Falsifiable Predictions Research Articles

Registering Theory-Based Predictions in Political Science

ABSTRACT How can political scientists rigorously evaluate the predictive power of theories? Many peer-reviewed political science articles include predictions about future outcomes, and scholars make predictions on social media and other public forums. The prevalence of predictions suggests that scholars recognize the utility of leveraging theories for this purpose; however, the predictions often are not made in a manner that allows for rigorously evaluating their accuracy. Building on the increasing popularity of study preregistration in the social sciences, this article proposes “prediction registration” as a means for scholars to publish falsifiable, systematic, and verifiable theory-based predictions. Increasing the rigor of predictive theory testing can advance often-circular debates about accuracy and presents a “win-win” for scholars who aim to test the predictive power of theories. With a more rigorous approach, correct predictions would better demonstrate a theory’s ability to forecast outcomes, and missed predictions would reveal information that can be used to calibrate the theory.

Conjecture : The Theory of Everything is Embodied by Fundamental Replicators (Femes)

This paper conjectures that fundamental reality, taken to be an interacting system composed of discrete information, embodies replicating information structures called femes. These femes are posited to interact to cause the laws of physics. We therefore extend Universal Darwinism to propose the existence of four abstract replicators: femes, genes, memes, and temes. We firstly consider the problem of fine-tuning, and problems with current solutions. A detailed background section outlines key principles from physics, computation, evolutionary theory, and constructor theory. The conjecture is then provided in detail, along with five falsifiable predictions.

Modeling the consequences of an L1 grammar for L2 production: simulations, variation, and predictions

IntroductionThis paper presents a constraint-based grammar of Mandarin low vowel + nasal coda (loVN) sequences first as acquired by L1 learners, and then as transferred to L2 English.MethodsWe simulate phonological learning in Harmonic Grammar using a gradual, error-driven GLA learner, drawing on evidence from L1 Mandarin speakers' perceptual data to support our initial state assumptions. We then compare our simulation results with L2 English production (both anecdotal and ultrasound data), as well as evidence from Mandarin loanword phonology.ResultsOur results align with multiple patterns in the previous empirical literature, including an asymmetry among surface repairs for VN sequences, and we show how these emerge from our assumptions about both the L1 Mandarin grammar and the grammar's evaluation method (i.e., weighted constraints).DiscussionWe discuss the extent to which these results derive from our somewhat novel analysis of place contrasts in L1 Mandarin, and the variability in loVN outputs that we encode directly into the L1 grammar, which are then transferred to the L2 context. Ultimately we discuss how this type of modeling can make falsifiable predictions about phonological development, in both L1 and L2 contexts.

Falsifiable analog model predictions of W mass in CDF II and ATLAS

The CDF II measurement of [Formula: see text] boson mass is at significant ([Formula: see text]) tension with the Standard Model, and moderate ([Formula: see text]) tension with recent results from ATLAS. Any attempt to interpret this as a signature of new physics requires high-precision, robust, falsifiable predictions. The Classical Analogue to the Standard Model In pseudo-Riemannian space–time (CASMIR) is an analogue model predicting [Formula: see text] and [Formula: see text] boson masses of [Formula: see text] and [Formula: see text], respectively. During baryon collisions satisfying [Formula: see text], CASMIR predicts a color-mediated enhancement of [Formula: see text] and [Formula: see text] boson masses, becoming [Formula: see text] and [Formula: see text], respectively. The unenhanced masses are consistent with ATLAS data collected at [Formula: see text], and the enhanced masses are consistent with CDF II data collected at [Formula: see text]. According to CASMIR, operation of the Large Hadron Collider (LHC) at center-of-mass energies small compared with [Formula: see text] but large enough for [Formula: see text] boson formation should permit ATLAS to replicate the result from CDF II.

Virtual brain model‐guided optimization of transcranial direct current stimulation in Alzheimer’s disease

