Attitudes Toward Artificial General Intelligence: Results from American Adults in 2021 and 2023
Author: Jason Jeffrey Jones and Steven Skiena
Date: February, 2024
Area: Social Sciences, Technology
Text: PDF, Substack
A compact, inexpensive repeated survey on American adults’ attitudes toward Artificial General Intelligence (AGI) revealed a stable ordering but changing magnitudes of agreement toward three statements. Contrasting 2021 to 2023, American adults increasingly agreed AGI was possible to build. Respondents agreed more weakly that AGI should be built. Finally, American adults mostly disagree that an AGI should have the same rights as a human being; disagreeing more strongly in 2023 than in 2021.
The Economics of Time Travel
Author: Stuart Mills
Date: December, 2023
Area: Social Sciences
Text: PDF, Substack
The lack of time travellers visiting us may be seen as evidence that time travel is not possible. In this article, I argue an alternative explanation is that we are not economically important enough to our descendants to justify the costs of time travel. Using a cost-benefit analysis, I elaborate on this argument. I suggest that the major cost of time travel is likely to be the energy cost, whilst the largest benefit of time travel is knowledge which the present possesses, but the future has lost. Focusing on this benefit, I argue it is extremely unlikely that we possess a piece of knowledge which is sufficiently important to a future civilisation (system critical), but also has been lost by said civilisation. This is to say, we may not have been visited by time travellers because we are not important enough.
Scientific Theories and Their Psychological Corollaries: The Ecological Crisis as a Case Study in the Need for Synthesis
Author: Arnold Schroder
Date: December, 2023
Area: Metascience, Scientific Ethics, Social Sciences
Text: PDF, Substack
That policy makers will ever rationally respond to scientific warnings about the ecological crisis should be treated as a falsifiable hypothesis. After more than five decades of such warnings, there is a strong case for skepticism. Climate and other ecological tipping points constitute the quantitative thresholds beyond which current political systems can definitively be said to have failed. This presents a mandate to generate broad consensus on where tipping points lie, and at what proximity to them new strategies should be pursued. Central to any new strategy should be an understanding of why the old one failed—an understanding of why those in power almost exclusively derive from academic backgrounds other than physical science, and the psychological differences between those who issued or received so many warnings of collapse. To that end, a psychological trait syndrome relevant to political power is proposed, based on correlations between academic specialization, psychometric results, and the behavior of powerful people across a wide range of societies. This proposed syndrome consists of four covarying dimensions of individual difference. These are perceptions of hierarchy vs. egalitarianism, established knowledge vs. open inquiry, physical vs. symbolic action, and schematic vs. particular knowledge.
Visualizing researchers’ scientific contributions with radar plots
Author: Manh-Toan Ho
Date: December, 2023
Area: Metascience
Text: PDF, Substack
The essay advocates for diverse approaches in presenting a researcher's scientific contributions in a project. Taking inspiration from sports journalism and its visualization of football players' data, the essay suggests that a radar plot, incorporating CRediT contributor role data, enables multiple authors of a scientific paper to illustrate their contributions in a more specific manner. The suggested method, though subject to bias reporting, pays credit to different aspects of a research project, from conceptualization, analysis, administration, to writing and revising. It not only enables both academics and lay readers to better understand the considerable amount of work required in every project but also calls for the need to employ diverse viewpoints in science.
Cilia Disorders in the Genomics Era: Historical Overview and Commentary on Ciliopathy Diagnostics
Author: Sani Eskinazi
Date: October, 2023
Area: Biology
Text: PDF, Substack
Motile and sensory (primary) cilia are organelles that are found on the surface of almost all cells. Defects in cilia cause a number of multi-organ diseases known as ciliopathies, which have clinically heterogeneous symptoms. This heterogeneity makes diagnosing cilia disorders challenging and clinicians often rely on genetic sequencing to delineate ciliopathies from other diseases. However, there is not a consensus on which sequencing tools are most optimal for ciliopathy diagnosis and research. This review demonstrates the increasing body of knowledge on cilia genomics and highlights that next-generation sequencing will be integral towards optimizing diagnostics for these heterogeneous and debilitating group of disorders.
