Model Selection and Model Averaging for Mixed-Effects Models with Crossed Random Effects for Subjects and Items

Linear Mixed-Effects Models for Dependent Data: Power and Accuracy in Parameter Estimation

Linear mixed-effects models have been increasingly used to analyze dependent data in psychological research. Despite their many advantages over ANOVA, critical issues in their analyses remain. Due to increasing random effects and model complexity, estimation computation is demanding, and convergence becomes challenging. Applied users need help choosing appropriate methods to estimate random effects. The present Monte Carlo simulation study investigated the impacts when the restricted maximum likelihood (REML) and Bayesian estimation models were misspecified in the estimation. We also compared the performance of Akaike information criterion (AIC) and deviance information criterion (DIC) in model selection. Results showed that models neglecting the existing random effects had inflated Type I errors, unacceptable coverage, and inaccurate R-squared measures of fixed and random effects variation. Furthermore, models with redundant random effects had convergence problems, lower statistical power, and inaccurate R-squared measures for Bayesian estimation. The convergence problem is more severe for REML, while reduced power and inaccurate R-squared measures were more severe for Bayesian estimation. Notably, DIC was better than AIC in identifying the true models (especially for models including person random intercept only), improving convergence rates, and providing more accurate effect size estimates, despite AIC having higher power than DIC with 10 items and the most complicated true model.

Association between chronotype and metabolic-associated fatty liver disease in employed adults: A longitudinal study in Southwestern China

The circadian system is an essential physiological regulator of mammals, and sleep chronotype may be associated with the risk of metabolic disorders. However, evidence regarding the role of sleep chronotype in the development of metabolic-associated fatty liver disease (MAFLD) is scarce, particularly in employed adults. We conducted a longitudinal study of 1,309 employed adults in Southwestern China with a five-year follow-up from 2017 to 2021. MAFLD was assessed by the presence of hepatic steatosis using abdominal ultrasonography, overweight/obese status, diabetes mellitus, metabolic dysregulation, or elevation of high-sensitivity C-reactive protein. Chronotype was assessed by the Morning and Evening Questionnaire-5 (MEQ-5). The logistic random effects model was applied to analyze the 5-year panel data to estimate the association between chronotype and MAFLD, and the potential effect modification of demographics on such association. The MAFLD prevalence of participants was 38.6% at baseline and showed an increasing trend during follow-up (p for trends < 0.05). Compared with morning chronotype, evening chronotype was positively associated with MAFLD (OR = 2.19, 95%CI: [1.09, 4.40]) after controlled for covariates. Age, sex, ethnicity, and educational level did not modify the association between chronotype and MAFLD. These findings suggest that improving circadian rhythms could reduce the risk of MAFLD and chronic disease burden among employed adults.

Are valence and arousal related to the development of amodal representations of words? A computational study

In this study, we analyzed the relationship between the amodal (semantic) development of words and two popular emotional norms (emotional valence and arousal) in English and Spanish languages. To do so, we combined the strengths of semantics from vector space models (vector length, semantic diversity, and word maturity measures), and feature-based models of emotions. First, we generated a common vector space representing the meaning of words at different developmental stages (five and four developmental stages for English and Spanish, respectively) using the Word Maturity methodology to align different vector spaces. Second, we analyzed the amodal development of words through mixed-effects models with crossed random effects for words and variables using a continuous time metric. Third, the emotional norms were included as covariates in the statistical models. We evaluated more than 23,000 words, whose emotional norms were available for more than 10,000 words, in each language separately. Results showed a curve of amodal development with an increasing linear effect and a small quadratic deceleration. A relevant influence on the amodal development of words was found only for emotional valence (not for arousal), suggesting that positive words have an earlier amodal development and a less pronounced semantic change across early lifespan.

Strong versus weak embodiment: Spatial iconicity in physical, abstract, and social semantic categories

BACKGROUND

Perceptual and action systems seem to be related to complex cognitive processes, but the scope of grounded or embodied cognition has been questioned. Zwaan and Yaxley (2003) proposed that cognitive processes of making semantic relatedness judgments can be facilitated when word pairs are presented in ways that their referents maintain their iconic configuration rather than their reverse-iconic configuration (the spatial iconicity effect). This effect has been observed in different semantic categories using specific experiments, but it is known that embodiment is highly dependent on task demands.

METHOD

The present study analyzed the spatial iconicity effect in three semantic categories (physical, abstract, and social) using the same experimental criteria to determine the scope of embodied cognition. In this reaction-time experiment, 75 participants judged the semantic relatedness of 384 word pairs whose experimental items were presented in their iconic or reverse-iconic configurations.

