Skip to main content
NCBI home page
The Journal of the Acoustical Society of America logoLink to The Journal of the Acoustical Society of America
. 2009 Jan;125(1):324–327. doi: 10.1121/1.3021308

Pitch discrimination interference: The role of ear of entry and of octave similarity

Hedwig E Gockel 1,a), Ervin R Hafter 2, Brian C J Moore 3
PMCID: PMC2677271  EMSID: UKMS4967  PMID: 19173419

Abstract

Gockel et al. [(2004). J. Acoust. Soc. Am. 116, 1092–1104] reported that discrimination of the fundamental frequency (F0) of two sequentially presented complex tones (the target) was impaired when an additional complex tone (the interferer) was presented simultaneously with and to the same ear as the target, even though the target and interferer were filtered into separate frequency regions. This pitch discrimination interference (PDI) was greatest when the target and interferer had similar F0s. The current study examined the role of relative ear of entry of the target and interferer and whether the dependence of the PDI effect on the relative F0 of target and interferer is based on pitch height (F0 as such) or pitch chroma (the musical note). Sensitivity (d′) was measured for discrimination of the F0 of a target with a nominal F0 of 88 Hz, bandpass filtered from 1375 to 1875 Hz. The interferer was bandpass filtered from 125 to 625 Hz. The contralateral interferer produced marked PDI, but smaller than for ipsilateral presentation. PDI was not larger when the interferer’s F0 was twice the nominal target F0 than when it was a factor of 1.9 or 2.1 higher.

INTRODUCTION

In the auditory system, complex sounds are filtered into frequency channels or bands by the action of the basilar membrane. It is commonly assumed that it is possible to attend selectively to single channels or combinations of channels so as to achieve good performance in detection and discrimination tasks (Fletcher, 1940; Moore et al., 1997). However, under some conditions, it appears to be difficult or impossible for human listeners to process the outputs of some channels while ignoring others. This is manifested in effects such as across-channel masking (Schooneveldt and Moore, 1987; Moore et al., 1990), modulation detection interference for amplitude and∕or frequency modulated tones (Yost and Sheft, 1989, 1990; Moore et al., 1991), pitch discrimination interference (PDI) (Gockel et al., 2004, 2005), interference with detection of an increment in level of a tone by amplitude modulation of a spectrally distant tone (Gallun and Hafter, 2006), and disruption of intensity discrimination by a spectrally remote sound that is roved in level (Buss, 2008).

In the case of PDI, discrimination of the fundamental frequency (F0) of a target complex tone that is filtered into a restricted spectral region can be impaired by the presence of an interfering complex tone with fixed F0 that is filtered into a different spectral region (Gockel et al., 2004, 2005). The effect is greatest when the F0 of the interferer is close to the (mean) F0 of the target. Furthermore, the effect is greater when the interferer has a high pitch salience than when it has a low pitch salience. For example, a strong PDI effect is obtained when the target contains only unresolved components, while the interferer contains resolved components (Gockel et al., 2004). Part of the PDI effect may be explicable in terms of perceptual grouping of the target and interferer, which would be stronger when their F0s were similar (Broadbent and Ladefoged, 1957; Brokx and Nooteboom, 1982). This could explain the importance of the relative pitch salience of the target and interfering sounds. If the two sounds are perceptually fused, the higher the pitch salience of the interferer with fixed F0 the more it will “dilute” the perceived pitch change produced by the change in F0 of the target, leading to impaired F0 discrimination of the target. However, perceptual grouping probably cannot explain the whole of the PDI effect since PDI occurs even under conditions where strong cues are available to promote perceptual segregation of the target and interferer. For example, PDI is reduced, but not eliminated, by gating the interferer on before and off after the target, or by presenting the interferer continuously (Gockel et al., 2004).

In the present paper, two factors that might affect PDI are explored. The first factor is the ear of entry of the target and interferer. Presentation of two sounds, one to each ear, can provide a powerful cue for perceptual segregation of the sounds (Cherry, 1953; Hartmann and Johnson, 1991), especially when the sounds do not overlap spectrally (Darwin and Hukin, 2004). The target and interferer used here were filtered into two spectral regions separated by one octave, and their F0 was never identical. When these stimuli were presented dichotically (target to one ear and interferer to the other), subjects reported hearing two independent sound sources, one at each ear. In contrast, when two tone complexes have the same F0 and the complexes are filtered into two separate spectral regions with overlapping filter slopes, listeners report a single sound source coming from the center of the head (Broadbent and Ladefoged, 1957). If perceptual grouping is important for PDI, then presentation of the target and interferer to opposite ears (called the dichotic condition) should lead to a strong reduction of PDI relative to the case where the target and interferer are presented to the same ear. However, if some PDI persisted in the dichotic condition, this would provide evidence to support the idea that PDI does not depend entirely on perceptual grouping. Furthermore, the presence of PDI in the dichotic condition would indicate that at least part of the PDI effect occurs at a level in the auditory system where there is significant binaural interaction.