Taxonomies of Intelligence: A Comprehensive Guide to the Universe of Minds
Author: Roman V. Yampolskiy
Date: October, 2023
Areas: Biology, Psychology, Technology
Text: PDF, Substack
This paper explores the landscape of potential mind architectures by initially conceptualizing all minds as software. Through rigorous analysis, we establish intriguing properties of this intellectual space, including its infinite scope, variable dimensions of complexity, and representational intricacies. We then provide an extensive review of existing taxonomies for mind design. Building on this foundation, the paper introduces 'Intellectology' as a new field dedicated to the systematic study of diverse forms of intelligence. A compendium of open research questions aimed at steering future inquiry in this nascent discipline is also presented.
Forager Facts
Author: David Youngberg and Robin Hanson
Date: August, 2023
Areas: Biology, Psychology, Social Sciences
Text: PDF, Substack
Using an anthropology database that details many groups, we summarize how our forager ancestors likely lived on a variety of metrics. Though we have long since ceased to live as hunter-gatherers, its psychological shadow likely still shapes us, and so we would try to understand that lifestyle as fully as possible.
We See The Sacred From Afar, To See It The Same
Author: Robin Hanson
Date: June, 2023
Areas: Psychology, Social Sciences
Text: PDF, Substack
68 reported correlates of treating things as “sacred” are listed, and collected into seven themes. Most can be plausibly explained via two hypotheses. The first, taken from Durkheim, is that treating things as sacred mainly functions to bind groups together via a shared view of it. The second hypothesis, suggested by psychology’s construal level theory, is that humans acquired a habit of seeing sacred things as if from afar, even when they are close, to more consistently see those things the same as others in their groups.
Principles of Categorization: A Synthesis
Author: Davood Gozli
Date: June, 2023
Areas: Metascience, Psychology, Social Sciences
Text: PDF
The present article explores the nature of categorization and its role in shaping our relationship with reality. Drawing on Jens Mammen's distinction between sense categories and choice categories, and Eleanor Rosch's principles of categorization, I examine how our attitudes and modes of engagement with categories can reveal important insights relevant not only to psychology but other scientific fields as well. Furthermore, I argue that the connection between sense and choice categories can be traced by examining atypical instances and non-basic-level categories, which highlight the role of subjects embedded in particular situations. In general, categorization is an active process, influenced by our interests and commitments, even though it does not always appear as such. By correcting biases in our treatment of concepts and categories, we can ultimately correct our biases in scientific practices, thus revealing the entanglement of categorization with broader epistemological issues.
Perspective: Focused-Ultrasound Guided Neuropeptide Delivery as a Novel Therapeutic Approach in Psychiatry
Authors: Manjushri Karthikeyan, Ahaana Shrivastava, Andrew Neff
Date: April, 2023
Areas: Biology, Psychology
Text: PDF, Substack
Although drugs are a critical component of mental healthcare, most have modest benefits and significant side effects. One way to develop a superior intervention would be to administer drugs with the spatial and temporal precision that better replicates natural diversity within neurotransmitter systems. A technology called focused-ultrasound (FU) may be able to safely and transiently disrupt the blood-brain barrier with spatial precision, permitting the site-specific delivery of molecules that do not conventionally cross the blood-brain barrier. If this method is proven to be safe and effective in larger human trials, it may trigger a paradigm shift in biopsychology research where the level of precision with which neurotransmitter systems can be influenced is massively increased. In this article, we use the example of oxytocin in the treatment of Autism. We propose that intranasal administration is not highly effective because it leads to oxytocin’s wide dispersion throughout the brain, failing to specifically stimulate oxytocin’s prosocial effects in specific regions. Consequently, we hypothesize that site-specific delivery of oxytocin, particularly in brain regions such as the Nucleus Accumbens and Ventral Tegmental Area, would lead to more consistent benefits.
How to Escape From the Simulation
Author: Roman Yampolskiy
Date: March, 2023
Areas: Physical Sciences, Scientific Ethics, Technology
Text: PDF, Substack
Many researchers have conjectured that humankind is simulated along with the rest of the physical universe – a Simulation Hypothesis. In this paper, we do not evaluate evidence for or against such a claim, but instead ask a computer science question, namely: Can we hack the simulation? More formally the question could be phrased as: Could generally intelligent agents placed in virtual environments find a way to jailbreak out of them? Given that the state-of-the-art literature on AI containment answers in the affirmative (AI is uncontainable in the long-term), we conclude that it should be possible to escape from the simulation, at least with the help of superintelligent AI. By contraposition, if escape from the simulation is not possible, containment of AI should be. Finally, the paper surveys and proposes ideas for hacking the simulation and analyzes ethical and philosophical issues of such an undertaking.
