Incorrect Pairs Research Articles

Effect of poly(dA) stretch in a template sequence on the rate and yield of polymerase chain reaction with electroactive modified 2′-deoxyuridine-5′-triphosphates

The effective incorporation of electroactive labeled nucleotides into DNA during its enzymatic amplification provides the ground for developing assays based on direct electrochemical detection of amplification products. Yet, whether the template sequence can affect the incorporation efficiency of modified nucleotides is still poorly understood. Here we examined the effects of a poly(dA) stretch in template sequence on the rate and yield of polymerase chain reaction (PCR) in the presence of modified 2′-deoxyuridine-5′-triphosphates (dUTP) in the reaction mixture. Four dUTP derivatives carrying 4-hydroxyphenyl (tyrosine; dUTP-Y1 and dUTP-Y2), 4-nitrophenyl (dUTP-N1), or indolyl (tryptophan; dUTP-W1) aromatic groups were tested as substrates for Taq and KTN (lacking 3′-5′ exonuclease activity) polymerases at 100 % substitution of 2′-deoxythymidine-5′-triphosphate (dTTP) in the PCR mixture. The 120 base pair long synthetic DNA templates contained poly(dA)·poly(dT) tracts of various length (dAn·dTn, where n = 2–10). In all cases, full-size amplicons have been obtained. Overall, in the presence of modified dUTP, the PCR yield was higher with KTN polymerase than that with Taq polymerase. The PCR yield slightly (for dUTP-Y1 and dUTP-Y2) or significantly (for dUTP-N1 and dUTP-W1) decreased with an increase of n in the dAn·dTn tract. In most cases, in the presence of modified dUTP, the PCR rate substantially decreased with n ≥ 4 in dAn·dTn. No changes in amplicon sequences due to potential incorrect nucleotide pairing during amplification with the complete substitution of dTTP with dUTP-Y2 were revealed. Electrochemical activity of tyrosine labeled PCR products dsDNA-Y2 differing in the length of stretches of modified nucleotides was similar. The findings demonstrate that the incorporation efficiency for base-modified nucleotides can in general depend on template sequence, in particular on the presence of long poly(dA) stretches when the modified dUTP are used. That has to be taken into account when developing electrochemical or other DNA assays with a specific modified nucleotide.

An algorithm for decoy-free false discovery rate estimation in XL-MS/MS proteomics.

Cross-linking tandem mass spectrometry (XL-MS/MS) is an established analytical platform used to determine distance constraints between residues within a protein or from physically interacting proteins, thus improving our understanding of protein structure and function. To aid biological discovery with XL-MS/MS, it is essential that pairs of chemically linked peptides be accurately identified, a process that requires: (i) database search, that creates a ranked list of candidate peptide pairs for each experimental spectrum and (ii) false discovery rate (FDR) estimation, that determines the probability of a false match in a group of top-ranked peptide pairs with scores above a given threshold. Currently, the only available FDR estimation mechanism in XL-MS/MS is the target-decoy approach (TDA). However, despite its simplicity, TDA has both theoretical and practical limitations that impact the estimation accuracy and increase run time over potential decoy-free approaches (DFAs). We introduce a novel decoy-free framework for FDR estimation in XL-MS/MS. Our approach relies on multi-sample mixtures of skew normal distributions, where the latent components correspond to the scores of correct peptide pairs (both peptides identified correctly), partially incorrect peptide pairs (one peptide identified correctly, the other incorrectly), and incorrect peptide pairs (both peptides identified incorrectly). To learn these components, we exploit the score distributions of first- and second-ranked peptide-spectrum matches for each experimental spectrum and subsequently estimate FDR using a novel expectation-maximization algorithm with constraints. We evaluate the method on ten datasets and provide evidence that the proposed DFA is theoretically sound and a viable alternative to TDA owing to its good performance in terms of accuracy, variance of estimation, and run time. https://github.com/shawn-peng/xlms.

