Evaluating auditory stream segregation of SAM tone sequences by subjective and objective psychoacoustical tasks, and brain activity

Abstract

Auditory stream segregation refers to a segregated percept of signal streams with different acoustic features. Different approaches have been pursued in studies of stream segregation. In psychoacoustics, stream segregation has mostly been investigated with a subjective task asking the subjects to report their percept. Few studies have applied an objective task in which stream segregation is evaluated indirectly by determining thresholds for a percept that depends on whether auditory streams are segregated or not. Furthermore, both perceptual measures and physiological measures of brain activity have been employed but only little is known about their relation. How the results from different tasks and measures are related is evaluated in the present study using examples relying on the ABA- stimulation paradigm that apply the same stimuli. We presented A and B signals that were sinusoidally amplitude modulated (SAM) tones providing purely temporal, spectral or both types of cues to evaluate perceptual stream segregation and its physiological correlate. Which types of cues are most prominent was determined by the choice of carrier and modulation frequencies (f mod) of the signals. In the subjective task subjects reported their percept and in the objective task we measured their sensitivity for detecting time-shifts of B signals in an ABA- sequence. As a further measure of processes underlying stream segregation we employed functional magnetic resonance imaging (fMRI). SAM tone parameters were chosen to evoke an integrated (1-stream), a segregated (2-stream), or an ambiguous percept by adjusting the f mod difference between A and B tones (Δf mod). The results of both psychoacoustical tasks are significantly correlated. BOLD responses in fMRI depend on Δf mod between A and B SAM tones. The effect of Δf mod, however, differs between auditory cortex and frontal regions suggesting differences in representation related to the degree of perceptual ambiguity of the sequences.

Keywords: BOLD response; amplitude modulation; auditory scene analysis; fMRI; temporal and spectral cues; time shift detection.

Figure 1

Schematic view of the ABA-triplets presented in the objective task that relied on the detection of a time shifted B signal. In the third ABA- triplet a black arrow indicates the shift of the B signal, whereas the dashed line indicates the former position the un-shifted B signal. Top: Schematic temporal view of the ABA- triplets that were sinusoidally amplitude modulated (SAM) tones. A and B SAM tones had the same carrier frequency (f_c) but different modulation frequencies (f_mod). The f_mod of the B SAM tone was always larger than the f_mod of the A SAM tone (f_{mod A}). Here the f_mod difference between A and B SAM tones (Δf_mod) is schematically shown (see Table 1 for exact values). Bottom: Schematic spectral view of the SAM ABA- triplets.

Figure 2

Psychometric function of one subject for one stimulus condition (i.e., exp. 2, f_c = 4 kHz). The d'-value is plotted in relation to the shift of the B-signal in ms (x-axes). The differently colored lines and symbols show the different Δf_mod conditions tested (see legend).The threshold criterion of d' = 1.8 is indicated by the dotted gray line. The shift detection threshold (d' = 1.8) was interpolated between data points lying above and below that d'-value. The slight differences in largest d' values are due to different false alarm rates for the different Δf_mod conditions.

Figure 3

Proportions of a 2-stream percept (mean and s.e.m.) are shown for the f_{mod A} of 100 Hz (orange) and 300 Hz (gray) for the measurements in quiet (lighter coloring: n = 6) and during fMRI (darker coloring: n = 13) for all Δf_mod conditions.

Figure 4

Shift detection thresholds of the B SAM tone (n = 6; mean and SEM) are shown for the Δf_{mod A} of 100 Hz (orange) and 300 Hz (gray) for all Δf_mod conditions.

Figure 5

Group average activation maps (13 subjects) and BOLD signal time courses within regions of interest in Experiment 1. The maps depict all brain regions showing positive or negative deflections of the BOLD signal in at least one of the three Δf_mod conditions compared to the baseline (t = 4.5, p < 0.002) for each of the two f_{mod A} (100, 300 Hz). Several regions that showed significant differences between conditions and f_{mod A} variants are labeled and the respective averaged BOLD signal time courses are assigned. Error bars represent SEM.

Figure 6

Proportions of a 2-stream percept (mean and s.e.m.) are shown for the f_c of 1 kHz (blue) and 4 kHz (brown) for the measurements in quiet (lighter coloring: n = 6) and during fMRI (darker coloring: n = 9) for all Δf_mod conditions.

Figure 7

Shift detection thresholds of the B SAM tone (n = 6; mean and SEM) are shown for the f_c of 1 kHz (blue) and 4 kHz (brown) for all Δf_mod conditions.

Figure 8

Group average activation maps (10 subjects) and BOLD signal time courses within regions of interest in Experiment 2. The maps depict all brain regions with positive deflections of the BOLD response in at least one of the three Δf_mod conditions compared to the baseline (t = 4.5, p < 0.002) for each f_c (1, 4 kHz). Several regions that showed significant differences between conditions are labeled and the respective averaged BOLD signal time courses are assigned. Error bars represent SEM.

Figure 9

Relationship between tasks and measures of stream segregation. The graph in (A) represents the proportion of a 2-stream percept for all tested Δf_mod stimulus conditions obtained in the subjective psychophysical task (y-axis) and the matching shift detection threshold obtained in the objective psychophysical task (x-axis). (B) For the purpose of comparison and as an example, the proportion of a 2-stream percept for all tested Δf_mod stimulus conditions obtained during fMRI (y-axis) is related to the strength of BOLD responses (beta weights) in left Heschl's gyrus (x-axis). The symbols represent the three possible combinations of f_c, f_{mod A}, and Δf_mod (see legend for values). Generally, the shading of the symbol represent the Δf_mod stimulus condition; The darkest shading represent values for the small Δf_mod stimulus condition, whereas the lightest shading represent values for the large Δf_mod stimulus condition. Mean and error bars (SEM) are presented.