Research > Audition

Auditory computation

I am interested in spike-based computation, especially in the auditory system. To understand what functional role synchrony might play, I introduced the concept of "synchrony receptive field" (SRF) in two neurons, which is the set of stimuli that elicit synchronous spiking in these two neurons (9). This is motivated by the observation that neurons are extremely sensitive to coincidences in their inputs (7). These synchrony fields represent structural properties of stimuli and have many interesting computational properties. For example, the SRF of two monaural neurons on opposite sides (left/right) is a spatial field, and detecting these location-specific synchrony patterns forms the basis of a simple and accurate model of sound localization (2). In this model, sound location is indicated by the activation of a specific assembly of binaural neurons, and the mapping from assembly to location can be determined by Hebbian learning (3). This hypothesis implies that the filtering properties of monaural inputs to a binaural neuron are precisely matched: this has been confirmed in the barn owl, and can emerge through spike-timing-dependent plasticity (6).

I have shown that to encode auditory features in realistic situations, neurons should have heterogeneous properties (1). This heterogeneity is indeed a feature in this theory of synchrony-based computation (9), in contrast with other theories of neural computation where variability is averaged in populations ("neural masses").

An intriguing observation in vivo is that the preferred interaural delay of binaural neurons depends on their preferred frequency. We have shown that this relationship is in fact expected from Hebbian learning, because of the temporal correlations in their inputs (4).

One practical issue when implementing spiking models of auditory function is that the timing of spikes depends on input level. I showed how to solve this issue with an adaptive threshold (8).

To design auditory models, we developed an auditory toolbox for the Brian simulator (5).

Relevant publications (chronological order):

1. Brette R (2010) On the interpretation of sensitivity analyses of neural responses, JASA 128(5), 2965-2972.
2. Goodman DF and R Brette (2010). Spike-timing-based computation in sound localization. PLoS Comp Biol 6(11): e1000993. doi:10.1371/journal.pcbi.1000993.
3. Goodman DF and R Brette (2010). Learning to localise sounds with spiking neural networks, Advances in Neural Information Processing Systems 23, 784-792.
4. Fontaine B and Brette R (2011). Neural development of binaural tuning through Hebbian learning predicts frequency-dependent best delays. J Neurosci 31(32):11692–11696.
5. Fontaine B, Goodman DFM, Benichoux F, Brette R (2011). Brian Hears: online auditory processing using vectorisation over channels. Front Neuroinf 5:9. doi: 10.3389/fninf.2011.00009.
6. Fischer BJ, Steinberg LJ, Fontaine B, Brette R, Peña JL (2011). Effect of instantaneous frequency glides on ITD processing by auditory coincidence detectors. PNAS 108(44): 18138-18143.
7. Rossant C, Leijon S, Magnusson AK, Brette R (2011). Sensitivity of noisy neurons to coincident inputs. J Neurosci 31(47):17193-17206.
8. Brette R (2012). Spiking models for level-invariant encoding. Front Comput Neurosci. 5:63. doi: 10.3389/fncom.2011.00063.
9. Brette R (2012). Computing with neural synchrony. PLoS Comp Biol. 8(6): e1002561. doi:10.1371/journal.pcbi.1002561.

is there!