Efficiency of Human Visual Signal Discrimination

Burgess, Arthur E.; Wagner, R. F.; Jennings, Robert J.; Barlow, H. B.

doi:10.1126/science.7280685

Cited by 397 publications

(342 citation statements)

References 10 publications

Supporting

Mentioning

327

Contrasting

Order By: Relevance

“…Observer efficiency is a well-known metric that quantifies a human observer's detection sensitivity with respect to an ideal detector by taking the squared ratio between human and ideal contrast thresholds or d′ values (62,63). Similarly, group efficiency has been used to quantify group performance with respect to an ideal detector (64) or ideal group performance (25) by taking the squared ratio between the relevant contrast thresholds or d′ values.…”

Section: Resultsmentioning

confidence: 99%

“…The SNR is the distance in SD units between an ideal observer's respective decision variable distributions for target-present images and target-absent images. For white noise, it can be calculated from the signal and noise as follows (63,95): SNR = root signal energy=noise SD = ffiffiffi E p =σ. The SD of the white noise was the same in both tasks (σ = 8.61 cd/m 2 ).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

The wisdom of crowds for visual search

Juni

Eckstein

2017

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

Decision-making accuracy typically increases through collective integration of people's judgments into group decisions, a phenomenon known as the wisdom of crowds. For simple perceptual laboratory tasks, classic signal detection theory specifies the upper limit for collective integration benefits obtained by weighted averaging of people's confidences, and simple majority voting can often approximate that limit. Life-critical perceptual decisions often involve searching large image data (e.g., medical, security, and aerial imagery), but the expected benefits and merits of using different pooling algorithms are unknown for such tasks. Here, we show that expected pooling benefits are significantly greater for visual search than for single-location perceptual tasks and the prediction given by classic signal detection theory. In addition, we show that simple majority voting obtains inferior accuracy benefits for visual search relative to averaging and weighted averaging of observers' confidences. Analysis of gaze behavior across observers suggests that the greater collective integration benefits for visual search arise from an interaction between the foveated properties of the human visual system (high foveal acuity and low peripheral acuity) and observers' nonexhaustive search patterns, and can be predicted by an extended signal detection theory framework with trial to trial sampling from a varying mixture of high and low target detectabilities across observers (SDT-MIX). These findings advance our theoretical understanding of how to predict and enhance the wisdom of crowds for real world search tasks and could apply more generally to any decision-making task for which the minority of group members with high expertise varies from decision to decision.group decision rules | signal detection theory | ideal observer analyses | wisdom of crowds G roups of insects (1-4), fish (5-7), birds (8-10), mammals (11)(12)(13)(14), and primates (15-18) have been shown to aggregate their individual judgments into group decisions for various tasks (19,20). Although some groups seem to have leaders who make decisions alone on behalf of their groups (17,(21)(22)(23), it is difficult for individuals to outperform even simple aggregations of the entire group's individual judgments (4,7,9,10,19,(24)(25)(26). Perhaps that is why humans often make important decisions as a group (27-29), even if the only expedient (30, 31) but effective (24, 31-34) group decision mechanism is to use the simple majority voting rule (35).Previous human studies have shown that combining people's judgments into group decisions can lead to accuracy benefits in various domains, such as estimation (36-38), detection (34,(39)(40)(41)(42)(43)(44), identification (45-47), and prediction (46, 48-52), a phenomenon known as the wisdom of crowds (53). For artificial tasks, where perceptual decisions are limited only by noise that is internal to each observer's brain (i.e., no external noise), the maximum wisdom of crowd benefits are specified by the idealized signal dete...

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

The wisdom of crowds for visual search

Juni

Eckstein

2017

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

show abstract

“…3 and 4)? To address this question, we first note that, for detection in noise, humans never reach the absolute levels of performance of the MT (ideal) observer, and thus the relative performance of human and ideal observers is often compared by introducing an overall efficiency parameter η, which effectively scales up the variance of the MT responses or, equivalently, scales up all of the MT observer's thresholds by a constant (7,12,28,30),…”

Section: Signal Detection Analysis Of Detection Under Blockedmentioning

confidence: 99%

“…If the template response R exceeds a decision criterion value γ, then the observer reports that the target is present; otherwise, the observer reports that it is absent. The MT observer is the optimal (ideal) observer when the target is known (as it is here) and the background is Gaussian white noise (7,27,28), and for "narrow band" targets, it often is not too far from optimal if the background is Poisson white noise or is correlated Gaussian noise. Although natural image backgrounds have a more complex statistical structure, the MT observer (although not optimal) is a simple, principled signal detection model, and hence a good starting point.…”

mentioning

confidence: 99%

Constrained sampling experiments reveal principles of detection in natural scenes

Sebastian

Abrams

Geisler

2017

Proc. Natl. Acad. Sci. U.S.A.