The second factor explored in the present paper is concerned with the relative F0s of the target and interferer. Gockel et al. (2004) argued that similarity in pitch of the target and interferer is an important factor influencing the amount of PDI. However, pitch has two dimensions, pitch height and pitch chroma (Bachem, 1950). Two tones whose F0s are separated by an octave have different pitch heights but the same pitch chroma. The question addressed here was: is the similarity of pitch that governs PDI based on pitch height, pitch chroma, or a combination of the two? To address this question, PDI was measured for a wide range of F0 relationships between the target and interferer, including an octave relationship.

METHOD AND STIMULI

Stimuli

In a two-interval two-alternative forced choice (2AFC) task, subjects had to discriminate between the F0s of two sequentially presented target complex tones with a nominal F0 of 88 Hz and a fixed difference, ΔF0, between the F0s of the two tones within a trial. In one, randomly chosen, interval, the target complex had an F0 equal to F0−ΔF0∕2, while in the other interval its F0 was F0+ΔF0∕2. The value of ΔF0 was selected and fixed for each subject based on pilot data, so that in the easiest condition, i.e., in the absence of an interferer, the sensitivity index, d′, was between 1.5 and 2.0. The following values for ΔF0 were used: 3.5% for two subjects, 5.0% for three subjects, and 7.0% for one subject. The target was presented either alone or with a synchronously gated harmonic complex (the interferer), which had an F0 that was the same in the two intervals of a trial. The F0 of the interferer was either equal to the arithmetic mean of the actual F0s of the two target complexes in the two intervals or increased by various amounts (15%, 45%, 90%, 100%, or 110%). Thus, the interferer’s F0 was never identical to the F0 of the simultaneously presented target complex. The target was always presented to the left ear, while the interferer was presented either ipsilaterally or contralaterally to the target. Each harmonic complex was bandpass filtered (slopes of 48 dB∕octave) into one of two frequency regions. The target was bandpass filtered into a “MID” region with nominal cutoff frequencies (3-dB down points) of 1375 and 1875 Hz, while the interferer was filtered into a “LOW” region with nominal cutoff frequencies of 125 and 625 Hz. Thus, the interferer would contain resolved harmonics (Plomp, 1964; Plomp and Mimpen, 1968; Moore and Ohgushi, 1993; Shackleton and Carlyon, 1994; Bernstein and Oxenham, 2003), while the target would contain only unresolved harmonics. These frequency regions were chosen following Carlyon and Shackleton (1994), and the stimuli were similar to those used in experiment 1 of Gockel et al. (2004), except that in the previous study the target complex was bandpass filtered between 1375 and 15 000 Hz. In the present study, the bandwidth of the target was chosen to be identical to that of the interferer because we wanted to produce a large PDI effect (when the target and interferer had equal F0s) to maximize the chance of observing differences between PDIs for interferers whose F0s were widely separated from the nominal target F0.

The nominal F0 was randomly varied between trials over the range of ±10%. The filter cutoffs for both the LOW and the MID region were varied by the same factor in order to keep fixed the degree of resolvability of the components. This F0 randomization encouraged subjects to compare the pitch of the two target complexes in each trial and discouraged them from basing their decision on a long-term memory representation of the sound. All complex tones had a root-mean-square (rms) level of 53 dB sound pressure level (SPL), and components were added in sine phase. The stimulus duration was 400 ms, including 5-ms raised-cosine onset and offset ramps. The silent interval between the two stimuli within a trial was 500 ms. To mask possible distortion products, a continuous pink background noise, spectrally limited between 2 and 1512 Hz (corresponding to 1.1×1375 Hz), was presented diotically in all conditions. Its overall rms level was 49.2 dB SPL. Its rms level in the region from 125 to 625 Hz, the nominal frequency band covered by the interferer, was 10 dB below that of the interferer.

Tones were generated and bandpass filtered digitally. They were played using a Turtle Beach Santa Cruz sound card (20-bit digital-to-analog conversion) with a sampling rate of 44.1 kHz. Stimuli were presented using Sennheiser HD580 headphones. Subjects were seated individually in a sound-attenuating booth.

