1,721,028 research outputs found
Perceptual multistability as Markov Chain Monte Carlo inference
Full text linkWhile many perceptual and cognitive phenomena are well described in terms of Bayesian inference, the necessary computations are intractable at the scale of real-world tasks, and it remains unclear how the human mind approximates Bayesian computations algorithmically. We explore the proposal that for some tasks, humans use a form of Markov Chain Monte Carlo to approximate the posterior distribution over hidden variables. As a case study, we show how several phenomena of perceptual multistability can be explained as MCMC inference in simple graphical models for low-level vision
Recognition alters the spatial pattern of fMRI activation in early retinotopic cortex
Full text linkEarly retinotopic cortex has traditionally been viewed as containing a veridical representation of the low-level properties of the image, not imbued by high-level interpretation and meaning. Yet several recent results indicate that neural representations in early retinotopic cortex reflect not just the sensory properties of the image, but also the perceived size and brightness of image regions. Here we used functional magnetic resonance imaging pattern analyses to ask whether the representation of an object in early retinotopic cortex changes when the object is recognized compared with when the same stimulus is presented but not recognized. Our data confirmed this hypothesis: the pattern of response in early retinotopic visual cortex to a two-tone “Mooney” image of an object was more similar to the response to the full grayscale photo version of the same image when observers knew what the two-tone image represented than when they did not. Further, in a second experiment, high-level interpretations actually overrode bottom-up stimulus information, such that the pattern of response in early retinotopic cortex to an identified two-tone image was more similar to the response to the photographic version of that stimulus than it was to the response to the identical two-tone image when it was not identified. Our findings are consistent with prior results indicating that perceived size and brightness affect representations in early retinotopic visual cortex and, further, show that even higher-level information—knowledge of object identity—also affects the representation of an object in early retinotopic cortex
Explaining human multiple object tracking as resource-constrained approximate inference in a dynamic probabilistic model
Full text linkMultiple object tracking is a task commonly used to investigate the architecture of human visual attention. Human participants show a distinctive pattern of successes and failures in tracking experiments that is often attributed to limits on an object system, a tracking module, or other specialized cognitive structures. Here we use a computational analysis of the task of object tracking to ask which human failures arise from cognitive limitations and which are consequences of inevitable perceptual uncertainty in the tracking task. We find that many human performance phenomena, measured through novel behavioral experiments, are naturally produced by the operation of our ideal observer model (a Rao-Blackwelized particle filter). The tradeoff between the speed and number of objects being tracked, however, can only arise from the allocation of a flexible cognitive resource, which can be formalized as either memory or attention
Discovering Structure in the Space of fMRI Selectivity Profiles
No full textWe present a method for discovering patterns of selectivity in fMRI data for experiments with multiple stimuli/tasks. We introduce a representation of the data as profiles of selectivity using linear regression estimates, and employ mixture model density estimation to identify functional systems with distinct types of selectivity. The method characterizes these systems by their selectivity patterns and spatial maps, both estimated simultaneously via the EM algorithm. We demonstrate a corresponding method for group analysis that avoids the need for spatial correspondence among subjects. Consistency of the selectivity profiles across subjects provides a way to assess the validity of the discovered systems. We validate this model in the context of category selectivity in visual cortex, demonstrating good agreement with the findings based on prior hypothesis-driven methods.McGovern Institute Neurotechnology (MINT) ProgramNational Institutes of Health (U.S.) (Grant NIBIB NAMIC U54-EB005149)National Institutes of Health (U.S.) (Grant NCRR NAC P41-RR13218)National Eye Institute (grant 13455)National Science Foundation (U.S.) (grant CAREER 0642971)Collaborative Research in Computational Neuroscience (IIS/CRCNS 0904625)Deshpande Center for Technological Innovation (MIT HST Catalyst grant)American Society for Engineering Education. National Defense Science and Engineering Graduate Fellowshi
Recommended from our members
Principles underlying human physical prediction
Full text linkOur days are filled with instances of reasoning about the physics of the world, from simple tasks such as stacking dishes in a way that keeps them stable, to life-and-death decisions such as not crossing the street because we presume an oncoming car would hit us if we did. Yet the process we use to make inferences about physical events is not well understood. Here I argue that these interactions are based on a rich, approximately accurate simulation of physical events, but we must account for uncertainty about the current properties of objects in the world. In this thesis I investigate the structure of this simulation process and how it relates to other facets of cognition, including (1) demonstrating that the principles underlying interactions with the world are based on accurate physics, even if our explanations of those same principles are idiosyncratic and erroneous, (2) mapping out the types of uncertainty that this process accounts for, and demonstrating that the simulations themselves are therefore stochastic, and (3) explaining how physical predictions are updated over time due to changing evidence from evolving simulations. This provides a framework for understanding how people form and update representations of both the current and future state of the world based on rich, structured, probabilistic reasoning
