A semantic memory sentence verification model based on relative judgment theory

A subjective referent model of sentence verification in semantic memory tasks based on the relative judgment theory of Link and Heath (1975), together with the derivation of a discriminability index, are presented in this paper. An attractive feature of the model is its consideration of both error rates and response times (RTs) in the calculation of the discriminability index. The model is also able to account for the frequent finding in semantic memory tasks that error RTs are longer than correct RTs. A partial replication of Experiment 2 of McCloskey and Glucksberg's (1979) sentence verification context effect studies, in which we employed 44 subjects and 28 categories, and controlled for item familiarity, revealed that error RTs were consistently longer than correct RTs--a finding inconsistent with the McCloskey and Glucksberg property comparison model, but in accord with the subjective referent model. An important fortuitous result was the detection of a context effect by the discriminability measure, an effect not detected by the RT data alone. The discriminability measures yielded a near perfect correlation with estimates of the mean step size of the random walk obtained by application of the parameter estimation program FITTRW (Heath, 1983).     

The comparative judgements of generally and specifically gifted children

Different groups of children were compared on sentence verification tasks. The children were either academically, musically or artistically gifted, and there were two forms of the task. In one, a picture was followed by a sentence, and in another, one sentence was followed by another. Subjects had to decide as quickly as possible whether or not the second proposition logically confirmed the first. In the picture-sentence condition results from all groups could be fitted to the constituent comparison model for sentence verification proposed by Carpenter and Just (1975). For the sentence-sentence condition, however, the observed results diverged from those predicted by the model. The results are explained in terms of different degrees of linguistic processing capacities of the subjects, and they demonstrate the importance which verbal-logical congruence has for children. Artistically able children had difficulties in processing subject/object incongruence in sentence pairs whereas musically able children had more problems in processing above/below incongruence.     

A Method of Correcting Estimation Failure in Latent Differential Equations with Comparisons to Kalman Filtering

Abstract

Studies have used the latent differential equation (LDE) model to estimate the parameters of damped oscillation in various phenomena, but it has been shown that correct, non-zero parameter estimates are only obtained when the latent series exhibits little or no process noise. Consequently, LDEs are limited to modeling deterministic processes with measurement error rather than those with random behavior in the true latent state. The reasons for these limitations are considered, and a piecewise deterministic approximation (PDA) algorithm is proposed to treat process noise outliers as functional discontinuities and obtain correct estimates of the damping parameter. Comprehensive, random-effects simulations were used to compare results with those obtained using a state-space model (SSM) based on the Kalman filter. The LDE with the PDA algorithm (LDEPDA) successfully recovered the simulated damping parameter under a variety of conditions when process noise was present in the latent state. The LDEPDA had greater precision and accuracy than the SSM when estimating parameters from data with sparse jump discontinuities, but worse performance for diffusion processes overall. All three methods were applied to a sample of postural sway data. The basic LDE estimated zero damping, while the LDEPDA and SSM estimated moderate to high damping. The SSM estimated the smallest standard errors for both frequency and damping parameter estimates.     

Polytomous multilevel testlet models for testlet‐based assessments with complex sampling designs

A pplications of standard item response theory models assume local independence of items and persons. This paper presents polytomous multilevel testlet models for dual dependence due to item and person clustering in testlet‐based assessments with clustered samples. Simulation and survey data were analysed with a multilevel partial credit testlet model. This model was compared with three alternative models – a testlet partial credit model (PCM), multilevel PCM, and PCM – in terms of model parameter estimation. The results indicated that the deviance information criterion was the fit index that always correctly identified the true multilevel testlet model based on the quantified evidence in model selection, while the Akaike and Bayesian information criteria could not identify the true model. In general, the estimation model and the magnitude of item and person clustering impacted the estimation accuracy of ability parameters, while only the estimation model and the magnitude of item clustering affected the item parameter estimation accuracy. Furthermore, ignoring item clustering effects produced higher total errors in item parameter estimates but did not have much impact on the accuracy of ability parameter estimates, while ignoring person clustering effects yielded higher total errors in ability parameter estimates but did not have much effect on the accuracy of item parameter estimates. When both clustering effects were ignored in the PCM, item and ability parameter estimation accuracy was reduced.     

