Fuzzy clustering of gradient‐echo functional MRI in the human visual cortex. Part II: Quantification

Comment on “The physiology of developmental changes in BOLD functional imaging signals” by Harris, Reynell, and Attwell

Neural responses to visual stimuli are modulated by spatial and temporal context. For example, in primary visual cortex (V1), responses to an oriented target stimulus will be suppressed when embedded within an oriented surround stimulus. This suppression is orientation-specific, with the largest suppression observed when stimuli in the neuron's classical receptive field and surround are of similar orientation. In human psychological experiments, the tilt illusion and tilt aftereffect demonstrate an effect of context on perceived orientation of a target stimulus. Similar to the neurophysiological data, the strength of these effects is modulated by the orientation difference between the target stimulus and context. It has been hypothesized that the neural mechanism underlying both the tilt illusion and tilt aftereffect involves orientation-tuned inhibition in V1. However, to date there is no direct evidence linking human perception of these illusions with measurements of inhibition from human visual cortex. Here, we measured context-induced suppression of neural responses in human visual cortex using functional magnetic resonance imaging (fMRI). In the same participants, we also measured magnitudes of their tilt illusion and tilt aftereffect. Our data revealed a significant relationship between the magnitude of neural suppression in V1 and size of the tilt illusion and tilt aftereffect. That is, participants who showed stronger blood oxygenation level dependent (BOLD) suppression in V1 also perceived stronger shifts in illusory tilt. This agreement between perception and neural responses in human V1 suggests a shared inhibitory mechanism that mediates both spatial and temporal effects of context in human perception.

Nonlinear statistical properties of fMRI signals in human visual cortex during resting-state

Dynamic metabolic changes were investigated by functional magnetic resonance spectroscopy (fMRS) during sustained stimulation of human primary visual cortex. Two established paradigms, consisting of either a full-field or a small-circle flickering checkerboard, were employed to generate wide-spread areas of positive or negative blood oxygenation level-dependent (BOLD) responses, respectively. Compared to baseline, the glutamate concentration increased by 5.3% (p = 0.007) during activation and decreased by −3.8% (p = 0.017) during deactivation. These changes were positively correlated with the amplitude of the BOLD response (R = 0.60, p = 0.002) and probably reflect changes of tricarboxylic acid cycle activity. During deactivation, the glucose concentration decreased by −7.9% (p = 0.025) presumably suggesting increased consumption or reduced glucose supply. Other findings included an increased concentration of glutathione (4.2%, p = 0.023) during deactivation and a negative correlation of glutathione and BOLD signal changes (R = −0.49, p = 0.012) as well as positive correlations of aspartate (R = 0.44, p = 0.035) and N-acetylaspartylglutamate (R = 0.42, p = 0.035) baseline concentrations with the BOLD response. It remains to be shown in future work if the observed effects on glutamate and glucose levels deviate from the assumption of a direct link between glucose utilization and regulation of blood flow or support previous suggestions that the hemodynamic response is mainly driven by feedforward release of vasoactive messengers.

Functional magnetic resonance imaging (fMRI) studies of early sensory cortex often measure stimulus-driven increases in the blood oxygenation level-dependent (BOLD) signal. However, these positive responses are frequently accompanied by reductions in the BOLD signal in adjacent regions of cortex. Although this negative BOLD response (NBR) is thought to result from neuronal suppression, the precise relationship between local activity, suppression and perception remains unknown. By measuring BOLD signals in human primary visual cortex while varying the baseline contrast levels in the region affected by the NBR, we tested three physiologically-plausible computational models of neuronal modulation which could explain this phenomenon: a subtractive model, a response gain model and a contrast gain model. We also measured the ability of isoluminant contrast to generate an NBR. We show that the NBR can be modeled as a pathway-specific contrast gain modulation that is strongest outside the fovea. We found a similar spatial bias in a psychophysical study using identical stimuli, although these data indicated a response- rather than a contrast-gain mechanism. We reconcile these findings by proposing 1) that the NBR is associated with a long-range suppressive mechanism that hyperpolarizes a subset of magnocellularly-driven neurons at the input to V1; 2) that this suppression is broadly-tuned to match the spatial features of the mask region; 3) that increasing the baseline contrast in the suppressed region drives all neurons in the input layer, reducing the relative contribution of the suppressing subpopulation in the fMRI signal.

