Initial Bias Research Articles

On the feasibility of resolving Android GNSS carrier-phase ambiguities

High-precision navigation using low-cost handsets has profound potential for mass-market applications, which has been being boosted by the release of raw GNSS data from Google Android smart devices. However, integer ambiguity fixing for centimeter-level GNSS positioning is prevented by the unaligned chipset initial phase biases (IPBs) found within Android carrier-phase data. In this study, we thus investigate the temporal behaviors of those chipset IPBs using zero baselines where smart devices are linked to external survey-grade antennas, and find that the IPBs are generally stable over time as the mean standard deviation of single-epoch IPB estimates derived from continuous carrier-phase data is as low as 0.04 cycles for all satellites. Unfortunately, these chipset IPBs differ randomly among satellites and change unpredictably if carrier-phase signals are re-tracked, discouragingly suggesting that the chipset IPBs cannot be pre-calibrated or even calibrated on the fly. We therefore have to presumably correct for them in a post-processing manner with the goal of inspecting the potential of Android GNSS ambiguity resolution if hopefully the IPBs can be gone. For a vehicle-borne Nexus 9 tablet with respect to a survey-grade receiver located 100–2000 m away, we achieve the first ambiguity-fixed solution within 321 s and finally 51.6% of all epochs are resolved; the ambiguity-fixed epochs can achieve a positioning accuracy of 1.4, 2.2 and 3.6 cm for the east, north and up components, respectively, showing an improvement of 30–80% compared to the ambiguity-float solutions. While all smart devices above are connected to external survey-grade antennas, we find that a Xiaomi 8 smartphone can be coupled effectively with a miniaturized portable patch antenna, and then achieve commensurate carrier-phase tracking and ambiguity-fixing performance to those of a commercial μ-blox receiver with its dedicated patch antenna. This is encouraging since a compact and inexpensive patch antenna paired with smart devices can promote the democratization of high-precision GNSS.

A quantitative description of the transition between intuitive altruism and rational deliberation in iterated Prisoner\u2019s Dilemma experiments

What is intuitive: pro-social or anti-social behaviour? To answer this fundamental question, recent studies analyse decision times in game theory experiments under the assumption that intuitive decisions are fast and that deliberation is slow. These analyses keep track of the average time taken to make decisions under different conditions. Lacking any knowledge of the underlying dynamics, such simplistic approach might however lead to erroneous interpretations. Here we model the cognitive basis of strategic cooperative decision making using the Drift Diffusion Model to discern between deliberation and intuition and describe the evolution of the decision making in iterated Prisoner’s Dilemma experiments. We find that, although initially people’s intuitive decision is to cooperate, rational deliberation quickly becomes dominant over an initial intuitive bias towards cooperation, which is fostered by positive interactions as much as frustrated by a negative one. However, this initial pro-social tendency is resilient, as after a pause it resets to the same initial value. These results illustrate the new insight that can be achieved thanks to a quantitative modelling of human behavior.

Fearful Temperament and the Risk for Child and Adolescent Anxiety: The Role of Attention Biases and Effortful Control.

Fearful temperament represents one of the most robust predictors of child and adolescent anxiety; however, not all children with fearful temperament unvaryingly develop anxiety. Diverse processes resulting from the interplay between automatic processing (i.e., attention bias) and controlled processing (i.e., effortful control) drive the trajectories toward more adaptive or maladaptive directions. In this review, we examine the associations between fearful temperament, attention bias, and anxiety, as well as the moderating effect of effortful control. Based on the reviewed literature, we propose a two-mechanism developmental model of attention bias that underlies the association between fearful temperament and anxiety. We propose that the sub-components of effortful control (i.e., attentional control and inhibitory control) play different roles depending on individuals' temperaments, initial automatic biases, and goal priorities. Our model may help resolve some of the mixed findings and conflicts in the current literature. It may also advance our knowledge regarding the cognitive mechanisms linking fearful temperament and anxiety, as well as facilitate the continuing efforts in identifying and intervening with children who are at risk. Finally, we conclude the review with a discussion on the existing limitations and then propose questions for future research.

