focus on relations between perceptual inputs, rather than the attributes of individual inputs. We review a family of models that employ this approach to induce abstractions in a data-efficient manner, emphasizing their potential as candidate models for the acquisition of abstract concepts in the human mind and brain.

Remotely controlled drug release in deep brain regions of non-human primates.

Many areas of science and medicine would benefit from selective release of drugs in specific regions. Nanoparticle drug carriers activated by focused ultrasound-remotely applied, depth-penetrating energy-may provide such selective interventions. Here, we developed stable, ultrasound-responsive nanoparticles that can be used to release drugs effectively and safely in non-human primates. The nanoparticles were used to release propofol in deep brain visual regions. The release reversibly modulated the subjects' visual choice behavior and was specific to the targeted region and to the released drug. Gadolinium-enhanced MR imaging suggested an intact blood-brain barrier. Blood draws showed normal clinical chemistry and hematology. In summary, this study provides a safe and effective approach to release drugs on demand in selected deep brain regions at levels sufficient to modulate behavior.

Remotely controlled drug release in deep brain regions of non-human primates.

Many areas of science and medicine would benefit from selective release of drugs in specific regions of interest. Nanoparticle drug carriers activated by focused ultrasound-remotely applied, depth-penetrating energy-may provide such selective interventions. Here, we developed stable, ultrasound-responsive nanoparticles that can be used to release drugs effectively and safely in non-human primates. The nanoparticles were used to release propofol in deep brain visual regions. The release reversibly modulated the subjects' visual choice behavior and was specific to the targeted region and to the released drug. Gadolinium-enhanced MRI imaging suggested an intact blood-brain barrier. Blood draws showed normal clinical chemistry and hematology. In summary, this study provides a safe and effective approach to release drugs on demand in selected deep brain regions at levels sufficient to modulate behavior.

The Moralizing Effect: self-directed emotions and their impact on culpability attributions.

Introduction: A general trend in the psychological literature suggests that guilt contributes to morality more than shame does. Unlike shame-prone individuals, guilt-prone individuals internalize the causality of negative events, attribute responsibility in the first person, and engage in responsible behavior. However, it is not known how guilt- and shame-proneness interact with the attribution of responsibility to others. Methods: In two Web-based experiments, participants reported their attributions of moral culpability (i.e., responsibility, causality, punishment and decision-making) about morally ambiguous acts of killing in different conditions. In Study 1 the vignettes were presented in the 1st person, while in Study 2 in the 3rd person. To test proneness to guilt and shame, we utilized the GASP scale, which differentiates between the affective and behavioral components of each emotion. Statistical analyses were performed in Matlab and R. Results: We found that guilt- and shame-proneness were associated with the severity of attributions in both the first and the third person, but the effect was strong only in the guilt case (both subtypes) and shame-affect case, and not in the shame-behavior case. We call this the Moralizing Effect. Discussion: We wonder whether our finding that guilt-prone people tend to attribute a higher degree of culpability to others is really consistent with the view that guilt motivates people to choose the "moral paths in life". This echoes views about the harmful aspects of guilt, which have been expressed historically in philosophy, for example, by Friedrich Nietzsche.

Zero-shot visual reasoning through probabilistic analogical mapping.

Human reasoning is grounded in an ability to identify highly abstract commonalities governing superficially dissimilar visual inputs. Recent efforts to develop algorithms with this capacity have largely focused on approaches that require extensive direct training on visual reasoning tasks, and yield limited generalization to problems with novel content. In contrast, a long tradition of research in cognitive science has focused on elucidating the computational principles underlying human analogical reasoning; however, this work has generally relied on manually constructed representations. Here we present visiPAM (visual Probabilistic Analogical Mapping), a model of visual reasoning that synthesizes these two approaches. VisiPAM employs learned representations derived directly from naturalistic visual inputs, coupled with a similarity-based mapping operation derived from cognitive theories of human reasoning. We show that without any direct training, visiPAM outperforms a state-of-the-art deep learning model on an analogical mapping task. In addition, visiPAM closely matches the pattern of human performance on a novel task involving mapping of 3D objects across disparate categories.

Correlation of Ultraviolet Radiation Levels With the Incidence of Cutaneous Squamous Cell and Merkel Cell Carcinomas in Non-sunbelt Locales in the United States: 2010-2017.