Associations of computer gaming with incident dementia, cognitive functions, and brain structure: a prospective cohort study and Mendelian randomization analysis

BackgroundComputer gaming has recently been suggested to be associated with benefits for cognition, but its impact on incident dementia remains uncertain. We aimed to investigate the observational associations of playing computer games with incident dementia, cognitive functions, and brain structural measures, and further explore the genetic associations between computer gaming and dementia.MethodsWe included 471,346 White British participants without dementia at baseline based on the UK Biobank, and followed them until November 2022. We estimated the risk of dementia using Cox proportional hazard models, and assessed the changes of cognitive functions and brain structural measures using logistic regression models and linear regression models. Mendelian randomization (MR) analyses were performed to examine the association between genetically determined computer gaming and dementia.ResultsHigh frequency of playing computer games was associated with decreased risk of incident dementia (HR, 0.81 [95% CI: 0.69, 0.94]). Individuals with high frequency of playing computer games had better performance in prospective memory (OR, 1.46 [1.26, 1.70]), reaction time (beta, -0.195 [-0.243, -0.147]), fluid intelligence (0.334 [0.286, 0.382]), numeric memory (0.107 [0.047, 0.166]), incorrect pairs matching (-0.253 [-0.302, -0.203]), and high volume of gray matter in hippocampus (0.078 [0.023, 0.134]). Genetically determined high frequency of playing computer games was associated with a low risk of dementia (OR, 0.37 [0.15, 0.91]).ConclusionsComputer gaming was associated with a decreased risk of dementia, favorable cognitive function, and better brain structure, suggesting that computer gaming could modulate cognitive function and may be a promising target for dementia prevention.

Identification and quantification of chain-pairing variants or mispaired species of asymmetric monovalent bispecific IgG1 monoclonal antibody format using reverse-phase polyphenyl chromatography coupled electrospray ionization mass spectrometry

Developing a knob-into-hole asymmetric bispecific IgG1 monoclonal antibody (mAb) poses manufacturing challenges due to the expression of chain pairing variants, also called mispaired species, in the desired product. The incorrect pairing of light and heavy chains could result in heterogeneous mispaired species of homodimers, heterodimers, light chain swapping, and low molecular weight species (LMWS). Standard chromatography, capillary electrophoretic, or spectroscopic methods poorly resolve these from the main variants. Here, we report a highly sensitive reverse-phase polyphenyl ultra-high-performance liquid chromatography (RP-UHPLC) method to accurately measure mispaired species of Duet mAb format, an asymmetric IgG1 bispecific mAb, for both process development and quality control analytical tests. Coupled with electrospray ionization mass spectrometry (ESI-MS), it enabled direct online characterization of mispaired species. This single direct assay detected diverse mispaired IgG-like species and LMWS. The method resolved eight disulfide bonds dissociated LMWS and three mispaired LMWS. It also resolved three different types of IgG-like mispaired species, including two homodimers and one heterodimer. The characterization and quantification simultaneously enabled the cell line selection that produces a lesser heterogeneity and lower levels of mispaired species with the desired correctly paired product. The biological activity assessment of samples with increased levels of these species quantified by the method exhibited a linear decline in potency with increasing levels of mispaired species in the desired product. We also demonstrated the utility of the technique for testing in-process intermediate materials to determine and assess downstream purification process capability in removing diverse mispaired IgG-like species and LMWS to a certain level during the downstream purification process. Our investigation demonstrates that adopting this method was vital in developing asymmetric bispecific mAb from the initial stage of cell line development to manufacturing process development. Therefore, this tool could be used in the control strategy to monitor and control mispaired species during manufacturing, thus improving the quality control of the final product.