Experimental Studies on the Performance of Prestressed Concrete Slabs with Parabolic Tendon Profile Shapes

Slab without beams cause bending cracks in the tensile region at a low loading level. One way to overcome these weaknesses is to provide a pre-concentric or eccentric initial bias style. In this study, we wanted to compare the performance of prestressed concrete slabs in terms of cross-sectional strength capacity and rigidity of concrete slab without prestressing through experiments. Objects that were tested without prestressing were made 1 piece, namely P1, and objects that were tested with prestressing were 3 piecesnamely P2 with 2 tendons, P3 with 3 tendons, and P4 with 4 tendons. The dimensions of each object tested are 100 × 40 × 8 cm and all of them are designed with concrete qualityf'c = 30 Mpa. The tendon usedis 8 mm plains teel that is pulled by tightening the nuts at both ends, with a parabolic profile shape and eccentricity value in the middle span of 1.5 cm. From the test results obtained a very significant increase in the crosssectional strength capacity of specimens P2, P3 and P4 against P1. The increase in peak load that can be borne by 192.18% is for P2, 286.54% is for P3, and 383.00% is for P4. It also happened to increase the peak bending moment which can be borne by 161.57% for P2, 241.05% for P3, and 322.68% for P4. Keywords: Concrete, Moment . DOI : 10.7176/CER/11-10-05 Publication date: November 30 th 2019

Depressive Symptoms in Late Pregnancy Disrupt Attentional Processing of Negative-Positive Emotion: An Eye-Movement Study.

This study investigated biases for negative–positive information in component processes of visual attention (initial shift vs. maintenance of gaze) among women in late pregnancy with or without depressive symptoms. Eye movements were recorded while participants viewed a series of picture pairs depicting negative, positive, and neutral scenes. Initial orienting (latency and percentage of first fixation) and gaze duration were computed. Compared with neutral pictures, the group with major depressive symptoms (MDS) were less able to sense the positive emotion-related pictures and were over-responsive to negative emotion-related pictures. The group with suspicious depressive symptoms (SDS) had an attention bias toward both positive and negative emotion-related pictures. The group with no depressive symptoms (NDS) had an attention bias toward positive emotion-related pictures and had an initial attention avoidance tendency for negative emotion-related pictures. The initial gaze direction bias score for negative emotion-related pictures was positively correlated with the severity of depressive symptoms. Therefore, women with a risk of perinatal depression have a significant bias toward negative stimuli. Hypervigilant emotion processing during pregnancy may increase a woman’s susceptibility to depression during late pregnancy. Attention away from negative information or attention toward positive information may provide a way of buffering emotional responses.

Two hidden layer neural network-based rotation forest ensemble for hyperspectral image classification

Decision tree-based Rotation Forest could generate satisfactory but lower classification accuracy for a given training sample set and image data, owing to the inherent disadvantages in decision trees, namely myopic, replication and fragmentation problem. To improve performance of Rotation Forest technique, we propose to utilize two-hidden-layered-feedforward neural network as base classifier instead of decision tree. We examine the classification performance of proposed model under two situations, namely when free network parameters are maintained the same across all ensemble components and otherwise. The proposed model, where each component is initialized with different pair of initial weights and bias, performs better than decision tree-based Rotation Forest on three different Hyperspectral sensor datasets – AVIRIS, ROSIS and Hyperion. Improvements in classification accuracy are above 2% and up to 3% depending upon dataset. Also, the proposed model achieves improvement in accuracy over Random Forest in the range 4.2–8.8%.

Why Is There So Much More Research on Vision Than on Any Other Sensory Modality?

Why is there so much more research on vision than on any other sensory modality? There is a seemingly easy answer to this question: It is because vision is our most important and most complex sense. Although there are arguments in favor of this explanation, it can be challenged in two ways: by showing that the arguments regarding the importance and complexity of vision are debatable and by demonstrating that there are other aspects that need to be taken into account. Here, I argue that the explanation is debatable, as there are various ways of defining “importance” and “complexity” and, as there is no clear consensus that vision is indeed the most important and most complex of our senses. Hence, I propose two additional explanations: According to the methodological-structural explanation, there is more research on vision because the available, present-day technology is better suited for studying vision than for studying other modalities – an advantage which most likely is the result of an initial bias toward vision, which reinforces itself. Possible reasons for such an initial bias are discussed. The cultural explanation emphasizes that the dominance of the visual is not an unchangeable constant, but rather the result of the way our societies are designed and thus heavily influenced by human decision-making. As it turns out, there is no universal hierarchy of the senses, but great historical and cross-cultural variation. Realizing that the dominance of the visual is socially and culturally reinforced and not simply a law of nature, gives us the opportunity to take a step back and to think about the kind of sensory environments we want to create and about the kinds of theories that need to be developed in research.

Experiential effects in speech perception: Do they arise from a level playing field?