Is a Qualitative Metric of Falsifiability Possible?
Author: Dan James
Date: March, 2023
Area: Metascience
Text: PDF, Substack
There is an ever-increasing number of quantitative metrics, most of which are intended to act as proxies of quality for either authors or journals in current scholarly publishing. In contrast, this paper presents a more directly qualitative paper-level metric that adds a falsifiability dimension to the existing methods used to assess scholarly research. This new metric, the "F-index", is derived from a "Falsifiability Statement" (FS) (examples of both are applied self-referentially in Annex A). An FS is a discrete metalevel statement provided by the author(s) outlining how their research or assumptions can be foreseeably falsified, and the F-index is a numerical estimate of how clear and practical the steps are to falsify the research or stated assumptions as outlined in the FS. Though the F-index is particularly suited to hypothesis or theory-driven fields, it is also relevant to any empirical inquiry that relies on propositions or assumptions that can be potentially falsified. An F-index is qualitative in that a high F-index number provides a good indication of how novel or original a paper is. Four candidate mechanisms for obtaining an F-index from a Falsifiability Statement are evaluated: a peer reviewer assessed metric, an author or self-reporting metric, a propositional density metric, and an NLP derived metric. This evaluation concludes that a FS is currently a practical proposition, and that the derivation of a meaningful F-Index is an achievable goal.
The Muscle Readers, a Historical Sketch
Author: Leverage Research
Date: February, 2023
Area: Psychology
Text: PDF, Substack
The notion that subtle nonverbal cues play an important role in social interaction is relatively uncontroversial. What the upper bounds of this capacity might be, however—how much and what kind of information can be conveyed through these channels—remains unclear and, at present, under-explored. In the present work, we consider possible answers to the question and ways in which it could be addressed by considering an historical line of investigation known as muscle reading. Spurred by public interest in mentalism and the specific popularity of thought-readers, researchers in the late 19th and early 20th centuries began investigating the possibility that information about our thoughts and inclinations could be “read” from muscle tension, unconscious vocalizations, and other subtle cues. While covering some of the same ground as contemporary research on nonverbal communication, the literature of this era contains many reports that go well beyond this. Feats such as locating a hidden object, guessing the suit of a card, and determining words or names held in another’s mind were said to be achieved in controlled conditions or by academic researchers themselves. In some cases, subtle but telling movements were also said to be captured by early biometric apparatuses. While we believe that such claims should be interpreted with caution, we contend that the reports of these early researchers should not be dismissed merely because of their age and that a better understanding of this literature offers important leads for investigators today.
Why Proposal Review Should Be More Like Meteorology
Author: Stuart Buck
Date: January, 2023
Area: Metascience
Text: PDF, Substack
The process of evaluating research proposals for funding is often based on subjective assessments of the "goodness" or "badness" of a proposal. However, this method of evaluation is not precise and does not provide a common language for reviewers to communicate with each other. In this paper, we propose that science funding agencies ask reviewers to assign quantitative probabilities to the likelihood of a proposal reaching a particular milestone or achieving technical goals. This approach would encourage reviewers to be more precise in their evaluations and could improve both agency-wide and individual reviewer calibration over time. Additionally, this method would allow funding agencies to identify skilled reviewers and allow reviewers to improve their own performance through consistent feedback. While this method may not be suitable for all types of research, it has the potential to enhance proposal review in a variety of fields. [abstract generated by ChatGPT]
Will general antiviral protocols always be science fiction?
Author: Rick Sheridan
Date: December, 2022
Area: Biology
Text: PDF, Substack
Currently, antiviral drugs are targeted at specific viruses, requiring extensive research to develop de novo before a trial molecule is approved for a single viral target. However, it is possible that a strategic asset for mainstreaming antivirals has been hiding in plain sight. In our recent article published in Frontiers in Pharmacology, we explored the scope for plant-derived polyphenols, such as flavonoids, to be applied against infections spanning phylogenetically unrelated virus families. Beyond a long history of safe use in ordinary diet, common polyphenols also feature promiscuous binding behavior, a quality echoed by laboratory evidence that many are effective in vitro against viruses from diverse genera. Moreover, we mapped the research base undergirding the inflammatory response’s tendency to selectively return serum polyphenols from their latent, metabolized form back into their active, promiscuously-binding form uniquely at sites of inflammation -- such as those routinely established by viral infections. Verifying efficacy of flavonoids to constitute an annually-updated general antiviral protocol will require mobilizing a spectrum of animal model studies and clinical trials for initial confirmation.
