Smaller Discrepancy Research Articles

Weighting non-IID batches for out-of-distribution detection

A standard network pretrained on in-distribution (ID) samples could make high-confidence predictions on out-of-distribution (OOD) samples, leaving the possibility of failing to distinguish ID and OOD samples in the test phase. To address this over-confidence issue, the existing methods improve the OOD sensitivity from modeling perspectives, i.e., retraining it by modifying training processes or objective functions. In contrast, this paper proposes a simple but effective method, namely Weighted Non-IID Batching (WNB), by adjusting batch weights. WNB builds on a key observation: increasing the batch size can improve the OOD detection performance. This is because a smaller batch size may make its batch samples more likely to be treated as non-IID from the assumed ID, i.e., associated with an OOD. This causes a network to provide high-confidence predictions for all samples from the OOD. Accordingly, WNB applies a weight function to weight each batch according to the discrepancy between batch samples and the entire training ID dataset. Specifically, the weight function is derived by minimizing the generalization error bound. It ensures that the weight function assigns larger weights to batches with smaller discrepancies and makes a trade-off between ID classification and OOD detection performance. Experimental results show that incorporating WNB into state-of-the-art OOD detection methods can further improve their performance.

Reducing discrepancies between actual and ideal affect across adulthood: the roles of activity flow conduciveness, pleasantness, and familiarity

ABSTRACT Previous findings demonstrate that people often do not feel how they want to feel, supporting the distinction between “actual affect” and “ideal affect.” But are there certain activities that reduce the discrepancy between actual and ideal affect? Based on flow theory and socioemotional selectivity theory, we examined whether the discrepancy between people’s actual and ideal positive affect would be smaller during activities that were more conducive to flow (a state of intense absorption and concentration), pleasant, and familiar. In Study 1, U.S. participants aged 17–79 (N = 393) reported their ideal affect and how they felt during activities with varying degrees of challenges and skills. For both low-arousal positive affect (LAP) and high-arousal positive affect (HAP), participants reported smaller actual-ideal affect discrepancies during flow-conducive activities (when skills matched challenges). Study 2 was a 14-day experience sampling study, in which Hong Kong participants aged 18–83 (N individual = 109) reported their momentary actual and ideal affect, and how pleasant and familiar their activities were (N experience = 3,815). Greater activity familiarity was associated with smaller discrepancies in actual-ideal LAP, while greater activity pleasantness was associated with smaller discrepancies in actual-ideal HAP. These findings provide insights on the activities that help people achieve their ideal affect more easily.

What drives startup valuations?

We find that valuations of Venture Capital-backed startup companies invested by mutual funds are driven by their peer valuations rather than their fundamentals, such as operating performances. As startups oftentimes experience substantial valuation changes at their Initial Public Offerings (IPO), we also investigate the share pricing transition of startups from private to public companies. We find that more active pricing of shares prior to the IPO correlates with smaller discrepancies between private and public valuations as well as lower IPO mispricing, suggesting a link between private and public market pricing efficiency. Jointly, our results help investors and regulators achieve a better understanding of the dynamics behind misvaluation in private assets in general, and startups in particular.

Managing Uncertainty in Urban Road Traffic Emissions Associated with Vehicle Fleet Composition: From the Perspective of Spatiotemporal Sampling Coverage

With pronounced differences in emission factors among vehicle types and marked spatiotemporal heterogeneity of vehicle fleet composition, extrapolating fleet composition from insufficient sample hour periods and road segments will introduce significant uncertainty in calculating regional daily road traffic emissions. We proposed a framework to manage uncertainty in urban road traffic emissions associated with vehicle fleet composition from the perspective of spatiotemporal sampling coverage. Initially, the respective relationships of the temporal and spatial sampling coverages of fleet composition with the resulting regional daily road traffic emission uncertainties were determined, using the core area of a typical small and medium-sized city in China with the widely-used International Vehicle Emissions (IVE) model as example. Subsequently, function models were developed to explore the determination of the spatiotemporal sampling coverage of fleet composition. These results of emission uncertainties and function models implied that gases with larger emission factor discrepancies between vehicle types, such as NOx, required greater spatiotemporal sampling coverage than gases with smaller discrepancies, such as CO2, under the same uncertainties target. Therefore, sampling efforts should be prioritized for gases with larger emission factor discrepancies. Additionally, increasing sampling coverage in one dimension (either spatial or temporal) can reduce the minimum required coverage in the other dimension. To further reduce uncertainty, enhancing both spatial and temporal sampling coverage of the fleet composition is more effective than enhancing one type of coverage alone. The framework and results proposed in this work can reduce the uncertainty of emissions calculations caused by insufficient sampling coverage and contribute to more accurate transport emission reduction policy formulation.