Much evidence has accrued indicating that language experience from infancy shapes our categorization and discrimination of consonants and vowels, differentiating perception of native versus non-native speech contrasts. The Perceptual Assimilation (PAM: Best), Speech Learning (SLM: Flege), Native Language Magnet (NLM: Kuhl) and Second Language Linguistic Perception (L2LP: Escudero) models each offer theoretical accounts of this attunement but do not explicitly consider whether it begins with perceptual equivalence across all speech contrasts. Other findings suggest the initial perceptual playing field Is not level -- some contrasts are intrinsically easier than others to discriminate. This paper considers two theoretical accounts of this uneven initial state, the Natural Referent Vowel framework (NRV: Polka and Bohn) and the Articulatory Organ Hypothesis (AOH: Goldstein), in light of additional findings of discrimination biases. We will discuss how initial biases may interact with language experience to modulate perceptual attunment to the native language.Much evidence has accrued indicating that language experience from infancy shapes our categorization and discrimination of consonants and vowels, differentiating perception of native versus non-native speech contrasts. The Perceptual Assimilation (PAM: Best), Speech Learning (SLM: Flege), Native Language Magnet (NLM: Kuhl) and Second Language Linguistic Perception (L2LP: Escudero) models each offer theoretical accounts of this attunement but do not explicitly consider whether it begins with perceptual equivalence across all speech contrasts. Other findings suggest the initial perceptual playing field Is not level -- some contrasts are intrinsically easier than others to discriminate. This paper considers two theoretical accounts of this uneven initial state, the Natural Referent Vowel framework (NRV: Polka and Bohn) and the Articulatory Organ Hypothesis (AOH: Goldstein), in light of additional findings of discrimination biases. We will discuss how initial biases may interact with language experience to mo...

Design of plasmonic directional antennas via evolutionary optimization.

We demonstrate inverse design of plasmonic nanoantennas for directional light scattering. Our method is based on a combination of full-field electrodynamical simulations via the Green dyadic method and evolutionary optimization (EO). Without any initial bias, we find that the geometries reproducibly found by EO work on the same principles as radio-frequency antennas. We demonstrate the versatility of our approach by designing various directional optical antennas for different scattering problems. EO-based nanoantenna design has tremendous potential for a multitude of applications like nano-scale information routing and processing or single-molecule spectroscopy. Furthermore, EO can help to derive general design rules and to identify inherent physical limitations for photonic nanoparticles and metasurfaces.

From alternation to repetition: Spatial attention biases contribute to sequential effects in a choice reaction-time task

ABSTRACTObservers often take longer to respond to a visual target when it appears at a recently stimulated location than when it appears at a new location in the visual field. This behavioral impairment — known as inhibition of return (IOR) — is mirrored by a reduction of an event-related potential (ERP) component called the N2pc that has been associated with attentional selection. Together, these findings indicate that the mechanism underlying IOR operates to bias covert attention against re-visiting the most recently attended location. The goal of the present study was to determine how this inhibitory attention bias evolves across successive trials of a two-item search task. Initially, targets appearing at previously attended locations were associated with behavioral IOR and a concomitant reduction of the N2pc. After several successive trials, this initial inhibitory bias was superseded by expectancy-based biases associated with “predictable” inter-trial patterns of location repeats or location changes, in some cases leading to faster responses and a larger N2pc when the target location repeated (facilitation of return). These results provide evidence that biases in the covert deployment of attention are updated dynamically according to the recent selection history and contribute to well-known sequential effects in serial choice reaction-time tasks.

An integrating genetic algorithm and modified Newton method for tracking control and vibration suppression

With the trend toward taller and more flexible building structures, a mass-damper shaking table system has been considered as means for vibration suppression to external excitation and disturbances in recent years. However, there are few researches on the control of nonlinear structure using active mass damper under earthquake excitation, especially for high-rise building. This study presents a model combining the advantages of adaptive genetic algorithm and modified Newton method is developed for system identification and vibration suppression of a building structure with an active mass damper. The genetic algorithm with adaptive reproduction, crossover, and mutation operators is to search for initial weight and bias of the neural network, while the modified Newton method, similar to BFGS algorithm, is to increase network training performance. Experimental results show that the controller performance is strongly influenced by the accuracy of system identification. The controller is also shown to be robust to variations in system parameters.

