Functional magnetic resonance imaging (fMRI) is increasingly used to study functional connectivity in large-scale brain networks that support cognitive and perceptual processes. We face serious ...conceptual, statistical and data analysis challenges when addressing the combinatorial explosion of possible interactions within high-dimensional fMRI data. Moreover, we need to know, and account for, the physiological mechanisms underlying our signals. We argue here that (i) model selection procedures for connectivity should include consideration of more than just a few brain structures, (ii) temporal precedence – and causality concepts based on it – are essential in dynamic models of connectivity and (iii) undoing the effect of hemodynamics on fMRI data (by deconvolution) can be an important tool. However, it is crucially dependent upon assumptions that need to be verified.
Can we decipher speech content ("what" is being said) and speaker identity ("who" is saying it) from observations of brain activity of a listener? Here, we combine functional magnetic resonance ...imaging with a data-mining algorithm and retrieve what and whom a person is listening to from the neural fingerprints that speech and voice signals elicit in the listener's auditory cortex. These cortical fingerprints are spatially distributed and insensitive to acoustic variations of the input so as to permit the brain-based recognition of learned speech from unknown speakers and of learned voices from previously unheard utterances. Our findings unravel the detailed cortical layout and computational properties of the neural populations at the basis of human speech recognition and speaker identification.
Columnar arrangements of neurons with similar preference have been suggested as the fundamental processing units of the cerebral cortex. Within these columnar arrangements, feed-forward information ...enters at middle cortical layers whereas feedback information arrives at superficial and deep layers. This interplay of feedforward and feedback processing is at the core of perception and behavior. Here we provide in vivo evidence consistent with a columnar organization of the processing of sound frequency in the human auditory cortex. We measure submillimeter functional responses to sound frequency sweeps at high magnetic fields (7 tesla) and show that frequency preference is stable through cortical depth in primary auditory cortex. Furthermore, we demonstrate that—in this highly columnar cortex—task demands sharpen the frequency tuning in superficial cortical layers more than in middle or deep layers. These findings are pivotal to understanding mechanisms of neural information processing and flow during the active perception of sounds.
Visual face identification requires distinguishing between thousands of faces we know. This computational feat involves a network of brain regions including the fusiform face area (FFA) and anterior ...inferotemporal cortex (aIT), whose roles in the process are not well understood. Here, we provide the first demonstration that it is possible to discriminate cortical response patterns elicited by individual face images with high-resolution functional magnetic resonance imaging (fMRI). Response patterns elicited by the face images were distinct in aIT but not in the FFA. Individual-level face information is likely to be present in both regions, but our data suggest that it is more pronounced in aIT. One interpretation is that the FFA detects faces and engages aIT for identification.
Multi-Voxel Pattern Analysis (MVPA) is a well established tool to disclose weak, distributed effects in brain activity patterns. The generalization ability is assessed by testing the learning model ...on new, unseen data. However, when limited data is available, the decoding success is estimated using cross-validation. There is general consensus on assessing statistical significance of cross-validated accuracy with non-parametric permutation tests. In this work we focus on the false positive control of different permutation strategies and on the statistical power of different cross-validation schemes.
With simulations, we show that estimating the entire cross-validation error on each permuted dataset is the only statistically valid permutation strategy. Furthermore, using both simulations and real data from the HCP WU-Minn 3T fMRI dataset, we show that, among the different cross-validation schemes, a repeated split-half cross-validation is the most powerful, despite achieving slightly lower classification accuracy, when compared to other schemes. Our findings provide additional insights into the optimization of the experimental design for MVPA, highlighting the benefits of having many short runs.
Recent functional MRI (fMRI) studies have highlighted differences in responses to natural sounds along the rostral-caudal axis of the human superior temporal gyrus. However, due to the indirect ...nature of the fMRI signal, it has been challenging to relate these fMRI observations to actual neuronal response properties. To bridge this gap, we present a forward model of the fMRI responses to natural sounds combining a neuronal model of the auditory cortex with physiological modeling of the hemodynamic BOLD response. Neuronal responses are modeled with a dynamic recurrent firing rate model, reflecting the tonotopic, hierarchical processing in the auditory cortex along with the spectro-temporal tradeoff in the rostral-caudal axis of its belt areas. To link modeled neuronal response properties with human fMRI data in the auditory belt regions, we generated a space of neuronal models, which differed parametrically in spectral and temporal specificity of neuronal responses. Then, we obtained predictions of fMRI responses through a biophysical model of the hemodynamic BOLD response (P-DCM). Using Bayesian model comparison, our results showed that the hemodynamic BOLD responses of the caudal belt regions in the human auditory cortex were best explained by modeling faster temporal dynamics and broader spectral tuning of neuronal populations, while rostral belt regions were best explained through fine spectral tuning combined with slower temporal dynamics. These results support the hypotheses of complementary neural information processing along the rostral-caudal axis of the human superior temporal gyrus.
