Tag Archives: localization

Lost in the gardens of Versailles

Like a lot of students, I used this past weekend as a chance to visit places easily accessible from Paris. On Saturday, Jamie, Genevieve, and I boarded the RER-C and headed to see the Château de Versailles. We went straight to the gardens, a gorgeous, intricate maze of hedges filled with sculptures and fountains. After aimlessly wandering far into the gardens, we heard classical music playing. We knew that it must have been the soundtrack to one of the weekend fountain shows, when the flow of water is set to follow the rhythm of music.

Map of the palace and gardens of Versailles

Just moments after the music started playing, we managed to find our way to the fountain show just by following the sound of the music. The water spouted and spun to the rhythm of the music and the slight mist was refreshingly cool in the middle of a hot summer day. Later that day, I was fascinated by how we managed to find the fountain show based only on the sound of the music. We had never been to the gardens before, we did not have a map, and we did not even know where we were going.

The fountain show

While it is often unconscious, determining where a sound is coming from is a remarkable ability. Figuring out where a sound originates can help us with everything from avoiding oncoming traffic to turning towards our friends in a crowded room (Dobreva et al., 2011; Middlebrooks JC, 2015). Part of this ability comes from having two ears because we can decide which direction a sound is coming from by comparing what we hear in our left and right ears (Joris and Yin, 2007).

However, in our daily lives, there are a lot of hidden challenges that make this task harder. Sound waves from a single source bounce off people and objects and ultimately hit our ears from several angles and directions.Both the unhindered sound itself and the reflections of that sound eventually reach our ears. The sound itself is known as the lead because, since it follows the most direct path, it hits the ear first (Brown et al., 2015). The ability to respond to the lead and not the subsequent reflections of that sound (the lags) is known as the precedence effect (Wallach et al. 1949).

The precedence effect is crucial for the accuracy of our sound localization. As we walked through the gardens of Versailles our ears were struck by soundwaves from both the music itself and reflections of the music bouncing off the greenery. The precedence effect is what allowed us to fuse these sounds together and find the fountain show rather than accidentally ending up at a tree the music had bounced off.

While theories about the precedence effect have been around for decades, the biological mechanisms underlying it were still unclear. Some scientists argued that this effect occurred within the brain while the sounds were being processed through synaptic inhibition (Pecka et al. 2007; Xia et al. 2010). Synaptic inhibition is when interconnected neurons send excitatory and inhibitory signals to amplify certain information (here, the lead sound) and depress other information (the lagging sounds).

Other scientists have argued that the effect could be a mechanical result of the cochlea, the inner ear where sounds are converted to electrical signals. These researchers note that once the lead hits the cochlea that sound continues to resonate. They contend that the lag cannot be communicated as strongly because the cochlea is still passing information about the lead (e.g. Bianchi et al., 2013).

In the past, there was limited evidence to supported one of these ideas over the other because it is technically difficult to impair the auditory structures of the ear or the auditory areas of the brain without impairing both. In their recent work, Brown et al. examined these theories by comparing normal hearing subjects to deaf subjects with cochlear implants, which directly stimulate the brain in response to sound. These deaf subjects had two implants and could still perceive sounds on either side of them but did not have functioning cochlea.

A cochlear implant

The researchers exposed subjects to lead-lag pairs of stimuli that mimicked a sound and the reflection of that sound. They asked subjects to indicate if they heard one sound or two and where the sound(s) originated. In normal hearing subjects these pairs were acoustic clicks. In deaf patients, they were electrical impulses sent directly into the cochlear implants. To measure the precedence effect, the researchers measured the subjects’ ability to recognize the two stimuli as one sound (termed “fusion”) and to determine the origin of the sound (“localization dominance”).

The authors found that both normal hearing and deaf patients could fuse the paired stimuli together and perceive them as one sound, although this ability was marginally weaker in the deaf patients. Furthermore, while there were idiosyncratic differences between individuals, whether subjects had cochlear implants did not affect their ability to determine the origin of sounds. This study presents evidence that people without cochlea can demonstrate the precedence effect at about the same levels as people with normal hearing. Since this effect can be seen in subjects without cochlea, this effect cannot be due to the mechanical features of the cochlea. This suggests that features of auditory neurons can account for the precedence effect that allows us to accurately localize sound.

This was a clever study but of course it is important to remember that it is not conclusive. There were small differences in deaf subjects’ ability to fuse the stimuli into a single sound, which could indicate that the cochlea at least contributes to the precedence effect. Also, mechanical aspects of structures beside the cochlea could be crucial. While this study is not conclusive it does highlight the importance of synaptic inhibition. This provides a launching pad for the continued study of the biological mechanisms underlying the precedence effect, which could help with everything from more immersive virtual reality to better treatment for hearing loss.

References

Bianchi F, Verhulst S, Dau T (2013). Experimental evidence for a cochlear source of the precedence effect. Journal of the Association for Research in Otolaryngology JARO, 14(5):767–779.