Class War in the Voting Booth: Bias Against High‐Income Congressional Candidates

Do Americans care how much money congressional candidates earn? We conducted three experiments to examine how candidates' incomes affect voters' perceptions of the candidates' traits and ultimately their vote intention. Subjects evaluated otherwise identical candidates with annual incomes randomly varying between $75,000, $3 million, and a candidate with no income information provided. Results from the three experiments are remarkably similar. Subjects viewed the $3 million earner as significantly more intelligent than the candidate with no income information provided, but this benefit of high income was overshadowed by significant biases against the $3 million candidate. Subjects consistently viewed the $3 million earner as less honest, less caring, and less representative of them than the other candidates. Ultimately, subjects were less likely to say they would vote for the $3 million candidate. These findings demonstrate that the campaign advantages that high‐income candidates enjoy are somewhat offset by voters' initial bias against them.

Can Energy Storage Pack Hybridization Improve the Best Theoretically-Achievable Battery State Estimation Accuracy?

This article examines the degree to which the creation of a series hybrid ultracapacitor-battery string can improve achievable battery state estimation accuracy. The article is motivated by the existence of fundamental information-theoretic limits on the degree to which one can use input/output current/voltage data to estimate a battery's internal states (e.g., state of charge, state of health, etc.). Previous literature quantifies these limits rigorously using Fisher information theory [1] – [5]. One conclusion of this quantification is that it is particularly difficult to obtain accurate battery state estimates during phase transition processes, such as the phase transition that occurs in lithium iron phosphate electrodes during charge/discharge. The problematic effect, in this context, is the very small dV/dQ slope, or effective electrical elastance, of the battery. Poor battery elastance complicates battery state estimation accuracy by directly contributing to the worsening of the best-achievable battery state estimation accuracy. The literature already shows this through the use of Fisher information analysis which shows the dependence of the best-achievable battery state estimation accuracy and the effective battery elastance. Building on the above literature, the key conjecture motivating this article is that the creation of hybrid series ultracapacitor-battery strings can improve battery state estimation accuracy substantially by furnishing a significant dV/dQ system behavior. We derive this conclusion analytically, using Fisher information analysis. Then consider estimating the lithium-ion battery state of charge using input current and output battery voltage, where the current measurement is biased and the voltage sensor has zero-mean, random, Gaussian noise. By comparison between the Cramer-Rao lower bound of both the battery-only and ultracapacitor-battery hybrid, we demonstrate improvements in theoretically-achievable initial charge accuracy as well as current bias, and in turn improve the battery SOC estimation. This conclusion is validated using a Monte Carlo simulation study. Performing least square method to estimate both initial charge in the battery and current sensor bias to achieve point clouds with and without an ultracapacitor in series, as shown in Figure 1a. One can conclude that there is: (i) a dependence between the estimated current bias and initial charge; (ii) a large condition number of the Fisher information metric which is reflected in the thin shape of the cloud ellipse; (iii) significant decrease in the estimation error after adding a series ultracapacitor into the system. Finally, the article presents an experimental ultracapacitor-battery validation study, as shown in Figure 1b. The full experiment setup includes a system of three parallel 400F ultracapacitors in series with a single 0.6Ah LFP cell. Experimental data is used to solve the optimization problem of minimizing the mean squared error of physical and simulation system outputs with respect to estimated initial charge and current bias. Results show clear improvement on the estimation accuracy. Figure 1. (a) Monte Carlo simulation results comparison (above) (b) Experiment setup structure (below) This article answers the title question by providing quantifiable evidence of improved state estimation through theory, simulation and experiment. Figure 1

Temporal attention causes systematic biases in visual confidence

Temporal attention enhances the perceptual representation of a stimulus at a particular point in time. The number of possible attentional episodes in a given period is limited, but whether observers’ confidence reflects such limitations is still unclear. To investigate this issue, we adapted an “Attentional Blink” paradigm, presenting observers with a rapid visual stream of letters containing two targets cued for subsequent perceptual reports and confidence judgments. We found three main results. First, when two targets fell within the same attentional episode, the second target underwent a strong under-confidence bias. In other words, confidence neglected that a single attentional episode can benefit to both targets. Second, despite this initial bias, confidence was strongly correlated with response probability. Third, as confidence was yoked to the evidence used in perceptual reports, it remains blind to delays in response selection for the second target. Notably, the second target was often mistaken with a later item associated with higher confidence. These results suggest that confidence does not perfectly evaluate the limits of temporal attention in challenging situations.