BACKGROUND: Non-melanoma skin cancer (NMSC) is the most common human malignancy worldwide, with increasing incidence in the United States (US). Recent environmental data have shown that ultraviolet radiation (UVR) levels have increased in the US, particularly in the higher latitudes, but the potential impact of this on NMSC incidence is not well known, despite estimates that 90% of NMSC is due to sun exposure. Our exploratory study synthesizes environmental data with demographic and clinical data to determine whether UV indices (UVIs) and non-sunbelt (non-SB) locale (latitudes >40 degrees, which comprises most of the US) might contribute to incidence rates of two types of NMSC: cutaneous squamous cell and Merkel cell carcinomas. METHODS: UVIs from 2010 to 2017 were obtained from the National Oceanic and Atmospheric Administration database and meshed with corresponding locales in the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) database (version 8.4.0.1). Four SB and five NSB locales contained sufficient data for analysis. Linear mixed modeling was performed with the outcome variable of the age-adjusted incidence of NMSC cancer (comprised of cutaneous squamous cell carcinoma of the head and neck (CSCCHN) and Merkel cell carcinoma (MCC)), the two most common types of NMSC contained within SEER). Non-SB locale and percent of days with UVI >3 were independent variables. RESULTS: Percent of days with UVI >3 increased during this period, as did the overall NMSC (combined CSCCHN and MCC) skin cancer incidence, though MCC incidence alone did not increase during our study period. Environmental factors that significantly contributed to the age-adjusted overall NMSC (combined CSCCHN and MCC) cancer incidence (per 100,000 individuals) included NSB locale (b=1.227, p=0.0019) and percent of days with UVIs >3 (b=0.028, p<0.0001), as well as clinical factors of percent white race and percent male, by linear mixed modeling. CONCLUSIONS: Our results are limited by the completeness of the NOAA and SEER databases, and do not include basal cell carcinoma. Nevertheless, our data demonstrate that environmental factors, such as latitude in NSB locale and UVI indices, can affect the age-adjusted overall NMSC (defined as CSCCHN and MCC in this study) incidence even in this relatively short period of time. Prospective studies over longer time periods are needed to identify the extent to which these findings are clinically significant so that increased educational efforts to promote sun-safe behaviors can be maximally effective.

Emergent analogical reasoning in large language models.

The recent advent of large language models has reinvigorated debate over whether human cognitive capacities might emerge in such generic models given sufficient training data. Of particular interest is the ability of these models to reason about novel problems zero-shot, without any direct training. In human cognition, this capacity is closely tied to an ability to reason by analogy. Here we performed a direct comparison between human reasoners and a large language model (the text-davinci-003 variant of Generative Pre-trained Transformer (GPT)-3) on a range of analogical tasks, including a non-visual matrix reasoning task based on the rule structure of Raven's Standard Progressive Matrices. We found that GPT-3 displayed a surprisingly strong capacity for abstract pattern induction, matching or even surpassing human capabilities in most settings; preliminary tests of GPT-4 indicated even better performance. Our results indicate that large language models such as GPT-3 have acquired an emergent ability to find zero-shot solutions to a broad range of analogy problems.

Natural statistics support a rational account of confidence biases.

Previous work has sought to understand decision confidence as a prediction of the probability that a decision will be correct, leading to debate over whether these predictions are optimal, and whether they rely on the same decision variable as decisions themselves. This work has generally relied on idealized, low-dimensional models, necessitating strong assumptions about the representations over which confidence is computed. To address this, we used deep neural networks to develop a model of decision confidence that operates directly over high-dimensional, naturalistic stimuli. The model accounts for a number of puzzling dissociations between decisions and confidence, reveals a rational explanation of these dissociations in terms of optimization for the statistics of sensory inputs, and makes the surprising prediction that, despite these dissociations, decisions and confidence depend on a common decision variable.

Remote targeted electrical stimulation.

Objective:The ability to generate electric fields in specific targets remotely would transform manipulations of processes that rest on electrical signaling.Approach:This article shows that focal electric fields are generated from distance by combining two orthogonal, remotely applied energies-magnetic and focused ultrasonic fields. The effect derives from the Lorentz force equation applied to magnetic and ultrasonic fields.Main results:We elicited this effect using standard hardware and confirmed that the generated electric fields align with the Lorentz equation. The effect significantly and safely modulated human peripheral nerves and deep brain regions of non-human primates.Significance:This approach opens a new set of applications in which electric fields are generated at high spatiotemporal resolution within intact biological tissues or materials, thus circumventing the limitations of traditional electrode-based procedures.

