Response Time Model Research Articles

Bayesian Procedures for Identifying Aberrant Response-Time Patterns in Adaptive Testing

In order to identify aberrant response-time patterns on educational and psychological tests, it is important to be able to separate the speed at which the test taker operates from the time the items require. A lognormal model for response times with this feature was used to derive a Bayesian procedure for detecting aberrant response times. Besides, a combination of the response-time model with a regular response model in an hierarchical framework was used in an alternative procedure for the detection of aberrant response times, in which collateral information on the test takers’ speed is derived from their response vectors. The procedures are illustrated using a data set for the Graduate Management Admission Test® (GMAT®). In addition, a power study was conducted using simulated cheating behavior on an adaptive test.

A Measurement Model for Likert Responses That Incorporates Response Time

This article describes a model for response times that is proposed as a supplement to the usual factor-analytic model for responses to graded or more continuous typical-response items. The use of the proposed model together with the factor model provides additional information about the respondent and can potentially increase the accuracy of the individual trait estimates. First, the rationale of the model is discussed in relation to previous developments in binary responses. Second, procedures for fitting the model and for assessing model-data fit at both overall and individual level (person-fit) are proposed. Third, the usefulness of the model and its potential applications in the typical-response domain are discussed. All the proposed developments are used in 2 empirical applications in the personality domain. The first application analyzes 2 scales from a Big Five questionnaire. The second example analyzes a sociability scale developed from Eysenck's questionnaires.

An Integrated Model of Discrete Choice and Response Time

Computer and web-based interviewing tools have made response times ubiquitous in marketing research. These data are used as an indicator of data quality by practitioners, and of latent processes related to memory, attributes and decision making by academics. We investigate a Poisson race model with choice and response times as dependent variables. The model facilitates inference about respondent preference for choice alternatives, their diligence in providing responses, and the accessibility of attitudes/the speed of thinking. Thus, the model distinguishes respondents who are quick to think versus those who react quickly but without much thought. Empirically, we find support for the endogenous nature of response times and demonstrate that models that treat response times as exogenous variables may result in misleading inferences.

A Hierarchical Framework for Modeling Speed and Accuracy on Test Items

Current modeling of response times on test items has been strongly influenced by the paradigm of experimental reaction-time research in psychology. For instance, some of the models have a parameter structure that was chosen to represent a speed-accuracy tradeoff, while others equate speed directly with response time. Also, several response-time models seem to be unclear as to the level of parametrization they represent. A hierarchical framework for modeling speed and accuracy on test items is presented as an alternative to these models. The framework allows a plug-and-play approach with alternative choices of models for the response and response-time distributions as well as the distributions of their parameters. Bayesian treatment of the framework with Markov chain Monte Carlo (MCMC) computation facilitates the approach. Use of the framework is illustrated for the choice of a normal-ogive response model, a lognormal model for the response times, and multivariate normal models for their parameters with Gibbs sampling from the joint posterior distribution.

Mathematical modeling of the LCD response time

Abstract— Techniques to reduce LCD motion blur are extensively used in industry and they depend on an inherent LCD parameter: response time. However, normative response time is not a sufficient reference to improve LCD performance and all the gray‐to‐gray response‐time quantities are required to obtain good improvement. However, measuring and gathering all the gray‐to‐gray transitions takes an excessive amount of time. Consequently, we propose a novel LCD model to simulate as well as compute gray‐to‐gray transitions (response time and behavior) from a reduced measurement set in order to decrease the response‐time measurement.

Server Allocation Algorithms for Tiered Systems

Many web-based systems have a tiered application architecture, in which a request needs to transverse all the tiers before finishing its processing. One of the most important QoS metrics for these applications is the expected response time for the user. Since the expected response time in any tier depends upon the number of servers allocated to this tier, and a request's total response time is the sum of the response times over all the tiers, many different configurations (number of servers allocated to each tier) can satisfy the expected response-time requirement. Naturally, one would like to find the configuration that minimizes the total system cost while satisfying the total response-time requirement. This is modeled as a non-linear optimization problem using an open-queuing network model of response time, which we call the server allocation problem for tiered systems (SAPTS). In this paper we study the computational complexity of SAPTS and design efficient algorithms to solve it. For a variable number of tiers, we show that the decision version of SAPTS is NP-complete. Then we design a simple two-approximation algorithm and a fully polynomial-time approximation scheme (FPTAS). If the number of tiers is a constant, we show that SAPTS is polynomial-time solvable. Furthermore, we design a fast polynomial-time exact algorithm to solve the important two-tier case. Most of our results extend to the general case in which each tier has an arbitrary response-time function.

