Results for most listeners are uniquely consistent with the edge model of spatial hearing. My own sample results, over 10 repetitions, are available here or under the ‘Plot ≡’ menu.

Notes:

• • Previous similar experiments have typically adjusted binaural independence indices until specific measures of interaural correlation were obtained.  For comparison, correlation values obtained for a variety of indices can be found here.
• • That results are consistent with edge model of spatial hearing is not surprising since fluctuations of interaural phase and level  broaden as binaural independence indices decline from 1 to ~0.5.

   Additional notes:

• • The main purpose of the first 100 ms noise-burst is to ‘center’ a listener’s sense of auditory space before listening to the second 200 ms test sound, played after a 1.9 s delay. The presence of the first noise-burst, which always has a binaural independence index of 0, may nevertheless also increase the the number of trials during which subjects report hearing 2 edges or images. 
• • For intermediate indices between 0 and 1, the left and right noise-bursts (L and R) are described by the equations: L = a∙(1-index)+b∙(index) and R = a∙(1-index)+c∙(index), where a, b, and c are statistically independent noises and the resulting noise-bursts (L and R) are scaled to have amplitudes equivalent to those of a, b, and c.
• • All noise-bursts are generated with ^1/frequency spectra (‘pink’).
• • Large individual differences are typical in psychoacoustic experiments exploring binaural independence. So much so that some subjects may always report hearing two edges or images while other subjects may never report more than one.
• • Open-ear headphones may increase binaural independence across all sounds and therefore prompt slightly different results when compared to closed-ear headphones or earphones.
• • Listeners are less likely to hear two edges or two images when shorter test sounds are played (< 200 ms) and more likely when longer sounds are played. This is because recurring spatial cues available during the onset of a shorter sound contribute more to spatial hearing in comparison to spatial cues that fluctuate during the ongoing portion, when the binaural independence index is non-zero. In contrast, recurring cues available during the onset of a longer sound contribute less to spatial hearing in comparison to spatial cues that may fluctuate during the relatively long ongoing portion.
• • Binaural independence indices corresponding to scaled and summed noise-bursts differ from those achieved by scaling and summing phase and frequency values obtained using the Fourier transformation.
• • Reporting where edges or images are heard is optional. If adjusted during the test, reported azimuths can be viewed by clicking or tapping on the circle markers indicating percentages of trials when 2 edges or images  were indicated.
• • One should be able to copy response data, percentages, or reported azimuths to the clipboard (requires a modern browser). Copy and paste one of these data sets onto the above interface to view sample results
• • I would much appreciate receiving copies of results to summarize.
• • Tumult Hype, javascript, and Howler interface tested using Safari (MacOS) and iPhone (iOS 14).

Contact: Brian