AbstractBackgroundNon‐invasive brain stimulation techniques such as transcranial direct current stimulation (tDCS) can potentially counteract disrupted local and distributed brain network activity in Alzheimer’s disease (AD) with a beneficial effect on cognition. However, results of tDCS studies in AD up till now are inconsistent, probably partly due to different methodological choices such as electrode placement. We therefore explored tDCS optimization using a virtual brain network model of AD. Our aim was to compare a large, representative set of virtual tDCS intervention setups, to identify and potentially explain the optimal theoretical tDCS electrode positions for reducing or restoring functional network disruption.MethodWe simulated 20 tDCS setups in a computational macroscopic dynamic network model of 78 neural masses, coupled according to human structural topology. AD network damage was simulated using an activity‐dependent degeneration algorithm, which produces typical gradual oscillatory slowing and loss of connectivity. Current flow modeling was used to estimate tDCS‐targeted cortical regions for different electrode positions, and excitability of the pyramidal neurons of the corresponding neural masses was modulated to simulate tDCS. Outcome measures were relative power spectral density in the lower (8‐10 Hz) and upper (8‐13 Hz) alpha bands, total spectral power, posterior alpha peak frequency, and connectivity measures phase lag index (PLI) and amplitude envelope correlation (AEC).ResultThe performance of different tDCS strategies varied considerably, with best setups temporarily improving all spectral and connectivity outcome measures, while other setups had little to no effect, or even caused network deterioration. The best performing setup involved right parietal anodal stimulation, with a contralateral supraorbital cathode. A clear correlation between the network role of a region (e.g. hub) and its involvement in successful tDCS was not observed.ConclusionModeling tDCS produces specific, falsifiable predictions on how to counter disrupted brain dynamics in AD. In follow‐up studies, our general predictions will be compared to those of personalized virtual models, and then validated with tDCS‐magnetoencephalography (MEG) in a clinical AD patient cohort. This modeling‐informed approach can guide and perhaps accelerate tDCS therapy development, as well as enhance our understanding of tDCS effects.

Detecting cosmic voids via maps of geometric-optics parameters

ABSTRACT Curved space–time geometric-optics maps derived from a deep photometric survey should contain information about the 3D matter distribution and thus about cosmic voids in the survey, despite projection effects. We explore to what degree sky-plane geometric-optics maps can reveal the presence of intrinsic 3D voids. We carry out a cosmological N-body simulation and place it further than a gigaparsec from the observer, at redshift 0.5. We infer 3D void structures using the watershed algorithm. Independently, we calculate a surface overdensity map and maps of weak gravitational lensing and geometric-optics scalars. We propose and implement a heuristic algorithm for detecting (projected) radial void profiles from these maps. We find in our simulation that given the sky-plane centres of the 3D watershed-detected voids, there is significant evidence of finding corresponding void centres in the surface overdensity Σ, the averaged weak-lensing tangential shear $\overline{{\gamma }_{\perp }}$, the Sachs expansion θ, and the Sachs shear modulus |σ|. Recovering the centres of the 3D voids from the sky-plane information alone is significant given the Sachs expansion θ, or the Sachs shear |σ|, mildly significant given the weak-lensing shear $\overline{{\gamma }_{\perp }}$, and not significant for the surface overdensity Σ. Void radii are uncorrelated between 3D and 2D voids; our algorithm is not designed to distinguish voids that are nearly concentric in projection. This investigation shows preliminary evidence encouraging observational studies of gravitational lensing through individual voids, either blind or with spectroscopic/photometric redshifts. The former case – blind searches – should generate falsifiable predictions of intrinsic 3D void centres.

Peripheral and central sensation: multisensory orienting and recognition across species

Attentional bottlenecks force animals to deeply process only a selected fraction of sensory inputs. This motivates a unifying central-peripheral dichotomy (CPD), which separates multisensory processing into functionally defined central and peripheral senses. Peripheral senses (e.g., human audition and peripheral vision) select a fraction of the sensory inputs by orienting animals' attention; central senses (e.g., human foveal vision) allow animals to recognize the selected inputs. Originally used to understand human vision, CPD can be applied to multisensory processes across species. I first describe key characteristics of central and peripheral senses, such as the degree of top-down feedback and density of sensory receptors, and then show CPD as a framework to link ecological, behavioral, neurophysiological, and anatomical data and produce falsifiable predictions.

Model-based experimental manipulation of probabilistic behavior in interpretable behavioral latent variable models.

Interpretable latent variable models that probabilistically link behavioral observations to an underlying latent process have increasingly been used to draw inferences on cognition from observed behavior. The latent process usually connects experimental variables to cognitive computation. While such models provide important insights into the latent processes generating behavior, one important aspect has often been overlooked. They may also be used to generate precise and falsifiable behavioral predictions as a function of the modeled experimental variables. In doing so, they pinpoint how experimental conditions must be designed to elicit desired behavior and generate adaptive experiments. These ideas are exemplified on the process of delay discounting (DD). After inferring DD models from behavior on a typical DD task, the models are leveraged to generate a second adaptive DD task. Experimental trials in this task are designed to elicit 9 graded behavioral discounting probabilities across participants. Models are then validated and contrasted to competing models in the field by assessing the ouf-of-sample prediction error. The proposed framework induces discounting probabilities on nine levels. In contrast to several alternative models, the applied model exhibits high validity as indicated by a comparably low prediction error. We also report evidence for inter-individual differences with respect to the most suitable models underlying behavior. Finally, we outline how to adapt the proposed method to the investigation of other cognitive processes including reinforcement learning. Inducing graded behavioral frequencies with the proposed framework may help to highly resolve the underlying cognitive construct and associated neuronal substrates.

