Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Oct 3;38(40):8574-8587.
doi: 10.1523/JNEUROSCI.0587-18.2018. Epub 2018 Aug 20.

Active Sound Localization Sharpens Spatial Tuning in Human Primary Auditory Cortex

Kiki van der Heijden  1 , Josef P Rauschecker  2 , Elia Formisano  1   3 , Giancarlo Valente  1 , Beatrice de Gelder  4   5
Affiliations

Active Sound Localization Sharpens Spatial Tuning in Human Primary Auditory Cortex

Kiki van der Heijden et al. J Neurosci. .

Abstract

Spatial hearing sensitivity in humans is dynamic and task-dependent, but the mechanisms in human auditory cortex that enable dynamic sound location encoding remain unclear. Using functional magnetic resonance imaging (fMRI), we assessed how active behavior affects encoding of sound location (azimuth) in primary auditory cortical areas and planum temporale (PT). According to the hierarchical model of auditory processing and cortical functional specialization, PT is implicated in sound location ("where") processing. Yet, our results show that spatial tuning profiles in primary auditory cortical areas (left primary core and right caudo-medial belt) sharpened during a sound localization ("where") task compared with a sound identification ("what") task. In contrast, spatial tuning in PT was sharp but did not vary with task performance. We further applied a population pattern decoder to the measured fMRI activity patterns, which confirmed the task-dependent effects in the left core: sound location estimates from fMRI patterns measured during active sound localization were most accurate. In PT, decoding accuracy was not modulated by task performance. These results indicate that changes of population activity in human primary auditory areas reflect dynamic and task-dependent processing of sound location. As such, our findings suggest that the hierarchical model of auditory processing may need to be revised to include an interaction between primary and functionally specialized areas depending on behavioral requirements.SIGNIFICANCE STATEMENT According to a purely hierarchical view, cortical auditory processing consists of a series of analysis stages from sensory (acoustic) processing in primary auditory cortex to specialized processing in higher-order areas. Posterior-dorsal cortical auditory areas, planum temporale (PT) in humans, are considered to be functionally specialized for spatial processing. However, this model is based mostly on passive listening studies. Our results provide compelling evidence that active behavior (sound localization) sharpens spatial selectivity in primary auditory cortex, whereas spatial tuning in functionally specialized areas (PT) is narrow but task-invariant. These findings suggest that the hierarchical view of cortical functional specialization needs to be extended: our data indicate that active behavior involves feedback projections from higher-order regions to primary auditory cortex.

Keywords: cortical functional specialization; fMRI; human auditory cortex; sound localization.

