Robert D. Sorkin

Systems with human monitors: a signal detection analysis

Automated factories, the flightdecks of commercial aircraft, and the control rooms of power plants are examples of decision-making environments in which a human operator performs an alerted-monitor role. These human-machine systems include automated monitor or alerting subsystems operating in support of a human monitor. The automated monitor subsystem makes preprogrammed decisions about the state of the underlying process based on current inputs and expectations about normal/abnormal operating conditions. When alerted by the automated monitor subsystem, the human monitor may analyze input data, confirm or disconfirm the decision made by the automated monitor, and take appropriate further action. In this paper, the combined automated monitor-human monitor system is modeled as a signal detection system in which the human operator and the automated component monitor partially correlated noisy channels. The signal detection analysis shows that overall system performance is highly sensitive to the interaction between the humans monitoring strategy and the decision parameter, Ca, of the automated monitor subsystem. Usual design practice is to set Ca to a value that optimizes the automated monitors detection and false alarm rates. Our analysis shows that this setting will not yield optimal performance for the overall human-machine system. Furthermore, overall system performance may be limited to a narrow range of realizable detection and error rates. As a result, large gains in system performance can be achieved by manipulating the parameters of the automated monitor subsystem in light of the workload characteristics of the human operator.

Extension of the Theory of Signal Detectability to Matching Procedures in Psychoacoustics

The theory of signal detectability is used to analyze experiments where the observers task is to state, after two signal presentations, whether the signals were the same or different. A model is suggested for predicting human performance in auditory “matching” tasks using data from detection and discrimination experiments. Three experiments are discussed which lend support to the models application. The first, a two‐alternative forced‐choice, pulsed‐carrier experiment, serves to determine each observers efficiency. The efficiency η is assumed constant provided certain observation interval and signal characteristics are held fixed, as they were during the entire experimental sequence. The second was a simple matching task, where the first observation interval always contained specified signal and the second sometimes contained a signal identical to the first. In the third experiment the signal appearing in the first interval was randomly chosen from a set of possible signals, thus removing the long‐term...

Simultaneous two‐channel signal detection. III. 630‐ and 1400‐Hz signals

Performance in a simultaneous two‐frequency detection task was studied under monaural and dichotic conditions of signal presentation. In the basic task a 630‐Hz tone, a 1400‐Hz tone, neither tone, or both tones occurred with equal likelihood on an experimental trial. The observer had to specify which of these events had occured. Performance was examined: (a) when the signal could occur only in the right earphone channel, and (b) when the 630‐ and 1400‐Hz signals were confined, respectively, to the left and right earphone channels. Single‐frequency comparison conditions were also run. A performance decrement was observed in both of the two‐frequency conditions compared to the single‐frequency condition. However, there was no consistent difference between the monaural and dichotic two‐frequency conditions. An analysis of the conditional detection probabilities revealed the presence of a cross‐frequency interference effect on trials when a signal was present at the other frequency. This effect was sufficient to account for the differences between single‐frequency and two‐frequency detection performance.

OBSERVER SENSITIVITY TO ELEMENT RELIABILITY IN A MULTIELEMENT VISUAL DISPLAY

This study tested whether selected visual display conditions can aid observers in prioritizing information from display elements that differ in reliability. The observers task was to decide whether the information conveyed by an array of nine graphical elements represented a signal or noise input. The value displayed by each element was an independent sample from the signal or noise distribution. The reliability of a display element was manipulated by controlling the variance of these distributions (high reliability = low variance). Observers tended to weight high-reliability sources slightly more than low-reliability sources. The observers-detection performance and weighting efficiency were highest when the reliability of the source was cued by the luminance of the display element. Thus observers can prioritize visual information sources based on statistical properties such as variability if luminance cues are available during the presentation of the information.