A data-driven unsupervised learning such as an independent component analysis was gainfully applied to bloodoxygenation- level-dependent (BOLD) functional magnetic resonance imaging (fMRI) data compared to a model-based general linear model (GLM). This is due to an ability of this unsupervised learning method to extract a meaningful neuronal activity from BOLD signal that is a mixture of confounding non-neuronal artifacts such as head motions and physiological artifacts as well as neuronal signals. In this study, we support this claim by identifying neuronal underpinnings of cigarette craving and cigarette resistance. The fMRI data were acquired from heavy cigarette smokers (n = 14) while they alternatively watched images with and without cigarette smoking. During acquisition of two fMRI runs, they were asked to crave when they watched cigarette smoking images or to resist the urge to smoke. Data driven approaches of group independent component analysis (GICA) method based on temporal concatenation (TC) and TCGICA with an extension of iterative dual-regression (TC-GICA-iDR) were applied to the data. From the results, cigarette craving and cigarette resistance related neuronal activations were identified in the visual area and superior frontal areas, respectively with a greater statistical significance from the TC-GICA-iDR method than the TC-GICA method. On the other hand, the neuronal activity levels in many of these regions were not statistically different from the GLM method between the cigarette craving and cigarette resistance due to potentially aberrant BOLD signals.

Comparison of two exploratory data analysis methods for fMRI: fuzzy clustering vs. principal component analysis

A paired-stimuli paradigm combined with fMRI was utilized to study the effect of gradient acoustic noise on fMRI response in the human primary visual cortex (V1) in terms of the auditory-visual cross-modal neural interaction. The gradient noise generated during the fMRI acquisition was used as the primary stimulus, and a single flashing light was used as the secondary stimulus. An interstimulus interval (ISI) separated the two. Six tasks were designed with different ISIs ranging from 50 to 700 ms. Both BOLD signal intensity and the number of activated pixels in V1 were analyzed and examined, and they showed a significant reduction when the gradient noise preceded the flashing light by approximately 300 ms. These results indicate that the gradient acoustic noise generated during fMRI acquisitions does interfere with neural behavior and the BOLD signal in the human visual cortex. This interference is modulated by the delay between the gradient noise and visual stimulation, and it can be studied quantitatively when the stimulation paradigm is designed appropriately. This study provides evidence of the auditory-visual interaction during fMRI studies, and the results should have an impact on fMRI applications.

Functional MR imaging was performed in sixteen healthy human subjects measuring both regional cerebral blood flow (CBF) and blood oxygen level dependent (BOLD) signal when visual and auditory stimuli were presented to subjects in the presence or absence of anesthesia. During anesthesia, 0.25 mean alveolar concentration (MAC) sevoflurane was administrated. We found that low-dose sevoflurane decreased the task-induced changes in both BOLD and CBF. Within the visual and auditory regions of interest inspected, both baseline CBF and the task-induced changes in CBF decreased significantly during anesthesia. Low-dose sevoflurane significantly altered the task-induced CBF-BOLD coupling; for a unit change of CBF, a larger change in BOLD was observed in the anesthesia condition than in the anesthesia-free condition. Low-dose sevoflurane was also found to have significant impact on the spatial nonuniformity of the task-induced coupling. The alteration of task-induced CBF-BOLD coupling by low-dose sevoflurane introduces ambiguity to the direct interpretation of functional MRI (fMRI) data based on only one of the indirect measures-CBF or BOLD. Our observations also indicate that the manipulation of the brain with an anesthetic agent complicates the model-based quantitative interpretation of fMRI data, in which the relative task-induced changes in oxidative metabolism are calculated by means of a calibrated model given the relative changes in the indirect vascular measures, usually CBF and BOLD.