124

View full text Add to dashboard Cite

A fundamental everyday visual task is to detect target objects within a background scene. Using relatively simple stimuli, vision science has identified several major factors that affect detection thresholds, including the luminance of the background, the contrast of the background, the spatial similarity of the background to the target, and uncertainty due to random variations in the properties of the background and in the amplitude of the target. Here we use an experimental approach based on constrained sampling from multidimensional histograms of natural stimuli, together with a theoretical analysis based on signal detection theory, to discover how these factors affect detection in natural scenes. We sorted a large collection of natural image backgrounds into multidimensional histograms, where each bin corresponds to a particular luminance, contrast, and similarity. Detection thresholds were measured for a subset of bins spanning the space, where a natural background was randomly sampled from a bin on each trial. In low-uncertainty conditions, both the background bin and the amplitude of the target were fixed, and, in high-uncertainty conditions, they varied randomly on each trial. We found that thresholds increase approximately linearly along all three dimensions and that detection accuracy is unaffected by background bin and target amplitude uncertainty. The results are predicted from first principles by a normalized matched-template detector, where the dynamic normalizing gain factor follows directly from the statistical properties of the natural backgrounds. The results provide an explanation for classic laws of psychophysics and their underlying neural mechanisms.natural scene statistics | detection | masking | normalization | Weber's law V isual systems are the result of evolution by natural selection, and, as a consequence, their design is strongly constrained by the properties of natural visual stimuli and by the specific visual tasks performed to survive and reproduce. Thus, to understand the human visual system, it is critical to characterize natural visual stimuli and performance in natural visual tasks (1, 2).Perhaps the most fundamental visual task is to identify target objects in the natural backgrounds that surround us. It is known that the specific properties of a background can have a strong influence on detectability (Fig. 1). For example, the detectability of a target pattern with given amplitude decreases with increases in background luminance (3, 4), background contrast (5-7), and the similarity of the spatial properties of the background to those of the target (8-12). In addition to the direct effects of such background properties, there are other factors that affect detection performance. Specifically, under natural conditions, the strength (amplitude) and location of the target often randomly vary on every occasion, and the target typically appears against a different background scene on every occasion. The uncertainty created by the random amplitude and location of the target ("target...

show abstract

“…Thus, performance difference between valid and invalid (or neutral) trials may reflect both "attention" effects and stimulus uncertainty losses. Contrast of External Noise (%) different perceptual tasks (Burgess et al, 1981;Parish & Sperling, 199i;Pelli, 1981Pelli, , 1990. The external noise paradigm and noisy observer models have a long history in visual psychophysics extending over the past half century (see Burgess, Shaw, & Lubin, 1999, for a review).…”

Section: Ptmmentioning

confidence: 99%

Spatial attention: Different mechanisms for central and peripheral temporal precues?

Lü¹,

Dosher²

2000

Journal of Experimental Psychology: Human Perception and Perfor

120

204

View full text Add to dashboard Cite

The external noise paradigm (Z.-L. Lu & B. A. Dosher, 1998) was applied to investigate mechanisms of spatial attention in location precuing. Observers were precued or simultaneously cued to identify 1 of 4 pseudocharacters embedded in various amounts of external noise. The cues were either central or peripheral. Both central and peripheral precuing significantly reduced threshold in the presence of high external noise (16% and 17.5%). Only peripheral precuing significantly reduced threshold in the presence of low, or no, external noise (11%). A perceptual template model identified different mechanisms of attention for central and peripheral precuing, external noise exclusion for central precuing, and a combination of external noise exclusion and stimulus enhancement (or equivalently, internal additive noise reduction) for peripheral cuing.In a high workload environment, a human operator might need to monitor a large number of information sources simultaneously. Whereas this might be accomplished by saccades, the eye movement system is limited by the relatively long time (about 200 ms) each saccade takes (Hallett, 1986). On the other hand, the existence of a faster spatial attention system was proposed about a century ago (Helmholtz, 186671911;Wundt, 1902), selecting regions of the visual field for further processing without eye movements. This view has been established by the more recent literature

show abstract

Efficiency of Human Visual Signal Discrimination

Cited by 397 publications

References 10 publications

The wisdom of crowds for visual search

The wisdom of crowds for visual search

Constrained sampling experiments reveal principles of detection in natural scenes

Spatial attention: Different mechanisms for central and peripheral temporal precues?

Contact Info

Product

Resources

About