RESULTS

Two mixed-effects models with crossed random effects revealed that the interaction between word meaning and spatial position was present only for physical concepts but neither for abstract nor social concepts.

CONCLUSIONS

Within the framework of strong and weak embodiment theories, the data support weak embodiment theory as the most explicative one.

The wrong horse was bet on: the effects of argument structure versus argument adjacency on the processing of idiomatic sentences

Introduction

Psycholinguistic research remains puzzled about the circumstances under which syntactically transformed idioms keep their figurative meaning. There is an abundance of linguistic and psycholinguistic studies that have examined which factors may determine why some idioms are more syntactically fixed than others, including transparency, compositionality, and syntactic frozenness; however, they have returned inconclusive, sometimes even conflicting, results. This is the first study to examine argument structure (i.e., the number of arguments a verb takes) and argument adjacency (i.e., the position of the critical arguments relative to the verb) and their effects on the processing of idiomatic and literal sentences in German. Our results suggest that neither the traditional models of idiom processing (according to which idioms are stored as fixed entries) nor more recent hybrid theories (which concede some compositional handling in addition to a fixed entry) adequately account for the effects of argument structure or argument adjacency. Therefore, this study challenges existing models of idiom processing.

Methods

In two sentence-completion experiments, participants listened to idiomatic and literal sentences in both active and passive voice without the sentence-final verb. They indicated which of three visually-presented verbs best completed the sentence. We manipulated the factor argument structure within experiments and argument adjacency across experiments. In Experiment 1, passivized three-argument sentences had the critical argument adjacent to the verb while two-argument sentences had the critical argument non-adjacent to the verb, and vice versa in Experiment 2.

Results

In both experiments, voice interacted with argument structure. Active sentences-both literal and idiomatic-showed equivalent processing of two- and three-argument sentences. However, passive sentences returned contrasting effects. In Experiment 1, three-argument sentences were processed faster than two-argument sentences and vice versa in Experiment 2. This pattern corresponds to faster processing when critical arguments are adjacent than non-adjacent.

Discussion

The results point to the dominant role of argument adjacency over the number of arguments in the processing of syntactically transformed sentences. Regarding idiom processing, we conclude that the adjacency of the verb to its critical arguments determines whether passivized idioms keep their figurative meaning and present the implications of this finding for relevant models of idiom processing.

The long-term effects of a polygenetic predisposition to general cognition on healthy cognitive ageing: evidence from the English Longitudinal Study of Ageing

BACKGROUND

As an accelerated cognitive decline frequently heralds onset of severe neuropathological disorders, understanding the source of individual differences in withstanding the onslaught of cognitive ageing may highlight how best cognitive abilities may be retained into advanced age.

METHODS

Using a population representative sample of 5088 adults aged •50 years from the English Longitudinal Study of Ageing, we investigated relationships of polygenic predisposition to general cognition with a rate of change in cognition during a 10-year follow-up period. Polygenic predisposition was measured with polygenic scores for general cognition (GC-PGS). Cognition was measured employing tests for verbal memory and semantic fluency.

RESULTS

The average baseline memory score was 11.1 (s.d. = 2.9) and executive function score was 21.5 (s.d. = 5.8). An increase in GC-PGS by one standard deviation (1-s.d.) was associated with a higher baseline verbal memory by an average 0.27 points (95% CI 0.19-0.34, p < 0.001). Similarly, 1-s.d. increase in GC-PGS was associated with a higher semantic fluency score at baseline in the entire sample (β = 0.45, 95% CI 0.27-0.64, p < 0.001). These associations were significant for women and men, and all age groups. Nonetheless, 1-s.d. increase in GC-PGS was not associated with decreases in verbal memory nor semantic fluency during follow-up in the entire sample, as well stratified models by sex and age.

CONCLUSION

Although common genetic variants associated with general cognition additively are associated with a stable surplus to cognition in adults, a polygenic predisposition to general cognition is not associated with age-related cognitive decline during a 10-year follow-up.

Multi-scale goal distance representations in human hippocampus during virtual spatial navigation

Goal-directed navigation relies on both coarse and fine-grained coding of spatial distance between the current position of a navigating subject and a goal destination. However, the neural signatures underlying goal distance coding remain poorly understood. Using intracranial EEG recordings from the hippocampus of drug-resistant epilepsy patients who performed a virtual spatial navigation task, we found that the right hippocampal theta power was significantly modulated by goal distance and decreased with goal proximity. This modulation varied along the hippocampal longitudinal axis such that theta power in the posterior hippocampus decreased more strongly with goal proximity. Similarly, neural timescale, reflecting the duration across which information can be maintained, increased gradually from the posterior to anterior hippocampus. Taken together, this study provides empirical evidence for multi-scale spatial representations of goal distance in the human hippocampus and links the hippocampal processing of spatial information to its intrinsic temporal dynamics.