Construction of IgG–Fab2 bispecific antibody via intein-mediated protein trans-splicing reaction

A bispecific antibody (bsAb) is a class of engineered antibody molecules that simultaneously binds to two different antigens by having two kinds of antigen-binding domains. One of the major obstacles for the bsAb production is the incorrect chain-pairing problem, wherein each heavy and light chain should form pairings with the correct counterpart’s chains, but the structural similarity of the incorrect partners also forms the incorrect pairings. This study aimed to demonstrate a bsAb construction method using intein-mediated protein trans-splicing to create IgG–Fab2–type bsAbs, which is a modified antibody with a structure in which two additional Fabs are linked to the N-terminus of the heavy chain of an IgG molecule. The chain-paring problem between a heavy chain and a light chain is circumvented by separate expression and purification of the IgG part and the Fab part. We found that the deletion of a possible glycosylation residue improved the reaction yield and side-reaction cleavage in the protein ligation step. The resulting bsAb, IgG–Fab2 (Her2/CD3), demonstrated target binding activity and cytotoxicity mediated by activated T cells. These results indicate that the use of the protein ligation to produce the IgG–Fab2 type bsAb will expand the bsAb production method.

Backbone Conformational Equilibrium in Mismatched DNA Correlates with Enzyme Activity.

T:G mismatches in mammals arise primarily from the deamination of methylated CpG sites or the incorporation of improper nucleotides. The process by which repair enzymes such as thymine DNA glycosylase (TDG) identify a canonical DNA base in the incorrect pairing context remains a mystery. However, the abundant contacts of the repair enzymes with the DNA backbone suggest a role for protein-phosphate interaction in the recognition and repair processes, where conformational properties may facilitate the proper interactions. We have previously used 31P NMR to investigate the energetics of DNA backbone BI-BII interconversion and the effect of a mismatch or lesion compared to canonical DNA and found stepwise differences in ΔG of 1-2 kcal/mol greater than equivalent steps in unmodified DNA. We have currently compared our results to substrate dependence for TDG, MBD4, M. HhaI, and CEBPβ, testing for correlations to sequence and base-pair dependence. We found strong correlations of our DNA phosphate backbone equilibrium (Keq) to different enzyme kinetics or binding parameters of these varied enzymes, suggesting that the backbone equilibrium may play an important role in mismatch recognition and/or conformational rearrangement and energetics during nucleotide flipping or other aspects of enzyme interrogation of the DNA substrate.

Automatic Resource Augmentation for Machine Translation in Low Resource Language: EnIndic Corpus

Parallel corpus is the primary ingredient of machine translation. It is required to train the statistical machine translation (SMT) and neural machine translation (NMT) systems. There is a lack of good quality parallel corpus for Hindi to English. Comparable corpora for a given language pair are comparatively easy to find, but this cannot be used directly in SMT or NMT systems. As a result, we generate a parallel corpus from the comparable corpus. For this purpose, the sentences (which are translations of each other) are mined from the comparable corpus to prepare the parallel corpus. The proposed algorithm uses the length of the sentence and word translation model to align sentence pairs that are translations of each other. Then, the sentence pairs that are poor translations of each other (measured by a similarity score based on IBM model 1 translation probability) are filtered out. We apply this algorithm to comparable corpora, which are crawled from speeches of the President and Vice-President of India, and mined parallel corpora out of them. The prepared parallel corpus contains good quality aligned sentences (with 96.338% f-score). Subsequently, incorrect sentence pairs are filtered out manually to make the corpus in qualitative practical use. Finally, we gather various sentences from different sources to prepare the EnIndic corpus, which comprises 1,656,207 English-Hindi sentence pairs (miscellaneous domain). We have deployed this prepared largest English-Hindi parallel corpus at https://github.com/debajyoty/EnIndic.git and the source code at https://github.com/debajyoty/EnIndicSourceCode.git.

Highly mismatch-tolerant homology testing by RecA could explain how homology length affects recombination.