Cognitive and neural modeling of dynamics of trust in competitive trustees

Trust dynamics can be modeled in relation to experiences. In this paper two models to represent human trust dynamics are introduced, namely a model on a cognitive level and a neural model. These models include a number of parameters, providing the possibility to express certain relations between trustees. The behavior of each of the models is further analyzed by means of simulation experiments and formal verification techniques. Thereafter, both models have been compared to see whether they can produce patterns that are comparable. As each of the models has its own specific set of parameters, with values that depend on the type of person modeled, such a comparison is non-trivial. To address this, a special comparison approach is introduced, based on mutual mirroring of the models in each other. More specifically, for a given parameter values set for one model, by an automated parameter estimation procedure the most optimal values for the parameter values of the other model are determined in order to show the same behavior. Roughly spoken the results are that the models can mirror each other up to an accuracy of around 90%.     

Eye fixations and cognitive processes

This paper presents a theoretical account of the sequence and duration of eye fixation during a number of simple cognitive tasks, such as mental rotation, sentence verification, and quantitative comparison. In each case, the eye fixation behavior is linked to a processing model for the task by assuming that the eye fixates the referent of the symbol being operated on.     

Sampling plans for fitting the psychometric function

Research on estimation of a psychometric function psi has usually focused on comparing alternative algorithms to apply to the data, rarely addressing how best to gather the data themselves (i.e., what sampling plan best deploys the affordable number of trials). Simulation methods were used here to assess the performance of several sampling plans in yes-no and forced-choice tasks, including the QUEST method and several variants of up-down staircases and of the method of constant stimuli (MOCS). We also assessed the efficacy of four parameter estimation methods. Performance comparisons were based on analyses of usability (i.e., the percentage of times that a plan yields usable data for the estimation of all the parameters of psi) and of the resultant distributions of parameter estimates. Maximum likelihood turned out to be the best parameter estimation method. As for sampling plans, QUEST never exceeded 80% usability even when 1000 trials were administered and rendered accurate estimates of threshold but misestimated the remaining parameters. MOCS and up-down staircases yielded similar and acceptable usability (above 95% with 400-500 trials) and, although neither type of plan allowed estimating all parameters with optimal precision, each type appeared well suited to estimating a distinct subset of parameters. An analysis of the causes of this differential suitability allowed designing alternative sampling plans (all based on up-down staircases) for yes-no and forced-choice tasks. These alternative plans rendered near optimal distributions of estimates for all parameters. The results just described apply when the fitted psi has the same mathematical form as the actual psi generating the data; in case of form mismatch, all parameters except threshold were generally misestimated but the relative performance of all the sampling plans remained identical. Detailed practical recommendations are given.     

高中英语阅读测验中题组模型的选择与应用

本研究通过高中英语阅读测验实测数据,对比分析双参数逻辑斯蒂克模型 (2PL-IRT)和加入不同数量题组的双参数逻辑斯蒂克模型 (2PL-TRT), 探究题组数量对参数估计及模型拟合的影响。结果表明：(1) 2PL-IRT模型对能力介于-1.50到0.50的被试,能力参数估计偏差较大;(2)将题组效应大于0.50的题组作为局部独立题目纳入模型,会导致部分题目区分度参数的低估和大部分题目难度参数的高估;(3)题组效应越大,将其当作局部独立题目纳入模型估计项目参数的偏差越大。     

Dynamics Versus Development in Numerosity Estimation: A Computational Model Accurately Predicts a Developmental Reversal

Perceptual judgments result from a dynamic process, but little is known about the dynamics of number-line estimation. A recent study proposed a computational model that combined a model of trial-to-trial changes with a model for the internal scaling of discrete numbers. Here, we tested a surprising prediction of the model—a situation in which children's estimates of numerosity would be better than those of adults. Consistent with the model simulations, task contexts led to a clear developmental reversal: children made more adult-like, linear estimates when to-be-estimated numbers were descending over trials (i.e., backward condition), whereas adults became more like children with logarithmic estimates when numbers were ascending (i.e., forward condition). In addition, adults’ estimates were subject to inter-trial differences regardless of stimulus order. In contrast, children were not able to use the trial-to-trial dynamics unless stimuli varied systematically, indicating the limited cognitive capacity for dynamic updates. Together, the model adequately predicts both developmental and trial-to-trial changes in number-line tasks.     

Similarity,multiple estimations,and the anchoring effect

This paper compares the Selective Accessibility and Scale Distortion theories of anchoring as explanations for anchoring tasks involving (1) perceived dissimilarity between comparison and estimation objects and (2) successive estimation tasks. We begin by describing the two theories of anchoring and what each would predict for these conditions. Two studies are presented in which multiple estimates are made following a single comparison task and the effect sizes of these estimates are correlated to operationalizations of similarity. In the first study, the stimuli varied with respect to how well they fit within an existing category reasonably familiar to the participant population: aircraft. In the second study, the stimuli varied with respect to external features that did not define the category: the brand and location of hotels. In both studies, we find that the anchoring effect size has a positive correlation with the semantic similarity between the comparison and estimation objects, a finding consistent with Selective Accessibility.     