Visual behavior characteristics of historical landscapes based on eye-tracking technology

ABSTRACT The subjective encounter with architecture encompasses the particular presentation of architectural elements or the overall structure to individuals, taking into account their perception, cognition, and thought processes. The purpose of this research is to investigate the intricate relationship between visual attention directed towards historical architecture and the subjective experiences engendered during the process of observation. Utilizing eye-tracking technology within the realm of virtual reality, this study delves into the observation patterns exhibited by individuals when confronted with historical architecture, specifically focusing on the traditional courtyard residences found in Beijing’s hutongs. The panoramic images, meticulously modeled and rendered, are divided into six distinct areas of interest: base, ground, window&doors, walls, roof, and eaves. The eye-tracking data of 81 participants, who engaged with 10 virtual scenes through the employment of VR headsets, along with their responses to architectural style questionnaires, were systematically gathered. Through comprehensive analysis, encompassing the examination of total fixation duration (TFD), fixation count (FC), first fixation duration (FFD), and time to first fixation (TFF) across the areas of interest (AOIs), in addition to deviations in scores among different architectural areas and styles, notable insights emerged. Results indicate that both professionals and non-professionals allocate heightened attention, as evidenced by TFD, FC, TFF, and FFD, to elements that exhibit greater score deviations in the questionnaires as opposed to those with smaller discrepancies. Moreover, the TFD and FC metrics pertaining to the Windows&doors AOI collectively constitute 40.20% and 40.71%, respectively. Undoubtedly, these figures signify the preeminent focal points within all AOIs. This underscores the pivotal role played by doors and windows in shaping individuals’ judgments pertaining to historical architectural styles. This research provides designers with valuable understanding of the cognitive patterns exhibited by individuals in engaging with aspects of history, which enable them to identify and preserve key elements associated with historical styles. Additionally, it establishes a fundamental cornerstone for future quantitative research endeavors centered around the preservation of historical architectural styles.

Black carbon instrument responses to laboratory generated particles

Accurate measurement of black carbon (BC) particles is vital for climate models as well as air quality assessments. While the need for BC particle measurement has been recognized, standardization of instruments and procedures for ambient measurement is still underway. In this study, we used laboratory generated soot particles to assess nine instruments targeting BC mass concentration measurement. The measurement matrix included different BC concentrations (ranging from atmospheric levels to combustion emission levels), different particle coatings, two particle sources (gas burner and spark generator) and two dilution methods. The nine instruments included six different models: aethalometers AE33 and MA200, thermo-optical OC-EC analysis, multi-angle absorption photometer MAAP 5012, photoacoustic instrument MSS, and soot particle aerosol mass spectrometer SP-AMS. The main discrepancy we observed was that the SP-AMS results were systematically lower, approximately only half of the BC measured by other instruments. A portion of this is explained by particle losses in the aerodynamic lens of the SP-AMS and the parameters used in the data analysis. Some smaller discrepancies were identified for the other instruments, but overall, the median values from were within 25 % of each other. Instruments’ operation principles and covered concentration ranges need to be carefully considered especially in emission measurements where the aerosols can have high temporal variation as well as high BC concentrations. In general, the results can decrease the uncertainties in climate and air quality studies by providing tools for more accurate and comparable BC measurements and when the existing BC data is interpreted.

What are the real populations of Chinese cities? On mistakes in city border defi nition and data interpretation in popular internet sources

Economic growth and intensive metropolisation processes in the PRC have increased interest in information on the size of the country’s cities. Chinese institutions apply population data to administrative units which is considerably larger than its urban area. The publicly available data are often divergent; according to Gibson & Li (2017) hundreds of studies in economics misinterpret China’s subnational population and over 80% of articles use these data erroneously. Few specialists are able to use data directly from original Chinese sources (see Chan, 2007; Chan & Wan, 2017), most of them use publicly available sources. Scientists and other users often have at their disposal estimates published by international institutions. A comparative analysis of those data exposes marked dissimilarities. The article is to identify differences between particular sources and to establish their causes. I assume that the reason for the discrepancy in the assessment of the city's population is not a difference in method of defining the city, but rather the fact that the authors of the studies do not apply established criteria and do not provide accurate calculations. Finally, guidelines are created that, if followed, should result in smaller discrepancies between data published by various sources.