Brown AD, Stecker GC, Tollin DJ (2015) The Precedence Effect in Sound Localization JARO, 16(1): 1-28

Brown AD, Jones HG, Kan A, Thakkar T, Stecker GC, Goupell MJ, Litovsky RY (2015). Evidence for a neural source of the precedence effect in sound localization. Journal of neurophysiology 114(5): 2991–3001

Dobreva MS, O’Neill WE, Paige GD (2011). Influence of aging on human sound localization. Journal of neurophysiology, 105(5): 2471–2486

Joris P, Yin TCT, (2007) A matter of time: internal delays in binaural processing, Trends in Neurosciences, 30(2): 70-78

Middlebrooks, JC (2015) Chapter 6 – Sound localization, Handbook of Clinical Neurology, 129: 99-116.

Pecka M,  Zahn TP, Saunier-Rebori B, Siveke I,  Felmy F, Wiegrebe L,  Klug A, Pollak GD, Grothe B (2007) Inhibiting the Inhibition: A Neuronal Network for Sound Localization in Reverberant Environments Journal of Neuroscience, 27(7):1782-1790

Wallach H, Newman EB, Rosenzweig R (1949) The precedence effect in sound localization. Am J Psychiatr 62:315–336

Xia J, Brughera A, Colburn HS, Shinn-Cunningham B (2010). Physiological and psychophysical modeling of the precedence effect. Journal of the Association for Research in Otolaryngology : JARO, 11(3), 495–513

 Diagram

Cochlear Implant: “Ryan-Funderburk-1.jpg” by Rfunderburk90 is licensed under CC PDM 1.0

Baroque Music at Sainte Chapelle

 

Last week, we took a group trip to Sainte Chapelle, a Gothic church in the heart of Paris. Though many visit the church to admire the magnificent collection of 13th century stained glass and extensive collection of Christian relics, we were there to hear an authentic Baroque music concert. Much of the hour-long concert featured a trio of harpsichord, violin, and recorder, but at one point, the recorder player took a solo and switched from playing the soprano version of the instrument to the higher-pitched sopranino. After the recorder player began his solo on the stage, he proceeded to wander throughout the entire chapel—through the audience, along the walls of the chapel, and at one point even directly behind the seating area.

When I said "heart of Paris," I meant it.

As the soloist walked around far behind me, I noticed that I was able to identify where the sound was coming from even though I was facing the opposite direction. I immediately remembered a lecture from neurobiology class that covered the neural mechanisms involved in sound localization. The process depends on the fact that our brain has the ability to compare inputs from both ears. When sound emanates from a source, it usually is not exactly the same distance away from each ear, so when the sound reaches both ears, the difference is calculated and converted into an angle to the direction of the sound (Grothe, 2003). Oftentimes, the difference between arrival at each ear can be as little as 10 microseconds (that’s 1/100,000th of a second)!. This calculation of interaural time difference (ITD) occurs in the medial superior olive, a collection of cells in the brainstem.

Basis of interaural time difference

I was curious, though, if the same process was taking place during the concert. I remembered that the ITD calculations are particularly effective for low frequency sounds, certainly lower than the high-pitched sopranino recorder (Devore and Delguette, 2010). Additionally, I imagined that the tall ceiling and compartmentalized roof, typical of Gothic architecture, would create echoes that further complicate the use of ITD to localize sound. I looked into the matter, and found an article that attempted to explore how the brain’s ability to localize sound is affected by reverberation (Devore and Delguette, 2010). The researchers thought that the lateral superior olive, which uses interaural level difference (ILD) to localize sound rather than ITD, would be preferentially involved in sounds distorted by reverberation. In order to test this hypothesis, the experimenters used a technique called in-vivo electrophysiology, which places a collection of small recording electrodes into the brain and is able to isolate activity from single brain cells in awake, behaving subjects. Once the electrodes were in place, the subjects (in this case, rabbits) were exposed to a variety of auditory stimuli. The stimuli varied in direction, pitch, and reverberant nature so that the researchers could determine which pathways were in use in different situations. They found that the brain relies more heavily on ILD than ITD in reverberant situations, particularly for high-frequency pitches.

Beautiful? Yes. Reverberant? Absolutely.

While researching sound localization, I stumbled across another article that I found to be particularly interesting. I have always been amazed by conductors’ abilities to identify a single individual who misses a note in a 70-person ensemble. Researchers found that the constant exposure to wide field multi source sound environments experienced by conductors actually changes the way that conductors’ brains process the information (Münte et al., 2001). Not surprisingly, conductors outperformed both non-musical controls and pianists in accuracy of sound localization, but interestingly, recordings of brain activity during the task revealed that all groups utilized identical regions of the brain. This finding led the researchers to suggest that the intensive training and exposure that conductors receive simply trains their brain to more effectively use the sound localization pathways that we all use everyday. Take home message: the phrase “mental exercise” has a good deal of validity to it.

 – Max Farina

 

Works Cited:

Devore S, Delguette B (2010) Effects of reverberation on the directional sensitivity of auditory neurons across the tonotopic axis: influences of interaural time and level differences. Journal of Neuroscience 30: 7826-7837.

Grothe B (2003) New roles for synaptic inhibition in sound localization. Nature Reviews Neuroscience 4: 540-550.

Münte TF, Kohlmetz C, Nager W, Altenmüller E (2001) Neuroperception: Superior auditory spatial tuning in conductors. Nature 409.