Ma Lab @ CNS NYU

Publication filters

Filter by type:

Filter by year:

Filter by topic:

In press

Dongjae Kim, Heiko H Schuett, Ma WJ (in press)
Reward prediction error neurons implement an efficient code for reward
Nature Neuroscience
bioRxiv

Paul Bays, Sebastian Schneegans, Ma WJ, Timothy Brady
Representation and computation in working memory
Nature Human Behaviour
PsyArxiv

Allen KR, Brändle F, Botvinick M, Fan J, Gershman SJ, Gopnik A, Griffiths TL, Hartshorne J, Hauser TU, Ho MK, de Leeuw J, Ma WJ, Murayama K, Nelson JD, van Opheusden B, Pouncy HT, Rafner J, Rahwan I, Rutledge R, Sherson J, Simsek O, Spiers H, Summerfield C, Thalmann M, Velez N, Watrous A, Tenenbaum J, Schulz E (in press)
Using Games to Understand the Mind
Nature Human Behaviour
PsyArXiv

Preprints

Ionatan Kuperwajs, Mark Ho, Wei Ji Ma
Heuristics for meta-planning from a normative model of information search
PsyArXiv | Twitter thread by Ionatan Kuperwajs
We propose to abstract planning as an information search problem to produce heuristics for meta-planning, or to determine which action to plan for. Specifically, we model a metacognitive process where evaluating candidate actions is viewed as gaining noisy measurements of the value of each action. This statistical estimate is then combined with prior experience in a Bayesian manner to decide whether and in which direction to continue sampling. This Bayesian meta-planner makes intuitive predictions across a range of parameters and acts as a more valuable, informed method for guiding search when compared to best-first and breadth-first search. Additionally, the meta-planner qualitatively accounts for response time trends in a complex planning task.

Stephan Pohl, Edgar Y. Walker, David L. Barack, Jennifer Lee, Rachel N. Denison, Ned Block, Florent Meyniel, Wei Ji Ma
Desiderata of evidence for representation in neuroscience
arXiv | Twitter thread by David Barack
This paper develops a systematic framework for the evidence neuroscientists use to establish whether a neural response represents a feature. Researchers try to establish that the neural response is (1) sensitive and (2) specific to the feature, (3) invariant to other features, and (4) functional, which means that it is used downstream in the brain. We formalize these desiderata in information-theoretic terms. This formalism allows us to precisely state the desiderata while unifying the different analysis methods used in neuroscience under one framework. We discuss how common methods such as correlational analyses, decoding and encoding models, representational similarity analysis, and tests of statistical dependence are used to evaluate the desiderata. In doing so, we provide a common terminology to researchers that helps to clarify disagreements, to compare and integrate results across studies and research groups, and to identify when evidence might be missing and when evidence for some representational conclusion is strong.

Aditi Singh, Andra Mihali, Wen-Chuang Chou, Wei Ji Ma
A Computational Approach to Search in Visual Working Memory
PsyArXiv
Search in visual working memory (VWM) is an important process in everyday life, but it has so far not been investigated sufficiently using psychophysical and computational approaches. Here, we examine whether computational process models from the study of VWM and of visual search also apply to search in VWM. Participants performed a two-alternative VWM-based color localization task in which set size and color similarity were varied. We tested models with a resource-limited encoding stage and a Bayesian decision stage. We found that a variable-precision model fit the data well and that adding an item limit did not improve the fit. Next, we took a resource-rational approach, in which mean precision at each set size is set by optimizing a combination between behavioral performance and encoding cost. We found that a resource-rational model with an ad-hoc cost function captures behavior well, but not better than the best non-resource-rational model. Our results, albeit unsurprising, contribute to the evidence that search in VWM is very similar to search in perception.