Recovering crossed random effects in mixed-effects models using model averaging

Random effects contain crucial information to understand the variability of the processes under study in mixed-effects models with crossed random effects (MEMs-CR). Given that model selection makes all-or-nothing decisions regarding to the inclusion of model parameters, we evaluated if model averaging could deal with model uncertainty to recover random effects of MEMs-CR. Specifically, we analyzed the bias and the root mean squared error (RMSE) of the estimations of the variances of random effects using model averaging with Akaike weights and Bayesian model averaging with BIC posterior probabilities, comparing them with two alternative analytical strategies as benchmarks: AIC and BIC model selection, and fitting a full random structure. A simulation study was conducted manipulating sample sizes for subjects and items, and the variance of random effects. Results showed that model averaging, especially Akaike weights, can adequately recover random variances, given a minimum sample size in the modeled clusters. Thus, we endorse using model averaging to deal with model uncertainty in MEMs-CR. An empirical illustration is provided to ease the usability of model averaging.

Mapping Pitch Accents to Memory Representations in Spoken Discourse Among Chinese Learners of English: Effects of L2 Proficiency and Working Memory

We examined L2 learners' interpretation of pitch accent cues in discourse memory and how these effects vary with proficiency and working memory (WM). One hundred sixty-eight L1-Chinese participants learning L2-English listened to recorded discourses containing pairs of contrastive alternatives and then took a later recognition memory test. Their language proficiency and WM were measured through standard tests and the participants were categorized into low, medium, advanced, and high advanced language proficiency groups. We analyzed recognition memory task performance using signal detection theory to tease apart response bias (an overall tendency to affirm memory probes) from sensitivity (the ability to discern whether a specific probe statement is true). The results showed a benefit of contrastive L + H* pitch accents in rejecting probes referring to items unmentioned in a discourse, but not contrastive alternatives themselves. More proficient participants also showed more accurate memory for the discourses overall, as well as a reduced overall bias to affirm the presented statements as true. Meanwhile, that the benefit of L + H* accents in rejecting either contrast probes or unmentioned probes was modulated for people with greater working memory. Participants with higher WM were quite sure that it did not exist in the memory trace as this part of discourse wasn't mentioned. The results support a contrast-uncertainty hypothesis, in which comprehenders recall the contrast set but fail to distinguish which is the correct item. Further, these effects were influenced by proficiency and by working memory, suggesting they reflect incomplete mapping between pitch accent and discourse representation.

Insights Into the Processing of Collocations During L2 English Reading: Evidence From Eye Movements

We report an eye movement experiment that investigates the effects of collocation strength and contextual predictability on the reading of collocative phrases by L2 English readers. Thirty-eight Chinese English as foreign language learners (EFL) read 40 sentences, each including a specific two-word phrase that was either a strong (e.g., black coffee) or weak (e.g., bitter coffee) adjective-noun collocation and was either highly predictable or unpredictable from the previous sentence context. Eye movement measures showed that L2 reading times for the collocative phrases were sensitive to both collocation strength and contextual predictability. However, an interaction effect between these factors, which appeared relatively late in the eye movement record, additionally revealed that contextual predictability more strongly influenced time spent reading weak compared with strong collocations. This was most likely because the greater familiarity of strong collocations facilitated their integration, even in the absence of strong contextual constraint. We discuss the findings in terms of the value of collocations in second language learning.

Tracking Object-State Representations During Real-Time Language Comprehension by Native and Non-native Speakers of English

The present "visual world" eye-tracking study examined the time-course of how native and non-native speakers keep track of implied object-state representations during real-time language processing. Fifty-two native speakers of English and 46 non-native speakers with advanced English proficiency joined this study. They heard short stories describing a target object (e.g., an onion) either having undergone a substantial change-of-state (e.g., chop the onion) or a minimal change-of-state (e.g., weigh the onion) while their eye movements toward competing object-states (e.g., a chopped onion vs. an intact onion) and two unrelated distractors were tracked. We found that both groups successfully directed their visual attention toward the end-state of the target object that was implied in the linguistic context. However, neither group showed anticipatory eye movements toward the implied object-state when hearing the critical verb (e.g., "weigh/chop"). Only native English speakers but not non-native speakers showed a bias in visual attention during the determiner ("the") before the noun (e.g., "onion"). Our results suggested that although native and non-native speakers of English largely overlapped in their time-courses of keeping track of object-state representations during real-time language comprehension, non-native speakers showed a short delay in updating the implied object-state representations.