The production and verification tasks in mental addition: An empirical comparison

We respond to A. Baroody's comment (1984, Developmental Review, 4, 148–156) with an empirical comparison of the production and verification tasks. With the exception of performance at the first grade level, the two tasks yield essentially identical conclusions. The results of an adjunct task, in which the rate of mental counting was assessed, suggest that children as young as second grade are relying on memory retrieval to a significant degree. In contrast to Baroody's speculation, there appear to be no widespread difficulties associated with results from the verification task. Furthermore, the task permits a more analytic examination of performance and underlying mental process than is afforded by the production task. We conclude by reiterating the empirical support for a model of fact retrieval, and suggesting that accessibility of the arithmetic facts is the basic factor which underlies both fact retrieval and procedural knowledge performance.     

Hierarchical coding in the perception and memory of spatial layouts

Two experiments were performed to investigate the organization of spatial information in perception and memory. Participants were confronted with map-like configurations of objects which were grouped by color (Experiment 1) or shape (Experiment 2) so as to induce cognitive clustering. Two tasks were administered: speeded verification of spatial relations between objects and unspeeded estimation of the Euclidean distance between object pairs. In both experiments, verification times, but not distance estimations, were affected by group membership. Spatial relations of objects belonging to the same color or shape group were verified faster than those of objects from different groups, even if the spatial distance was identical. These results did not depend on whether judgments were based on perceptually available or memorized information, suggesting that perceptual, not memory processes were responsible for the formation of cognitive clusters. Received: 7 October 1999 / Accepted: 17 February 2000     

Revisiting dispersion in count data item response theory models: The Conway–Maxwell–Poisson counts model

Count data naturally arise in several areas of cognitive ability testing, such as processing speed, memory, verbal fluency, and divergent thinking. Contemporary count data item response theory models, however, are not flexible enough, especially to account for over- and underdispersion at the same time. For example, the Rasch Poisson counts model (RPCM) assumes equidispersion (conditional mean and variance coincide) which is often violated in empirical data. This work introduces the Conway–Maxwell–Poisson counts model (CMPCM) that can handle underdispersion (variance lower than the mean), equidispersion, and overdispersion (variance larger than the mean) in general and specifically at the item level. A simulation study revealed satisfactory parameter recovery at moderate sample sizes and mostly unbiased standard errors for the proposed estimation approach. In addition, plausible empirical reliability estimates resulted, while those based on the RPCM were biased downwards (underdispersion) and biased upwards (overdispersion) when the simulation model deviated from equidispersion. Finally, verbal fluency data were analysed and the CMPCM with item-specific dispersion parameters fitted the data best. Dispersion parameter estimates indicated underdispersion for three out of four items. Overall, these findings indicate the feasibility and importance of the suggested flexible count data modelling approach.     

Task bias and the accuracy of judgment: Setting a baseline for expected group performance

When a judgment task evokes unbiased estimates (i.e. the errors in individual judgments are distributed randomly around the true value), mathematical aggregation of individual estimates, even by a simple arithmetic mean, often will outperform all group members. However, when a task evokes biased estimates, mathematical aggregation does not perform so well. In this study, simulated data were accumulated to specify the expected' accuracy of mathematical aggregation relative to the accuracy of observed judgment of individual group members under varying conditions of task bias. Three types of judgment tasks were employed: (1) single-estimate, holistic tasks, (2) multiple-estimate, ranking tasks, and (3) multi-cue, decomposed tasks. Findings indicated across all task types that a large percentage of judgment-making group estimates formed strictly by computing the arithmetic mean of individual estimates performed better than their most capable members when a judgment task evoked little or no bias, a result particularly pronounced for ranking tasks. When the task was more greatly bias-evoking, a large percentage of parallel groups performed more poorly than average (or median) members, again a pattern more starkly evident for ranking tasks. These results suggest that the extent to which a judgment task evokes bias in a population of prospective group members is an important explanatory variable deserving much greater attention in the study of group performance. For example, an assertion about the efficacy of a particular group intervention based on a reliable demonstration of group performance as accurate as the most capable members may be unfounded when a task evokes no bias, since the baseline standard under such conditions should be much higher. By selecting tasks and populations that jointly produced highly biased estimates, researchers can lower the performance floor enough to detect (with reasonably small samples of groups) experimental effects should they occur.     

