In the auditory detection of a sinusoidal signal in the presence of random noise, it has previously been demonstrated that the presence of a “pedestal” or background sinusoid of the same frequency and phase as the signal increases detectability. This increase was confirmed in a two-interval forced choice experiment in which a 1000-cps sinusoid was present in one of two 0.1-sec intervals, and the noise plus pedestal were present in both intervals. Pedestals of moderate intensity in phase with the signal increased detectability. Pedestals in quadrature (90° out of phase) with the signal did not improve detectability. Very large pedestals decreased detectability irrespective of phase. A simple energy-detection model was developed and its performance compared with the experimental results. The model consists of a bandpass filter, a square-law element, and an integrator. The probability of correct detection varies with signal level in a manner similar to the auditory results. Detectability is improved by an in-phase pedestal and approaches that of a correlation detector in the limit of infinite pedestal. Detectability is not improved by quadrate pedestals. The model can be modified to predict the decreasing detectability with large pedestals.

No abstract

Twelve reproducible noises were used as stimuli in a 2-interval forced-choice signal-detection experiment. The noises were stored numerically in a PB250 computer that converted thein to sound during the experiment by means of a digital-to-analog converter. The 240 numbers specifying a noise were sampled at a rate of 2500 numbers/sec, generating a 96-msec stimulus. A sinusoidal signal with a period of 8 samples/cycle was added to one noise on half the trials. Spectral analyses were computed for all stimuli. On nonsignal trials, biases to particular noises were found that could be explained in part, but not entirely, by differences between the noise pairs in energy around the signal frequency. Performance on signal trials was related to the energy difference between the stimuli in the region near the signal frequency, but was not entirely accounted for by this variable. Special characteristics of certain noises appear to affect the subject's response when these noises appear in either signal or nonsignal trials. Except for bias effects, detection on signal trials with the same noise in both intervals was similar to that on trials with different noises. Implications for theories of signal detection are discussed.

The detectability of auditory signals in reproducible random noise was studied under two conditions: a single noise used throughout a block of 288 trials, and 12 noises occurring at random, but with equal frequency, throughout a block of trials. Both two-interval forced-choice judgments and judgments of the presence or absence of the signal in single-noise samples were obtained in separate blocks of trials. On individual trials of the forced-choice judgments, the same noise appeared in both intervals. Signal detectability was found to be significantly better when a single noise was present in a block of trials. Introducing variability in the stimulus by altering the number of different signal levels presented during a block of trials did not affect detection. The results support the importance of memory for the noise from trial to trial in the detection process.

In three experiments the free recall of rapidly presented word strings was studied. The word strings were shown either in grammatical order or a scrambled order. It was found that the invariance of recall with various presentation times for grammatical sequences breaks down at presentation times between 175 and 250 msec. The order of recall, however, remains consistent with the type of sequence presented irrespective of presentation time. The amount of time required to utilize syntactic information fully in free recall is of approximately the same order of magnitude as that required to read the words.In a number of studies it has been shown that the recall of individual words depends primarily on the total viewing time, not on how that time is distributed (Cooper & Pantle, 1967). One of the major exceptions to the total time hypothesis is the recall of grammatical word sequences. King (1971) has shown that when grammatical sequences are presented, a word at a time, the exposure time per word does not affect recall, over a range from 5.0 to 1.0 sec. One explanation which has been offered for this difference is that the processing employed to learn grammatical sequences differs from that for random sequences, and that the farmer is completed in less time than the latter. If this is true, as observation time is decreased, at some point the processes by which grammatical sequences are learned must also be affected by time constraints. This point should represent the minimum time necessary to utilize syn tactic information fully in the learning of sequentially presented word lists.The order of recall is also of interest in determining what process is being employed. The free recall of unrelated word strings shows characteristic seria! position effects (Murdock, 1962). These effects are substantially modified in the case of grammatical word sequences (Deese & Kaufman, 1957). It is of interest to ask whether the grammatical pattern of free recall shifts in the direction of that for random lists when time pressures begin to affect recall. Such a shift would suggest that Ss change their strategy to compensate for the loss of syntactic information. In the following experiments, the free recall of grammatical and scrambled word sequences was studied under conditions in which each word in the stimulus list was presented very briefly; the longest presentation time was 0.5 sec per word. A fixed number of learning trials was used for the word sequences under each presentation time. The presentation times were varied to determine the effect on free recall for a range of very brief times. EXPERIMENT I Method DesignSs were shown four sets of eight words each. Each set was presented, one word at a time, for five trials. Two of the word sets were shown in a grammatical sequence, and two were shown scrambled. Two presentation times were used for different word sets, in combination with both scrambled and grammatical orders. A Greco-Latin square design was used to counterbalance time, word sets, and the order in which the word sets w...