Procedure

Subjects had to indicate the interval in which the target sound had the higher F0. They were requested to focus attention on the target and to ignore the interferer as much as possible when it was present. Each interval was marked by a light, and visual feedback was provided after each trial. If the interferer was present, it was present in both intervals of the 2AFC task. The amount by which the interferer’s F0 was increased above that of the nominal target F0 was fixed within a block of 105 trials; the first five trials in each block were considered as “warm-up” trials and were discarded. The upward shift of the interferer’s F0 was pseudorandomized between blocks. One block was run for each condition in turn before additional blocks were run in any other condition, with one exception: usually a block of “target alone” trials was run first, so that subjects were familiar with the timbre of the target. This was followed by a block with the interferer present. The relative ear of presentation of the interferer was fixed within a session; ipsilateral and contralateral conditions were run in alternating repeated sessions. The total duration of a single session was about 2 h, including rest times. For all subjects, at least 500 trials were run per condition. The data collection proper (after 6–10 h of practice) took between 15 and 22 h. To familiarize subjects with the procedure and equipment, they participated in at least three practice sessions, and up to two more if practice effects were seen. The practice covered conditions both with and without the interferer.

Subjects

Six subjects with normal hearing participated, one of whom was the first author. They ranged in age from 18 to 46 years. All of the subjects except one had some musical training. One had participated in earlier studies on PDI. All subjects participated in all conditions.

RESULTS AND DISCUSSION

The filled symbols in Fig. 1 show the mean d′ values and the corresponding standard errors averaged across five subjects as a function of the ratio between the interferer’s F0 and the nominal target F0. The data of one “aberrant” subject are plotted separately (open symbols) and will be described below. The upward and downward pointing triangles show results for contralateral and ipsilateral presentation of the interferer, respectively. For clarity, data for the contralateral interferer are plotted toward the left of their true values on the x-axis. Performance was best, with a d′ value of 1.86, in the absence of an interferer (filled circle in the left upper corner, condition None). Performance was worst, with a d′ value of 0.32, for the ipsilateral interferer with an F0 corresponding to the nominal target F0. The size of the PDI effect (d′ without interferer minus d′ in the presence of the interferer) for this condition was 1.54, which is substantially larger than the PDI of about 1.0 observed by Gockel et al. (2004) for the corresponding condition of their experiment 1. For the ipsilateral interferer whose F0 was 1.45 times the nominal target F0, PDI was about 0.7, which is substantially larger than the PDI of about 0.2 observed by Gockel et al. (2004) in their experiment 1 for an F0 ratio of 1.3 (the largest they employed). The larger PDI effects in the present study are probably due to the fact that here the target and interferer had equal bandwidths and overall levels, while in experiment 1 of the earlier study the level per component was the same for the target and interferer, and the target’s bandwidth was a factor of 27 larger than that of the interferer.

Figure 1.

Figure 1

Open in a new tab

Performance (d′) in an F0-discrimination task, plotted as a function of the ratio between the interferer’s F0 and the nominal target F0, which was 88 Hz. The target was bandpass filtered from 1375 to 1875 Hz, while the interferer was filtered from 125 to 625 Hz. The filled symbols connected by solid lines show the mean d′ values for five subjects who produced very similar results and the associated standard errors across subjects. The open symbols connected by dashed lines show the mean d′ values for one aberrant subject, who showed no tuning effect, and the associated standard errors across five blocks with 100 trials each. The downward and the upward pointing triangles show performance in the presence of ipsilaterally and contralaterally presented interferers, respectively. The filled and the open circles in the upper-left corner show performance in the absence of any interferer (condition None) for the five subjects and the single aberrant subject.

In the presence of a contralateral interferer, performance was clearly impaired relative to that without an interferer. However, the PDI effect was smaller than for an ipsilateral interferer. For both ipsi- and contralateral interferers, performance improved as the interferer’s F0 increased. Performance leveled out when the interferer’s F0 was more than 1.9 times that of the target, and there was no sign of a local increase in PDI (decrease in performance) when the interferer’s F0 was twice that of the target.

One subject showed a different pattern of results from all the others. The open symbols show the mean d′ and the corresponding standard errors across five blocks with 100 trials each, for this subject. While his performance was better than that of the other subjects when the interferer’s F0 was equal to the nominal target F0, he showed no tuning effect for either side of presentation; i.e., his performance was independent of the F0 of the interferer relative to that of the target. This lack of tuning of the PDI effect has not been observed among the 12 subjects tested in this and previous studies.