In E. coli, double strand breaks (DSBs) are resected and loaded with RecA protein. The genome is then rapidly searched for a sequence that is homologous to the DNA flanking the DSB. Mismatches in homologous partners are rare, suggesting that RecA should rapidly reject mismatched recombination products; however, this is not the case. Decades of work have shown that long lasting recombination products can include many mismatches. In this work, we show that in vitro RecA forms readily observable recombination products when 16% of the bases in the product are mismatched. We also consider various theoretical models of mismatch-tolerant homology testing. The models test homology by comparing the sequences of Ltest bases in two single-stranded DNAs (ssDNA) from the same genome. If the two sequences pass the homology test, the pairing between the two ssDNA becomes permanent. Stringency is the fraction of permanent pairings that join ssDNA from the same positions in the genome. We applied the models to both randomly generated genomes and bacterial genomes. For both randomly generated genomes and bacterial genomes, the models show that if no mismatches are accepted stringency is ∼ 99% when Ltest = 14 bp. For randomly generated genomes, stringency decreases with increasing mismatch tolerance, and stringency improves with increasing Ltest. In contrast, in bacterial genomes when Ltest ∼ 75 bp, stringency is ∼ 99% for both mismatch-intolerant and mismatch-tolerant homology testing. Furthermore, increasing Ltest does not improve stringency because most incorrect pairings join different copies of repeats. In sum, for bacterial genomes highly mismatch tolerant homology testing of 75 bp provides the same stringency as homology testing that rejects all mismatches and testing more than ∼75 base pairs is not useful. Interestingly, in vivo commitment to recombination typically requires homology testing of ∼ 75 bp, consistent with highly mismatch intolerant testing.

A Technique for Determining Three-Dimensional Storm Cloud-Top Locations Using Stereo Optical Lightning Pulses Observed from Orbit

AbstractWe have developed a technique to estimate the three-dimensional (3D) location of lightning optical pulses based on the stereo view of common lightning pulses from two different orbital instruments. The technique only requires the satellite position and the look vector to the lightning optical source. An example dataset of the Geostationary Lightning Mappers (GLMs) onGOES-16andGOES-17from 10 June 2019 is used to illustrate the technique. For this dataset, we find that the values for the stereo determination of cloud-top altitudes are on average lower by 740 m than the ones calculated from the lightning ellipsoid that is currently applied during geolocation. When we compare the locations to the Advanced Baseline Imager (ABI) Cloud Height Algorithm (ACHA), we find that our technique also produces slightly lower altitude values by 240 m. There is greater spread in our technique than either the lightning ellipsoid or the ABI cloud-top height that is likely due to the incorrect pairing of groups between the two GLMs and the 8–14-km resolution in the group locations. Based on GLM location errors derived from comparisons to ground truth sources, the uncertainty in the radial location determined by the stereo location technique is 5.2 km, while the altitude uncertainty is 4.0 km. The technique can be used to 3D map lightning or other optical sources such as bolides and other upper-atmospheric optical phenomena from any two orbital sensors with overlapping fields of view.

Flexible constraint hierarchy during the visual encoding of tool-object interactions.

Tools and objects are associated with numerous action possibilities that are reduced depending on the task-related internal and external constraints presented to the observer. Action hierarchies propose that goals represent higher levels of the hierarchy while kinematic patterns represent lower levels of the hierarchy. Prior work suggests that tool-object perception is heavily influenced by grasp and action context. The current study sought to evaluate whether the presence of action hierarchy can be perceptually identified using eye tracking during tool-object observation. We hypothesize that gaze patterns will reveal a perceptual hierarchy based on the observed task context and grasp constraints. Participants viewed tool-objects scenes with two types of constraints: task-context and grasp constraints. Task-context constraints consisted of correct (e.g., frying pan-spatula) and incorrect tool-object pairings (e.g., stapler-spatula). Grasp constraints involved modified tool orientations, which requires participants to understand how initially awkward grasp postures can help achieve the task. The visual scene contained three areas of interests (AOIs): the object, the functional tool-end (e.g., spoon handle) and the manipulative tool-end (e.g., spoon bowl). Results revealed two distinct processes based on stimuli constraints. Goal-oriented encoding, the attentional bias towards the object and manipulative tool-end, was demonstrated when grasp did not lead to meaningful tool-use. In images where grasp postures were critical to action performance, attentional bias was primarily between the object and functional tool-end, which suggests means-related encoding of the graspable properties of the object. This study expands from previous work and demonstrates a flexible constraint hierarchy depending on the observed task constraints.

