CompCogSci Darmstadt

Materials with complex appearances, like textiles and foodstuffs, pose challenges for conventional theories of vision. How does the brain learn to see properties of the world—like the glossiness of a surface—that cannot be measured by any other senses? Recent advances in unsupervised deep learning may help shed light on material perception. I will show how an unsupervised deep neural network trained on an artificial environment of surfaces that have different shapes, materials and lighting, spontaneously comes to encode those factors in its internal representations. Most strikingly, the model makes patterns of errors in its perception of material that follow, on an image-by-image basis, the patterns of errors made by human observers. Unsupervised deep learning may provide a coherent framework for how many perceptual dimensions form, in material perception and beyond.

Human vision is full of puzzles. Observers can grasp the essence of a scene in an instant, yet when probed for details they are at a loss. People have trouble finding their keys, yet they may be quite visible once found. How does one explain this combination of marvelous successes with quirky failures? I will describe our attempts to develop a unifying theory that brings a satisfying order to multiple phenomena. One key is to understand peripheral vision. A visual system cannot process everything with full fidelity, and therefore must lose some information. Peripheral vision must condense a mass of information into a succinct representation that nonetheless carries the information needed for vision at a glance. We have proposed that the visual system deals with limited capacity in part by representing its input in terms of a rich set of local image statistics, where the local regions grow — and the representation becomes less precise — with distance from fixation. This scheme trades off computation of sophisticated image features at the expense of spatial localization of those features. What are the implications of such an encoding scheme? Critical to our understanding has been the use of methodologies for visualizing the equivalence classes of the model. These visualizations allow one to quickly see that many of the puzzles of human vision may arise from a single encoding mechanism. They have suggested new experiments and predicted unexpected phenomena. Furthermore, visualization of the equivalence classes has facilitated the generation of testable model predictions, allowing us to study the effects of this relatively low-level encoding on a wide range of higher-level tasks. Peripheral vision helps explain many of the puzzles of vision, but some remain. By examining the phenomena that cannot be explained by peripheral vision, we gain insight into the nature of additional capacity limits in vision. In particular, I will suggest that decision processes face general-purpose limits on the complexity of the tasks they can perform at a given time.

One of the most important questions in psychology and neuroscience is understanding how the outside world maps to internal representations. Classical psychophysics approaches to this problem have a number of limitations: they mostly study low dimensional perpetual spaces, and are constrained in the number and diversity of participants and experiments. As ecologically valid perception is rich, high dimensional, contextual, and culturally dependent, these impediments severely bias our understanding of perceptual representations. Recent technological advances—the emergence of so-called “Virtual Labs”— can significantly contribute toward overcoming these barriers. Here I present a number of specific strategies that my group has developed in order to probe representations across a number of dimensions. 1) Massive online experiments can increase significantly the amount of participants and experiments that can be carried out in a single study, while also significantly diversifying the participant pool. We have developed a platform, PsyNet, that enables “experiments as code,” whereby the orchestration of computer servers, recruiting, compensation of participants, and data management is fully automated and every experiment can be fully replicated with one command line. I will demonstrate how PsyNet allows us to recruit thousands of participants for each study with a large number of control experimental conditions, significantly increasing our understanding of auditory perception. 2) Virtual lab methods also enable us to run experiments that are nearly impossible in a traditional lab setting. I will demonstrate our development of adaptive sampling, a set of behavioural methods that combine machine learning sampling techniques (Monte Carlo Markov Chains) with human interactions and allow us to create high-dimensional maps of perceptual representations with unprecedented resolution. 3) Finally, I will demonstrate how the aforementioned methods can be applied to the study of perceptual priors in both audition and vision, with a focus on our work in cross-cultural research, which studies how perceptual priors are influenced by experience and culture in diverse samples of participants from around the world.