Suitability of Computerised and Pencil‐and‐Paper Neuropsychological Assessments in Healthy Older Culturally and Linguistically Diverse Australians: The CogSCAN Study

AbstractBackgroundComputerised neuropsychological assessments (CNAs) are proposed as more accessible and culturally appropriate alternatives to traditional pencil‐and‐paper neuropsychological assessments (PnPAs) for diagnosing dementia in diverse populations. Whilst some PnPAs have demonstrated cultural and linguistic bias, research investigating the suitability of CNAs for culturally and linguistically diverse (CALD) individuals is limited, especially outside of the United States.MethodThis study compared cognitive performance in a heterogenous sample of older Australian CALD English‐speakers and an L1 English‐speaking background (ESB) sample. Cross‐sectional data were drawn from the ongoing CogSCAN Study. Participants were 212 community‐dwelling individuals aged 60‐95 years without a dementia diagnosis (159 ESB/53 CALD). Participants completed a PnPA and/or CNAs, sessions were counterbalanced accordingly. CNA sessions involved two of four commercially‐available CNAs in a randomised order. Differences in the level and pattern of cognitive performance between groups on two CNAs (CANTAB/Cogstate) and a comprehensive PnPA in English, controlling for sociodemographic and health factors, are reported here. Selection of key variables for examination were based on previous findings (PnPA) and test developer recommendations (CNAs).ResultSample characteristics are summarised in Table 1. CALD participants’ cognitive performance on individual tests (Figure 1) was largely comparable to ESB participants on PnPA and each CNA. Significant differences between groups, after Bonferroni correction for multiple comparisons, were observed for the Boston Naming Test (PnPA; a language test widely criticised for use with CALD populations due to culturally biased stimuli) and Rapid Visual Information Processing (CANTAB; a test of attention/processing speed). Among participants who completed PnPA and Cogstate, a significant interaction between group and assessment type on global cognitive performance was observed; ESB participants outperformed CALD participants on PnPA, but not on Cogstate (Figure 2). No significant main effects or interaction were observed in PnPA versus CANTAB.ConclusionThese results showed fewer and smaller discrepancies across groups and assessment types than has been found in previous literature with other CALD samples, perhaps reflecting higher levels of English proficiency, acculturation, and education in English than previous cohorts. Work in progress is investigating the influence of linguistic and acculturation factors on cognitive performance in a larger and more diverse sample.

Auditory phonological identification impairment in primary progressive aphasia

ObjectiveTo examine the audiological characteristics and neuroanatomical regions associated with auditory phonological identification impairment in primary progressive aphasia (PPA). MethodsTwenty-seven patients with PPA [13 non-fluent/agrammatic variant PPA (nfvPPA), three logopenic variant PPA (lvPPA), seven semantic variant PPA (svPPA), and four mixed type PPA] were included in the study. Neuropsychological, language, audiological, and neuroradiological examinations were also performed. Auditory function examinations consisted of a pure-tone threshold test, a phonological identification task, and temporal auditory acuity tests, such as click counting or fusion. As an evaluation value of phonological identification ability, we calculated the discrepancy scores, which were the smaller discrepancy (left or right ear) in phonological identification ability scores between measured and expected values from the pure-tone threshold. In the neuroradiological examination, we evaluated the regional cerebral blood flow using 123I-iodoamphetamine single-photon emission computed tomography. ResultsEight of the 27 patients were allocated to the impaired phonological identification group, and four were considered to have significant impairment on further analysis. Two of these patients, one with lvPPA and one with mixed type of lvPPA and nfvPPA, showed apparent phonological identification deficits that could be observed in daily life. The discrepancy scores were not significantly related to the results of neuropsychological, language, or any other auditory examinations, except for the click counting score in the left ear. Voxel-based correlation analyses revealed that regional cerebral blood flow in the bilateral superior temporal gyrus and bilateral primary auditory cortex was significantly and positively correlated with phonological identification ability. ConclusionsOur results suggest that progressive dysfunction of the bilateral superior temporal gyrus and bilateral primary auditory cortex due to neurodegenerative diseases leads to phonological identification impairment in PPA syndrome.