Understanding the brain is aided by visualizing neural activity over time. The most popular method for doing so in humans is functional magnetic resonance imaging (fMRI)—a method that tracks blood oxygenation as a proxy for neural activity. fMRI relies on neurovascular coupling, the brain’s capacity to increase its blood supply locally and on demand. Apart from humans, fMRI can be also applied to experimental animals and thereby plays an essential role in translating findings across species. Additionally, the combination of animal fMRI with electrophysiological and optical methods is crucial for uncovering the neural correlates of the observed blood-oxygen-level-dependent (BOLD) fMRI signal. Since fMRI necessitates immobility, animals must be either restrained or anesthetized. Most researchers take the latter approach, for both practical and ethical reasons. However, anesthesia confounds the results of fMRI experiments by profoundly altering neural activity and by interfering with neurovascular coupling. This conundrum, which can be viewed both as a challenge and an opportunity, motivated the three studies presented in this thesis. The challenge lies in choosing the right anesthesia for animal fMRI experiments. The ideal anesthetic protocol must provide sufficient sedation, guarantee immobility, and crucially, preserve a degree of neural responsiveness and neurovascular coupling. Anesthetic protocols based on the continuous infusion of the sedative medetomidine exhibit these qualities and have thus become a popular choice for rats—the most widely used animal fMRI model. Despite this, it has not yet been established how fMRI readouts evolve over several hours of medetomidine anesthesia and how they are affected by variations in timing, dose, and route of administration. In my first study (Chapter 2), I used four different protocols of medetomidine administration to anesthetize rats for up to six hours and repeatedly evaluated stimulus-evoked responses and fMRI measures of functional connectivity. I found that the temporal evolution of fMRI readouts varied between administration schemes. Based on the results, I made recommendations regarding the administration of medetomidine and the timing of fMRI experiments. These factors are important for obtaining reproducible results and should be considered for the design and interpretation of future rat fMRI studies. The opportunity lies in exploiting anesthesia’s effects on fMRI to better understand large-scale phenomena in the anesthetized brain. The case in point is burst-suppression, a poorly understood pattern of neural activity that appears in deep anesthesia and coma. In animals anesthetized with isoflurane, burst-suppression has been associated with the widespread synchronization of brain areas. In the second study (Chapter 3), I used fMRI data from four species—humans, macaques, marmosets, and rats—to precisely describe the fMRI signatures of anesthesia-induced burst-suppression and to map their distribution across the brain. I discovered a marked difference between primates and rodents. In rats the entire neocortex engaged in burst-suppression, while in the three primate species certain cortical areas were excluded—most notably the visual cortex. Based on the fMRI data alone, I could not determine the underlying cause of this exclusion. I concluded that answering this question would necessitate direct recordings of neural activity in the visual cortex of both primates and rodents. In the third study (Chapter 4), I aimed to develop the methods required for such direct neural recordings. Specifically, I conducted a series of pilot experiments in isoflurane-anesthetized rats and demonstrated the feasibility of in vivo two-photon calcium imaging through chronically implanted cranial windows. I was able to record the activity of hundreds of layer 2/3 neurons in the rat somatosensory and visual cortex and confirm my previous findings regarding the pancortical distribution of burst-suppression. I also examined the effects of varying the isoflurane dose on spontaneous activity and stimulus-evoked responses, thereby reproducing several known properties of burst-suppression in rodents. The developed methods can be easily adapted to record from the marmoset visual cortex, with the aim of understanding the primate-rodent difference described in Chapter 3. The above studies showcase that anesthetizing animals for functional neuroimaging experiments should not be viewed as a necessary evil. Anesthetic protocols can be optimized to allow for a host of neuroscientific questions to be asked. Moreover, such experiments can shed light on the functional organization of the anesthetized brain and on elusive anesthetic mechanisms of actions.

Functional magnetic resonance imaging (fMRI), with blood oxygenation level-dependent (BOLD) contrast, is a widely used technique for studying the human brain. However, it is an indirect measure of underlying neuronal activity and the processes that link this activity to BOLD signals are still a topic of much debate. In order to relate findings from fMRI research to other measures of neuronal activity it is vital to understand the underlying neurovascular coupling mechanism. Currently, there is no consensus on the relative roles of synaptic and spiking activity in the generation of the BOLD response. Here we designed a modelling framework to investigate different neurovascular coupling mechanisms. We use Electroencephalographic (EEG) and fMRI data from a visual stimulation task together with biophysically informed mathematical models describing how neuronal activity generates the BOLD signals. These models allow us to non-invasively infer the degree of local synaptic and spiking activity in the healthy human brain. In addition, we use Bayesian model comparison to decide between neurovascular coupling mechanisms. We show that the BOLD signal is dependent upon both the synaptic and spiking activity but that the relative contributions of these two inputs are dependent upon the underlying neuronal firing rate. When the underlying neuronal firing is low then the BOLD response is best explained by synaptic activity. However, when the neuronal firing rate is high then both synaptic and spiking activity are required to explain the BOLD signal.