Lifetime of actin-dependent protein nanoclusters

Protein nanoclusters (PNCs) are dynamic collections of a few proteins that spatially organize in nanometer-length clusters. PNCs are one of the principal forms of spatial organization of membrane proteins, and they have been shown or hypothesized to be important in various cellular processes, including cell signaling. PNCs show remarkable diversity in size, shape, and lifetime. In particular, the lifetime of PNCs can vary over a wide range of timescales. The diversity in size and shape can be explained by the interaction of the clustering proteins with the actin cytoskeleton or the lipid membrane, but very little is known about the processes that determine the lifetime of the nanoclusters. In this paper, using mathematical modeling of the cluster dynamics, we model the biophysical processes that determine the lifetime of actin-dependent PNCs. In particular, we investigated the role of actin aster fragmentation, which had been suggested to be a key determinant of the PNC lifetime, and we found that it is important only for a small class of PNCs. A simple extension of our model allowed us to investigate the kinetics of protein-ligand interaction near PNCs. We found an anomalous increase in the lifetime of ligands near PNCs, which agrees remarkably well with experimental data on RAS-RAF kinetics. In particular, analysis of the RAS-RAF data through our model provides falsifiable predictions and novel hypotheses that will not only shed light on the role of RAS-RAF kinetics in various cancers, but also will be useful in studying membrane protein clustering in general.

The Equations for Consciousness: A Reply to “Tracking the Travels,” a Review of Journey of the Mind

The Equations for Consciousness: A Reply to “Tracking the Travels,” a Review of <i>Journey of the Mind</i>

Deduction of signaling mechanisms from cellular responses to multiple cues

Cell signaling networks are complex and often incompletely characterized, making it difficult to obtain a comprehensive picture of the mechanisms they encode. Mathematical modeling of these networks provides important clues, but the models themselves are often complex, and it is not always clear how to extract falsifiable predictions. Here we take an inverse approach, using experimental data at the cell level to deduce the minimal signaling network. We focus on cells’ response to multiple cues, specifically on the surprising case in which the response is antagonistic: the response to multiple cues is weaker than the response to the individual cues. We systematically build candidate signaling networks one node at a time, using the ubiquitous ingredients of (i) up- or down-regulation, (ii) molecular conversion, or (iii) reversible binding. In each case, our method reveals a minimal, interpretable signaling mechanism that explains the antagonistic response. Our work provides a systematic way to deduce molecular mechanisms from cell-level data.

Hot new early dark energy: Towards a unified dark sector of neutrinos, dark energy and dark matter

Hot new early dark energy describes a supercooled, first-order phase transition that takes place at sub-eV temperatures in the dark sector. It lowers the sound horizon, which provides a possible solution to the Hubble tension, and, at the same time, it can explain the neutrino masses through the inverse seesaw mechanism by making a set of sterile Majorana fermions massive. First, we argue that this scenario strengthens existing cosmological bounds on the heaviest neutrino mass. This, in turn, constrains the dark sector temperature, which provides us in total with two falsifiable predictions. In a second step, we discuss the phenomenological consequences of embedding hot new early dark energy in a larger gauge group that is partially broken above the TeV scale. This novel theory, which could even be motivated independently of the Hubble tension, completes the high-energy corner of the inverse seesaw mechanism and explains the mass of a dark matter candidate that can be produced through gravitational interactions at high energies.

