Perceived auditory continuity with alternately rising and falling frequency transitions.

Six experiments were conducted investigating perceived auditory continuity with alternately rising and falling frequency glides, in which the glides were perceived as continuous when deleted portions were replaced by white noise bursts. The first three experiments showed that perceptual continuity could be obtained when the deleted portion came either in the middle of the glide, or at the top and bottom of the glides; continuity was actually better for the latter condition. Also, it was found that as glide duration increased, the threshold between perceived continuity and discontinuity increased; there was a similar increase as the difference between highest and lowest frequencies increased. It was also found, in Experiments IV-VI, that when the peak was deleted and replaced with noise, there was no perceptual extrapolation of the incomplete glides; rather, there seemed to be considerable rounding off of the trajectory of the glide. An interesting auditory phenomenon which has recently been investigated is perceived auditory continuity, or auditory induction (Warren, Obusek, & Ackroff, 1972). Basically, the phenomenon is one in which a soft, intermittent sound is perceived as continuous when the breaks are filled in with a louder sound. This phenomenon has been observed using a variety of stimulus materials, such as two tones of different frequencies and amplitudes (Thurlow, 1957; Thurlow & Elfner, 1959; Elfner & Homick, 1967b); soft noise bursts with loud tones as the interpolated items (Elfner & Caskey, 1965; Elfner & Homick, ig66; Elfner & Marsella, 1966; Elfner & Homick, 1967a; Elfner, 1969; Elfner, 1971); a soft tone with loud noise bursts as the interpolated signal (Warren, Obusek, & Ackroff, 1972); and soft speech stimuli with louder sounds filling in gaps in the speech (Miller & Licklider, 1950; Cherry & Wiley, 1967; Holloway, 1970; Warren & Obusek, 1971; Warren & Sherman, 1974). What all of these studies suggest is that there seems to be some sort of general mechanism which allows the auditory system to generate missing auditory stimuli under certain conditions. One implication of studies such as that by Cherry and Wiley (1967) is that missing formant transitions must be regenerated for the speech to sound smooth and continuous. The importance of frequency transitions in speech has recently been emphasized by Cole and Scott (1973). They showed that when the transition in a cv syllable is removed, subjects hear the consonant perceptually segregated from the vowel. Rather than hearing a repeating cv syllable, subjects heard a repeating vowel with a consonant-like noise in the background; the cv syllable was no longer perceived as a unit. Cole and Scott suggested that the important role which frequency transitions play in speech perception is to hold the speech stream together, to prevent the sounds from becoming grouped on the basis of frequency. Bregman and Dannen*Experiments i-m of this research were presented at the 1974 meeting of the Eastern Psychological Association, Philadelphia, Pa., and Experiments iv—vi were presented at the 1974 meeting of the Canadian Psychological Association, Windsor, Ontario. This paper is based on part of a PH D thesis by the author. The work was supported by the following giants from the National Research Council of Canada awarded to Dr Albert Bregman: A-0127, E-3221, E-3338, and E-3413. Additional support was provided by Defence Research Board grant no. 9401-40 to Dr Albeit Bregman, giants from the FCAC program of the Quebec Ministry of Education, the McGill University Graduate Faculty, and a National Research Council of Canada Postgraduate Scholarship awarded to the author. The author wishes to thank Dr Albert Bregman for his advice throughout this research. Canad. J. Psychol./Rev. Canad. Psychol., 1976, 30 (2) 99 bring (1973) have shown that frequency transitions do seem to play such a role with non-speech sounds. They found that a rapidly presented sequence of high and low tones tends to split into two streams based on frequency, a high stream and a low stream. However, adding frequency transitions between high and low tones, or adding partial transitions which merely 'pointed' at the neighbouring tones, reduced the tendency for segregation to occur, and enabled subjects to more accurately report the correct order of the tones. Thus, it appears that frequency transitions are important in allowing the listener to correctly perceive an ongoing stream of auditory material. In addition, it seems probable that the auditory system must be able to generate frequency transitions which are missing or masked. The present studies, therefore, have been designed as a preliminary investigation of the ability of the auditory system to perceptually generate missing portions of frequency transitions. A total of six experiments were conducted. The first three experiments looked at perceived auditory continuity when the tonal stimuli were linear frequency transitions, varying the extent of the frequency change (A/), the rate of change, and the location of the noise burst relative to the sine tone transitions. Experiments iv, v, and vi investigated the nature of the perceptually continuous signal which occurs when noise bursts are located at the top and bottom of the frequency glides. The numbering of the experiments indicates the historical order in which they were conducted. E X P E R I M E N T S I—III Experiment 1 was designed to investigate perceived auditory continuity with rising and falling frequency glides which had noise bursts located in the middle of each glide. In Experiment 11, steady state tones were used to look at the basic effect of increasing tone duration. Finally, in Experiment in, the stimuli were again like those in Experiment 1, except that the noise bursts were located at the top and bottom of the transitions rather than in the middle.