Goal-directed navigation requires precise estimates of spatial relationships between current position and future goal, as well as planning of an associated route or action. While neurons in the hippocampal formation can represent the animal’s position and nearby trajectories, their role in determining the animal’s destination or action has been questioned. We thus hypothesize that brain regions outside the hippocampal formation may play complementary roles in navigation, particularly for guiding goal-directed behaviours based on the brain’s internal cognitive map. In this seminar, I will first describe a subpopulation of neurons in the retrosplenial cortex (RSC) that increase their firing when the animal approaches environmental boundaries, such as walls or edges. This boundary coding is independent of direct visual or tactile sensation but instead depends on inputs from the medial entorhinal cortex (MEC) that contains spatial tuning cells, such as grid cells or border cells. However, unlike MEC border cells, we found that RSC border cells encode environmental boundaries in a self-centred egocentric coordinate frame, which may allow an animal for efficient avoidance from approaching walls or edges during navigation. I will then discuss whether the brain can possess a precise estimate of remote target location during active environmental exploration. Such a spatial code has not been described in the hippocampal formation. However, we found that neurons in the rat orbitofrontal cortex (OFC) form spatial representations that persistently point to the animal’s subsequent goal destination throughout navigation. This destination coding emerges before navigation onset without direct sensory access to a distal goal, and are maintained via destination-specific neural ensemble dynamics. These findings together suggest key roles for extra-hippocampal regions in spatial navigation, enabling animals to choose appropriate actions toward a desired destination by avoiding possible dangers.

In Europe, the number of people aged 65 and older is increasing dramatically, and research related to ageing is more crucial than ever. The main research dedicated to age-related changes concentrates on cognitive or sensory deficits. This is also the case in vision research. However, the majority of older adults ages without major cognitive or optical or deficits. These are foremost good news, but even in the absence of neurodegenerative or eye diseases changes in visual perception occur. It has been suggested that age-related changes are due to a general decline of cognitive, perceptual and sensory functions. However, more recent studies reveal large individual differences within the ageing population and whereas some functions show age-related deterioration, others are surprisingly unaffected. Overall, it becomes increasingly apparent that perceptual changes in healthy ageing cannot be attributed to one single underlying factor. I will present studies from various areas of visual perception that challenge the view that age-related changes are primarily related to decline. Instead, our findings suggest that age-related changes are the result of visual experience, such that the brain ages optimally given the input it receives.

Several studies have showed that face stimuli elicit extremely fast and involuntary saccadic responses toward them, relative to other categories of visual stimuli. In the talk, I will mainly focus on a quite recent research done in our team that investigated to what extent face stimuli influence the programming and execution of saccades. In this research, two experiments were performed using a saccadic choice task: two images (one with a face, one with a vehicle) were simultaneously displayed in the left and right visual fields of participants who had to execute a saccade toward the image (Experiment 1) or toward a cross added in the center of the image (Experiment 2) containing a target stimulus (a face or a vehicle). As expected participants were faster to execute a saccade toward a face than toward a vehicle and did less errors. We also observed shorter saccades toward vehicle than face targets, even if participants were explicitly asked to perform their saccades toward a specific location (Experiment 2). Further analyses, that I will detailed in the talk, showed that error saccades might be interrupted in mid-fight to initiate a concurrently programmed corrective saccade.

Even young children seem to have an early understanding of the world around them, and the people in it. Before children can reliably say "ball", "wall", or "Saul", they expect balls to not go through walls, and for Saul to go right for a ball (if there's no wall). What is the formal conceptual structure underlying this commonsense reasoning about objects and agents? I will raise several possibilities for models underlying core intuitive physics as a way of talking about models of core knowledge and intuitive theories more generally. In particular, I will present some recent ML work trying to capture early expectations about object solidly, cohesion, and permanence, that relies on a rough-derendering approach.