SLA as complex, dynamical and predictable: A Processability Theory perspective

This article enters the debate about the complex and dynamical nature of second language acquisition (SLA) by discussing and commenting on Pallotti’s critique of Complex Dynamic Systems Theory (CDST). Pallotti’s critique brings to the fore the argument that, due to its anti-reductionist stance, CDST research fails to observe three fundamental research criteria, namely that research should (1) construct models, (2) make generalizations and falsifiable predictions and (3) adopt clear empirical approaches. While Pallotti does not explore research already moving SLA beyond anti-reductionist approaches to complexity and dynamism, the debate can be advanced by examining research doing exactly this. This article aims to evaluate the extent to which Processability Theory (PT) – a theory of second language (L2) processing that advocates reductionism for the sake of theory construction – meets Pallotti’s criteria for researching L2 complexity and dynamics. By discussing Pallotti’s – and PT’s – criticisms of CDST, the article outlines the implied research criteria and delineates how PT meets these criteria by (1) its bidimensional design, (2) its generalizations and falsifiable predictions for development and variation, and (3) its operationalization of constructs. Findings from a longitudinal study of L2 English are presented. These findings elucidate that, by meeting the criteria, PT can reveal the predictable, implicational development of processing complexity in eight learners and the patterned, dynamical interaction between development and variation in the trajectories of two of these learners. The article concludes that PT fulfils Pallotti’s (2022) criteria and, by a reductionist approach which models SLA as multidimensional, it can generate what CDST cannot; dynamical and complex L2 systems which progress in predictable stages, with well-defined constraints on variation, including DST-defined constraints. The article closes by discussing the implications of PT’s approach for future research on L2 complexity and dynamics.

The impact of two massive early accretion events in a Milky Way-like galaxy: repercussions for the buildup of the stellar disc and halo

ABSTRACT We identify and characterize a Milky Way-like realization from the Auriga simulations with two consecutive massive mergers $\sim 2$ Gyr apart at high redshift, comparable to the reported Kraken and Gaia-Sausage-Enceladus. The Kraken-like merger (z = 1.6, $M_{\rm Tot}=8\times 10^{10}\, \rm {M_{\odot }}$) is gas-rich, deposits most of its mass in the inner $10\,$ kpc, and is largely isotropic. The Sausage-like merger (z = 1.14, $M_{\rm Tot}=1\times 10^{11}\, \rm {M_{\odot }}$) leaves a more extended mass distribution at higher energies, and has a radially anisotropic distribution. For the higher-redshift merger, the stellar mass ratio of the satellite to host galaxy is high (1:3). As a result, the chemistry of the remnant is indistinguishable from contemporaneous in situ populations, making it challenging to identify through chemical abundances. This naturally explains why all abundance patterns attributed so far to Kraken are in fact fully consistent with the metal-poor in situ so-called Aurora population and thick disc. However, our model makes a falsifiable prediction: if the Milky Way underwent a gas-rich double merger at high redshift, then this should be imprinted on its star formation history with bursts about $\sim 2\,$ s apart. This may offer constraining power on the highest-redshift massive mergers.

An expert judgment model to predict early stages of the COVID-19 pandemic in the United States.

From February to May 2020, experts in the modeling of infectious disease provided quantitative predictions and estimates of trends in the emerging COVID-19 pandemic in a series of 13 surveys. Data on existing transmission patterns were sparse when the pandemic began, but experts synthesized information available to them to provide quantitative, judgment-based assessments of the current and future state of the pandemic. We aggregated expert predictions into a single "linear pool" by taking an equally weighted average of their probabilistic statements. At a time when few computational models made public estimates or predictions about the pandemic, expert judgment provided (a) falsifiable predictions of short- and long-term pandemic outcomes related to reported COVID-19 cases, hospitalizations, and deaths, (b) estimates of latent viral transmission, and (c) counterfactual assessments of pandemic trajectories under different scenarios. The linear pool approach of aggregating expert predictions provided more consistently accurate predictions than any individual expert, although the predictive accuracy of a linear pool rarely provided the most accurate prediction. This work highlights the importance that an expert linear pool could play in flexibly assessing a wide array of risks early in future emerging outbreaks, especially in settings where available data cannot yet support data-driven computational modeling.

Geometry of neural computation unifies working memory and planning

Real-world tasks require coordination of working memory, decision-making, and planning, yet these cognitive functions have disproportionately been studied as independent modular processes in the brain. Here, we propose that contingency representations, defined as mappings for how future behaviors depend on upcoming events, can unify working memory and planning computations. We designed a task capable of disambiguating distinct types of representations. In task-optimized recurrent neural networks, we investigated possible circuit mechanisms for contingency representations and found that these representations can explain neurophysiological observations from the prefrontal cortex during working memory tasks. Our experiments revealed that human behavior is consistent with contingency representations and not with traditional sensory models of working memory. Finally, we generated falsifiable predictions for neural data to identify contingency representations in neural data and to dissociate different models of working memory. Our findings characterize a neural representational strategy that can unify working memory, planning, and context-dependent decision-making.

