Listening to speech at two levels at once

Monitoring performance was compared when subjects were asked to listen simultaneously for two different targets embedded in a sentence and when they listened only for one. The effects of different combinations of targets (phonemes, words, or meanings) were investigated and an attempt was made to separate the effects of particular combinations from the effects of difficulty of the two component targets. Performance was worse when either of two targets might occur than when subjects knew which single type to listen for. However there were no differences attributable to which particular types of targets were being combined. Listening for a phonemic target was as difficult to combine with listening for a semantic target as for another phonemic target. The result is consistent with a central speech analysing system whose limited processing capacity must be shared (at least in this monitoring task) both between and within all relevant levels of linguistic analysis, rather than with relatively independent subsystems for each type of listening.     

Sensory scaling and the psychophysical law

Sensory scaling is analysed as a sensori-motor task for which two relations must be determined, the first, S = f(I), being the relation between I, the stimulus intensity, and S, a measure of the neural effect of the stimulus, and the second being the relation between the subject's response, R, and S. It is shown that for any scaling procedure the relation between R and S can be arbitrarily specified, and according to the choice made S = f(I) will be found to be either a power function or a logarithmic function; these two “psycho- physical laws” do not reflect empirical differences between scaling procedures but conventional differences in the assumptions made when interpreting these procedures. A model is then elaborated in which S = f(I) is fixed and the relation between R and S depends on the demands of the psychophysical task and the effects of previous learning. With this more complex model the empirical implications of the power function and log function differ; it is shown that if the latter is accepted explanations can be given for two major problems in scaling theory, the disparity between the functions given by category scaling and the direct methods, and the difference between prothetic and metathetic continua.     

Book reviews

Nervous and Hormonal Mechanisms of Integration: Symposia of the Society for Experimental Bilogy, No. 20. Edited by G. M. Hughes. London: Cambridge University press. 1966. Pp. viii + 553. 75s.

Essential Works of Pavlov. Edited with an Introduction by Muchael Kaplan. New York: bantam Books. 1966. Pp. xiii + 392. $1.25.

Psycholinguistic Papers. Proceedings of the Edinburgh Conference 1966. Edited by J. Lyons and R. J. Wales. Edinburgh University Press. 1966. Pp. 243. 42s.

On Human Communication. By Colin Cherry. 2nd edition. London and Cambridge, Mass.: M.I.T. Press. 1966. Pp. xiv + 337. 80s.

Psycholigical Principles Appropvite to Socal and Cinical Problems. By John C. Raven. London: H. K. Lewis. 1966. Pp. xx + 187. 25s.     

A statistical decision model for sensory discrimination which predicts Weber’s law and other sensory laws: Some results of a computer simulation

A model for visual intensity threshold discrimination is described. Simplified assumptions represent the main features of the visual afferent paths. Discriminative responses are selected centrally by a statistical decision procedure, limited by the noise level. Noise arises from the irreducible physical variability of light, from spontaneous firing, and from variations in transmission in the afferent paths, These variations will tend to be positively correlated. The model was simulated on a computer: it correctly predicts the form of the Weber function (the relation between difference threshold and background intensity) and the features of spatial and temporal “summation”. It also shows that the function relating the central effect of a stimulus to its physical intensity is unlikely to be logarithmic or a power function with a small exponent.     

Lowering of an auditory threshold by a near threshold warning signal

Following previous experiments on the effect of suprathreshold “warning” signals in lowering the threshold for stimuli in another modality, we have used visual “warning” signals which are themselves within the threshold range. These near threshold “warnings” appear to act in exactly the same way as suprathreshold “warnings” provided they are seen. Very weak “warnings” have a reduced effect, probably because they are not always seen. The experiment is of interest in that it supports the theory that the effect is due to the “warning” signal reducing the subject's uncertainty about when the threshold stimulus will occur, and secondly because it indicates that under these conditions at least, the subject can attend effectively to two simultaneous near-threshold stimuli.     