Sustained modulation of primate deep brain circuits with focused ultrasonic waves.

BACKGROUND: Transcranial focused ultrasound has the potential to noninvasively modulate deep brain circuits and impart sustained, neuroplastic effects. OBJECTIVE: Bring the approach closer to translations by demonstrating sustained modulation of deep brain circuits and choice behavior in task-performing non-human primates. METHODS: Low-intensity transcranial ultrasound of 30 s in duration was delivered in a controlled manner into deep brain targets (left or right lateral geniculate nucleus; LGN) of non-human primates while the subjects decided whether a left or a right visual target appeared first. While the animals performed the task, we recorded intracranial EEG from occipital screws. The ultrasound was delivered into the deep brain targets daily for a period of more than 6 months. RESULTS: The brief stimulation induced effects on choice behavior that persisted up to 15 minutes and were specific to the sonicated target. Stimulation of the left/right LGN increased the proportion of rightward/leftward choices. These effects were accompanied by an increase in gamma activity over visual cortex. The contralateral effect on choice behavior and the increase in gamma, compared to sham stimulation, suggest that the stimulation excited the target neural circuits. There were no detrimental effects on the animals' discrimination performance over the months-long course of the stimulation. CONCLUSION: This study demonstrates that brief, 30-s ultrasonic stimulation induces neuroplastic effects specifically in the target deep brain circuits, and that the stimulation can be applied daily without detrimental effects. These findings encourage repeated applications of transcranial ultrasound to malfunctioning deep brain circuits in humans with the goal of providing a durable therapeutic reset.

Remus: System for remote deep brain interventions.

Transcranial-focused ultrasound brings personalized medicine to the human brain. Ultrasound can modulate neural activity or release drugs in specific neural circuits but this personalized approach requires a system that delivers ultrasound into specified targets flexibly and on command. We developed a remote ultrasound system (Remus) that programmatically targets deep brain regions with high spatiotemporal precision and in a multi-focal manner. We validated these functions by modulating two deep brain nuclei-the left and right lateral geniculate nucleus-in a task-performing nonhuman primate. This flexible system will enable researchers and clinicians to diagnose and treat specific deep brain circuits in a noninvasive yet targeted manner, thus embodying the promise of personalized treatments of brain disorders.

Virtual Town Hall Meetings to Convey Emergency Medicine Residency Program Information to Students.

BACKGROUND: Applying to emergency medicine (EM) residency programs as a medical student is challenging and complicated in a normal year, but the 2020/2021 application cycle was further complicated by the COVID-19 pandemic. Due to the decrease of in-person opportunities for students to connect with residency programs, virtual "town-hall" meetings were developed. In this study our primary objective was to determine whether attendance at a virtual residency program information session improved the perceived knowledge of curriculum information and program exposure to medical students applying to an EM residency. METHODS: Four study sites hosted a total of 12 virtual events consisting of residents, faculty, or both. Standardized pre-event/post-event surveys were conducted to capture medical student perceptions before/after each of the virtual sessions. Apart from measuring the improvement in students' perceived knowledge of a program by gauging their responses to each question, we used a 10-question composite score to compare pre- vs post-event improvement among the participants. RESULTS: The pre-event survey was completed by 195 attendees, and the post-event survey was completed by 123 attendees. The median and mean composite score to this 10-question survey improved from 32.19 to 45, and 31.45 to 44.2, respectively, in the pre- to post-event survey. CONCLUSION: This study showed improvement of medical students' perceived knowledge of residency programs (reflected as increased agreement from pre- to post-event survey). The data demonstrates through question responses that students not only obtained information about the programs but also were able to gain exposure to the culture and "feel" of a program. In a non-traditional application season in which students are unable to pursue their interest in a program through audition rotations, virtual town hall events, along with other asynchronous events, may be a reasonable approach to increasing medical student understanding and awareness of a program and its culture.

Comparison between MR and CT imaging used to correct for skull-induced phase aberrations during transcranial focused ultrasound.