To examine the statistical significance of the results, a repeated measures two-way analysis of variance (ANOVA) with factors side of presentation of the interferer and interferer’s F0 was calculated, using the mean PDI value for each subject and condition as input. Note that the results of the ANOVA were basically unchanged when the data from the aberrant subject were excluded. The ANOVA showed a highly significant main effect of the F0 of the interferer [F(5,25)=23.57, p<0.001], a significant main effect of side of presentation of the interferer [F(1,5)=11.95, p<0.05], and a significant interaction between side and F0 [F(5,25)=5.31, p<0.05]. To check for an “octave effect,” two separate ANOVAs were calculated, one for the contralateral and one for the ipsilateral interferer, using only the PDI values for the three highest F0 values (F0 ratios of 1.9, 2.0, and 2.1). The results of both ANOVAs showed no significant main effect of the interferer’s F0. Thus, PDI did not locally increase when the interferer’s F0 was an octave above the nominal target F0. Planned t-tests for the three highest F0 values showed that for all cases but one (ipsilateral interferer with frequency ratio of 2.1), the observed PDI was significantly larger than zero (one-tailed, 5% error level).

In the present study, the relationship between the interferer’s F0 and the nominal target F0 was varied by increasing (rather than decreasing) the interferer’s F0 above that of the target. This was done to ensure that all components of the interferer were resolved in all conditions. Therefore, the current findings apply only to the situation in which the interferer’s F0 is one octave above that of the target. However, there is no obvious reason why an interferer with an F0 one octave below the nominal target F0 should have a special status, differing from that of an interferer with an F0 above the nominal target F0.

SUMMARY AND CONCLUSIONS

F0 discrimination between two target tones containing only unresolved harmonics was clearly impaired by the presence of an additional tone complex containing resolved harmonics. The impairment was less strong when the interferer was presented to the ear opposite to the ear receiving the target. However, in spite of the fact that ear of entry provides a strong segregation cue, a marked PDI effect was observed for the contralateral interferer. Furthermore, a small but significant PDI effect was observed even when the interferer was presented contralaterally to the target and had an F0 that was as much as a factor of 2.1 above the nominal target F0. These findings give further support to the idea that PDI can occur under conditions that are likely to lead to perceptual segregation of the target and interferer. The finding of significant PDI when the interferer was presented to the opposite ear to the target suggests that at least part of the PDI effect originates at a level of the auditory system where there is binaural interaction. No local increase in PDI was observed when the interferer’s F0 was an octave above that of the target complex. Thus, the similarity in pitch which has been found to influence PDI seems to be based on pitch height rather than on pitch chroma.

ACKNOWLEDGMENTS

This work was supported by NIH Grant No. DCD 00087. We thank Anastasios Sarampalis for help with calibration and generous “sharing” of his booth and office. We would also like to thank Richard Freyman and two anonymous reviewers for helpful comments on an earlier version of this paper.

