Proportion Of False Alarms Research Articles

Performance Assessment of GPM IMERG Products at Different Time Resolutions, Climatic Areas and Topographic Conditions in Catalonia

Quantitative Precipitation Estimates (QPEs) from the Integrated Multisatellite Retrievals for GPM (IMERG) provide crucial information about the spatio-temporal distribution of precipitation in semiarid regions with complex orography, such as Catalonia (NE Spain). The network of automatic weather stations of the Meteorological Service of Catalonia is used to assess the performance of three IMERG products (Early, Late and Final) at different time scales, ranging from yearly to sub-daily periods. The analysis at a half-hourly scale also considered three different orographic features (valley, flat and ridgetop), diverse climatic conditions (BSk, Csa, Cf and Df) and five categories related to rainfall intensity (light, moderate, intense, very intense and torrential). While IMERG_E and IMERG_L overestimate precipitation, IMERG_F reduces the error at all temporal scales. However, the calibration to which a Final run is subjected causes underestimation regardless in some areas, such as the Pyrenees mountains. The proportion of false alarms is a problem for IMERG, especially during the summer, mainly associated with the detection of false precipitation in the form of light rainfall. At sub-daily scales, IMERG showed high bias and very low correlation values, indicating the remaining challenge for satellite sensors to estimate precipitation at high temporal resolution. This behaviour was more evident in flat areas and cold semi-arid climates, wherein overestimates of more than 30% were found. In contrast, rainfall classified as very heavy and torrential showed significant underestimates, higher than 80%, reflecting the inability of IMERG to detect extreme sub-daily precipitation events.

Facilitating Sustained Attention: Is Mere Presence Sufficient?

Abstract Research on sustained attention regularly reports declines in task performance as a function of time on task, a phenomenon commonly called the vigilance decrement. Therefore, previous work from theoretical and applied fields has focused on attenuating this performance decrement. Recently, aspects of social facilitation have been used to attain this goal. Social facilitation, or improving task performance through the social presence of an individual, has been found to improve performance on sustained attention tasks. However, the extent to which the forms of social facilitation can mitigate performance are unclear. Thus, the impetus of the present research was to explore the effects of two prevalent forms of social facilitation (i.e., mere presence and evaluative presence) on a sustained attention (i.e., vigilance) task. In sum, 132 observers completed a 24-min cognitive-based vigilance task. The results indicated that an evaluative presence produced the strongest effects on performance, such that the proportion of false alarms was significantly reduced relative to no social presence. Moreover, the mere presence of an individual did not significantly improve vigilance task performance. The results of the present experiment suggest that an evaluative context may be necessary for social presence to improve detection performance on tasks that require sustained attention.

Extended Abstract: Utilizing Event Rate As a Manipulation of Task Difficulty in Vigilance

Vigilance refers to the ability of an observer to maintain attention over extended periods of time, and to respond to critical signals that occur (Davies & Parasuraman, 1982). Vigilance has been ex...

Sensor-Assisted Face Tracking

Generally, face detection and tracking focus only on visual data analysis. In this paper, we propose a novel method for face tracking in camera video. By making use of the context metadata captured by wearable sensors on human bodies at the time of video recording, we could improve the performance and efficiency of traditional face tracking algorithms. Specifically, when subjects wearing motion sensors move around in the field of view (FOV) of a camera, motion features collected by those sensors help to locate frames most probably containing faces from the recorded video and thus save large amount of time spent on filtering out faceless frames and cut down the proportion of false alarms. We conduct extensive experiments to evaluate the proposed method and achieve promising results.

Response Inhibition Moderates the Association between Drug Use and Risky Sexual Behavior

Background: HIV infection is problematic among all drug users, not only injection drug users. Drug users are at risk for contracting HIV by engaging in risky sexual behaviors. Objective: The present study sought to determine whether inhibitory processes moderate the relationship between problematic drug use and HIV-risk behaviors (unprotected sex and multiple sex partners). Methods: One hundred ninety-six drug offenders enrolled in drug education programs were administered a battery of computer-based assessments. Measures included a cued go/no-go assessment of inhibitory processes, the Drug Abuse Screening Test (DAST) assessment of problematic drug use, and self-report assessment of condom use and multiple sex partners. Results: Findings revealed that response inhibition assessed by the proportion of false alarms on the cued go/no-go moderated the relationship between problematic drug use and an important measure of HIV risk (condom nonuse) among drug offenders. However, response inhibition did not moderate the relationship between problematic drug use and another measure of HIV risk: multiple sex partners. Conclusions: Among this sample of drug offenders, we have found a relationship between problematic drug use and condom nonuse, which is exacerbated by poor control of inhibition. These findings have implications for the development of HIV intervention components among high-risk populations.