Transcranial focused ultrasound with the InSightec Exablate system uses thermal ablation for the treatment of movement and mood disorders and blood brain barrier disruption for tumor therapy. The system uses computed tomography (CT) images to calculate phase corrections that account for aberrations caused by the human skull. This work investigates whether magnetic resonance (MR) images can be used as an alternative to CT images to calculate phase corrections. Phase corrections were calculated using the gold standard hydrophone method and the standard of care InSightec ray tracing method. MR binary image mask, MR-simulated-CT (MRsimCT), and CT images of three ex vivo human skulls were supplied as inputs to the InSightec ray tracing method. The degassed ex vivo human skulls were sonicated with a 670 kHz hemispherical phased array transducer (InSightec Exablate 4000). 3D raster scans of the beam profiles were acquired using a hydrophone mounted on a 3-axis positioner system. Focal spots were evaluated using six metrics: pressure at the target, peak pressure, intensity at the target, peak intensity, positioning error, and focal spot volume. Targets at the geometric focus and 5 mm lateral to the geometric focus were investigated. There was no statistical difference between any of the metrics at either target using either MRsimCT or CT for phase aberration correction. As opposed to the MRsimCT, the use of CT images for aberration correction requires registration to the treatment day MR images; CT misregistration within a range of ± 2 degrees of rotation error along three dimensions was shown to reduce focal spot intensity by up to 9.4%. MRsimCT images used for phase aberration correction for the skull produce similar results as CT-based correction, while avoiding both CT to MR registration errors and unnecessary patient exposure to ionizing radiation.

Improving Transcranial Acoustic Targeting: The Limits of CT-Based Velocity Estimates and the Role of MR.

Transcranial magnetic resonance-guided focused ultrasound (tcMRgFUS) enables the noninvasive treatment of the deep brain. This capacity relies on the ability to focus acoustic energy through the in-tact skull, a feat that requires accurate estimates of the acoustic velocity in individual patient skulls. In current practice, these estimates are generated using a pretreatment computed tomography (CT) scan and then registered to a magnetic resonance (MR) dataset on the day of the treatment. Treatment safety and efficacy can be improved by eliminating the need to register the CT data to the MR images and by improving the accuracy of acoustic velocity measurements. In this study, we examine the capacity of MR to supplement or replace CT as a means of estimating velocity in the skull. We find that MR can predict velocity with less but comparable accuracy to CT. We then use micro-CT imaging to better understand the limitations of Hounsfield unit (HU)-based estimates of velocity, demonstrating that the macrostructure of pores in the skull contributes to the acoustic velocity of the bone. We find evidence that detailed T2 measurements provide information about pore macrostructure similar to the information obtained with micro-CT, offering a potential clinical mechanism for improving patient-specific estimates of acoustic velocity in the human skull.

Acoustic properties across the human skull.

Transcranial ultrasound is emerging as a noninvasive tool for targeted treatments of brain disorders. Transcranial ultrasound has been used for remotely mediated surgeries, transient opening of the blood-brain barrier, local drug delivery, and neuromodulation. However, all applications have been limited by the severe attenuation and phase distortion of ultrasound by the skull. Here, we characterized the dependence of the aberrations on specific anatomical segments of the skull. In particular, we measured ultrasound propagation properties throughout the perimeter of intact human skulls at 500 kHz. We found that the parietal bone provides substantially higher transmission (average pressure transmission 31 ± 7%) and smaller phase distortion (242 ± 44 degrees) than frontal (13 ± 2%, 425 ± 47 degrees) and occipital bone regions (16 ± 4%, 416 ± 35 degrees). In addition, we found that across skull regions, transmission strongly anti-correlated (R=-0.79) and phase distortion correlated (R=0.85) with skull thickness. This information guides the design, positioning, and skull correction functionality of next-generation devices for effective, safe, and reproducible transcranial focused ultrasound therapies.

The Ethnic Identity Scale: Affirmation, really?