Remote Testing of the Familiar Word Effect With Non-dialectal and Dialectal German-Learning 1-2-Year-Olds

Variability is pervasive in spoken language, in particular if one is exposed to two varieties of the same language (e.g., the standard variety and a dialect). Unlike in bilingual settings, standard and dialectal forms are often phonologically related, increasing the variability in word forms (e.g., German Fuß "foot" is produced as [fus] in Standard German and as [fs] in the Alemannic dialect). We investigate whether dialectal variability in children's input affects their ability to recognize words in Standard German, testing non-dialectal vs. dialectal children. Non-dialectal children, who typically grow up in urban areas, mostly hear Standard German forms, and hence encounter little segmental variability in their input. Dialectal children in turn, who typically grow up in rural areas, hear both Standard German and dialectal forms, and are hence exposed to a large amount of variability in their input. We employ the familiar word paradigm for German children aged 12-18 months. Since dialectal children from rural areas are hard to recruit for laboratory studies, we programmed an App that allows all parents to test their children at home. Looking times to familiar vs. non-familiar words were analyzed using a semi-automatic procedure based on neural networks. Our results replicate the familiarity preference for non-dialectal German 12-18-month-old children (longer looking times to familiar words than vs. non-familiar words). Non-dialectal children in the same age range, on the other hand, showed a novelty preference. One explanation for the novelty preference in dialectal children may be more mature linguistic processing, caused by more variability of word forms in the input. This linguistic maturation hypothesis is addressed in Experiment 2, in which we tested older children (18-24-month-olds). These children, who are not exposed to dialectal forms, also showed a novelty preference. Taken together, our findings show that both dialectal and non-dialectal German children recognized the familiar Standard German word forms, but their looking pattern differed as a function of the variability in the input. Frequent exposure to both dialectal and Standard German word forms may hence have affected the nature of (prelexical and/or) lexical representations, leading to more mature processing capacities.

Variability in cTBS Aftereffects Attributed to the Interaction of Stimulus Intensity With BDNF Val66Met Polymorphism

Objective: To evaluate whether a common polymorphism (Val66Met) in the gene for brain-derived neurotrophic factor (BDNF)-a gene thought to influence plasticity-contributes to inter-individual variability in responses to continuous theta-burst stimulation (cTBS), and explore whether variability in stimulation-induced plasticity among Val66Met carriers relates to differences in stimulation intensity (SI) used to probe plasticity. Methods: Motor evoked potentials (MEPs) were collected from 33 healthy individuals (11 Val66Met) prior to cTBS (baseline) and in 10 min intervals immediately following cTBS for a total of 30 min post-cTBS (0 min post-cTBS, 10 min post-cTBS, 20 min post cTBS, and 30 min post-cTBS) of the left primary motor cortex. Analyses assessed changes in cortical excitability as a function of BDNF (Val66Val vs. Val66Met) and SI. Results: For both BDNF groups, MEP-suppression from baseline to post-cTBS time points decreased as a function of increasing SI. However, the effect of SI on MEPs was more pronounced for Val66Met vs. Val66Val carriers, whereby individuals probed with higher vs. lower SIs resulted in paradoxical cTBS aftereffects (MEP-facilitation), which persisted at least 30 min post-cTBS administration. Conclusions: cTBS aftereffects among BDNF Met allele carriers are more variable depending on the SI used to probe cortical excitability when compared to homozygous Val allele carriers, which could, to some extent, account for the inconsistency of previously reported cTBS effects. Significance: These data provide insight into the sources of cTBS response variability, which can inform how best to stratify and optimize its use in investigational and clinical contexts.

Subject Gaps Revisited: Complement Clauses and Complementizer-Trace Effects

This study investigates how filler-gap dependencies associated with subject position are formed in online sentence comprehension. Since Crain and Fodor (1985), "filled-gap" studies have provided evidence that the parser actively seeks to associate a wh-filler with a gap in direct object position of a sentence wherever possible; the evidence that this same process applies for subject position, is, however, more limited (Stowe, 1986; Lee, 2004). We examine the processing of complement clauses, finding that wh dependency formation is actively attempted at embedded subject position (e.g., Kate in Who did Lucy think Kate could drive us home to?), unless, however, the embedded clause contains a complementizer (e.g., Who did Lucy think that Kate … .?). The absence of the dependency formation in the latter case demonstrates that the complementizer-trace effect (cf., ∗Who did Lucy think that could drive us home to mom?; Perlmutter, 1968) is, like syntactic island constraints (Ross, 1967; Keshev and Meltzer-Asscher, 2017), immediately operative in online structure building.