References

  1. Bachem, A. (1950). “Tone height and tone chroma as two different pitch qualities,” Acta Psychol. 10.1016/0001-6918(50)90004-7 7, 80–88. [DOI] [Google Scholar]
  2. Bernstein, J. G., and Oxenham, A. J. (2003). “Pitch discrimination of diotic and dichotic tone complexes: Harmonic resolvability or harmonic number?,” J. Acoust. Soc. Am. 10.1121/1.1572146 113, 3323–3334. [DOI] [PubMed] [Google Scholar]
  3. Broadbent, D. E., and Ladefoged, P. (1957). “On the fusion of sounds reaching different sense organs,” J. Acoust. Soc. Am. 10.1121/1.1909019 29, 708–710. [DOI] [Google Scholar]
  4. Brokx, J. P. L., and Nooteboom, S. G. (1982). “Intonation and the perceptual separation of simultaneous voices,” J. Phonetics 10, 23–36. [Google Scholar]
  5. Buss, E. (2008). “Across-channel interference in intensity discrimination: The role of practice and listening strategy,” J. Acoust. Soc. Am. 10.1121/1.2816569 123, 265–272. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Carlyon, R. P., and Shackleton, T. M. (1994). “Comparing the fundamental frequencies of resolved and unresolved harmonics: Evidence for two pitch mechanisms?,” J. Acoust. Soc. Am. 10.1121/1.409971 95, 3541–3554. [DOI] [PubMed] [Google Scholar]
  7. Cherry, E. C. (1953). “Some experiments on the recognition of speech with one and two ears,” J. Acoust. Soc. Am. 10.1121/1.1907229 25, 975–979. [DOI] [Google Scholar]
  8. Darwin, C. J., and Hukin, R. W. (2004). “Limits to the role of a common fundamental frequency in the fusion of two sounds with different spatial cues,” J. Acoust. Soc. Am. 10.1121/1.1760794 116, 502–506. [DOI] [PubMed] [Google Scholar]
  9. Fletcher, H. (1940). “Auditory patterns,” Rev. Mod. Phys. 10.1103/RevModPhys.12.47 12, 47–65. [DOI] [Google Scholar]
  10. Gallun, F. J., and Hafter, E. R. (2006). “Amplitude modulation sensitivity as a mechanism for increment detection,” J. Acoust. Soc. Am. 10.1121/1.2200136 119, 3919–3930. [DOI] [PubMed] [Google Scholar]
  11. Gockel, H., Carlyon, R. P., and Moore, B. C. J. (2005). “Pitch discrimination interference: The role of pitch pulse asynchrony,” J. Acoust. Soc. Am. 10.1121/1.1898084 117, 3860–3866. [DOI] [PubMed] [Google Scholar]
  12. Gockel, H., Carlyon, R. P., and Plack, C. J. (2004). “Across-frequency interference effects in fundamental frequency discrimination: Questioning evidence for two pitch mechanisms,” J. Acoust. Soc. Am. 10.1121/1.1766021 116, 1092–1104. [DOI] [PubMed] [Google Scholar]
  13. Hartmann, W. M., and Johnson, D. (1991). “Stream segregation and peripheral channeling,” Music Percept. 9, 155–184. [Google Scholar]
  14. Moore, B. C. J., Glasberg, B. R., and Baer, T. (1997). “A model for the prediction of thresholds, loudness and partial loudness,” J. Audio Eng. Soc. 45, 224–240. [Google Scholar]
  15. Moore, B. C. J., Glasberg, B. R., Gaunt, T., and Child, T. (1991). “Across-channel masking of changes in modulation depth for amplitude- and frequency-modulated signals,” Q. J. Exp. Psychol. 43A, 327–347. [DOI] [PubMed] [Google Scholar]
  16. Moore, B. C. J., Glasberg, B. R., and Schooneveldt, G. P. (1990). “Across-channel masking and comodulation masking release,” J. Acoust. Soc. Am. 10.1121/1.399416 87, 1683–1694. [DOI] [PubMed] [Google Scholar]
  17. Moore, B. C. J., and Ohgushi, K. (1993). “Audibility of partials in inharmonic complex tones,” J. Acoust. Soc. Am. 10.1121/1.405625 93, 452–461. [DOI] [PubMed] [Google Scholar]
  18. Plomp, R. (1964). “The ear as a frequency analyzer,” J. Acoust. Soc. Am. 10.1121/1.1919256 36, 1628–1636. [DOI] [PubMed] [Google Scholar]
  19. Plomp, R., and Mimpen, A. M. (1968). “The ear as a frequency analyzer II,” J. Acoust. Soc. Am. 10.1121/1.1910894 43, 764–767. [DOI] [PubMed] [Google Scholar]
  20. Schooneveldt, G. P., and Moore, B. C. J. (1987). “Comodulation masking release (CMR): Effects of signal frequency, flanking-band frequency, masker bandwidth, flanking-band level, and monotic versus dichotic presentation of the flanking band,” J. Acoust. Soc. Am. 10.1121/1.395639 82, 1944–1956. [DOI] [PubMed] [Google Scholar]
  21. Shackleton, T. M., and Carlyon, R. P. (1994). “The role of resolved and unresolved harmonics in pitch perception and frequency modulation discrimination,” J. Acoust. Soc. Am. 10.1121/1.409970 95, 3529–3540. [DOI] [PubMed] [Google Scholar]
  22. Yost, W. A., and Sheft, S. (1989). “Across-critical-band processing of amplitude-modulated tones,” J. Acoust. Soc. Am. 10.1121/1.397556 85, 848–857. [DOI] [PubMed] [Google Scholar]
  23. Yost, W. A., and Sheft, S. (1990). “A comparison among three measures of cross-spectral processing of amplitude modulation with tonal signals,” J. Acoust. Soc. Am. 10.1121/1.398899 87, 897–900. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of the Acoustical Society of America are provided here courtesy of Acoustical Society of America

RESOURCES