Some models of observer behavior in two-channel auditory signal detection

Five models of 0 behavior in a two-channel exclusive-or (XOR) detection task were evaluated. The models included two types of single-channel and three types of two-channel O. Only the most efficient two-channel model adequately described human performance in a set of monaural and dichotic XOR conditions. Detectability measures derived from the XOR task matched those obtained from separate single-channel control conditions. We concluded that. in this two-channel task. the O’s performance was not limited by any inability to monitor signals arriving simultaneousl.v in two different earphone or two different frequency channels. The implications of this result for two-channel information processing and for multicomponent and sequential signal detectmn are discussed.

Simultaneous Two‐Channel Signal Detection. II. Correlated and Uncorrelated Signals

Two‐channel auditory signal detection was investigated with narrow‐band noise signals masked by independent, binaurally uncorrelated noise. The two‐channel tasks included conditions where the signals in each channel were correlated, uncorrelated, or opposite in phase. No significant differences were observed between the single‐channel conditions and the two‐channel correlated signal condition, but performance in the uncorrelated and 180° phase conditions was significantly poorer. A comprehensive analysis of cross‐channel interaction indicated a cross‐channel inhibitory or masking effect, which may be counteracted by the presence of lateralization cues. Constraints on signal detectability due to a limited processing capacity appeared to be negligible.

Decision interaction between auditory channels

Observers detected signals presented in a two-channel, simultaneous detection task at 630 and 1,400 Hz. The familiar pattern of interference between the frequency channels was observed: detectability in one channel was depressed on trials when a signal or a yes response occurred in the other channel. On each trial, measures were made of the energy within specified frequency bands around each signal frequency. The relationships between these measures and performance in each channel were studied. There was no consistent relationship between the magnitude of the measure in one channel and performance in the other. Interactions between the channels probably originate in the response process of the interfering channel

UNCERTAIN SIGNAL DETECTION WITH SIMULTANEOUS CONTRALATERAL CUES.

A series of auditory signal‐detection experiments was run under conditions of signal‐parameter uncertainty. This uncertainty was effected by allowing one signal parameter, either signal frequency or starting time, to vary randomly across the sequence of experimental trials. These experiments, run monaurally, employed a simple yes‐no detection procedure, signal plus noise occurring on half of the trials and noise alone occurring on the other half. A series of comparison experiments using the same observers was run under identical conditions, with the addition of a simultaneous cue signal in the contralateral ear. This cue was present on both signal and no‐signal trials and was identical, in all parameters except amplitude, to the signal that might have been presented to the detecting ear. The results demonstrated (a) the previously noted result—that a simultaneous contralateral cue degrades performance at relatively high signal‐to‐noise levels—and (b) a new result—that at low signal levels such a cue facil...

Perception of temporal patterns defined by tonal sequences

This experiment tested how listeners discriminate between the temporal patterns defined by two sequences of tones. Two arrhythmic sequences of n tones were played successively (n = 8, 12, or 16, tone duration = 35 ms, frequency = 1000 Hz), and the listener reported whether the sequences had the same or different temporal patterns. In the first sequence, the durations of the intertone gaps were chosen at random; in the second sequence, the gaps were either (a) the same as the first sequence or (b) chosen at random. Discrimination performance increased with the variability of the gap sequences and decreased with the size of the correlation between the sequences. A discrimination model based on computation of the sample correlation between the sequences of gaps, but limited by an internal variability of approximately 15 ms, described observer performance in a variety of conditions.

Simultaneous three-channel signal detection: Performance and criterion as a function of order of report

Performance on a task requiring detection of three sinusoidal signals occurring simultaneously and independently was compared to single-channel performance. The signal frequencies were 500, 810, and 1,320 Hz, and all signals were presented monaurally in noise. The data for each channel were analyzed conditional on the stimulus-response events occurring in the remaining channels and conditional on the order of report. The decrement in performance in any channel in the three-channel task was found to increase as the number of signals and/or “yes” responses present in the remaining channels increased. Increasing decrements and criterion shifts were noted conditional on the order of report. The data are consistent with a model attributing the decrements in multichannel listening tasks to the later or “cognitive” stages of processing rather than to the earlier “perceptual” processing stages.