Aging effects on functional auditory and visual processing using fMRI with variable sensory loading

In the human brain, the incoming light to the retina is transformed into meaningful representations that allow us to interact with the world. In a similar vein, the RGB pixel values are transformed by a deep neural network (DNN) into meaningful representations relevant to solving a computer vision task it was trained for. Therefore, in my research, I aim to reveal insights into the visual representations in the human visual cortex and DNNs solving vision tasks. In the previous decade, DNNs have emerged as the state-of-the-art models for predicting neural responses in the human and monkey visual cortex. Research has shown that training on a task related to a brain region’s function leads to better predictivity than a randomly initialized network. Based on this observation, we proposed that we can use DNNs trained on different computer vision tasks to identify functional mapping of the human visual cortex. To validate our proposed idea, we first investigate a brain region occipital place area (OPA) using DNNs trained on scene parsing task and scene classification task. From the previous investigations about OPA’s functions, we knew that it encodes navigational affordances that require spatial information about the scene. Therefore, we hypothesized that OPA’s representation should be closer to a scene parsing model than a scene classification model as the scene parsing task explicitly requires spatial information about the scene. Our results showed that scene parsing models had representation closer to OPA than scene classification models thus validating our approach. We then selected multiple DNNs performing a wide range of computer vision tasks ranging from low-level tasks such as edge detection, 3D tasks such as surface normals, and semantic tasks such as semantic segmentation. We compared the representations of these DNNs with all the regions in the visual cortex, thus revealing the functional representations of different regions of the visual cortex. Our results highly converged with previous investigations of these brain regions validating the feasibility of the proposed approach in finding functional representations of the human brain. Our results also provided new insights into underinvestigated brain regions that can serve as starting hypotheses and promote further investigation into those brain regions. We applied the same approach to find representational insights about the DNNs. A DNN usually consists of multiple layers with each layer performing a computation leading to the final layer that performs prediction for a given task. Training on different tasks could lead to very different representations. Therefore, we first investigate at which stage does the representation in DNNs trained on different tasks starts to differ. We further investigate if the DNNs trained on similar tasks lead to similar representations and on dissimilar tasks lead to more dissimilar representations. We selected the same set of DNNs used in the previous work that were trained on the Taskonomy dataset on a diverse range of 2D, 3D and semantic tasks. Then, given a DNN trained on a particular task, we compared the representation of multiple layers to corresponding layers in other DNNs. From this analysis, we aimed to reveal where in the network architecture task-specific representation is prominent. We found that task specificity increases as we go deeper into the DNN architecture and similar tasks start to cluster in groups. We found that the grouping we found using representational similarity was highly correlated with grouping based on transfer learning thus creating an interesting application of the approach to model selection in transfer learning. During previous works, several new measures were introduced to compare DNN representations. So, we identified the commonalities in different measures and unified different measures into a single framework referred to as duality diagram similarity. This work opens up new possibilities for similarity measures to understand DNN representations. While demonstrating a much higher correlation with transfer learning than previous state-of-the-art measures we extend it to understanding layer-wise representations of models trained on the Imagenet and Places dataset using different tasks and demonstrate its applicability to layer selection for transfer learning. In all the previous works, we used the task-specific DNN representations to understand the representations in the human visual cortex and other DNNs. We were able to interpret our findings in terms of computer vision tasks such as edge detection, semantic segmentation, depth estimation, etc. however we were not able to map the representations to human interpretable concepts. Therefore in our most recent work, we developed a new method that associates individual artificial neurons with human interpretable concepts. Overall, the works in this thesis revealed new insights into the representation of the visual cortex and DNNs...

Functional correspondences in the human and marmoset visual cortex during movie watching: Insights from correlation, redundancy, and synergy.

A group comparison in fMRI data using a semiparametric model under shape invariance