PubMed Disclaimer

Figures

Figure 1.
Figure 1.
Stimuli. A, Azimuth locations at which sound sources were presented. B, Example of a probe trial (top), a target trial for the sound localization task (middle), and a target trial for the sound identification task (bottom). A probe trial consisted of a block of five stimulus presentations at one azimuth location. In the sound localization task, the target trial consisted of five stimulus presentations as well, yet for the fourth (depicted here) or fifth repetition the azimuth location was changed. For target trials in the sound identification condition, azimuth location remained constant across the five stimulus repetitions but the fourth or fifth repetition was replaced by a deviant click train. C, Lines reflect the ITD (left) and ILD (right) for stimuli at a specific sound azimuth position, averaged across the binaural recordings of all participants. ILD was computed as the arithmetic difference in power (measured as root mean square) between the left and right channel of each binaural recording. To compute ITD, we first computed the interaural phase difference, which we subsequently converted to time differences. D, Plotted is the power spectrum of the left channel of the binaural recordings (i.e., the left ear) at specific azimuth positions, averaged across all participants. The difference in power in specific frequency bands dependent on sound azimuth location illustrates the availability of spectral cues in the recordings. Colors similar to C.
Figure 2.
Figure 2.
Estimating sound azimuth location with a maximum-likelihood population pattern decoder. Bottom row shows fMRI response (β value) to a sound presentation for individual voxels, with warmer colors (orange) indicating a weaker response (lower β value) and brighter colors (yellow) indicating a stronger response (higher β value). Small graphs show the log-likelihood function for each voxel for a given sound azimuth location (rows), with the fMRI response strength (β value) on the-x-axis, and the log-likelihood on the y-axis. Large graph on the right shows the resulting population log-likelihood function, which is the sum of the log-likelihood functions of the individual voxels at each location.
Figure 3.
Figure 3.
Parcellation of the human auditory cortex. A, The figure shows an enlarged view of the superior temporal plane in the right hemisphere, with a schematic overview of the parcellation used in the present study overlaid on top. B, Left and right superior temporal plane of a representative participant with the group map of frequency preference overlaid (top row; warm colors indicate a maximum response to low frequencies, cold colors to high frequencies), and frequency selectivity (bottom row; orange to green colors indicate broad tuning, blue to purple colors indicate progressively sharper tuning. C, Similar to A but displaying maps for a single representative participant.
Figure 4.
Figure 4.
Spatial selectivity across auditory cortical areas in humans. A, Boxplots show, for each cortical area, the distribution of the proportion of spatially selective voxels across participants (averaged across task conditions). B, Boxplots reflect the distribution of relative spatial tuning width (ERRF width, averaged across task conditions) across participants. The central circle of a box indicates the median of the distribution, the edges the 25th and 75th percentiles, and lines the full range of values. Circles represent outliers. Horizontal lines indicate a significant difference between areas at p < 0.05, FDR corrected for multiple comparisons at q < 0.05.
Figure 5.
Figure 5.
Task modulations of spatial selectivity in human auditory cortex. A, Boxplots show for each task condition the distribution of the proportion of voxels that exhibit a spatially selective response across participants. Black boxes indicate the passive listening condition, red boxes the sound identification condition, and blue boxes the sound localization condition. B, Boxplots reflect the distribution of relative spatial tuning width (ERRF width) across participants for each area and task condition. Colors similar to A. The central circle of a box indicates the median of the distribution, the edges the 25th and 75th percentiles, and lines the full range of values. Circles represent outliers. Horizontal lines with asterisks indicate a significant difference between areas at p < 0.05, FDR corrected for multiple comparisons at q < 0.05. C, Population RAFs are plotted for the spatially selective voxels within an area for the two active task conditions. RAFs are averaged across participants; blue lines indicate the sound identification condition, red lines the sound localization condition.
Figure 6.
Figure 6.
Sharper spatial selectivity during active sound localization is a result of response sharpening. Scatterplots show for each participant the average β value across voxels that exhibited sharper spatial selectivity (i.e., a decrease in ERRF width of 15% or more) during the sound localization condition (y-axis) than sound identification condition (x-axis) at the preferred (filled circles) and non-preferred location (open circles) for the left core region (left) and right CM (right). Circles below the diagonal reference line reflect a decrease in β value in the sound localization condition.
Figure 7.
Figure 7.
Decoding sound azimuth from population pattern activity in the core region and PT during a sound identification (“what”) and a sound localization (“where”) task. A, Lines reflect the average absolute error of the sound azimuth estimate resulting from the population pattern decoder (y-axis) as a function of actual sound azimuth (x-axis) for a particular cortical area and task condition. Light blue lines, Core region during sound identification task; dark blue lines, core region during sound localization task; light green lines, PT during sound identification task; dark green lines, PT during sound localization task. Error bars reflect the SEM. B, Boxplots of the absolute error of the sound azimuth estimates averaged across the seven sound azimuth positions. Colors similar to A. Horizontal black lines at the top of the figure indicate a significant difference in prediction error between cortical areas or task conditions [p < 0.05, q(FDR) < 0.05]. Horizontal red lines at the bottom of the figure indicate that the absolute error is below chance level [p < 0.05, q(FDR) < 0.05]. C, Lines reflect the performance of the population pattern decoder for PT controlled for the number of voxels. Similar to A, lines reflect the average absolute error. Solid lines are identical to those for area PT depicted in A. Dashed lines show the average absolute error across random samples (200 iterations) of voxels in PT. Specifically, for each participant we sampled a number of voxels from PT equal to the number of voxels included in the analysis for the core. Error bars reflect the SEM.
Figure 8.
Figure 8.
Decoding sound azimuth from population pattern activity across two hemispheres. A, Lines reflect the average absolute error of the sound azimuth estimate resulting from the population pattern decoder (y-axis) as a function of actual sound azimuth (x-axis) for a particular cortical area and task condition. Light blue lines: core region during sound identification task. Dark blue lines, Core region during sound localization task; light green lines, PT during sound identification task; dark green lines, PT during sound localization task. Error bars reflect the SEM. B, Boxplots of the absolute error of the sound azimuth estimates averaged across the seven sound azimuth positions. Colors similar to A. Gray boxes are identical to the boxes shown in Figure 7 and show the absolute error for the left hemisphere only (left-most gray box) and for the right hemisphere only (right-most gray box) for comparison. Horizontal black lines with asterisks at the top of the figure indicate a significant difference in prediction error between cortical areas or task conditions [p < 0.05, q(FDR) < 0.05]. Horizontal red lines at the bottom of the figure indicate that the absolute error is below chance level [p < 0.05, q(FDR) < 0.05].
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Ahissar M, Nahum M, Nelken I, Hochstein S (2009) Reverse hierarchies and sensory learning. Philos Trans R Soc Lond B Biol Sci 364:285–299. 10.1098/rstb.2008.0253 - DOI - PMC - PubMed
    1. Ahveninen J, Jääskeläinen IP, Raij T, Bonmassar G, Devore S, Hämäläinen M, Shinn-Cunningham BG, Witzel T, Belliveau JW (2006) Task-modulated “what” and “where” pathways in human auditory cortex. Proc Natl Acad Sci U S A 103:14608–14613. 10.1073/pnas.0510480103 - DOI - PMC - PubMed
    1. Alain C, Arnott SR, Hevenor S, Graham S, Grady CL (2001) “What” and “where” in the human auditory system. Proc Natl Acad Sci U S A 98:12301–12306. 10.1073/pnas.211209098 - DOI - PMC - PubMed
    1. Arnott SR, Binns MA, Grady CL, Alain C (2004) Assessing the auditory dual-pathway model in humans. Neuroimage 22:401–408. 10.1016/j.neuroimage.2004.01.014 - DOI - PubMed
    1. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 57:289–300.

Publication types

LinkOut - more resources