Notes on the Inexact Sciences
Author: Suspended Reason
Date: November, 2022
Areas: Metascience, Psychology, Social Sciences
Text: PDF, Substack
Popular wisdom warns us against premature optimization. And yet, in a quest for public legitimacy and tidy problem domains, many fields discourage vitally necessary descriptive and conceptual work in favor of statistical analysis and laboratory experiments. Topics of unprecedented complexity are tackled using rote, mechanical approaches, by researchers who routinely fail to realize how much linguistic and conceptual clarification is a precondition of headway. Meanwhile, sociological and professional incentives prevent the sorts of synthetic work that might de-provincialize researchers' theories, and initiate exactly those conceptual refactorings which would advance the discipline.
The Rise and Fall of the Dot-Probe Task: Opportunities for Metascientific Learning
Authors: Benjamin T. Sharpe, Monika Halls, Thomas E. Gladwin
Date: November, 2022
Area: Psychology
Text: PDF, Substack
Much of the extensive literature on spatial attentional bias is built on measurements using the dot-probe task. In recent years, concerns have been raised about the psychometric properties of bias scores derived from this task. The goal of the current paper is to look ahead and evaluate possible responses of the field to this situation from a metascientific perspective. Therefore, educated guesses are made on foreseeable but preventable future (repeats of) errors. We discuss, first, the issue of overreactions to the disappointing findings, especially in the context of the potential of a new generation of promising variations on the traditional dot-probe task; second, concerns with competition between tasks; and third, the misuse of rationales to direct research efforts. Alternative directions are suggested that may be more productive. We argue that more adequately exploring and testing methods and adjusting scientific strategies will be critical to avoiding suboptimal research and potentially failing to learn from mistakes. The current articulation of arguments and concerns may therefore be of use in discussions arising around future behavioural research into spatial attentional bias and more broadly in psychological science.
What are the Red Flags for Neural Network Suffering?
Authors: Marius Hobbhahn, Jan Kirchner
Date: September, 2022
Areas: Biology, Scientific Ethics, Technology
Text: PDF, Substack
Which kind of evidence would we need to see to believe that artificial neural networks can suffer? We review neuroscience literature, investigate behavioral arguments and propose high-level considerations that could shift our beliefs. Of these three approaches, we believe that high-level considerations, i.e. understanding under which circumstances suffering arises as an optimal training strategy, is the most promising. Our main finding, however, is that the understanding of artificial suffering is very limited and should likely get more attention.
Moral Weights of Animals, Considering Viewpoint Uncertainty
Authors: Richard Bruns, Jim Davies
Date: July, 2022
Areas: Biology, Scientific Ethics
Text: PDF, Substack
Many utilitarians would like a number to use to evaluate the moral impact of actions that affect animals. However, there is a great disagreement among scholars involved with animal ethics, both about how much different animals can suffer and how much that suffering morally matters. To illustrate this uncertainty, while showing as a proof of concept that it may be possible to produce useful estimates in spite of it, we ran a Monte Carlo simulation that samples the ranges of major viewpoints scholars hold in the field, to show a spread of uncertainty for how we should treat six representative animals: crickets, salmon, chickens, pigs, cows, and elephants. The results show that the uncertainty is very large, with a 90% confidence interval ranging between an animal having no value and being valued as much as a human being. More research, in the form of expert surveys and a thorough and rigorous literature review, would be required to produce better estimates, but as an illustration, we present 20% and 40% confidence intervals, as well as the median and geometric mean, based on weighting the theories according to our informal estimate of their prevalence in the literature.
