Measures and Interpretations of Vigilance Performance: Evidence against the Detection Criterion

1998 ◽  
Vol 40 (4) ◽  
pp. 601-623 ◽  
Author(s):  
J. D. Balakrishnan
Keyword(s):  
Signal Detection ◽  
False Alarm ◽  
Traffic Control ◽  
Signal Detection Theory ◽  
Vigilance Task ◽  
Detection Theory ◽  
Time On Task ◽  
Signal Rate ◽  
Detection Criterion ◽  
False Alarm Rates

Operators′ performance in a vigilance task is often assumed to depend on their choice of a detection criterion. When the signal rate is low this criterion is set high, causing the hit and false alarm rates to be low. With increasing time on task the criterion presumably tends to increase even further, thereby further decreasing the hit and false alarm rates. Virtually all of the empirical evidence for this simple interpretation is based on estimates of the bias measure β from signal detection theory. In this article, I describe a new approach to studying decision making that does not require the technical assumptions of signal detection theory. The results of this new analysis suggest that the detection criterion is never biased toward either response, even when the signal rate is low and the time on task is long. Two modifications of the signal detection theory framework are considered to account for this seemingly paradoxical result. The first assumes that the signal rate affects the relative sizes of the variances of the information distributions; the second assumes that the signal rate affects the logic of the operator's stopping rule. Actual or potential applications of this research include the improved training and performance assessment of operators in areas such as product quality control, air traffic control, and medical and clinical diagnosis.

scholarly journals Predicting individual false alarm rates and signal detection theory: A role for remembering

2000 ◽  
Vol 28 (8) ◽  
pp. 1347-1356 ◽  
Author(s):  
Ian G. Dobbins ◽  
Wayne Khoe ◽  
Andrew P. Yonelinas ◽  
Neal E. A. Kroll
Keyword(s):  
Signal Detection ◽  
False Alarm ◽  
Signal Detection Theory ◽  
Detection Theory ◽  
False Alarm Rates

Spesialisasie binne die bedryfsielkunde

1985 ◽  
Author(s):  
I. Van W Raubenheimer
Keyword(s):  
Signal Detection ◽  
Personnel Selection ◽  
Signal Detection Theory ◽  
Work Performance ◽  
Job Analysis ◽  
Vigilance Task ◽  
Detection Theory ◽  
Industrial Psychology ◽  
High Degree ◽  
Selection Instrument

Personnel selection is carried out by the personnel psychologist/ personnel practitioner responsible for the supply of human resources. Selection models and -technology have been refined to a high degree and are scientifically acceptable. With regards to the identification of critical attributes on which selection is based, it is suggested in the study that job analysis be performed in the light of insights and knowledge gained in other branches of Industrial Psychology. Specifically the study shows that high and low work performance operators (N=25) involved in a a vigilance task, can be differentiated on the basis of signal detection theory. The results of the study (t-test for differences between means) suggest that selection in the situation investigated, should be based on the ability to discriminate between signals and not on the response bias of operators. The results also indicate that the task, based on signal detection theory, which was designed for the study, could be used as a selection instrument. OpsommingPersoneelkeuring setel gewoonlik by die personeelsiel-kundige/ personeelpraktisyn wat vir die voorsiening van menslike hulpbronne verantwoordelik is. Keuringsmodelle en –tegnologie is in 'n hoë mate verfyn en vakkundig verantwoordbaar. Ten opsigte van die identifisering van kritieke attribute op grond waarvan keuring plaasvind, word daar in die studie betoog dat daar aanvullend tot werkontleding ook op die kennis en insig wat in ander vertakkinge van die Bedryfsielkunde gegenereer word, gesteun moet word. In besonder word daar in die studie (N=25) aangetoon dat 'n onderskeid tussen hoë en lae werkprestasie-operateurs, gemoeid met 'n waaksaamheidstaak, op grond van seinbespeuringsteorie gemaak kan word. Uit die resultate van die studie (t-toets vir gemiddeldes) blyk dit dat keuring in die betrokke situasie op diskriminasievermoë en nie op responsievooroordeel nie, afgestem behoort te word en dat die taak, voortspruitend uit seinbespeuringsteorie wat vir die studie ontwerp was, as keuringsinstrument gebruik kan word.