Implications of the cosmological 21-cm absorption profile for high-redshift star formation and deep JWST surveys

ABSTRACT Apart from its anomalously large depth, the cosmological 21-cm absorption signal measured by the EDGES collaboration also has a shape that is distinctly different from theoretical predictions. Models with non-traditional components such as super-adiabatic baryonic cooling or an excess radio background (ERB) explain the depth of the observed profile, but still conspicuously fail to explain its shape. In this paper, we quantify the requirements imposed by the EDGES measurement on sources of Ly α and X-ray photons in the presence of ERB at cosmic dawn. In extreme cases, the Ly α and X-ray emissivities require to be enhanced by up to an order of magnitude relative to traditional models. Furthermore, this enhancement needs to be active only for a short duration. We find that under conventional assumptions for the cosmic star formation rate density (SFRD), standard stellar populations are incapable of meeting these conditions. Only highly unusual models of massive metal-free stars seem to provide a possible mechanism. Conversely, if the sources of Ly α and X-ray photons are compelled to have standard properties, the EDGES measurement puts strong demands on the cosmic SFRD. This provides interesting falsifiable predictions for high-redshift galaxy surveys enabled by the James Webb Space Telescope (JWST). We derive predictions for galaxy UV luminosity functions and number densities, and show that a deep JWST survey with a limiting UV magnitude of mUV, lim = 32 would potentially be able to rule out the predictions enforced by the EDGES measurement.

Critical Tests of Leading Gamma Ray Burst Theories

It has been observationally established that supernovae (SNe) of Type Ic produce long duration gamma-ray bursts (GRBs) and that neutron star mergers generate short hard GRBs. SN-Less GRBs presumably originate in a phase transition of a neutron star in a high mass X-ray binary. How these phenomena actually generate GRBs is debated. The fireball and cannonball models of GRBs and their afterglows have been widely confronted with the huge observational data, with their defenders claiming success. The claims, however, may reflect multiple choices and the use of many adjustable parameters, rather than the validity of the models. Only a confrontation of key falsifiable predictions of the models with solid observational data can test their validity. Such critical tests are reviewed in this report.

The Indirect Genetic Effect Interaction Coefficient ψ: Theoretically Essential and Empirically Neglected.

The interaction effect coefficient ψ has been a much-discussed, fundamental parameter of indirect genetic effect (IGE) models since its formal mathematical description in 1997. The coefficient simultaneously describes the form of changes in trait expression caused by genes in the social environment and predicts the evolutionary consequences of those IGEs. Here, we report a striking mismatch between theoretical emphasis on ψ and its usage in empirical studies. Surveying all IGE research, we find that the coefficient ψ has not been equivalently conceptualized across studies. Several issues related to its proper empirical measurement have recently been raised, and these may severely distort interpretations about the evolutionary consequences of IGEs. We provide practical advice on avoiding such pitfalls. The majority of empirical IGE studies use an alternative variance-partitioning approach rooted in well-established statistical quantitative genetics, but several hundred estimates of ψ (from 15 studies) have been published. A significant majority are positive. In addition, IGEs with feedback, that is, involving the same trait in both interacting partners, are far more likely to be positive and of greater magnitude. Although potentially challenging to measure without bias, ψ has critically-developed theoretical underpinnings that provide unique advantages for empirical work. We advocate for a shift in perspective for empirical work, from ψ as a description of IGEs, to ψ as a robust predictor of evolutionary change. Approaches that “run evolution forward” can take advantage of ψ to provide falsifiable predictions about specific trait interactions, providing much-needed insight into the evolutionary consequences of IGEs.

The oxidatively damaged DNA and amyloid-β oligomer hypothesis of Alzheimer's disease

The amyloid-β (Aβ) oligomer hypothesis of Alzheimer's disease (AD) still dominates the field, yet the clinical trial evidence does not robustly support it. A falsifiable prediction of the hypothesis is that Aβ oligomer levels should be elevated in the brain regions and at the disease stages where and when neuron death and synaptic protein loss begin and are the most severe, but we review previous evidence to demonstrate that this is not consistently the case. To rescue the Aβ oligomer hypothesis from falsification, we propose the novel ad-hoc hypothesis that the exceptionally vulnerable hippocampus may normally produce Aβ peptides even in healthily aging individuals, and hippocampal oxidatively damaged DNA, pathogen DNA, and metal ions such as zinc may initiate and drive Aβ peptide aggregation into oligomers and spreading, neuron death, synaptic dysfunction, and other aspects of AD neurodegeneration. We highlight additional evidence consistent with the underwhelming efficacy of Aβ oligomer-lowering agents, such as aducanumab, and of antioxidants, such as vitamin E, versus the so far isolated case report that DNase-I treatment for 2 months resulted in a severe AD patient's Mini-Mental State Exam score increasing from 3 to 18, reversing his diagnosis to moderate AD, according to the Mini-Mental State Exam.