What does it mean to represent? Mental representations as falsifiable memory patterns
Authors: Eloy Parra-Barrero, Yulia Sandamirskaya
Date: May, 2022
Areas: Biology, Psychology
Text: PDF
Representation is a key notion in neuroscience and artificial intelligence (AI). However, a longstanding philosophical debate highlights that specifying what counts as representation is trickier than it seems. With this brief opinion paper we would like to bring the philosophical problem of representation into attention and provide an implementable solution. We note that causal and teleological approaches often assumed by neuroscientists and engineers fail to provide a satisfactory account of representation. We sketch an alternative according to which representations correspond to inferred latent structures in the world, identified on the basis of conditional patterns of activation. These structures are assumed to have certain properties objectively, which allows for planning, prediction, and detection of unexpected events. We illustrate our proposal with the simulation of a simple neural network model. We believe this stronger notion of representation could inform future research in neuroscience and AI.
The Cult Deficit: Analysis and Speculation
Author: Roger’s Bacon
Date: May, 2022
Areas: Psychology, Science Education, Social Sciences, Technology
Text: PDF, Substack
Using a dataset derived from the long-running “Cults” podcast by Parcast, I find that the number of new cults began to increase in the 50s, peaked in the 70s-80s, and has been in steady decline in recent decades. I discuss various factors (historical, technological, cultural, pharmacological) that may have played a role in the rise and fall of cults since the 1950s and speculate on what the future may hold.
Market Failures in Science
Author: Milan Cvitkovic
Date: April, 2022
Area: Metascience
Text: PDF, Substack
(No abstract — and that’s okay)
Building a Brain: An Introduction to Narrative Complexity, a language & internal dialogue-based theory of human consciousness
Author: R. Salvador Reyes
Date: April, 2022
Areas: Biology, Psychology
Text: PDF, Substack
Narrative Complexity is an internal dialogue-based looping model of consciousness & behavior—one that provides a framework for effectively defining the specific (and central) role of language and inner speech within consciousness & cognition, and in turn, within the mechanisms of emotion & decision-making that consciousness & cognition help to govern. This introduction to the theory lays out some of its primary premises and outlines the model enough to demonstrate that it is novel, plausible, systematically comprehensive, and worth exploring in greater detail.
On Scaling Academia
Author: Jan Hendrik Kirchner
Date: April, 2022
Area: Metascience
Text: PDF
Overcoming humanity's challenges will require a deep understanding of both the problem and the possible solutions. There are early indications that the scientific apparatus, which has traditionally been the primary tool for gaining deep understanding, might not be able to keep pace. In this essay, I outline a set of interventions that might help the scientific apparatus overcome existing bottlenecks, and I discuss limitations and possible implications. Centrally, I argue for systematization and automation of the research process to allow researchers to benefit from emerging technology like artificial intelligence.
The Prospect of Extracting Brain-Region-Specific Exosomes in the Human Bloodstream
Authors: Evan Yang, Andrew Neff
Date: November, 2021
Area: Biology
Text: PDF, Substack
“Exosomes are small vesicles, secreted by eukaryotic cells, containing molecular cargo that reflects the biochemical composition of the origin cell, including protein and RNA. Once secreted, exosomes can enter the circulatory system and be found in blood, urine, and saliva. It has been hypothesized that because exosomes contain transmembrane proteins unique to their cell of origin, specific populations of exosomes could be non-invasively extracted from the bloodstream. The protein L1CAM may serve as a marker of neuronal exosomes. However, although “neuron-derived-exosomes'' could offer some specific information about in-vivo molecular neurobiology, this population of exosomes still provides a relatively noisy signal, including data on protein expression from a variety of different neuronal subpopulations. We argue that it may be possible to isolate brain-region-specific exosomes, and that data derived from these exosomes would provide a superior diagnostic tool.”
Randomness in Science
Authors: Roger’s Bacon, Sergey Samsonau, Dario Krpan (example article)
Date: May, 2021
Areas: Metascience, Psychology
Text: PDF, Substack
“Could we improve science by exploring new ways to inject randomness into the research process?"
Stories as Technology: Past, Present, and Future
Authors: Roger’s Bacon, Sergey Samsonau, Dario Krpan (example article)
Date: May, 2021
Areas: Psychology, Technology
Text: PDF
“What is it about a good story that causes it to have life-changing effects on one person and not another? I wonder if future technologies will enable us to develop the type of truly deep and fine-grained understanding of stories as social, cognitive, and emotional technologies that might allow us to answer this question with a high-level of precision.”