Current impulse response of thin InP p+-i-n+ diodes using full band structure Monte Carlo method

A random response time model to compute the statistics of the avalanche buildup time of double-carrier multiplication in avalanche photodiodes (APDs) using full band structure Monte Carlo (FBMC) method is discussed. The effect of feedback impact ionization process and the dead-space effect on random response time are included in order to simulate the speed of APD. The time response of InP p+-i-n+ diodes with the multiplication region of 0.2μm is presented. Finally, the FBMC model is used to calculate the current impulse response of the thin InP p+-i-n+ diodes with multiplication lengths of 0.05 and 0.2μm using Ramo’s theorem [Proc. IRE 27, 584 (1939)]. The simulated current impulse response of the FBMC model is compared to the results simulated from a simple Monte Carlo model.

An EZ-diffusion model for response time and accuracy

The EZ-diffusion model for two-choice response time tasks takes mean response time, the variance of response time, and response accuracy as inputs. The model transforms these data via three simple equations to produce unique values for the quality of information, response conservativeness, and nondecision time. This transformation of observed data in terms of unobserved variables addresses the speed-accuracy trade-off and allows an unambiguous quantification of performance differences in two-choice response time tasks. The EZ-diffusion model can be applied to data-sparse situations to facilitate individual subject analysis. We studied the performance of the EZ-diffusion model in terms of parameter recovery and robustness against misspecification by using Monte Carlo simulations. The EZ model was also applied to a real-world data set.

Modeling of Responses and Response Times with the Packagecirt

In computerized testing, the test takers' responses as well as their response times on the items are recorded. The relationship between response times and response accuracies is complex and varies over levels of observation. For example, it takes the form of a tradeoff between speed and accuracy at the level of a fixed person but may become a positive correlation for a population of test takers. In order to explore such relationships and test hypotheses about them, a conjoint model is proposed. Item responses are modeled by a two-parameter normal-ogive IRT model and response times by a lognormal model. The two models are combined using a hierarchical framework based on the fact that response times and responses are nested within individuals. All parameters can be estimated simultaneously using an MCMC estimation approach. A R-package for the MCMC algorithm is presented and explained.

On the linear relation between the mean and the standard deviation of a response time distribution.

Although it is generally accepted that the spread of a response time (RT) distribution increases with the mean, the precise nature of this relation remains relatively unexplored. The authors show that in several descriptive RT distributions, the standard deviation increases linearly with the mean. Results from a wide range of tasks from different experimental paradigms support a linear relation between RT mean and RT standard deviation. Both R. Ratcliff's (1978) diffusion model and G. D. Logan's (1988) instance theory of automatization provide explanations for this linear relation. The authors identify and discuss 3 specific boundary conditions for the linear law to hold. The law constrains RT models and supports the use of the coefficient of variation to (a) compare variability while controlling for differences in baseline speed of processing and (b) assess whether changes in performance with practice are due to quantitative speedup or qualitative reorganization.

Aging and Comparative Search for Feature Differences

ABSTRACT In a comparative visual search experiment, two halves of a display contained visual primitives of various shapes and colors. These halves were identical (50% of trials) or contained a non-matching pair (50% of trials). Response time (RT), accuracy, and eye movements were measured in both young and older adults. There were Age Group × Display Size interactions found for RT, with older adult RT affected more than younger adult RT by increases in display size. This interaction was consistent with predictions generated by sequential-sampling models for RT. There were age group main effects on fixation number and fixation duration, but no age group main effects on accuracy, saccade amplitude, or measures of scan-path efficiency; this indicated that search strategies were similar across age groups. Overall, the results showed no special age group deficits for comparative visual search.

Current impulse response of thin InP p+–i–n+ diodes

The simulation of current impulse response using random response time model in avalanche photodiode (APD) is presented. A random response time model considers the randomness of times at which the primary and secondary carriers are generated in multiplication region. The dead-space effect is included in our model to demonstrate the impact on current impulse response of thin APDs. Current impulse response of homojunction InP p+–i–n+ diodes with the multiplication widths of 0.1 and 0.2μm are calculated. Our results show that dead-space gives a slower decay rate of current impulse response in thin APD, which may degrade the bit-error-rate of the optical communication systems.