Sensitivity and Bias in the Resolution of Stream-Bounce Stimuli

Perception ◽  
2016 ◽  
Vol 46 (2) ◽  
pp. 178-204 ◽  
Author(s):  
Mick Zeljko ◽  
Philip M. Grove
Keyword(s):  
Experimental Design ◽  
Signal Detection ◽  
False Alarm ◽  
Signal Detection Theory ◽  
Detection Theory ◽  
Correct Rejection ◽  
Visual Displays ◽  
Signal Discrimination ◽  
Motion Sequence ◽  
Significant Change

The audiovisual stream-bounce effect refers to the resolution of ambiguous motion sequences as streaming or bouncing depending on the presence or absence of a sound. We used a novel experimental design and signal detection theory (SDT) to determine its sensory or decisional origins. To account for issues raised by Witt et al. on the interpretation of SDT results, we devised a pure signal detection (as opposed to signal discrimination) paradigm and measured participants’ sensitivity and criterion when detecting a weak tone concurrent with objectively streaming or bouncing visual displays. We observed no change in sensitivity but a significant change in criterion with participants’ criterion more liberal with bouncing targets than for streaming targets with. In a second experiment, we tasked participants with detecting a weak tone in noise while viewing an ambiguous motion sequence. They also indicated whether the targets appeared to stream or bounce. Participants’ reported equivalent, mostly bouncing responses for hit and false alarm trials, and equivalent, mostly streaming responses for correct rejection and miss trials. Further, differences in participants’ sensitivity and criterion measures for detecting tones in subjectively streaming compared to subjectively bouncing targets were inconsistent with sensory factors. These results support a decisional account of the sound-induced switch from mostly streaming to mostly bouncing responses in audiovisual stream-bounce displays.

A Computer Program Demonstrating the Effect of the Payoff Matrix on the Signal Detection Measure, Beta

1994 ◽  
Vol 21 (2) ◽  
pp. 113-115 ◽  
Author(s):  
Janet D. Larsen
Keyword(s):  
Computer Program ◽  
Signal Detection ◽  
False Alarm ◽  
Response Bias ◽  
Signal Detection Theory ◽  
Visual Display ◽  
Payoff Matrix ◽  
Detection Theory ◽  
Basic Concepts ◽  
Hit Rates

The computer program described in this article demonstrates the basic concepts of signal detection theory and illustrates the effect of changes in the payoff matrix on beta, a measure of response bias. Observers attempt to identify the trials on which there was a signal in a noisy visual display. They earn or lose points for correct or incorrect answers. After three sets of 100 trials, with a different payoff matrix for each set, the program provides observers with their hit rates, false alarm rates, d' scores, and beta scores for the three sets of trials, as well as the total points they have earned for the 300 trials. The program runs on MS-DOS machines and does not require a computer with graphics capabilities.

On the Analysis and Interpretation of Vigilance: Some Remarks on Two Recent Articles by Craig

1978 ◽  
Vol 20 (4) ◽  
pp. 447-451 ◽  
Author(s):  
Michel Loeb
Keyword(s):  
Signal Detection ◽  
Recent Article ◽  
Signal Detection Theory ◽  
Detection Theory ◽  
Signal Frequency ◽  
Time On Task ◽  
Probability Matching ◽  
Vigilance Decrement ◽  
Response Frequency ◽  
Group Data

In a recent article, Craig (1977) has considered the question, first attacked by Broadbent and Gregory (1963), as to the applicability of signal detection theory (SDT) to the vigilance situation, and he has discussed the use of group data by Loeb and Binford (1964) in approaching this same question. In a still more recent article by Craig (1978) it is argued that a probability-matching model is appropriate for explaining the “vigilance decrement” with time on task. Here it is agreed that to a degree SDT may be applied to vigilance and that some matching of response frequency to signal frequency may occur, but it is argued that there are changes within and across sessions which may not be explained in these terms and which may involve learning about the differential characteristics of signal and non-signal stimuli.

Vigilance and Signal Detection Theory: An Empirical Evaluation of Five Measures of Response Bias

1997 ◽  
Vol 39 (1) ◽  
pp. 14-29 ◽  
Author(s):  
Judi E. See ◽  
Joel S. Warm ◽  
William N. Dember ◽  
Steven R. Howe
Keyword(s):  
Signal Detection ◽  
Response Bias ◽  
Signal Detection Theory ◽  
Empirical Evaluation ◽  
Vigilance Task ◽  
Parametric Model ◽  
Detection Theory ◽  
Nonparametric Model ◽  
Effective Measure ◽  
Bias Index

Three experiments were conducted to determine which of five response bias indices (β, c, B", B'H and B"D) defined by the theory of signal detection provides the most effective measure of the observer's willingness to respond in a vigilance task. The results indicated that the traditional parametric bias index β was an inadequate measure of response bias in every respect, whereas the newer parametric measure c was the most effective of all five indices. When the three nonparametric measures (B", B'H' and B"D) were examined separately, B"D emerged as the most effective nonparametric index. We recommend that vigilance researchers use c rather than β to measure bias when a parametric model is involved and B"D instead of B" and B' H when a nonparametric model is used.

So, Whatever Happened to Signal Detection Theory?

1995 ◽  
Vol 40 (10) ◽  
pp. 972-972
Author(s):  
Jerome R. Busemeyer
Keyword(s):  
Signal Detection ◽  
Signal Detection Theory ◽  
Detection Theory
Sign in / Sign up

Export Citation Format

Share Document