Vision Science seminars

In many butterflies, the ancestral trichromatic insect colour vision, based on UV-, blue- and green-sensitive photoreceptors, is extended with red-sensitive cells. Physiological evidence for red receptors has been missing in nymphalid butterflies, although some species can discriminate red hues well. In eight species from genera Archaeoprepona, Argynnis, Charaxes, Danaus, Melitaea, Morpho, Heliconius and Speyeria, we found a novel class of green-sensitive photoreceptors that have hyperpolarizing responses to stimulation with red light. These green-positive, red-negative (G+R–) cells are allocated to positions R1/2, normally occupied by UV and blue-sensitive cells. Spectral sensitivity, polarization sensitivity and temporal dynamics suggest that the red opponent units (R–) are the basal photoreceptors R9, interacting with R1/2 in the same ommatidia via direct inhibitory synapses. We found the G+R– cells exclusively in butterflies with red-shining ommatidia, which contain longitudinal screening pigments. The implementation of the red colour channel with R9 is different from pierid and papilionid butterflies, where cells R5–8 are the red receptors. The nymphalid red-green opponent channel and the potential for tetrachromacy seem to have been switched on several times during evolution, balancing between the cost of neural processing and the value of extended colour information.

Vision begins in the eye, and what the “retina tells the brain” is a major interest in visual neuroscience. To deduce what the retina encodes (“tells”), computational models are essential. The most important models in the retina currently aim to understand the responses of the retinal output neurons – the ganglion cells. Typically, these models make simplifying assumptions about the neurons in the retinal network upstream of ganglion cells. One important assumption is linear spatial integration. In this talk, I first define what it means for a neuron to be spatially linear or nonlinear and how we can experimentally measure these phenomena. Next, I introduce the neurons upstream to retinal ganglion cells, with focus on bipolar cells, which are the connecting elements between the photoreceptors (input to the retinal network) and the ganglion cells (output). This pivotal position makes bipolar cells an interesting target to study the assumption of linear spatial integration, yet due to their location buried in the middle of the retina it is challenging to measure their neural activity. Here, I present bipolar cell data where I ask whether the spatial linearity holds under artificial and natural visual stimuli. Through diverse analyses and computational models, I show that bipolar cells are more complex than previously thought and that they can already act as nonlinear processing elements at the level of their somatic membrane potential. Furthermore, through pharmacology and current measurements, I illustrate that the observed spatial nonlinearity arises at the excitatory inputs to bipolar cells. In the final part of my talk, I address the functional relevance of the nonlinearities in bipolar cells through combined recordings of bipolar and ganglion cells and I show that the nonlinearities in bipolar cells provide high spatial sensitivity to downstream ganglion cells. Overall, I demonstrate that simple linear assumptions do not always apply and more complex models are needed to describe what the retina “tells” the brain.

Vision provides us with a holistic interpretation of the world that is, with very few exceptions, coherent and consistent across multiple levels of abstraction, from scene to objects to features. In this talk I will present results from past and ongoing work in my laboratory that investigates the role top-down signals play in establishing such coherent perceptual experience. Based on the results of several psychophysical experiments I will introduce a theory of “self-consistent inference” and show how it can account for human perceptual behavior. The talk will close with a discussion of how the theory can help us understand more cognitive processes.

Common factors are omnipresent in everyday life, e.g., it is widely held that there is a common factor g for intelligence. In vision, however, there seems to be a multitude of specific factors rather than a strong and unique common factor. In my thesis, I first examined the multidimensionality of the structure underlying visual illusions. To this aim, the susceptibility to various visual illusions was measured. In addition, subjects were tested with variants of the same illusion, which differed in spatial features, luminance, orientation, or contextual conditions. Only weak correlations were observed between the susceptibility to different visual illusions. An individual showing a strong susceptibility to one visual illusion does not necessarily show a strong susceptibility to other visual illusions, suggesting that the structure underlying visual illusions is multifactorial. In contrast, there were strong correlations between the susceptibility to variants of the same illusion. Hence, factors seem to be illusion-specific but not feature-specific. Second, I investigated whether a strong visual factor emerges in healthy elderly and patients with schizophrenia, which may be expected from the general decline in perceptual abilities usually reported in these two populations compared to healthy young adults. Similarly, a strong visual factor may emerge in action video gamers, who often show enhanced perceptual performance compared to non-video gamers. Hence, healthy elderly, patients with schizophrenia, and action video gamers were tested with a battery of visual tasks, such as a contrast detection and orientation discrimination task. As in control groups, between-task correlations were weak in general, which argues against the emergence of a strong common factor for vision in these populations. While similar tasks are usually assumed to rely on similar neural mechanisms, the performances in different visual tasks were only weakly related to each other, i.e., performance does not generalize across visual tasks. These results highlight the relevance of an individual differences approach to unravel the multidimensionality of the visual structure.