Changes in threshold level produced by a signal preceding or following the threshold stimulus

In a previous experiment (Howarth and Treisman, 1958) it was shown that when a warning preceded a stimulus by a fixed interval, the threshold level for the stimulus was higher the longer the fixed interval. A model of this effect was proposed, assuming that the threshold criterion adopted by the subject at any moment was modified by his estimate of the probability of a stimulus at that moment.

Three predictions are derived and tested. It is found that: the standard deviation of the response is constant despite the shifts in threshold induced; the threshold continues to fall as the interval from warning to stimulus is decreased to zero, with the lowest threshold when the warning signal and the stimulus are simultaneous; and that a warning signal following the stimulus by an interval of less than 0 5 to 1 sec. also lowers the threshold.     

Repetition blindness: The survival of the grouped

The repetition blindness (RB) effect demonstrates that people often fail to detect the second presentation of an identical object (e.g., Kanwisher, 1987). Grouping of identical items is a well-documented perceptual phenomenon, and this grouping generally facilitates perception. These two effects pose a puzzle: RB impairs perception, while perceptual grouping improves it. Here, we combined these two effects and studied how they interact. In a series of three experiments, we presented repeated items in a simultaneous string, while manipulating the organization of the repeated items in groups within a string. We observed an interaction between RB and grouping that we summarize with a rule that we call “the survival of the grouped”: In essence, the ability to group repeated elements protects them from RB. These findings are discussed within the framework of the object file theory.     

Time perception and the internal clock: Effects of visual flicker on the temporal oscillator

Abstract

Our ability to estimate time intervals has sometimes been attributed to a biological source of temporal information. A model for a temporal oscillator that provides such information has recently been described (Treisman, Faulkner, Naish & Brogan, 1990). This predicts that an imposed stimulus rhythm at certain frequencies may interfere with the temporal oscillator so as to alter its frequency. This interference would cause perturbations in temporal judgements at certain frequencies of the imposed rhythm. The pattern of interference would depend on the frequency at which the temporal oscillator runs, and so would contain information about the oscillator frequency. Evidence for such a pattern was found when auditory clicks at different rates were presented concurrently with time intervals whose durations subjects estimated. The present study examines whether a similar interference pattern can be obtained if visual flicker is substituted for auditory clicks. On each trial, flicker was presented at a rate between 2.5 and 17.5 Hz, concurrently with a time interval to be estimated. A pattern of increased estimates at some rates and decreased estimates at others was obtained. This pattern showed similarities to interference patterns obtained using auditory clicks. This provides evidence that the entrainment of the internal clock predicted by the model can also be produced by visual inputs. Other theoretical implications are discussed.     

Perception of objects in natural scenes: is it really attention free?

Studies have suggested attention-free semantic processing of natural scenes in which concurrent tasks leave category detection unimpaired (e.g., F. Li, R. VanRullen, C. Koch, & P. Perona, 2002). Could this ability reflect detection of disjunctive feature sets rather than high-level binding? Participants detected an animal target in a rapid serial visual presentation (RSVP) sequence and then reported its identity and location. They frequently failed to identify or to localize targets that they had correctly detected, suggesting that detection was based only on partial processing. Detection of targets was considerably worse in sequences that also contained humans, presumably because of shared features. When 2 targets were presented in RSVP, a prolonged attentional blink appeared that was almost eliminated when both targets were detected without being identified. The results suggest rapid feature analysis mediating detection, followed by attention-demanding binding for identification and localization.     

The effect of signal probability on the slope of the receiver operating characteristic given by the rating procedure

Signal detection theory (SDT) requires that the slope of the receiver operating characteristic (ROC) is independent of the probability of signal and noise. But it has been shown that when the rating procedure is applied to detection, the slope of the receiver operating characteristic may increase as a function of the probability of the signal (Schulman & Greenberg, 1970). This presents a serious problem for signal detection theory. This problem is examined in relation to a recent theory of criterion setting (Treisman & Williams, 1984), and an explanation for the effect of signal probability on ROC slope is derived which is compatible with SDT. Further data on the relation between signal probability and ROC slope are reported.