Detecting oestrus by monitoring sows’ visits to a boar

This paper suggests a method for automatic detection of sows returning to oestrus in the gestation department. The detection is based on monitoring of sows’ visits to a boar, where the duration and frequency of visits are modelled separately and subsequently combined. The hypothesis is that it is possible to reduce the response time and the number of false alarms compared to previously published attempts. The duration of visits to a boar is defined as seconds per hour the sow is near the boar – logarithmically transformed. The duration is modelled with a multiprocess dynamic linear model with first order Markov probabilities. The indicator of oestrus is the probability of the model describing oestrus, P ( M O E ) , and it is monitored with a threshold value. The frequency of visits to a boar is defined as number of visits per 6 h. A dynamic generalised linear model with two built-in diurnal periods is applied. The indicator of oestrus is the relative deviation from the forecasted frequency, which is monitored with a threshold value. The probability, P ( M O E ) , and the relative deviation from the forecasted frequency are combined by means of Bayes Theorem. The combined probability of oestrus is monitored with a threshold value as well. Results indicate that the specificity is superior compared to previously published attempts. The model describing duration alone yields the most satisfactory specificity – 99.4% per sow day, which is considerably greater than previously published studies. Furthermore, this model detects 87.4% of the sows entering oestrus, which is slightly lower than previous attempts. The response time of the models is 1 h for the duration model and the combined model and 6 h for the frequency model. This is better than previous attempts. Even though the specificity is greater, the proportion of false alarms on a day-to-day basis is still too high (91.0%), which is due to the very large proportion of the sow days defined as non-oestrus. In order to improve the specificity of the detection method, it is suggested to combine the detection method in the present study with other information sources about oestrus.

Recollection- and familiarity-based decisions reflect memory strength

We used event-related fMRI to investigate whether recollection- and familiarity-based memory judgments are modulated by the degree of visual similarity between old and new art paintings. Subjects performed a flower detection task, followed by a Remember/Know/New surprise memory test. The old paintings were randomly presented with new paintings, which were either visually similar or visually different. Consistent with our prediction, subjects were significantly faster and more accurate to reject new, visually different paintings than new, visually similar ones. The proportion of false alarms, namely remember and know responses to new paintings, was significantly reduced with decreased visual similarity. The retrieval task evoked activation in multiple visual, parietal and prefrontal regions, within which remember judgments elicited stronger activation than know judgments. New, visually different paintings evoked weaker activation than new, visually similar items in the intraparietal sulcus. Contrasting recollection with familiarity revealed activation predominantly within the precuneus, where the BOLD response elicited by recollection peaked significantly earlier than the BOLD response evoked by familiarity judgments. These findings suggest that successful memory retrieval of pictures is mediated by activation in a distributed cortical network, where memory strength is manifested by differential hemodynamic profiles. Recollection- and familiarity-based memory decisions may therefore reflect strong memories and weak memories, respectively.

Signal detection behavior in humans and rats: a comparison with matched tasks.

Animal models of human cognitive processes are essential for studying the neurobiological mechanisms of these processes and for developing therapies for intoxication and neurodegenerative diseases. A discrete-trial signal detection task was developed for assessing sustained attention in rats; a previous study showed that rats perform as predicted from the human sustained attention literature. In this study, we measured the behavior of humans in a task formally homologous to the task for rats, varying two of the three parameters previously shown to affect performance in rats. Signal quality was manipulated by varying the increment in the intensity of a lamp. Trial rate was varied among values of 4, 7, and 10 trials/min. Accuracy of signal detection was quantified by the proportion of correct detections of the signal (P(hit)) and the proportion of false alarms (P(fa), i.e. incorrect responses on non-signal trials). As with rats, P(hit) in humans increased with increasing signal intensity whereas P(fa) did not. Like rats, humans were sensitive to the trial rate, though the change in behavior depended on the sex of the subject. These data show that visual signal detection behavior in rats and humans is controlled similarly by two important parameters, and suggest that this task assesses similar processes of sustained attention in the two species.