As a developmental outpocket of the brain, the retina exhibits features commonly found in most brain areas, including neurovascular interactions. In this presentation I will discuss various pathways that contribute to neurovascular interactions in the mammalian retina and present newly uncovered elements that likely participate in these pathways. Information obtained from retina could improve our understanding of neurovascular coupling pathways throughout the brain.

Visual comparisons are ubiquitous, and they can also be an important source for learning (e.g., Gentner et al., 2016; Kok et al., 2013). In science, technology, engineering, and math (STEM), key information is often conveyed through figures, graphs, and diagrams (Mayer, 1993). Comparing within and across visuals is critical for gleaning insight into the underlying concepts, structures, and processes that they represent. This talk addresses how people make visual comparisons and how visual comparisons can be best supported to improve learning. In particular, the talk will present a series of studies exploring the Spatial Alignment Principle (Matlen et al., 2020), derived from Structure-Mapping Theory (Gentner, 1983). Structure-mapping theory proposes that comparisons involve a process of finding correspondences between elements based on structured relationships. The Spatial Alignment Principle suggests that spatially arranging compared figures directly – to support correct correspondences and minimize interference from incorrect correspondences – will facilitate visual comparisons. We find that direct placement can facilitate visual comparison in educationally relevant stimuli, and that it may be especially important when figures are less familiar. We also present complementary evidence illustrating the preponderance of visual comparisons in 7th grade science textbooks.

Several lines of research have demonstrated that internally generated sensory experience - such as during memory, dreaming and mental imagery - activates similar neural representations as externally triggered perception. This overlap raises a fundamental challenge: how is the brain able to keep apart signals reflecting imagination and reality? In a series of online psychophysics experiments combined with computational modelling, we investigated to what extent imagination and perception are confused when the same content is simultaneously imagined and perceived. We found that simultaneous congruent mental imagery consistently led to an increase in perceptual presence responses, and that congruent perceptual presence responses were in turn associated with a more vivid imagery experience. Our findings can be best explained by a simple signal detection model in which imagined and perceived signals are added together. Perceptual reality monitoring can then easily be implemented by evaluating whether this intermixed signal is strong or vivid enough to pass a ‘reality threshold’. Our model suggests that, in contrast to self-generated sensory changes during movement, our brain does not discount self-generated sensory signals during mental imagery. This has profound implications for our understanding of reality monitoring and perception in general.

The brain represents the external world through the bottleneck of sensory organs. The network of hierarchically organized neurons is thought to recover the causes of sensory inputs to reconstruct the reality in the brain in idiosyncratic ways depending on individuals and their internal states. How can we understand the world model represented in an individual’s brain, or the neuroverse? My lab has been working on brain decoding of visual perception and subjective experiences such as imagery and dreaming using machine learning and deep neural network representations. In this talk, I will outline the progress of brain decoding methods and present how subjective experiences are externalized as images and how they could be shared across individuals via neural code conversion. The prospects of these approaches in basic science and neurotechnology will be discussed.

Our visual experience of living in a three-dimensional world is created from the information contained in the two-dimensional images projected into our eyes. The overlapping visual fields of the two eyes mean that their images are highly correlated, and that the small differences that are present represent an important cue to depth. Binocular neurons encode this information in a way that both maximises efficiency and optimises disparity tuning for the depth structures that are found in our natural environment. Neural network models provide a clear account of how these binocular neurons encode the local binocular disparity in images. These models can be expanded to multi-layer models that are sensitive to salient features of scenes, such as the orientations and discontinuities between surfaces. These deep neural network models have also shown the importance of binocular disparity for the segmentation of images into separate objects, in addition to the estimation of distance. These results demonstrate the usefulness of machine learning approaches as a tool for understanding biological vision.

Modeling neural responses to naturalistic stimuli has been instrumental in advancing our understanding of the visual system. Dominant computational modeling efforts in this direction have been deeply rooted in preconceived hypotheses. In contrast, hypothesis-neutral computational methodologies with minimal apriorism which bring neuroscience data directly to bear on the model development process are likely to be much more flexible and effective in modeling and understanding tuning properties throughout the visual system. In this study, we develop a hypothesis-neutral approach and characterize response selectivity in the human visual cortex exhaustively and systematically via response-optimized deep neural network models. First, we leverage the unprecedented scale and quality of the recently released Natural Scenes Dataset to constrain parametrized neural models of higher-order visual systems and achieve novel predictive precision, in some cases, significantly outperforming the predictive success of state-of-the-art task-optimized models. Next, we ask what kinds of functional properties emerge spontaneously in these response-optimized models? We examine trained networks through structural ( feature visualizations) as well as functional analysis (feature verbalizations) by running `virtual' fMRI experiments on large-scale probe datasets. Strikingly, despite no category-level supervision, since the models are solely optimized for brain response prediction from scratch, the units in the networks after optimization act as detectors for semantic concepts like `faces' or `words', thereby providing one of the strongest evidences for categorical selectivity in these visual areas. The observed selectivity in model neurons raises another question: are the category-selective units simply functioning as detectors for their preferred category or are they a by-product of a non-category-specific visual processing mechanism? To investigate this, we create selective deprivations in the visual diet of these response-optimized networks and study semantic selectivity in the resulting `deprived' networks, thereby also shedding light on the role of specific visual experiences in shaping neuronal tuning. Together with this new class of data-driven models and novel model interpretability techniques, our study illustrates that DNN models of visual cortex need not be conceived as obscure models with limited explanatory power, rather as powerful, unifying tools for probing the nature of representations and computations in the brain.