COMPARISON OF TWO MATHEMATICAL MODELS FOR FISHEYE CAMERAS APPLIED IN THE ORB-SLAM

Abstract. Fisheye lens cameras are becoming increasingly popular for vSLAM applications due to their wide field of view (FoV), providing more features to be tracked in a single image shot. However, the complex lens geometry involved in the image formation process still limits their full potential, especially when points in the hyperhemispherical field are unmodeled. In this paper, we compare two adaptations of ORB-SLAM for fisheye lens cameras, considering the use of the rigorous projection model (equisolid-angle) versus the use of the generic projection model (EUCM). The ORB-SLAM versions were adapted for real-time processing on the Nvidia Jetson TX2 board. The experiment was conducted using hyperhemispherical images obtained with a Ricoh Theta S camera. Our results showed that the trajectory estimated with the equisolid-angle ORB-SLAM had smaller discrepancies, compared to the reference trajectory, than the EUCM ORB-SLAM. This suggests that a rigorous photogrammetric model with a suitable treatment of hyperhemispherical points is beneficial for trajectory estimation.

Backcasting COVID-19: a physics-informed estimate for early case incidence.

It is widely accepted that the number of reported cases during the first stages of the COVID-19 pandemic severely underestimates the number of actual cases. We leverage delay embedding theorems of Whitney and Takens and use Gaussian process regression to estimate the number of cases during the first 2020 wave based on the second wave of the epidemic in several European countries, South Korea and Brazil. We assume that the second wave was more accurately monitored, even though we acknowledge that behavioural changes occurred during the pandemic and region- (or country-) specific monitoring protocols evolved. We then construct a manifold diffeomorphic to that of the implied original dynamical system, using fatalities or hospitalizations only. Finally, we restrict the diffeomorphism to the reported cases coordinate of the dynamical system. Our main finding is that in the European countries studied, the actual cases are under-reported by as much as 50%. On the other hand, in South Korea-which had a proactive mitigation approach-a far smaller discrepancy between the actual and reported cases is predicted, with an approximately 18% predicted underestimation. We believe that our backcasting framework is applicable to other epidemic outbreaks where (due to limited or poor quality data) there is uncertainty around the actual cases.

Charisma failure: Understanding differences in support for conservation of Asian elephants compared to tigers and African elephants

The disproportionately high support for conservation of charismatic species is well-documented. However, available data demonstrate substantial discrepancies in funding for conservation even among charismatic species, and these differences are poorly understood. We use two survey experiments to explore explanations for differences in demand for conservation between charismatic species by comparing Asian elephants (Elephas maximus) with tigers (Panthera tigris) in India and with African elephants (Loxodonta africana and Loxodonta cyclotis) globally. Through social media, we collect data from conservation enthusiasts to measure (a) relative demand for conservation, (b) awareness about conservation status, (c) relative favourability, and (d) perception of which species faces greater conservation challenges (“risk perception”). Overall, respondents allocated significantly more funds to the species/taxon they perceive to be more at risk and 41–54 % more to the species/taxon they say they like more (though about 70 % of respondents claimed to like both species/taxa presented equally). On average, respondents in India allocated 5.6 % more for tigers than Asian elephants, a far smaller discrepancy than found in public data from key real-life funders. Internationally, respondents allocated 14 % more funds for African than Asian elephants (despite there being fewer Asian elephants) and generally overestimated how endangered African elephants are compared to Asian elephants. Our study suggests (i) conservation enthusiasts wish to donate more to more endangered charismatic species but are sometimes misinformed about the level of endangerment of even very well-known species, and (ii) discrepancies in actual funding levels across charismatic species appear to be greater than conservation enthusiasts believe they should be.

Two source energy balance maize evapotranspiration estimates using close-canopy mobile infrared sensors and upscaling methods under variable water stress conditions