Genetics of premature ovarian insufficiency and the association with X-autosome translocations.

Premature ovarian insufficiency (POI) is the cessation of menstruation before the age of 40 and can result from different etiologies, including genetic, autoimmune, and iatrogenic. Of the genetic causes, single-gene mutations and cytogenetic alterations, such as X-chromosome aneuploidies and chromosome rearrangements, can be associated with POI. In this review, we summarize the genetic factors linked to POI and list the main candidate genes. We discuss the association of these genes with the ovarian development, the functional consequences of different mutational mechanisms and biological processes that are frequently disrupted during POI pathogenesis. Additionally, we focus on the high prevalence of X-autosome translocations involving the critical regions in Xq, known as POI1 and POI2, and ddiscuss in depth the main hypotheses proposed to explain this association. Although the incorrect pairing of chromosomes during meiosis could lead to oocyte apoptosis, the reason for the prevalence of X-chromosome breakpoints at specific regions remains unclear. In most cases, studies on genes disrupted by balanced structural rearrangements cannot explain the ovarian failure. Thus, the position effect has emerged as a putative explanation for genetic mechanisms as translocations possibly result in changes in overall chromatin topology due to chromosome repositioning. Given the tremendous impact of POI on women's quality of life, we highlight the value of investigations in to the interplay between ovarian function and gene regulation to deepen our understanding of the molecular mechanisms related to this disease, with the ultimate goal of improving patients' care and assistance.

Resolving issues of scaling for gramian-based input–output pairing methods

A key problem in process control is to decide which inputs should control which outputs. There are multiple ways to solve this problem, among them using gramian-based measures, which include the Hankel interaction index array, the participation matrix and the method. The gramian-based measures, however, have issues with input and output scaling. Generally, this is resolved by scaling all inputs and outputs to have equal range. However, we demonstrate how this can result in an incorrect pairing and examine alternative methods of scaling the gramian-based measures, using either row or column sums or by utilising the Sinkhorn-Knopp algorithm. To systematically analyse the benefits of the scaling schemes, a multiple-input multiple-output model generator is used to test the different schemes on a large number of systems. This assessment shows considerable benefits to be gained from the alternative scaling of the gramian-based measures, especially when using the Sinkhorn-Knopp algorithm.

Corrigendum] FNDC3B promotes epithelial-mesenchymal transition in tongue squamous cell carcinoma cells in a hypoxic microenvironment.

During the preparation of the above article, the authors inadvertently selected an incorrect pair of data panels for Fig.5B. Essentially, in the published version of Fig.5B, the images were presented incorrectly for the NC migration and KD#3 migration and invasion data panels associated with the TSCCA cell Transwell assay experiments. Furthermore, in Fig.7A, the GAPDH control bands were erroneously selected for the figure. Figs.5 and 7 as they should have appeared in the Journal are shown opposite, incorporating the correct data for Figs.5B and 7A. These errors did not affect either the results or the conclusions of this work. The authors all agree to this Corrigendum, and thank the Editor of OncologyReports for allowing them to have the opportunity to present their corrected data. Furthermore, the authors apologize to the readership for any inconvenience these errors may have caused. [the original article was published in Oncology Reports 39: 1853‑1859, 2018; DOI: 10.3892/or.2018.6231].

Sperm telomere length: Diagnostic and prognostic biomarker in male infertility (Review)

Telomere length is an important biomarker in a number of diseases, including male infertility. A decreased telomere length has been reported in several diseases and telomere shortening may occur due to aging, decreased telomerase activity, oxidative stress or cell division. In recent years, several studies have indicated that males with infertility have a shorter sperm telomere length than fertile males. Sperm telomere shortening is associated with male infertility through several mechanisms, including the apoptosis of spermatozoa, decreased motility, low sperm count, incorrect chromosomal pairing and movement during meiosis, and failed fertilization. The aim of this review was to compile current findings on sperm telomere length and discuss findings to compare sperm from infertile males with that of fertile males. Several studies reported shortening of sperm telomeres associated with infertility. Thus, sperm telomere length can be used as biomarker for the diagnosis and prognosis of male infertility, since fertile males have longer telomeres and the length decreases with age. However, there is no specific telomere length that is set as standard/recommended length.