The assumed data generating models (response distributions) of experimental or observational data in cognitive science have become increasingly complex over the past decades. This trend follows a revolution in model estimation methods and a drastic increase in computing power available to researchers. Today, higher-level cognitive functions can well be captured by and understood through computational cognitive models, a common example being drift diffusion models for decision processes. Such models are often expressed as the combination of two modeling layers. The first layer is the response distribution with corresponding distributional parameters tailored to the cognitive process under investigation. The second layer are latent models of the distributional parameters that capture how those parameters vary as a function of design, stimulus, or person characteristics, often in an additive manner. Such cognitive models can thus be understood as special cases of distributional regression models where multiple distributional parameters, rather than just a single centrality parameter, are predicted by additive models. Because of their complexity, distributional models are quite complicated to estimate, but recent advances in Bayesian estimation methods and corresponding software make them increasingly more feasible. In this talk, I will speak about the specification, estimation, and post-processing of Bayesian distributional regression models and how they can help to better understand cognitive processes.

I will begin by describing recent research taking a new, model-based approach to perceptual development. This approach uncovers fundamental changes in information processing underlying the protracted development of perception, action, and decision-making in childhood. For example, integration of multiple sensory estimates via reliability-weighted averaging – widely used by adults to improve perception – is often not seen until surprisingly late into childhood, as assessed by both behaviour and neural representations. This approach forms the basis for a newer question: the scope for the nervous system to deploy useful computations (e.g. reliability-weighted averaging) to optimise perception and action using newly-learned sensory signals provided by technology. Our initial model system is augmenting visual depth perception with devices translating distance into auditory or vibro-tactile signals. This problem has immediate applications to people with partial vision loss, but the broader question concerns our scope to use technology to tune in to any signal not available to our native biological receptors. I will describe initial progress on this problem, and our approach to operationalising what it might mean to adopt a new signal comparably to a native sense. This will include testing for its integration (weighted averaging) alongside the native senses, assessing the level at which this integration happens in the brain, and measuring the degree of ‘automaticity’ with which new signals are used, compared with native perception.

Humans have a remarkable ability to figure out what happened and why. In this talk, I will shed light on this ability from multiple angles. I will present a computational framework for modeling causal explanations in terms of counterfactual simulations, and several lines of experiments testing this framework in the domain of intuitive physics. The model predicts people's causal judgments about a variety of physical scenes, including dynamic collision events, complex situations that involve multiple causes, omissions as causes, and causal responsibility for a system's stability. It also captures the cognitive processes underlying these judgments as revealed by spontaneous eye-movements. More recently, we have applied our computational framework to explain multisensory integration. I will show how people's inferences about what happened are well-accounted for by a model that integrates visual and auditory evidence through approximate physical simulations.

In many societal problems, individuals exhibit a conflict between keeping resources (e.g., money, time or attention) to themselves or sharing them with another individual or group. The reasons motivating decisions in favor of others welfare can thereby vary from purely altruistic to completely strategic. Be it the stranger making an effort returning a lost valet to its rightful owner or a co-worker pitching in her fair share in a joint project. Actions like that create an environment that makes living together a pleasant experience. Hence, understanding how decisions determining the welfare of oneself and others are made is important for facilitating this behavior by building institutions that maximize the rate of cooperation in a society. To shed new light on such decision making processes I will present recent evidence from a set of process tracing experiments utilizing eye-tracking and economic games. Experiments will focus on the role of social preferences in the choice construction process and will identify mechanisms (i.e., search and processing depth, information weighting, and ignorance) through which they guide choice behavior. I will in particular focus on the differences and commonalitiesbetween strategic and altruistic decisions. Specifically, investigating to which extent people direct attention towards certain components of the decision situation in a context-dependent manner.

Theoretical linguistics postulates abstract structures that successfully explain key aspects of language. However, the precise relation between abstract theoretical ideas and empirical data from language use is not always apparent. Here, we propose to empirically test abstract semantic theories through the lens of probabilistic pragmatic modelling. We consider the historically important case of quantity words (e.g., `some', `all'). Data from a large-scale production study seem to suggest that quantity words are understood via prototypes. But based on statistical and empirical model comparison, we show that a probabilistic pragmatic model that embeds a strict truth-conditional notion of meaning explains the data just as well as a model that encodes prototypes into the meaning of quantity words.