Mobile infrared thermometers (IRTs) mounted on moving platforms provide one-time-of-day radiometric measurements (Tr), which can be used to calculate instantaneous actual evapotranspiration (ETa) using the two-source energy balance (TSEB) model. However, irrigation scheduling decisions utilize daily ETa estimates, hence the need for time scaling. This study evaluated different upscaling methods to calculate daily maize ETa using one-time-of day Tr under varying water stress conditions. Mobile IRTs were mounted on a high clearance mobile sensing platform and collected Tr in remote locations under full, deficit and rainfed conditions. Seven scaling methods via two pathways were employed to obtain daily ETa. First pathway was scaling one-time-of-day Tr (SC) whereas the second pathway involved use of six upscaling methods of instantaneous ETa including: original and modified evaporative factor ((EF)o, (EF)m) as well as crop coefficient ((Kc)o, (Kc)m), direct canopy resistance (Direct- rc), and solar radiation ratio (Rn/Rs); and all were compared to a neutron-based soil water balance (SWB) determined ETa. From the results, SC outperformed other methods in comparison to SWB ETa across all the selected treatments with smaller discrepancies and lower RMSE (0.9–1.7 mm d−1 vs. 0.7–4.3 mm d−1 for other methods). Furthermore, methods including SC, (EF)o, (EF)m, and Rn/Rs had their daily average ETa values in close agreement to SWB ETa with mean ETa differences ranging between 0.2 and 1.6 mm d−1. Overall, SC method performed better in fully irrigated maize (r2 = 0.52, RMSE = 0.9 mm d−1) than in deficit irrigated maize ( r2 = 0.48, RMSE = 1.4 mm d−1) but worst in rainfed maize (r2 = 0.16, RMSE = 1.7 mm d−1). This implies that SC is more suited for irrigated rather than rainfed settings. Importantly, the choice of any method depends on data requirements, irrigation water management strategy, and ETa estimation accuracy.

Comparing the values of the TMJ with different clinical methods

Nowadays computer-aided face bows can be used during the dental workflow, with them individual treatment plans can be provided. To compare the TMJ values with analog and digital devices and to measure the experienced and the inexperienced doctors’ work dynamic. One experienced(“control”) and three inexperienced (“sample”) doctors took part in both examinations. During the analog method, positional bite registrations were taken and according to them, the articulators were programmed. The results were compared in the KaVo and the Amann Girrbach systems. The Condylar Inclination and the Bennett Angl e were detected. Then the digital facebow ( AG Zebris For Ceramill Optic JMA, KaVo Arcus Digma ) was inserted, while the patient was doing TMJ movements, its angles were monitored. From the results, an average value was made for both the experienced and the inexperienced doctors who were compared in both systems. The values were written in a chart and the dispersion was counted statistically with Z-probe. The duration of the measurements was counted, so the workflow dynamics were also comparable. By using the analog methods the diversion was more than 15 degrees, in regard to all the angle values, by using the digital methods, this discrepancy was only 5 degrees. The duration of the examinations lasted at least 1.57seconds longer for the inexperienced doctors in both the analog and the digital measurements. With both methods, the TMJ values could be monitored, the digital method was faster and more reliable than the analog one, as its values had a smaller discrepancy.

Antecedents and Consequences of Discrepant Perceptions of Racial Socialization between Parents and Adolescents within Mexican-Origin Families.

Parents and adolescents often have different views regarding parental racial socialization practices; however, studies documenting such discrepancies remain scarce. Using a person-centered approach, this study investigated patterns of parent-adolescent discrepant views on racial socialization (i.e., cultural socialization, bias coping, bias awareness) as well as antecedents and consequences of the discrepancy profiles. Participants were 604 adolescents (54% female, Mage = 12.41, Rangeage = 11-15) and their mothers and fathers. The results showed distinct discrepancy patterns and suggested that more maternal/paternal warmth was associated with profiles that have smaller discrepancies or profiles in which adolescents reported higher socialization than parents. Adolescents who reported higher or similar socialization as parents demonstrated better adjustment. Implications for interventions aimed at strengthening parent-child relationships and communication about race and culture are discussed.

The Impact of FinTech on Discrimination in Mortgage Lending

The mortgage market has historically been plagued by racial discrimination, and recent data show ongoing disparities in the terms offered to minority borrowers by traditional lenders. The mortgage market has changed rapidly in the past decade, however, due to the rise of technology-based (FinTech) lenders. In this paper, I show that FinTech lenders show little to no gap in the terms provided to Black and Hispanic borrowers after adjusting for GSE credit-risk pricing determinants and loan size. This fact is shown using proprietary data on borrowers to better study racial bias and FinTech loans. The paper further shows that, in a matched analysis of observationally similar Black and Hispanic and non-Black or Hispanic borrowers, FinTech lenders provide statistically indistinguishable terms to the two groups. This stands in sharp contrast to the large discrepancies found under identical tests for traditional lenders. This matched analysis, wherein borrowers with similar characteristics such as income and age are compared, is supported by an analysis of preference and shopping heterogeneity in the National Survey of Mortgage Originations. Finally, I show that FinTech penetration in a region had consequences on legacy lender terms as well. Greater FinTech penetration in a zip code is associated with smaller discrepancies even among legacy lenders.

