Head-orienting behaviors during simultaneous speech detection and localization

Abstract

Head movement plays a vital role in auditory processing by contributing to spatial awareness and the ability to identify and locate sound sources. Here we investigate head-orienting behaviors using a dual-task experimental paradigm to measure: (a) localization of a speech source; and (b) detection of meaningful speech (numbers), within a complex acoustic background. Ten younger adults with normal hearing and 20 older adults with mild-to-severe sensorineural hearing loss were evaluated in the free field on two head-movement conditions: (1) head fixed to the front and (2) head moving to a source location; and two context conditions: (1) with audio only or (2) with audio plus visual cues. Head-tracking analyses quantified the target location relative to head location, as well as the peak velocity during head movements. Evaluation of head-orienting behaviors revealed that both groups tended to undershoot the auditory target for targets beyond 60° in azimuth. Listeners with hearing loss had higher head-turn errors than the normal-hearing listeners, even when a visual location cue was provided. Digit detection accuracy was better for the normal-hearing than hearing-loss groups, with a main effect of signal-to-noise ratio (SNR). When performing the dual-task paradigm in the most difficult listening environments, participants consistently demonstrated a wait-and-listen head-movement strategy, characterized by a short pause during which they maintained their head orientation and gathered information before orienting to the target location.

Keywords: dual-task; head movements; masking; spatial cues; speech detection; visual cues.

Figure 1

Audiogram showing mean pure tone thresholds with standard deviations for 10 normal-hearing (NH; solid lines) listeners and 20 listeners with hearing-loss (HL; dashed lines) for both right (red) and left (blue) ears in dB HL.

Figure 2

Experiment design. Audio and visual stimuli were presented to a participant seated at the center of the sound-field loudspeaker ring. Head movement was tracked in real time using a head tracking system composed of the head mount placed on the participant head (blue oval) and an infrared motion capture system (OptiTrack V120-Trio, pink oval). Thirteen 11.6″ video monitors (purple oval) are mounted at each loudspeaker (brown oval) location from −90° to +90° azimuth to provide visual cues. Participant’s digit detection response was collected via a Nintendo WiiMote (yellow oval).

Figure 3

Schematic of three head-movement conditions: (1) Fixed-head, (2) Moving (Audio + Visual), and (3) Moving (Audio Only). Open squares organized in a circle indicate a 24-loudspeaker array with 15-degree separation surrounding participant. Thirteen frontal hemifield loudspeakers had a video-monitor (black boxes) attached to them. Green boxes indicate video-monitor positions where visual cues were presented for condition-specific trials. An example target location is denoted by gray loud-speaker schematic for each condition, illustrating lack of useful information conveyed by the visual cues in the Moving (Audio Only) compared to the Moving (Audio + Visual) condition.

Figure 4

Example time-course data show yaw head-movement trajectory (purple solid line) for a single trial presentation. Audio waveform corresponding to the initial utterance shaded in gray, the mono-syllable speech in blue, and the interspersed target numbers in green. The reference point is noted at 0 degrees and the target in this trial was −75 degrees (yellow dashed line). The target stream began at 0 s followed by the initial head movement (black filled circle), peak velocity (black open circle), and maximum movement (black X marker).

Figure 5

Digit detection accuracy and accuracy difference. (A) Digit detection accuracy (%) across four different SNRs (−12, −6, 0, and + 6 dB) for three head-movement conditions [Fixed-head, Moving (Audio + Visual), and Moving (Audio Only)]. (B) Accuracy difference (%) between the fixed condition and each of the two moving conditions across SNRs (first and second columns) and between the two moving conditions across SNRs (third column). Mean and standard error of the mean plotted for normal-hearing listeners (solid lines with circle markers), and individuals with hearing loss (dashed lines with X markers). Asterisks indicate significant difference of post-hoc paired t-test at the 0.05 level (A) and significant difference of post-hoc one-sample t-test comparing to zero at the 0.05 level (B).

Figure 6

Panel (A–C) show peak head turn angle (in degrees) across target locations (in degrees) on three head-movement conditions. Panel (D) quantifies the difference between the two moving conditions. Normal-hearing group indicated with solid lines, hearing-loss group with dashed lines. Colors differentiate level of dB SNRs. In-figure texts indicate slopes for each group at each SNR. Panel (E): head turn angle (in degrees) for the −12 dB SNR of the hearing-loss group during the Moving (Audio Only) condition plotted as a function of target location (in degrees). Symbols indicate condition: open blue circles for no-digit detected trials and filled blue circles for ≥1-digit detected trials. Symbols correspond to the mean; error bars to standard error of mean. Dotted line represents ideal match between head orientation and target location. Asterisks denote target locations with a significant difference between head turn angle when 0-digit was detected compared to trials where at least 1-digit was detected (paired t-test).

Figure 7

Peak velocity (in degrees/s) across target locations for the two moving conditions (Moving: Audio + Visual; left panel, AudioOnly; right panel). Positive values indicate peak velocity to the right, negative values to the left, for normal-hearing (solid lines) and hearing-loss (dashed lines) groups. Colors differentiate the level of dB SNRs. In-figure texts denote slopes for each group at each SNR.

Figure 8

Dual-task temporal coherence between button presses and the initiation of head movements averaged from all 13 target-locations for each SNR for the Moving (Audio + Visual) condition (left panel) and for the Moving (Audio Only) condition (right panel). Solid line indicates normal-hearing listeners whereas dashed line indicates hearing-loss listeners. Asterisks denote significant differences at the 0.05 level, Bonferroni corrected.