OBJECTIVE: The Ethnic Identity Scale (EIS) was developed to distinguish between process and content components of ethnic-racial identity (ERI). However, the affirmation subscale is composed entirely of negatively worded items, measuring negative feelings about one's ethnic-racial group, rather than positive feelings as widely conceptualized. Addressing this gap, the present study examined the psychometric validity of a revised EIS with positively and negatively worded items to determine whether affirmation is best represented as a unidimensional construct, a bidimensional construct, or a combination of the two. METHOD: The sample consisted of 280 college students (75.5% female; Mage = 20.95 years; SD = 1.98 years). The largest ethnic-racial group consisted of Black or African Americans (68.2%), followed by Asian/Asian Americans (12.1%), Hispanic/Latinos (9.6%), and other ethnic-racial groups (10%). RESULTS: Confirmatory factor analysis provided evidence for both unidimensionality and multidimensionality. Indeed, although positively worded and negatively worded items of "affirmation" loaded onto a general factor representing affirmation, there was still a significant amount of variance captured by the negative ERI affect specific factor, indicating the presence of multidimensionality. In addition, results indicated that negative ERI affect, over and above the general ERI affirmation factor, predicted psychosocial functioning. CONCLUSIONS: The present study expands our understanding of the multidimensionality of ERI, highlighting the need for examination of how we measure ERI affect at the very least, and possibly how we conceptualize it within the broader ERI literature. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

Transcranial focused ultrasound phase correction using the hybrid angular spectrum method.

The InSightec Exablate system is the standard of care used for transcranial focused ultrasound ablation treatments in the United States. The system calculates phase corrections that account for aberrations caused by the human skull. This work investigates whether skull aberration correction can be improved by comparing the standard of care InSightec ray tracing method with the hybrid angular spectrum (HAS) method and the gold standard hydrophone method. Three degassed ex vivo human skulls were sonicated with a 670 kHz hemispherical phased array transducer (InSightec Exablate 4000). Phase corrections were calculated using four different methods (straight ray tracing, InSightec ray tracing, HAS, and hydrophone) and were used to drive the transducer. 3D raster scans of the beam profiles were acquired using a hydrophone mounted on a 3-axis positioner system. Focal spots were evaluated using six metrics: pressure at the target, peak pressure, intensity at the target, peak intensity, positioning error, and focal spot volume. For three skulls, the InSightec ray tracing method achieved 52 ± 21% normalized target intensity (normalized to hydrophone), 76 ± 17% normalized peak intensity, and 0.72 ± 0.47 mm positioning error. The HAS method achieved 74 ± 9% normalized target intensity, 81 ± 9% normalized peak intensity, and 0.35 ± 0.09 mm positioning error. The InSightec-to-HAS improvement in focal spot targeting provides promise in improving treatment outcomes. These improvements to skull aberration correction are also highly relevant for the applications of focused ultrasound neuromodulation and blood brain barrier opening, which are currently being translated for human use.

Creating an Automated Health Attestation System During the COVID-19 Global Pandemic Using Google's G Suite.

As of October 2020, the COVID-19 global pandemic has infected over 40,000,000 people and has claimed over 1,000,000 lives globally (Johns Hopkins University, 2020). To mitigate the spread of the virus and their own liability, organizations have adopted multiple strategies to protect their employees and consumers. In addition to mask wearing, social distancing, and contact tracing, health attestations are being adopted by organizations that depend on physical contact between employees and consumers. The purpose of this tutorial is to describe how an automated health attestation and notification system can be created using Google's G Suite platform at little to no cost to the organization. When combined with other mitigation strategies and strong organizational policies, health attestations may be an effective component for organizations to include in their response to the global pandemic. The benefits and limitations of including health attestations as a component within an organization's COVID-19 policies are discussed.

Acoustic Attenuation: Multifrequency Measurement and Relationship to CT and MR Imaging.

Transcranial magnetic resonance-guided focused ultrasound (tcMRgFUS) is gaining significant acceptance as a noninvasive treatment for motion disorders and shows promise for novel applications such as blood-brain barrier opening for tumor treatment. A typical procedure relies on CT-derived acoustic property maps to simulate the transfer of ultrasound through the skull. Accurate estimates of the acoustic attenuation in the skull are essential to accurate simulations, but there is no consensus about how attenuation should be estimated from CT images and there is interest in exploring MR as a predictor of attenuation in the skull. In this study, we measure the acoustic attenuation at 0.5, 1, and 2.25 MHz in 89 samples taken from two ex vivo human skulls. CT scans acquired with a variety of X-ray energies, reconstruction kernels, and reconstruction algorithms, and MR images acquired with ultrashort and zero echo time sequences are used to estimate the average Hounsfield unit value, MR magnitude, and T2* value in each sample. The measurements are used to develop a model of attenuation as a function of frequency and each individual imaging parameter.