Machine learning researchers frequently focus on human-level performance, in particular in games. However, in these applications human (or human-level) behavior is commonly reduced to a simple dot on a performance graph. Cognitive science, in particular theories of learning and decision making, could hold the key to unlock what is behind this dot, thereby gaining further insights into human cognition and the design principles of intelligent algorithms. However, cognitive experiments commonly focus on relatively simple paradigms such as restricted multi-armed bandit tasks. In this talk, I will argue that cognitive science can turn its lens to more complex scenarios to study exploration in real-world domains and online games. I will show in one large data set of online food delivery orders and across many online games how current cognitive theories of learning and exploration can describe human behavior in the wild, but also how these tasks demand us to expand our theoretical toolkit to describe a rich repertoire of real-world behaviors such as empowerment and fun.

Uncertainty plays a critical role in reinforcement learning and decision making. However, exactly how subjective uncertainty influences behaviour remains unclear. Multi-armed bandits are a useful framework to gain more insight into this. Paired with computational tools such as Kalman filters, they allow us to closely characterize the interplay between trial-by-trial value, uncertainty, learning, and choice. In this talk, I will present recent research where we also measured participants visual fixations on the options in a multi-armed bandit task. The estimated value of each option, and the uncertainty in these estimations, influenced what subjects looked at in the period before making a choice and their subsequent choice, as additionally did fixation itself. Uncertainty also determined how long participants looked at the obtained outcomes. Our findings clearly show the importance of uncertainty in learning and decision making.

Many psychological theories assume that qualitatively different cognitive processes can result in identical responses. Multinomial processing tree (MPT) models allow researchers to disentangle latent cognitive processes based on observed response frequencies. Recently, MPT models have been extended to explicitly account for participant and item heterogeneity. These hierarchical Bayesian MPT models provide the opportunity to connect two traditionally isolated disciplines. Whereas cognitive psychology has often focused on the experimental validation of MPT model parameters on the group level, psychometrics provides the necessary concepts and tools for measuring differences in MPT parameters on the item or person level. Moreover, MPT parameters can be regressed on covariates to model latent processes as a function of personality traits or other person characteristics.

How do people learn in real-world environments where the space of possible actions can be vast or even infinite? The study of human learning has made rapid progress in past decades, from discovering the neural substrate of reward prediction errors, to building AI capable of mastering the game of Go. Yet this line of research has primarily focused on learning through repeated interactions with the same stimuli. How are humans able to rapidly adapt to novel situations and learn from such sparse examples? I propose a theory of how generalization guides human learning, by making predictions about which unobserved options are most promising to explore. Inspired by Roger Shepard’s law of generalization, I show how a Bayesian function learning model provides a mechanism for generalizing limited experiences to a wide set of novel possibilities, based on the simple principle that similar actions produce similar outcomes. This model of generalization generates predictions about the expected reward and underlying uncertainty of unexplored options, where both are vital components in how people actively explore the world. This model allows us to explain developmental differences in the explorative behavior of children, and suggests a general principle of learning across spatial, conceptual, and structured domains.

Many regions of the human brain conduct highly specific functions, such as recognizing faces, understanding language, and thinking about other people’s thoughts. Why might this domain specific organization be a good design strategy for brains, and what is the origin of domain specificity in the first place? In this talk, I will present recent work testing whether the segregation of face and object perception in human brains emerges naturally from an optimization for both tasks. We trained artificial neural networks on face and object recognition, and found that networks were able to perform both tasks well by spontaneously segregating them into distinct pathways. Critically, networks neither had prior knowledge nor any inductive bias about the tasks. Furthermore, networks optimized on tasks which apparently do not develop specialization in the human brain, such as food or cars, and object categorization showed less task segregation. These results suggest that functional segregation can spontaneously emerge without a task-specific bias, and that the domain-specific organization of the cortex may reflect a computational optimization for the real-world tasks humans solve.