Sounds following a cue or embedded in a periodic rhythm are processed more effectively than sounds that are part of an aperiodic rhythm. One might also expect that a sound embedded in a periodic ...rhythm is processed more effectively than a sound following a single temporal cue. Such a finding would follow the theory that the entrainment of neural rhythmic activity by periodic stimuli renders the prediction of upcoming stimuli more efficient. We conducted a psychophysical experiment in which we tested the behavioral elements of this idea. Targets in periodic and aperiodic rhythms, if they occurred, always appeared at the same moment in time, and thus were fully predictable. In a first condition, participants remained unaware of this. In a second condition, an explicit instruction on the temporal location of the targets embedded in rhythms was provided. We assessed sensitivity and reaction times to the target stimuli in a difficult temporal detection task, and contrasted performance in this task to that obtained for targets temporally cued by a single preceding cue. Irrespective of explicit information about target predictability, target detection performance was always better in the periodic and temporal cue conditions, compared to the aperiodic condition. However, we found that the mere predictability of an acoustic target within a periodic rhythm did not allow participants to detect the target any better than in a condition where the target's timing was predicted by a single temporal cue. Only when participants were made aware of the specific moment in the periodic rhythm where the target could occur, did sensitivity increase. This finding suggests that a periodic rhythm is not automatically sufficient to provide perceptual benefits compared to a condition predictable yet not rhythmic condition (a cue). In some conditions, as shown here, these benefits may only occur in interaction with other factors such as explicit instruction and directed attention.
Ethological views of brain functioning suggest that sound representations and computations in the auditory neural system are optimized finely to process and discriminate behaviorally relevant ...acoustic features and sounds (e.g., spectrotemporal modulations in the songs of zebra finches). Here, we show that modeling of neural sound representations in terms of frequency-specific spectrotemporal modulations enables accurate and specific reconstruction of real-life sounds from high-resolution functional magnetic resonance imaging (fMRI) response patterns in the human auditory cortex. Region-based analyses indicated that response patterns in separate portions of the auditory cortex are informative of distinctive sets of spectrotemporal modulations. Most relevantly, results revealed that in early auditory regions, and progressively more in surrounding regions, temporal modulations in a range relevant for speech analysis (∼2–4 Hz) were reconstructed more faithfully than other temporal modulations. In early auditory regions, this effect was frequency-dependent and only present for lower frequencies (<∼2 kHz), whereas for higher frequencies, reconstruction accuracy was higher for faster temporal modulations. Further analyses suggested that auditory cortical processing optimized for the fine-grained discrimination of speech and vocal sounds underlies this enhanced reconstruction accuracy. In sum, the present study introduces an approach to embed models of neural sound representations in the analysis of fMRI response patterns. Furthermore, it reveals that, in the human brain, even general purpose and fundamental neural processing mechanisms are shaped by the physical features of real-world stimuli that are most relevant for behavior (i.e., speech, voice).
Multivariate pattern analysis (MVPA) in fMRI has been used to extract information from distributed cortical activation patterns, which may go undetected in conventional univariate analysis. However, ...little is known about the physical and physiological underpinnings of MVPA in fMRI as well as about the effect of spatial smoothing on its performance. Several studies have addressed these issues, but their investigation was limited to the visual cortex at 3T with conflicting results. Here, we used ultra-high field (7T) fMRI to investigate the effect of spatial resolution and smoothing on decoding of speech content (vowels) and speaker identity from auditory cortical responses. To that end, we acquired high-resolution (1.1mm isotropic) fMRI data and additionally reconstructed them at 2.2 and 3.3mm in-plane spatial resolutions from the original k-space data. Furthermore, the data at each resolution were spatially smoothed with different 3D Gaussian kernel sizes (i.e. no smoothing or 1.1, 2.2, 3.3, 4.4, or 8.8mm kernels). For all spatial resolutions and smoothing kernels, we demonstrate the feasibility of decoding speech content (vowel) and speaker identity at 7T using support vector machine (SVM) MVPA. In addition, we found that high spatial frequencies are informative for vowel decoding and that the relative contribution of high and low spatial frequencies is different across the two decoding tasks. Moderate smoothing (up to 2.2mm) improved the accuracies for both decoding of vowels and speakers, possibly due to reduction of noise (e.g. residual motion artifacts or instrument noise) while still preserving information at high spatial frequency. In summary, our results show that – even with the same stimuli and within the same brain areas – the optimal spatial resolution for MVPA in fMRI depends on the specific decoding task of interest.
In functional brain mapping, pattern recognition methods allow detecting multivoxel patterns of brain activation which are informative with respect to a subject's perceptual or cognitive state. The ...sensitivity of these methods, however, is greatly reduced when the proportion of voxels that convey the discriminative information is small compared to the total number of measured voxels. To reduce this dimensionality problem, previous studies employed univariate voxel selection or region-of-interest-based strategies as a preceding step to the application of machine learning algorithms.
Here we employ a strategy for classifying functional imaging data based on a multivariate feature selection algorithm, Recursive Feature Elimination (RFE) that uses the training algorithm (support vector machine) recursively to eliminate irrelevant voxels and estimate informative spatial patterns. Generalization performances on test data increases while features/voxels are pruned based on their discrimination ability.
In this article we evaluate RFE in terms of sensitivity of discriminative maps (Receiver Operative Characteristic analysis) and generalization performances and compare it to previously used univariate voxel selection strategies based on activation and discrimination measures. Using simulated fMRI data, we show that the recursive approach is suitable for mapping discriminative patterns and that the combination of an initial univariate activation-based (
F-test) reduction of voxels and multivariate recursive feature elimination produces the best results, especially when differences between conditions have a low contrast-to-noise ratio.
Furthermore, we apply our method to high resolution (2 × 2 × 2mm
3) data from an auditory fMRI experiment in which subjects were stimulated with sounds from four different categories. With these real data, our recursive algorithm proves able to detect and accurately classify multivoxel spatial patterns, highlighting the role of the superior temporal gyrus in encoding the information of sound categories. In line with the simulation results, our method outperforms univariate statistical analysis and statistical learning without feature selection.
PDF
