Structure and Function of the Auditory and Vestibular Systems (Fall 2014) Auditory Cortex (3) Prof. Xiaoqin Wang

Transcription

1 Structure and Function of the Auditory and Vestibular Systems (Fall 2014) Auditory Cortex (3) Prof. Xiaoqin Wang Laboratory of Auditory Neurophysiology Department of Biomedical Engineering Johns Hopkins University web1.johnshopkins.edu/xwang

2 Outline Lecture 1: Anatomy of auditory thalamus and cortex a) Subdivisions of medial geniculate body (MGB) b) Multiple fields of auditory cortex c) Thalamo-cortial and cortico-cortical connections Lecture 2: Tonotopic organization and stimulus selectivity a) Tonotopic organization of auditory cortex b) Firing patterns and tuning to preferred stimulus Lecture 3: Temporal processing a) Temporal-to-rate transformation in A1 b) Temporal-to-rate transformation outside A1 Lecture 4: Spectral, intensity and spatial processing a) Spectral processing b) Intensity processing c) Spatial processing Lecture 5: Complex sound processing a) Pitch and harmonicity processing b) Vocalization processing

3 What is unique about the auditory system? Auditory cortex (7) Thalamus (6) Inferior colliculus (5) 1) Longer subcortical pathway 2) Spectrally overlapping, time-varying input signal 3) Sounds entering the ear from anywhere at anytime 4) Hearing-speaking: sensorymotor processing Cochlear nucleus (3) Auditory-nerve (2) MSO, LSO (4) [binaural processing] Sounds Cochlea (1) Left Ear Right Ear

4 Temporal processing: How does auditory cortex represent timevarying signals?

5 Spectral and Temporal Characteristics of Speech! Spectrum! Formant" Temporal! Fine temporal structure ( carrier )" Coarse temporal structure ( envelop )"

6 Reduction of spike timing precision in CN: Fine temporal structure ( carrier ) Phase-locking AN Synchronization Index CN (primary-like cell) AN CN (chopper cell) 5 (Blackburn and Sachs, 1989)

7 Reduction of spike timing precision in CN: Coarse temporal structure ( envelop ) Modulation transfer function CN (primary-like cell) AN CN (chopper cell) Modulation Frequency (Hz) Modulation Frequency (Hz) Rohde and Greenberg (1994)

8 Further reduction of spike timing precision in IC: Modulation function changed from low-pass to band-pass Modulation Frequency (Hz) Langner and Schreiner (1988)

9 Thalamus (MGB) spiking faster than auditory cortex (AC) AC AC 10 Hz Increasing modulation frequency (Stim: SAM) MGB AC MGB AC MGB MGB AC AC 140 Hz MGB MGB Creutzfeldt et al. (1980)

10 Temporal modulation preference in A1 largely independent of spectral contents rbmf tbmf Max sync. frequency Time (msec) BMF: Best modulation frequency (rbmf: firing-rate based, tbmf: firing synchrony based) Modulation Frequency (Hz) Liang et al. (2002)

11 Responses to SAM, SFM stimuli in awake marmoset A1 SAM SFM Liang et al. (2002)

12 Modulation transfer functions in gerbil IC and A1 Modulation Frequency (Hz) Ter-Mikaelian et. al. (J Neurosci, 2007)

13 Distributions of tbmf in different auditory cortical areas of the cat R C Imig and Reale (1980) tbmf: R f0 +R 2f0 Schreiner and Urbas (1988)

14 Stimulus-locked responses are progressively slowed down along ascending auditory pathway Auditory cortex Slow Fast Thalamus Inferior colliculus Synchronization Index (Vector strength) Auditory cortex Thalamus Inferior colliculus Cochlear nucleus Auditory nerve Cochlear nucleus Auditory nerve (Characteristic Frequency) (AN: Johnson, 1980, CN: Blackburn and Sachs, 1990)

15 Slower stimulus-following responses in anesthetized condition? Goldstein, Kiang and Brown (1959)

16 Click trains produce largely synchronized discharges in anesthetized condition Anesthetized Cat A1 ~ 80% ~ 20% Lu and Wang (2000)

17 Click trains produce both synchronized and non-synchronized responses in awake condition Stimulus Synchronized Responses! Non-synchronized Responses! Lu et al. (2001)

18 Dual Temporal and Rate Representations in Auditory Cortex Synchronized response Non-synchronized response Lu, Liang, Wang (Nature Neurosci., 2001)

19 Cortical Representations in Different Time Scales: A two-stage mechanism Lu et al. (2001)

20 Why are auditory responses slowed down in cortex? For the purpose of multi-sensory integration. Other sensory systems (visual, tactile) are much slower at the periphery, but discharge synchrony rates are similar across sensory cortex. This slow-down allows auditory information to be integrated with information from other sensory modalities at the same clock rate. For the purpose of auditory object processing which requires temporal integration over longer time windows.