Experimental and finite-element study of convective accelerometer on CMOS

This paper addresses the design of CMOS thermal accelerometers. A test-chip including the sensor and a signal conditioning circuit has been designed, fabricated and characterized. The fabricated prototype exhibits good performance: the measured resolution is better than 30 mg (equivalent to 1.7° in inclination) with a sensitivity of 375 mV/g. The bandwidth is around 15 Hz. This test-chip is used to improve the modelling of heat transfer phenomenon for this family of devices. FEM is successfully used as a first modelling approach. Although thermal convective accelerometers have already been reported and even commercialised, few information regarding behavioural modelling, optimization issues, and their integration on CMOS technology is today available. In particular, no information about response time modelling was reported in the literature. Thanks to experimental results and FEM simulations, we provide in this study information concerning the integration of convection heat transfer accelerometers on CMOS. In addition, the effects of the packaging (i.e. the top and bottom cavities) on both the static sensitivity and the dynamic response of the sensor have been investigated.

Optimal Partitioning of Testing Time: Theoretical Properties and Practical Implications

This paper deals with optimal partitioning of limited testing time in order to achieve maximum total test score. Nonlinear optimization theory was used to analyze this problem. A general case using a generic item response model is first presented. A special case that applies a response time model proposed by Wang and Hanson (2005) is also presented. Theoretical properties of the optimal solution are derived. Their practical implications to optimal test-taking strategies are also discussed. The theoretical properties are in general agreement with the conventional advice to the examinees on pacing strategy.

Modelling memory processes and Internet response times: Weibull or power-law?

The Weibull distribution is proposed as a model for response times. Theoretical support is offered by classical results for extreme-value distributions. Fits of the Weibull distribution to response time data in different contexts show that this distribution (and the exponential distribution on small time-scales) perform better than the often-suggested power-law and logarithmic function. This study suggests that the power-law can be viewed as an approximation, at neural level, for the aggregate strength of superposed memory traces that have different decay rates in distinct parts of the brain. As we predict, this view does not find support at the level of induced response processes. The distinction between underlying and induced processes might also be considered in other fields, such as engineering, biology and physics.

Shared decision signal explains performance and timing of pursuit and saccadic eye movements

Each voluntary eye movement provides physical evidence of a visuomotor choice about where and when to look. Primates choose visual targets with two types of voluntary eye movements, pursuit and saccades, although the exact mechanism underlying their coordination remains unknown. Are pursuit and saccades guided by the same decision signal? The present study compares pursuit and saccadic choices using techniques borrowed from psychophysics and models of response time. Human observers performed a luminance discrimination task and indicated their choices with eye movements. Because the stimuli moved horizontally and were offset vertically, subjects' tracking responses consisted of combinations of both pursuit and saccadic eye movements. For each of two signal strengths, we constructed speed-accuracy curves for pursuit and saccades. We found that speed-accuracy curves for pursuit and saccades have the same shape, but are time-shifted with respect to one another. We argue that this pattern occurs because pursuit and saccades share a decision signal, but utilize different response thresholds and are subject to different motor processing delays.

Are unshifted distributional models appropriate for response time?

Van Breukelen (this issue) provides an approach to using both response time (RT) and accuracy for (1) measuring latent abilities of participants even when they may trade speed for accuracy, and for (2) providing insight into the psychological processes underlying task performance. In this commentary, I focus on the second of these aims and assess how useful this approach is for exploring cognition. The approach is based on formulating statistical rather than substantive models on accuracy and RT. I focus here on the appropriateness of the RT component model. This RT model joins a family of recent statistical approaches to RT including those of Glickman, Gray, and Morales (in press); Peruggia, Van Zandt, and Chen (2002); Rouder, Lu, Speckman, Sun, and Jiang (in press); Schnipke and Scrams (1997); and Wenger and Gibson (2004). The advantage of these statistical rather than substantive models is robust estimation and sound inference when data are collected across disparate individuals and items. It is not hard to see how these statistical developments will lead to better substantive theory testing and development in cognitive psychology. Van Breukelen’s model for the ith participant’s RT to the j th item is given by:

Psychometrics, Psychonomics, Psychonometrics: Rejoinder To Rouder And Wenger

Given the psychometric focus of this journal and of my paper, it is stimulating to have the comments of two experts in psychonomically oriented response time modeling who do make an explicit distinction between person and item parameters. Let me first of all express my gratitude for their thorough comments on my paper, and my agreement with several of their suggestions. I will of course elaborate on points of disagreement. Rouder (2005) focuses on my restriction to a lognormal RT model with two parameters, a scale parameter μ and a shape parameter σ , of which only the first one is treated as a function of person and item characteristics. A shift parameter is lacking and Rouder gives an example and references to show the need for such a parameter. With respect to the shape parameter, my focus on the scale parameter may have created a misunderstanding that the shape parameter was restricted to be the same for all persons and items. I did explore effects of angle and same/different on σ , the within-subject variance of the log RT, by allowing their regression parameters to vary and/or to covary with the intercept within persons (for details, see chapter 8 of Snijders and Bosker, 1999). None of these (co)variance estimates was significant (all p > 0.10) and none was substantial (all < 10% of the within-subject intercept variance). This was stated very briefly in sections 5.3 and 6.1 of my paper. With respect to the need for a shift parameter, I agree. However, extension of a two-parameter model to a three-parameter model may greatly complicate estimation and testing. This is well-known in the context of logistic IRT models. The guessing parameter of the three-parameter model plays a role similar to the role of a shift parameter in RT models, because it imposes a lower bound. This is considered to be a serious burden for IRT models. In the context of the lognormal model, similar statistical complications may arise (see, e.g., Cohen, 1988). It is therefore with much interest that I have read Rouder’s work on Bayesian inference for the three-parameter Weibull RT model by MCMC methods (e.g. Rouder, Sun, Speckman, Lu, and Zhou, 2003). Statistical and psychometric literature shows an increasing use of such methods and they do appear to become inevitable. However, I still remember Cox’s reaction to a lecture by Rubin involving Gibbs sampling 10 years ago, saying that he liked to keep it simple and to understand a method by writing it out on the back of a cigar box (for part of this discussion, see Cox, 1998). Old-fashioned as this attitude may seem, I have a similar one and I feel that even twenty-first century science cannot do without some of it. Before embracing MCMC estimation of many-parameter models for speed and accuracy, I need to know more about their identifiability, the behavior of MCMC estimation in studies with small numbers of items, say 20, and their added-value for person and item estimation in the context of intelligence testing. Independent of these philosophical considerations, most of the present findings may remain almost unaltered upon adding a shift parameter. For instance, item effects are clear from Fig. 2, and CAF estimation by logistic regression of response correctness on log RT does not require any assumption about the RT distribution. For

Low-Power Design of High-Speed A/D Converters

In this paper, low-power design techniques of high-speed A/D converters are reviewed and discussed. Pipeline and parallel-pipeline architectures are treated as these are dominant architectures when required high sampling rate and high resolution with reasonable power dissipation. A systematic approach to the power optimization of pipeline and parallel pipeline ADC's is introduced based on models of noise analysis and response time of a building block in the multiple-stage pipeline ADC. Finally, the theoretical minimum of required power as functions of the sampling rate, resolution and SNR is discussed. The analysis shows that, with the developments of new circuits and systems to approach to the minimum, the power can be further reduced by a factor of more than 1/10 without changing the basic architectures.

Parsing a cognitive task: a characterization of the mind's bottleneck.

Parsing a mental operation into components, characterizing the parallel or serial nature of this flow, and understanding what each process ultimately contributes to response time are fundamental questions in cognitive neuroscience. Here we show how a simple theoretical model leads to an extended set of predictions concerning the distribution of response time and its alteration by simultaneous performance of another task. The model provides a synthesis of psychological refractory period and random-walk models of response time. It merely assumes that a task consists of three consecutive stages—perception, decision based on noisy integration of evidence, and response—and that the perceptual and motor stages can operate simultaneously with stages of another task, while the central decision process constitutes a bottleneck. We designed a number-comparison task that provided a thorough test of the model by allowing independent variations in number notation, numerical distance, response complexity, and temporal asynchrony relative to an interfering probe task of tone discrimination. The results revealed a parsing of the comparison task in which each variable affects only one stage. Numerical distance affects the integration process, which is the only step that cannot proceed in parallel and has a major contribution to response time variability. The other stages, mapping the numeral to an internal quantity and executing the motor response, can be carried out in parallel with another task. Changing the duration of these processes has no significant effect on the variance.