Recommended from our members
Biases in Social Perception Arise from Rational Inference
Full text linkThe social information available to us at any given moment is, at best, ambiguous. Yet, remarkably, we are able to efficiently resolve this ambiguity and successfully navigate the social world. In this dissertation I use a rational inference framework to understand how we form rich, and largely accurate, social perceptions given this uncertain and underconstrained information. Our perceptions, of course, do not always perfectly align with reality, but – contrary to the classic perspective in social psychology – this is not evidence that we are irrational. In this dissertation I show how social biases can arise not as a failure of rationality, but as a consequence of making optimal use of statistical structure in the world. In Chapter 1, I demonstrate that our visual system’s strategy to circumvent resource limitations by capitalizing on redundancy in visual scenes can result in a bias to perceive faces in a crowd as more attractive. In Chapters 2 and 3, I show that two of the most well-known social biases – The Fundamental Attribution Error and Role-conferred Advantage – are not actually evidence of irrational reasoning. Although in these paradigms observers seem “bias” to systematically make attributions that are in a direction consistent with observed behavior, these judgments fall naturally out of optimal probabilistic inference
Recommended from our members
Representations of Hierarchical Structure in Visual Memory
Full text linkVisual working memory possesses a limited capacity for information but people can use objects’ statistical structure to help remember their features. If you know that your papers are scattered around your desk, for example, this constrains their possible locations (e.g. it is unlikely they are in the bathroom) and can help you remember specifically where each paper is on your desk. However, it is often uncertain what information visual working memory should summarize to aid recall later on. Is it sufficient to remember that the papers were near the desk? Or will you need to know where they were relative to each other? My dissertation investigates what statistical structure visual working memory seeks to encode by (Chapter 1) revealing what visuospatial groupings people expect and tend to use, (Chapter 2) examining how people use those expectations to form structured memories of objects’ groupings and (Chapter 3) evaluating the cost of using this grouping structure—what information is lost by encoding objects as components of groups. Overall, my dissertation reveals reveals that while exploiting the statistics of scenes introduces structured biases into memories, doing so enables visual memory to build accurate, multi-level representations of scenes
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Simulating bistable perception with periodically interrupted ambiguous stimulus using percept choice bifurcation with stochastic self-oscillator dynamics
Full text linkFormal modeling of cognitive bistability (e.g.[1][2]) is an interesting problem because a constant stimulus (e.g. the Necker cube) excites quasi periodic alternations between only two well defined perception states. Periodic stimulus–off switching (toff < 1 s, ton = 300 ms) was introduced by Orbach et al. [3] as experimental paradigm to get more insight into the underlying perceptual dynamics. Their Necker cube experiments showed a maximum of the percept reversal rate R at Rmax 36 min-1 and toff 200 ms which was confirmed by recent experiments [4]. Noest et al. [5] demonstrated with a low level neural activation model [6] that a bifurcation of the percept choice dynamics during the ambiguous-stimulus on-off switching dominates the statistics of the reversal time series. Our simulations based on a macroscopic (behavioral) dynamics model [7][8] (similar to [1]) support this finding and show that the measured R vs. toff-time characteristics can be fitted with only few model parameters: attention (= adaptive feedback gain) fatigue time constant = 1 – 2 s, feedback delay T = 40 ms, gain-noise power J. Synchronisation of attention fatigue induced self-oscillations (yielding inter-stimulus transition time TTr 4 – 5 T) with stimulus-onset induced percept bifurcation appears to determine the reversal rates and the toff-value at Rmax. A linear approximation allows for an estimate of the cognitive damping time constant (v ≈ 1 s) which by use of the Fluctuation-Dissipation theorem via Jdefines an index of cognitive inertia (suggested in [8]) as crucial parameter of the simulated dynamics.
[1] Ditzinger, T., Haken, H. (1989). Oscillations in the Perception of Ambiguous Patterns. Biol. Cybern. ( 61) 279-287
[2] Huys, R,, Jirsa, V.K. (2010): Nonlinear Dynamics in Human Behavior. Springer Verlag, Berlin, Heidelber.
[3] Orbach. J., Zucker, E., Olson, R. (1966). Reversibility of the Necker Cube: VII. Reversal rate as a function of figure-on and figure-off durations. Percept. Motor Skills (22), 615-618
[4] Kornmeier, J., Ehm, W. Bigalke, H., Bach, M. (2007): Discontinuous presentation of ambiguous figures: How interstimulus-interval durations affect reversal dynamics and ERP’s. Psychophysiology, 44, 552-560
[5] Noest, A.J., van Ee, R., Nijs, M.M., van Wezel, R.J.A. (2007) Percept-choice sequences driven by interrupted ambiguous stimuli: A low-level neural model. J of Vision 7, 1-14
[6] Amari, S. (1977): Dynamics of pattern formation in lateral-inhibition type neural fields. Biological Cybernetics vol. 27, 77-87
[7] Fürstenau, Norbert (2010). A nonlinear dynamics model for simulating long range correlations of cognitive multistability. Biol. Cybern., vol. 103. (3) 175-198
[8] Gao, J.B., Merk, I., Tung W. W., Billok V., White, K.D., Harris J G, Roychowdhury V P. (2006). Inertia and memory in visual perception. Cogn. Processing vol. 7 105-11
- …