Detection of visual signals by rats: effects of signal intensity, event rate, and task type

Animal models of human cognitive processes are being developed for investigating the neurobiological mechanisms of these processes and for identifying potential therapies for intoxication and neurodegenerative diseases. One promising method involves assessment of sustained attention in rats by use of a discrete-trial signal detection task. Several psychophysical and procedural factors have been identified in the literature that affect sustained attention in normal human subjects. Three key parameters that affect the level of performance, and whether that performance level deteriorates over time, include the quality of the signal, the event rate, and the type of task employed (simultaneous or successive discrimination). These three parameters were manipulated in this study to assess the degree of similarity in the behavior patterns engendered in rats by this signal detection task, in relation to the behavior observed in humans performing a variety of sustained attention tasks. Signal quality was manipulated by varying the increment in the intensity of a lamp (duration=300 ms), and event rate was varied among values of 4, 7, and 10 trials/min. The ‘standard’ detection task was used as a simultaneous discrimination and a successive discrimination task was designed in which a dim light flash was defined as a non-signal event and any of three brighter flashes were signal events. Accuracy of signal detection was quantified by the proportion of correct detections of the signal [P(hit)] and the proportion of false alarms [P(fa), i.e., incorrect responses on non-signal trials]. P(hit) fell with decreasing signal intensity, increasing event rate, and was lower in the discrimination task compared to the detection task. P(fa) increased with increasing event rate, but only in the detection task. A decrement in P(hit) across trial blocks was observed in the discrimination task primarily, and was most evident with a high event rate and dim signals. These data confirm that these tasks assess process(es) in rats that are very similar to those considered as sustained attention in humans.

Confounds of insensitivity and blind luck: statistical conclusion validity in stroke rehabilitation clinical trials.

We present theoretical analyses of some infrequently considered consequences of low power and quantitatively review power in 36 stroke rehabilitation clinical trials with multiple outcome variables. We demonstrate that low power affects the proportion of false alarms (type 1 errors) in the literature and probability of replication, not only the ability to detect true intervention effects; and that these consequences of low power are modulated by the probability that effective interventions have been preselected for study via theory and prior evidence. interventions have been preselected for study via theory and prior evidence. Rehabilitation trials showed low power to detect small, medium, or large treatment effects at alpha = 0.05 (0.09, 0.33, and 0.69, respectively). Effect sizes were generally small with adverse consequences on power. As expected, sample size was an important determinant of power. Power was lower for motor/reflex measures due to differences in sample sizes rather than effect sizes. These results and the conceptual analyses predict and address conflicting conclusions from stroke rehabilitation clinical trials.

The effect of adaptation on odor detection

The sensitivity of human Ss to a weak odor (hydrogen sulphide) following prior exposure to it varying in both duration (0–5 min) and concentration (4.05 × 10−7 to 24.40 × 10−7 mg/liter) was studied with a yes-no detection method and a specially designed olfactometer. The effect of adaptation was demonstrated, because the proportion of hits was lower the higher the concentration of the adaptation stimulus, when the response criterion, as measured by the proportion of false alarms, was fairly constant. By contrast, the proportion of hits was apparently independent of the duration of the exposure to the adaptation concentration. This finding suggests that the sense of smell is more stable than is usually believed. The paper ends with a discussion of the importance of considering adaptation in contemporary detection theory.

The human use of information--I: Signal detection for the case of the signal known exactly

A theory of visual detection is developed, based on the model provided by the theory of signal detectability,2 and, more generally, by the theory of statistical decision. Two experiments are reported which test some predictions of the theory for the case of the signal-known-exactly. These experiments demonstrate that the human observer tends toward optimum behavior, where optimum behavior is defined as that behavior which maximizes the expected gain from the decision. Their results show the proportion of correct detections to be dependent upon the proportion of false alarms; they indicate that neural activity is a power function of signal intensity. The data also demand a re-evaluation of the threshhold concept. Predictions are made for the data obtained using two different methods of response, forced-choice end yes-no, and the internal consistency of the theory is demonstrated. The predictions of the theory are compared with contrasting predictions of conventional sensory theory; the data are also related to conventional theory.