Nonreciprocal acoustic transmission in space-time modulated coupled resonators

Systems that break reciprocity offer new possibilities for controlling wave propagation. Here we study the scattering properties of coupled resonator systems that are under dynamic modulation. Strong linear nonreciprocal transmission is manifested in the acoustic regime by introducing an initial spatial phase bias to the space-time modulated coupled resonators. A theoretical model is developed to characterize the system and the results are in good agreement with experimental observations. Our work opens up new opportunities for designing compact nonreciprocal devices and developing acoustic topological insulators.

Spoken language identification based on optimised genetic algorithm–extreme learning machine approach

The determination and classification of a recognized spoken language based on certain contents and datasets is known as the process of language identification (LID). The common process in carrying out LID entails the mandatory processing of data which enables the extraction of the necessary features for the process. The extraction involves a mature process whereby the development of the standard LID features have been conducted much earlier by means of a mel-frequency cepstral coefficients, shifted delta cepstral, Gaussian mixture model and i-vector-based framework. Despite that, improvement or rather optimisation still needs to be done on the learning process based on the extracted features so as to obtain all the knowledge embedded within them. The classification and regression analysis can benefit tremendously from the use of the extreme learning machine (ELM) which is a particularly effective and useful learning model for training a single-hidden layer neural network. However, owing to the randomly selected weights embedded in the input’s hidden layers, the model’s learning process is rendered to be ineffective or not optimised in its entirety. In this study, the ELM is employed as the learning model for LID due to the standard feature extraction. In addition, this study proposes a new optimised genetic algorithm (OGA) with three different selection criteria (i.e., roulette wheel, K-tournament and random) to select the appropriate initial weights and biases of the input hidden layer of the ELM, thereby minimising the classification error and improving the general performance of the ELM for LID. Results show the excellent performance of the proposed OGA–ELM with three different selection criteria, namely, roulette wheel, K-tournament and random, with the highest accuracies of 99.50%, 100% and 99.38%, respectively.

Social inclusion increases with time for zero-tillage wheat in the Eastern Indo-Gangetic Plains.

Sustainable intensification (SI) approaches to agricultural development are urgently needed to meet the growing demand for crop staples while protecting ecosystem services and environmental quality. However, SI initiatives have been criticized for neglecting social welfare outcomes. A recent review found that better-off farmers benefitted disproportionately from SI and highlighted the dearth of studies assessing the equity of outcomes. In this study, we explore the social inclusiveness of zero-tillage (ZT) wheat adoption in Bihar, India. ZT is a proven SI technology for enhancing wheat productivity while boosting profitability and reducing greenhouse gas emissions from agricultural machinery in the densely populated Indo-Gangetic Plains. With an average landholding size of 0.39 ha, most farmers in Bihar depend on custom-hiring services to access the technology. While service provision models should foster inclusive growth by reducing financial barriers to technology adoption, early evidence suggested that smallholders remained at a disadvantage. Building on this previous research, we use a panel dataset from 961 wheat-growing households that spans a six-year period to analyze ZT adoption dynamics over time while accounting for the role of social networks and access to service provision. Using a heckprobit approach to correct for non-exposure bias, we compare determinants of ZT awareness and use in 2012 and 2015. We apply a multinomial logit model to identify determinants of early adoption, recent adoption, non-adoption, and dis-adoption. Furthermore, we explore the quality of ZT services as an additional dimension of socially-inclusive technology access. We find that the strong initial scale bias in ZT use declined substantially as awareness of the technology increased and the service economy expanded. Land fragmentation replaced total landholding size as a significant adoption determinant, which also affected the quality of ZT services received. Hence, farmers with small but contiguous landholdings appear to have gained a significant degree of access over time. We conclude that early-stage assessments of SI may be misleading, and that private sector-based service provision can contribute to socially inclusive development outcomes as markets mature.

No handedness effect on spatial orientation or ocular counter-roll during lateral head tilts.