21 How precise are spike timing in auditory cortex?

22 Spike timing is more precise at the first event than at successive events, more precise with sparse events than with densely packed events Lu and Wang (J. Neurophysiol., 2004)

23 Entropy computed from ISI distribution shows little temporal structure in non-synchronized responses Lu and Wang (J.Neurophysiol. 2004)

24 What about spike timing in auditory coryex? Auditory cortex marks sparse acoustic events (or onsets) with precise spike timing and transform rapidly occurring acoustic events into firing rate-based representations

25 The question is not whether spike timing is important, but under what stimulus conditions and for which population of cortical neurons. Synchronized Responses Non-Synchronized Responses Auditory cortex marks sparse acoustic events (or onsets) with precise spike timing and transform rapidly occurring acoustic events into firing rate-based representations Lu, Liang, Wang (Nature Neurosci., 2001)

26 The question is not whether spike timing is important, but under what stimulus conditions and for which population of cortical neurons. Synchronized Responses Non-Synchronized Responses Lu et al. (2001)

27 Spike timing is more sluggish in awake than in anesthetized condition in auditory cortex, but not so in IC A1 IC Ter-Mikaelian et. al. (J Neurosci, 2007)

28 Spike timing is more precise IC than in A1 Ter-Mikaelian et. al. (J Neurosci, 2007)

29 Summary: A1 representations to sequential events In awake condition: Synchronized population can explicitly represent slowly occurring sound sequences by their temporal discharge patterns. Non-synchronized population can implicitly represent rapidly occurring sound sequences by their firing rate. In anesthetized condition: Discharge synchronization rates are lower than those observed in awake condition. Non-synchronized responses are largely absent.

30 What about thalamus in awake condition?

31 Non-synchronized responses also observed in awake MGB Bartlett and Wang (J. Neurophysiol., 2007)

32 Greater extent of non-synchronized responses in A1 than in MGB MGB A1 Bartlett and Wang (J. Neurophysiol., 2007)

33 Progressive increase of non-synchronized responses from IC to A1 Bartlett and Wang (J. Neurophysiol., 2007)

34 Fewer synchronized responses in A1 than IC Fewer synchronized responses in awake A1 than anesthetized A1 Ter-Mikaelian et. al. (J Neurosci, 2007)

35 How does auditory cortex process sound sequences? Cortical processing of sound streams operates on a segment-by-segment basis rather than on a moment-by-moment basis as found in the auditory periphery Auditory cortex neurons mark sparse acoustic events (or onsets) with precise spike timing and transform rapidly occurring acoustic events into firing rate-based representations.

36 What else can firing rate-based representations encode? (There are always surprises )

37 Temporal-to-rate transformation at longer time scales Non-synchronized (Negative monotonic) Synchronized Non-synchronized (Positive monotonic) Inter Click Interval (ms) Thomas Lu (PhD Thesis, 2001) Lu et al. (Nat Neurosci. 2001)

38 Stimulus synchronization in the range of acoustic flutter Bendor and Wang (Nature Neuroscience 2007)

39 Firing rate-based representations of acoustic flutter All neurons (synchronized, mixed and unsynchronized) Bendor and Wang (Nature Neuroscience 2007)

40 Non-synchronizing firing also encodes low repetition rates Unsynchronized neurons Flutter Pitch Non-syncrhonized responses ( Lu et al. 2001) Bendor and Wang (Nat Neurosci. 2007)

41 Transformation of firing rate-based representations from A1 to rostral fields Bendor and Wang (Nature Neuroscience 2007)

42 Spectral/temporal integration pathways in primate auditory cortex Bendor and Wang (J. Neurophysiology, 2008)

43 A common coding strategy between the tactile and auditory system Synchronizing neuron in S1 Positive and Negative monotonic neurons in S2 Romo and Salinas (2003) Salinas et al. (2000)

44 What is the implication of these experiments? Non-synchronized responses are the results of temporal-to-rate transformations and represent processed (instead of preserved) information. The auditory cortex transforms stimulus features into internal representations that are no longer faithful replicas of acoustic dimensions.

45 Challenges in understanding cortical processing of acoustical information: 1) Transformation from isomorphic (faithful) to nonisomorphic representation of acoustic signals 2) Transformation from acoustical to perceptual dimension

46 Suggested readings: Lu, T., L. Liang and X. Wang. Temporal and rate representations of timevarying signals in the auditory cortex of awake primates. Nature Neuroscience 4: (2001) Bendor, D.A. and X. Wang. Differential neural coding of acoustic flutter within primate auditory cortex. Nature Neuroscience 10: (2007) Wang, X., T. Lu, D. Bendor and E.L. Bartlett. Neural coding of temporal information in auditory thalamus and cortex. Neuroscience 154: (2008) [Review]