To better understand human scene understanding, we extracted features from images using CLIP, a neural network model of visual concept trained with supervision from natural language. We then constructed voxelwise encoding models to explain whole brain responses arising from viewing natural images from the Natural Scenes Dataset (NSD) - a large-scale fMRI dataset collected at 7T. Our results reveal that CLIP, as compared to convolution based image classification models such as ResNet or AlexNet, as well as language models such as BERT, gives rise to representations that enable better prediction performance - up to a 0.86 correlation with test data and an r-square of 0.75 - in higher-level visual cortex in humans. Moreover, CLIP representations explain distinctly unique variance in these higher-level visual areas as compared to models trained with only images or text. Control experiments show that the improvement in prediction observed with CLIP is not due to architectural differences (transformer vs. convolution) or to the encoding of image captions per se (vs. single object labels). Together our results indicate that CLIP and, more generally, multimodal models trained jointly on images and text, may serve as better candidate models of representation in human higher-level visual cortex. The bridge between language and vision provided by jointly trained models such as CLIP also opens up new and more semantically-rich ways of interpreting the visual brain.

How do humans perceive daily objects of various features and categorize these seemingly intuitive and effortless mental representations? Prior literature focusing on the role of the inferotemporal region (IT) has revealed object category clustering that is consistent with the semantic predefined structure (superordinate, ordinate, subordinate). It has however been debated whether the neural signals in the IT regions are a reflection of such categorical hierarchy [Wen et al.,2018; Bracci et al., 2017]. Visual attributes of images that correlated with semantic and category dimensions may have confounded these prior results. Our study aimed to address this debate by building and comparing models using the DNN AlexNet, to explain the variance in representational dissimilarity matrix (RDM) of neural signals in the IT region. We found that mid and high level perceptual attributes of the DNN model contribute the most to neural RDMs in the IT region. Semantic categories, as in predefined structure, were moderately correlated with mid to high DNN layers (r = [0.24 - 0.36]). Variance partitioning analysis also showed that the IT neural representations were mostly explained by DNN layers, while semantic categorical RDMs brought little additional information. In light of these results, we propose future works should focus more on the specific role IT plays in facilitating the extraction and coding of visual features that lead to the emergence of categorical conceptualizations.

A recent approach to understanding the mammalian visual system is to show correspondence between the sequential stages of processing in the ventral stream with layers in a deep convolutional neural network (DCNN), providing evidence that visual information is processed hierarchically, with successive stages containing ever higher-level information. However, correspondence is usually defined as shared variance between brain region and model layer. We propose that task-relevant variance is a stricter test: If a DCNN layer corresponds to a brain region, then substituting the model’s activity with brain activity should successfully drive the model’s object recognition decision. Using this approach on three datasets (human fMRI and macaque neuron firing rates) we found that in contrast to the hierarchical view, all ventral stream regions corresponded best to later model layers. That is, all regions contain high-level information about object category. We hypothesised that this is due to recurrent connections propagating high-level visual information from later regions back to early regions, in contrast to the exclusively feed-forward connectivity of DCNNs. Using task-relevant correspondence with a late DCNN layer akin to a tracer, we used Granger causal modelling to show late-DCNN correspondence in IT drives correspondence in V4. Our analysis suggests, effectively, that no ventral stream region can be appropriately characterised as ‘early’ beyond 70ms after stimulus presentation, challenging hierarchical models. More broadly, we ask what it means for a model component and brain region to correspond: beyond quantifying shared variance, we must consider the functional role in the computation. We also demonstrate that using a DCNN to decode high-level conceptual information from ventral stream produces a general mapping from brain to model activation space, which generalises to novel classes held-out from training data. This suggests future possibilities for brain-machine interface with high-level conceptual information, beyond current designs that interface with the sensorimotor periphery.

Selection is the hallmark of attention: processing improves for attended items but is relatively impaired for unattended items. It is well known that visual spatial attention changes sensory signals and perception in this selective fashion. In the work I will present, we asked whether and how attentional selection happens across time. First, our experiments revealed that voluntary temporal attention (attention to specific points in time) is selective, resulting in perceptual tradeoffs across time. Second, we measured small eye movements called microsaccades and found that directing voluntary temporal attention increases the stability of the eyes in anticipation of an attended stimulus. Third, we developed a computational model of dynamic attention, which proposes specific mechanisms underlying temporal attention and its selectivity. Lastly, I will mention how we are testing predictions of the model with MEG. Altogether, this research shows how precisely timed voluntary attention helps manage inherent limits in visual processing across short time intervals, advancing our understanding of attention as a dynamic process.