Interference Effects of Emotional Arousal on Recognition Memory: Comparison of Within-object and Between-object Bindings

According to the Object-based Binding Theory, emotional arousal enhances within-object binding, but impairs or does not affect between-object memory binding. The present study aimed to investigate the effects of negative emotional arousal on memory updating of within-object and between-object associations. Negative emotional arousal was expected to cause a higher level of interference in intrinsic associations (within object condition) than in extrinsic associations (between object condition). Angry, sad and neutral facial expressions chosen for emotion conditions are paired with various hats as neutral objects. In the within-object condition, the hat was presented together with a face, while in the between-object condition, the hat was presented as a separate object distinct from the face. First, all participants learned a list of face-hat pairs, then learned a second list with the same faces paired with new hats. In the test phase, the second list of pairs was rearranged into half correct and half incorrect, and the participants decided whether the pairs were correct or incorrect. The data were analyzed on the basis of Signal Detection Theory. We found that higher false alarms were given to angry faces in the within-object condition than in the between-object condition. This high false alarm rate indicates that negative emotional arousal prevented memory updating of within-object associations. According to analysis of memory sensitivity, in all emotion conditions, correct and incorrect pairs were better discriminated in the between-object condition than in the within-object condition. When false alarm rates and sensitivity results are considered together, the hypotheses of the study are supported and it is seen that negative emotional arousal leads to proactive interference in the within-object condition and thus makes it difficult to update memory. These results are consistent with the OBT in that negative emotional arousal exerts an enhancing effect on remembering intrinsic features of objects.

Blood transcriptome analysis in a buck-ewe hybrid and its parents

Examples of living sheep-goat hybrids are rare, mainly due to incorrect chromosome pairing, which is thought to be the main cause for species incompatibility. This case represents the first report of a buck-ewe hybrid and the first mammalian hybrid to be analyzed with next generation sequencing. The buck-ewe hybrid had an intermediate karyotype to the parental species, with 57 chromosomes. Analysis of the blood transcriptomes of the hybrid and both parents revealed that gene expression levels differed between the hybrid and its parents. This could be explained in part by age-dependent differences in gene expression. Contribution to the geep transcriptome was larger from the paternal, compared to the maternal, genome. Furthermore, imprinting patterns deviated considerably from what is known from other mammals. Potentially deleterious variants appeared to be compensated for by monoallelic expression of transcripts. Hence, the data imply that the buck-ewe hybrid compensated for the phylogenetic distance between the parental species by several mechanisms: adjustment of gene expression levels, adaptation to imprinting incompatibilities, and selective monoallelic expression of advantageous transcripts. This study offers a unique opportunity to gain insights into the transcriptome biology and regulation of a hybrid mammal.

Removal of half antibody, hole-hole homodimer and aggregates during bispecific antibody purification using MMC ImpRes mixed-mode chromatography

During recombinant production of asymmetric IgG-like bispecific antibodies (bsAbs), various by-products are often observed due to unbalanced chain expression and incorrect chain pairing. Among them, half antibody and homodimer are found with high frequency. In this work, with a case study we demonstrated that Capto MMC ImpRes mixed-mode chromatography can effectively remove these two by-products as well as antibody aggregates under optimized conditions. This makes MMC ImpRes a powerful tool for bsAb purification.

Genetic and genomic basis of the mismatch repair system involved in Lynch syndrome.