How much variability is there in children's experiences with different educators?

Early childhood education and care settings consist of educators clustered into classrooms, clustered into centers. Yet, little empirical research has examined variance in the interaction styles of full-time educators who work in the same classroom. In this study we engage in preliminary work to examine variability in children's experiences with different educators. The sample consisted of 172 full-time educators from 86 classrooms. We conducted hierarchical linear modeling using the difference in the interaction styles of the two educators in each classroom as the outcome. Educators’ interaction styles were measured using the Responsive Interactions for Learning (RIFL-E) scale. Approximately 58% of the variance was accounted for at the classroom level and 42% at the center level. Educators with similar interaction styles tended to be grouped together in classrooms. Classrooms where educators had smaller discrepancies in interaction styles had higher overall quality and a higher percentage of educators with early childhood education degrees, but a bigger difference between educator's highest level of education. These findings highlight the importance of considering educator-level variance in conceptualizing and measuring quality. Implications for how quality of educator/child interaction is conceptualized and measured, and research and policy on oversight and quality improvement are discussed.

Experimental and modeling study on the auto-ignition properties of ammonia/methane mixtures at elevated pressures

Ammonia (NH3) is considered as a promising carbon free energy carrier for energy and transportation systems. However, its low flammability and high NOx emission potential inhibit the implementation of pure NH3 in these systems. On the other hand, methane is a favorable low emission fuel that can be used as a co-firing fuel in ammonia combustion to promote the reactivity and control the emission levels. However, knowledge of the ignition properties of NH3/CH4 mixtures at intermediate temperatures and elevated pressures is still scarce. This study reports ignition delay times of NH3/CH4/O2 mixtures diluted in Ar or Ar/N2 over a temperature range of 900–1100 K, pressures of 20 and 40 bar, and equivalence ratios of 0.5, 1.0, and 2.0. The results demonstrate that a higher CH4 mole fraction in the fuel mixture increases its reactivity, and that the reactivity decreases with increasing the fuel-oxygen equivalence ratio. The most recent mechanisms of Glarborg et al. (2018) and Li et al. (2019) were compared against the experimental data for validation purposes. Both mechanisms can predict the measurements fairly well, and key elementary reactions applied in both mechanisms were compared. A modified mechanism is provided, which can reproduce the measurements with smaller discrepancies in most cases. Detailed modeling for emissions indicated that adding CH4 to the fuel mixture increases the emission of NOx.

Evaluation of an empirical model used for deriving the fluidization velocity of binary mixtures of biomasses and sand

In a previous study, a statistical model was developed using the experimental planning technique for evaluating the influence of its variables on fluidization velocity. In this study, we investigated the Vasconcelos-statistical model (VSM) in data representation, considering fluidization with and without segregation. The methodology used was based on the simulation of the fluidization velocity of nine binary systems, comprising sand, and eight biomasses published by six authors. In addition, the results obtained using VSM were compared with those obtained using five other models, reported by different authors, but adjusted to the experimental data of these biomasses. The result obtained by the proposed models mainly indicated a discrepancy between the experimental and calculated fluidization velocities. VSM, using only three variables (particle size, particle diameter, and biomass mass fraction), yielded results of smaller discrepancy values in all simulations (2.23–12.51%), as opposed to the other comparative models, which presented more significant numbers of variables. Thus, VSM is defined as one of the most interesting models for predicting the fluidization velocity of several biomasses.

Client and therapist agreement in moment-to-moment helpfulness ratings in psychotherapy: A microprocess approach.

Clients (n = 16; Mage = 32.88 years; 81% female) and their therapists (n = 10; 80% female) from a training clinic provided moment-to-moment ratings while reviewing a video recording of a recent session. West and Kenny's Truth-and-Bias (T&B) model was used to examine agreement in ratings. Multilevel modeling was used to evaluate the agreement-alliance association while controlling for sessions completed. Consistent with the T&B model, client-therapist ratings were temporally congruent and were not discrepant overall. Greater congruence and smaller discrepancies were linked with stronger alliances. The within-session and between-dyad variability in helpfulness ratings highlight the utility of moment-to-moment rating methods. Findings suggest that therapists should attune to client perceptions on a within-session level and attunement may be a target for enhancing the alliance.