Although vestibular inputs are bilaterally represented within the cerebral hemispheres, the higher level vestibular functions exhibit hemispheric asymmetries. Previous studies have suggested that such asymmetries are associated with handedness. Here, we studied the impact of handedness (i.e., hemispheric lateralization) on spatial orientation using a subjective visual vertical (SVV) task. We tested 22 right‐handed and 22 left‐handed subjects in upright position, during prolonged lateral head tilts of 20° (~15 min), and after the head returned to upright position. The corresponding changes in torsional eye position were measured simultaneously using video‐oculography. During lateral head tilts, both right‐ and left‐handers had initial SVV biases in the opposite direction of the head tilt (right‐handers: left tilt 3.0 ± 1.3°, right tilt −4.7 ± 1.5°; left‐handers: left tilt 3.4 ± 1.1°, right tilt −4.1 ± 1.0°). The SVV subsequently drifted in the direction of the head tilt, and there was an aftereffect in the same direction when the head was brought back upright. The ocular torsion initially changed in the opposite direction of the head tilt (right‐handers: left tilt 3.8 ± 0.4°, right tilt −3.8 ± 0.4°; left‐handers: left tilt 4.2 ± 0.5°, right tilt −4.5 ± 0.5°), and there were also drift and aftereffect in the same direction as the head tilt. The changes in upright perception and ocular torsion did not differ between right‐ and left‐handers. These findings show no functional laterality, neither in the higher level neural mechanisms that maintain spatial orientation, nor in the lower level mechanisms that generate the ocular torsion response during lateral head tilt.

A two phase investment game for competitive opinion dynamics in social networks

We propose a setting for two-phase opinion dynamics in social networks, where a node’s final opinion in the first phase acts as its initial biased opinion in the second phase. In this setting, we study the problem of two camps aiming to maximize adoption of their respective opinions, by strategically investing on nodes in the two phases. A node’s initial opinion in the second phase naturally plays a key role in determining the final opinion of that node, and hence also of other nodes in the network due to its influence on them. However, more importantly, this bias also determines the effectiveness of a camp’s investment on that node in the second phase. In order to formalize this two-phase investment setting, we propose an extension of Friedkin–Johnsen model, and hence formulate the utility functions of the camps. We arrive at a decision parameter which can be interpreted as two-phase Katz centrality. There is a natural tradeoff while splitting the available budget between the two phases. A lower investment in the first phase results in worse initial biases in the network for the second phase. On the other hand, a higher investment in the first phase spares a lower available budget for the second phase, resulting in an inability to fully harness the influenced biases. We first analyze the non-competitive case where only one camp invests, for which we present a polynomial time algorithm for determining an optimal way to split the camp’s budget between the two phases. We then analyze the case of competing camps, where we show the existence of Nash equilibrium and that it can be computed in polynomial time under reasonable assumptions. We conclude our study with simulations on real-world network datasets, in order to quantify the effects of the initial biases and the weightage attributed by nodes to their initial biases, as well as that of a camp deviating from its equilibrium strategy. Our main conclusion is that, if nodes attribute high weightage to their initial biases, it is advantageous to have a high investment in the first phase, so as to effectively influence the biases to be harnessed in the second phase.

The Intention-to-Treat Effect of Bridging Treatments in the Setting of Milan Criteria-In Patients Waiting for Liver Transplantation.

In patients with hepatocellular carcinoma (HCC) meeting the Milan criteria (MC), the benefit of locoregional therapies (LRTs) in the context of liver transplantation (LT) is still debated. Initial biases in the selection between treated and untreated patients have yielded conflicting reported results. The study aimed to identify, using a competing risk analysis, risk factors for HCC-dependent LT failure, defined as pretransplant tumor-related delisting or posttransplant recurrence. The study was registered at www.clinicaltrials.gov (identification number NCT03723304). In order to offset the initial limitations of the investigated population, an inverse probability of treatment weighting (IPTW) analysis was used: 1083 MC-in patients (no LRT=182; LRT=901) were balanced using 8 variables: age, sex, Model for End-Stage Liver Disease (MELD) value, hepatitis C virus status, hepatitis B virus status, largest lesion diameter, number of nodules, and alpha-fetoprotein (AFP). All the covariates were available at the first referral. After the IPTW, a pseudo-population of 2019 patients listed for LT was analyzed, comparing 2 homogeneous groups of untreated (n=1077) and LRT-treated (n=942) patients. Tumor progression after LRT was the most important independent risk factor for HCC-dependent failure (subhazard ratio [SHR], 5.62; P<0.001). Other independent risk factors were major tumor diameter, AFP, MELD, patient age, male sex, and period of wait-list registration. One single LRT was protective compared with no treatment (SHR, 0.51; P<0.001). The positive effect was still observed when 2-3 treatments were performed (SHR, 0.66; P=0.02), but it was lost in the case of ≥4 LRTs (SHR, 0.80; P=0.27). In conclusion, for MC-in patients, up to 3 LRTs are beneficial for success in intention-to-treat LT patients, with a 49% to 34% reduction in failure risk compared with untreated patients. This benefit is lost if more LRTs are required. A poor response to LRT is associated with a higher risk for HCC-dependent transplant failure.