What are the shapes of response time distributions in visual search?

Palmer, Evan M.; Horowitz, Todd S.; Torralba, Antonio; Wolfe, Jeremy M.

doi:10.1037/a0020747

Cited by 162 publications

(162 citation statements)

References 65 publications

Supporting

Mentioning

153

Contrasting

Order By: Relevance

“…For example, the skew-t distribution had a lower μ for "Response box" than for "Keyboard" when six squares were shown, whereas for lower set sizes the situation was the reverse. Thus, we decided to use the ex-Gaussian distribution in our analysis, especially given the fact that this distribution has previously been shown to be useful in analyses of RTs in visual search tasks (Kristjánsson & Jóhannesson, 2014;Palmer, Horowitz, Torralba, & Wolfe, 2011). 1 Here and later, we report the results for a random-intercept model that did not include random effects for regression slopes.…”

Section: Response Times: Means and Standard Deviationsmentioning

confidence: 99%

Online versus offline: The Web as a medium for response time data collection

Chetverikov

Upravitelev²

2015

Behav Res

View full text Add to dashboard Cite

The Internet provides a convenient environment for data collection in psychology. Modern Web programming languages, such as JavaScript or Flash (ActionScript), facilitate complex experiments without the necessity of experimenter presence. Yet there is always a question of how much noise is added due to the differences between the setups used by participants and whether it is compensated for by increased ecological validity and larger sample sizes. This is especially a problem for experiments that measure response times (RTs), because they are more sensitive (and hence more susceptible to noise) than, for example, choices per se. We used a simple visual search task with different set sizes to compare laboratory performance with Web performance. The results suggest that although the locations (means) of RT distributions are different, other distribution parameters are not. Furthermore, the effect of experiment setting does not depend on set size, suggesting that task difficulty is not important in the choice of a data collection method. We also collected an additional online sample to investigate the effects of hardware and software diversity on the accuracy of RT data. We found that the high diversity of browsers, operating systems, and CPU performance may have a detrimental effect, though it can partly be compensated for by increased sample sizes and trial numbers. In sum, the findings show that Web-based experiments are an acceptable source of RT data, comparable to a common keyboard-based setup in the laboratory.

show abstract

Section: Response Times: Means and Standard Deviationsmentioning

confidence: 99%

Online versus offline: The Web as a medium for response time data collection

Chetverikov

Upravitelev²

2015

Behav Res

View full text Add to dashboard Cite

show abstract

“…The shape of the distribution of response time and its connection to complexity of the search (e.g. how difficult it is to discriminate between target and background) has recently been found to be a good way to disclose the efficiency and memory capacity of the putative internal tagging of previously scanned objects Palmer et al, 2011). For instance, a completely amnesic serial search with no internal tagging of already checked non-target objects would produce exponentially distributed response times, and partial tagging or a restricted memory for just a few previously attended items would translate into response time distributions becoming more skewed (e.g.…”

Section: Introductionmentioning

confidence: 99%

Visual search in hunting archerfish shares all hallmarks of human performance

Rischawy

Schuster

2013

Journal of Experimental Biology

View full text Add to dashboard Cite

SUMMARYArcherfish are renowned for shooting down aerial prey with water jets, but nothing is known about how they spot prey items in their richly structured mangrove habitats. We trained archerfish to stably assign the categories 'target' and 'background' to objects solely on the basis of non-motion cues. Unlike many other hunters, archerfish are able to discriminate a target from its background in the complete absence of either self-motion or relative motion parallax cues and without using stored information about the structure of the background. This allowed us to perform matched tests to compare the ways fish and humans scan stationary visual scenes. In humans, visual search is seen as a doorway to cortical mechanisms of how attention is allocated. Fish lack a cortex and we therefore wondered whether archerfish would differ from humans in how they scan a stationary visual scene. Our matched tests failed to disclose any differences in the dependence of response time distributions, a most sensitive indicator of the search mechanism, on number and complexity of background objects. Median and range of response times depended linearly on the number of background objects and the corresponding effective processing time per item increased similarlyapproximately fourfold -in both humans and fish when the task was harder. Archerfish, like humans, also systematically scanned the scenery, starting with the closest object. Taken together, benchmark visual search tasks failed to disclose any difference between archerfish -who lack a cortex -and humans.

show abstract

“…The findings of the kurtosis coefficient β > 0 ( Figures 1-4) presented the leptocurtic RT distributions and indicated a larger number of RTs around the RT mean. The shapes of the non-normal RT distribution revealed for both visual simple and choice reaction tests in the study correspond to the respected model of the nonGaussian RT distribution (Palmer & Horowitz, 2011;Whelan, 2008).…”

Section: Discussionmentioning

confidence: 99%

The visual reaction time distribution in the tasks with different demands on information processing

Psotta¹

2014

Acta Gymnica

View full text Add to dashboard Cite

Background: Reaction tests are a part of the neuropsychological assessment of the individual. The design of these tests and selection of the appropriate measures of reaction time (RT) should come out from the distribution qualities of RTs. According to the newer general theory of RT distribution, the quality of RTs also depends on the specific properties of a reaction task. Objective: The aim of the study was to examine the dependency of RT distribution on an amount of processed information in a visual reaction test, and to verify the stability of RT distribution between two series of RTs in male adolescents. Methods: 25 male adolescents, aged 17.1 ± 1.1 years, performed four different eye-hand visual reaction tests which differed in a number of choices -simple reaction test (RT 1 ), two, three and fourchoice RT tests (RT 2 , RT 3, RT 4 ). Each test involved 2 series with 20 reaction trials in each series, using the computerbased reactometer device. The first, second and third moments of the RT distribution were assessed -the measures of central tendency (mean, median), variability (% CV, median absolute deviation/median ratio -MAD/Mdn, interquartile range) and normality (Kolmogorov-Smirnov test, skewness and kurtosis coefficient α and β). Significance of differences of RTs was assessed by the Wilcoxon test (α = .05). Results: The analysis showed the non-Gaussian distributions of RTs (α = .01) with the skew to the right (α = 1.18-4.38) and leptocurtic distribution (β = 1.89-34.15) in all types of the RT tests. The measures of RT variability % CV and MAD/Mdn (%) were lower for the RT 1 test as compared to the RT 2 , RT 3 and RT 4 tests. No significant differences in RTs measured in the 1 st and 2 nd series of trials were found in the RT 1 test and RT 2 test in contrast to significantly shorter RTs in the 2 nd series of trials in the RT 3 test and RT 4 test (p = .006 and p < .001). RTs measured in the 2 nd series of trials of RT 1 , RT 2 and RT 4 tests manifested distribution with a higher skewness to the right and higher peakedness than the RTs in the 1 st series. Conclusions: The study supported the hypothesis on asymmetric distribution of RTs, specifically when measured in the both eye-hand visual simple and multiple-choice reaction tasks with two up to four choices. Thus, the nonparametric statistics show to be more appropriate for analysis of RTs than the parametric statistics. For both clinical and research purposes, the two series of twenty reaction trials in a computer-based test can provide a sufficient number of the RT data for reliable assessment of visual simple and choice-reaction abilities in the male adolescents.

show abstract

What are the shapes of response time distributions in visual search?

Cited by 162 publications

References 65 publications

Online versus offline: The Web as a medium for response time data collection

Online versus offline: The Web as a medium for response time data collection

Visual search in hunting archerfish shares all hallmarks of human performance

The visual reaction time distribution in the tasks with different demands on information processing

Contact Info

Product

Resources

About