Processing strategies in sentence comprehension

Response latencies in sentence-picture verification tasks were compared as a function of whether a mismatch was located in the logical subject (LS), verb (V), or logical object (LO) of the sentence. Sentences were presented auditorily and varied in voice and reversibility. The comparison process for nonreversibles was clearly serial self-terminating: latencies for both actives and passives were ordered LS < V < LO, or, after practice with a small number of mismatch types, LS < LO < V. Latencies for reversibles were ordered V < LS = LO, suggesting either a verb-first comparison process or an LS-V-LO comparison process which did not terminate with a subject-mismatch because of the confusability of the subject and object. The results attest to the importance of considering the "naturalness" of stimuli in sentence processing tasks, and the flexibility of subjects' encoding and comparison strategies both within and across task contexts.     

Where are the effects of frequency in visual word recognition tasks? Right where we said they were! Comment on Monsell, Doyle, and Haggard (1989)

Balota and Chumbley's studies led them to conclude that category verification, lexical decision, and pronunciation tasks involve combinations of processes that cause them to produce differing estimates of the relation between word frequency and ease of lexical identification. Monsell, Doyle, and Haggard challenged Balota and Chumbley's empirical evidence and conclusions, provided empirical evidence to support their challenge, and presented an alternative theoretical position. We show that Monsell et al.'s experiments, analyses, and theoretical perspective do not result in conclusions about the role of word frequency in category verification, lexical decision, and pronunciation that differ from those of Balota and Chumbley.     

Interval bias in 2AFC detection tasks: sorting out the artifacts

Proportion correct in two-alternative forced choice (2AFC) detection tasks often varies when the stimulus is presented in the first or in the second interval. Reanalysis of published data reveals that these order effects (or interval bias) are strong and prevalent, refuting the standard difference model of signal detection theory. Order effects are commonly regarded as evidence that observers use an off-center criterion under the difference model with bias. We consider an alternative difference model with indecision whereby observers are occasionally undecided and guess with some bias toward one of the response options. Whether or not the data show order effects, the two models fit 2AFC data indistinguishably, but they yield meaningfully different estimates of sensory parameters. Under indeterminacy as to which model governs 2AFC performance, parameter estimates are suspect and potentially misleading. The indeterminacy can be circumvented by modifying the response format so that observers can express indecision when needed. Reanalysis of published data collected in this way lends support to the indecision model. We illustrate alternative approaches to fitting psychometric functions under the indecision model and discuss designs for 2AFC experiments that improve the accuracy of parameter estimates, whether or not order effects are apparent in the data.     

The effect of different methods to construct non-symbolic stimuli in numerosity estimation and comparison

Numerosity estimation and comparison tasks are often used to measure the acuity of the approximate number system (ANS), a mechanism which allows extracting numerosity from an array of dots independently from several visual cues (e.g. area extended by the dots). This idea is supported by studies showing that numerosity can be processed while these visual cues are controlled for. Different methods to construct dot arrays while controlling their visual cues have been proposed in the past. In this paper, these methods were contrasted in an estimation and a comparison task. The way of constructing the dot arrays had little impact on estimation. In contrast, in the comparison task, participants' performance was significantly influenced by the method that was used to construct the arrays of dots, indicating better performance when the visual cues of the dot arrays (partly) co-varied with numerosity. The present study therefore shows that estimates of ANS acuity derived from comparison tasks are inconsistent and dependent on how the stimuli are constructed. This makes it difficult to compare studies which utilised different methods to construct the dot arrays in numerosity comparison tasks. In addition, these results question the currently held view of the ANS as capable of robustly extracting numerosity independently from visual cues.     

Studies of letter identification using qualitative error analysis: effects of speed stress, tachistoscopic presentation, and word context

In Experiment 1 subjects named letters under a response deadline chosen so that an appreciable number of errors would be produced. The stimulus confusions were analyzed via the same mathematical models of stimulus recognition that have been applied to the confusion matrices generated in tachistoscopic experiments. Both the Luce choice model and the informed guessing model (a new model having a simple and elegant process interpretation) provided excellent fits to the data. The parameter values of the informed guessing model changed in logical and interpretable ways with changes in the response deadline. In Experiment 2 a direct comparison was made of the types of errors produced in the data-limited tachistoscopic situation and the resource-limited response deadline situation. It was found that, relative to the response deadline task, identification in the tachistoscopic task is much more likely to be based on partial information. In Experiments 3 and 4 the same research methodology was applied to the problem of the effect of a word context on letter perception. The methodology allowed this problem to be addressed in the context of both response deadline and tachistoscopic tasks. Several advantages of the methodology for investigating other issues of interest to cognitive psychologists are discussed.