Lynch syndrome is a cancer-predisposing syndrome inherited in an autosomal-dominant manner, wherein colon cancer and endometrial cancer develop frequently in the family, it results from a loss-of-function mutation in one of four different genes (MLH1, MSH2, MSH6, and PMS2) encoding mismatch repair proteins. Being located immediately upstream of the MSH2 gene, EPCAM abnormalities can affect MSH2 and cause Lynch syndrome. Mismatch repair proteins are involved in repairing of incorrect pairing (point mutations and deletion/insertion of simple repetitive sequences, so-called microsatellites) that can arise during DNA replication. MSH2 forms heterodimers with MSH6 or MSH3 (MutSα, MutSβ, respectively) and is involved in mismatch-pair recognition and initiation of repair. MLH1 forms a complex with PMS2, and functions as an endonuclease. If the mismatch repair system is thoroughly working, genome integrity is maintained completely. Lynch syndrome is a state of mismatch repair deficiency due to a monoallelic abnormality of any mismatch repair genes. The phenotype indicating the mismatch repair deficiency can be frequently shown as a microsatellite instability in tumors. Children with germline biallelic mismatch repair gene abnormalities were reported to develop conditions such as gastrointestinal polyposis, colorectal cancer, brain cancer, leukemia, etc., and so on, demonstrating the need to respond with new concepts in genetic counseling. In promoting cancer genome medicine in a new era, such as by utilizing immune checkpoints, it is important to understand the genetic and genomic molecular background, including the status of mismatch repair deficiency.

Using collocation clusters to detect and correct English L2 learners’ collocation errors

In this article, we describe an online English collocation explorer developed to help English L2 learners produce correct and appropriate collocations. Our tool, which is able to visually represent relevant correct/incorrect collocations on a single webpage, was designed based on the notions of collocation clusters and intercollocability proposed by Cowie and Howarth. As they pointed out, in a collocation cluster L2 learners generally cannot distinguish true collocations (e.g., tell truth, state truth, and state fact) from impossible combinations (e.g., *say fact and *say truth). Accordingly, our tool applies natural language processing techniques to construct collocation clusters to enable learners to easily differentiate between correct and incorrect pairs. Relying on data from a reference corpus, our system instantaneously processes the collocability of users’ target combination (verb–noun or adj–noun) and all other relevant words and presents true/false collocations that L2 learners should master/avoid. To assess our tool, we investigated its performance in detecting and correcting learners’ V–N and A–N errors, with results comparable to those of most previous studies. Piloted using a sample of 13 intermediate- or upper-intermediate level English as a foreign language learners, our tool was found to help them self-correct their collocation errors effectively. Compared with similar tools or approaches, our tool requires much less data resources, but still demonstrates a remarkable capability to detect/correct errors and generate useful collocational knowledge in English.

Identifying Accurate and Inaccurate Stimulus Relations: Human and Computer Learning

In three experiments, we explore human and simulated participants’ potential for deriving and merging analogous forms of stimulus relations. In the first experiment, five human participants were exposed to compound stimuli (stimulus pairs) by way of an automated yes–no protocol. Participants received discrimination training focusing on four three-member stimulus classes, where only two of the four classes were correctly related algebraic expressions. Training was intended to establish generalized identification of novel correct stimulus pairs and generalized identification of novel incorrect stimulus pairs. In Experiment 2, we employed a three-layer connectionist model (CM) of a yes–no protocol aimed at training and testing an analogous set of stimulus relations. Our procedures were aimed at assessing a neural network’s ability to simulate derived stimulus relations consistent with the human performances observed in Experiment 1. In Experiment 3, we employed a four-layer CM to compute the number of training epochs required to attain mastery. As with our human participants, our neural network required specific training procedures to become proficient in identifying stimuli as being members or nonmembers of specific classes. Outcomes from Experiment 3 suggest that the number of training epochs required to attain mastery for our simulated participants corresponded closely with the number of training trials required of our human participants during Experiment 1. Moreover, generalization tests revealed that human and simulated participants exhibited analogous response patterns. We discuss the evolving potential for CMs to emulate and predict human training requirements for deriving and merging complex stimulus relations during generalization tests.