Ma I, Phaneuf C, van Opheusden B, Ma WJ, Hartley C
The component processes of complex planning follow distinct developmental trajectories
PsyArXiv
We use a complex planning task and a computational modeling framework to delineate the developmental trajectories of three component processes of complex planning. Participants (ages 8-25 years) played the Four-in-a-row task, presented as a 4x9 board game, against computer opponents. We found asynchronous age-related change in the component processes of complex planning, with the heuristic evaluation of relevant features rapidly improving from childhood into adolescence, and planning depth showing more protracted developmental improvements. Fewer attentional oversights predicted better planning but did not demonstrate age-related variability. These results provide evidence for the continued development of model-based decision strategies into adulthood, and contribute to a more nuanced understanding of the cognitive mechanisms underpinning planning ability within complex environments at different developmental stages.

Long Ni and Ma WJ
Disentangling sources of interference in the N-back task
PsyArxiv
The N-back task is super common, but exactly at what level the distractor interferes with the target is not known. Interference could happen at a sensory level (signals getting mixed), at the decision stage, or the distractor could get substituted for the target. We take inspiration from the visual crowding literature, which has done a great job in distinguishing sources of interference. We use a 2-back task with orientations instead of letters or numbers. We formulate a model for each level of interference, then compare the models. We cannot distinguish all models but hopefully it is a start.

Andra Mihali, Ma WJ
The psychophysics of visual search with heterogeneous distractors
bioRxiv | Data and Code (Github) | Video of virtual journal club on Nov 18, 2020 | Slides
One research tradition in visual search has used simple, parametric stimuli and a signal detection theory or Bayesian modeling framework. This tradition has mostly focused on homogeneous distractors, which are not very realistic. In a different tradition, Duncan and Humphreys (1989) conducted a landmark study on visual search with heterogeneous distractors. However, they used complex stimuli, making modeling and dissociation of component processes difficult. Here, we attempt to unify these research traditions by systematically examining visual search with heterogeneous distractors using simple, parametric stimuli and Bayesian modeling.

Ma WJ, Benjamin Peters
A neural network walks into a lab: towards using deep nets as models for human behavior
arxiv | Video of Virtual Journal Club on June 1, 2020
What might sound like the beginning of a joke has become an attractive prospect for many cognitive scientists: the use of deep neural network models (DNNs) as models of human behavior in perceptual and cognitive tasks. Although DNNs have taken over machine learning, attempts to use them as models of human behavior are still in the early stages. Can they become a versatile model class in the cognitive scientist's toolbox?

Ma WJ
Problematic usage of the Zhang and Luck mixture model
bioRxiv

De Silva N, Ma WJ
Optimal allocation of attentional resource to multiple items with unequal relevance
arxiv

2023

Zeming Fang, Ilona Bloem, Catherine Olsson, Wei Ji Ma, and Jonathan Winawer (in press)
Normalization by orientation-tuned surround in human V1-V3
PLoS Computational Biology 19 (12): e101170. DOI: 10.1371/journal.pcbi.1011704
PDF

planning
value-based decision-making
neural networks/AI
four-in-a-row
big data

Kuperwajs I, Schuett HH, Ma WJ (2023)
Using deep neural networks as a guide for modeling human planning
Scientific Reports 13, 20269. DOI: 10.1038/s41598-023-46850-1
PDF

Arnold SH, Bailey AH, Ma WJ, Shahade J, and Cimpian A (2023)
Checking gender bias: Parents and mentors perceive less chess potential in girls.
Journal of Experimental Psychology: General Advance online publication. DOI: 10.1037/xge0001466
PDF
My first foray into gender biases in chess, a community I know well. This foray was only possible thanks to Sophie Arnold and Andrei Cimpian, who led the study, and our wonderful collaborators April Bailey and Jennifer Shahade.

Edgar Walker, Stephan Pohl, Rachel Denison, David Barack, Jennifer Laura Lee, Ned Block, Ma WJ, Florent Meyniel
Studying the neural representations of uncertainty
Nature Neuroscience 26, 1857–1867. DOI: 10.1038/s41593-023-01444-ya>
PDF

van Opheusden B, Galbiati G, Kuperwajs I, Bnaya Z, Li Y, Ma WJ (2023)
Expertise increases planning depth in human gameplay
Nature DOI: doi.org/10.1038/s41586-023-06124-2
PDF | Data and code | Interactive
Research highlight in Nature Computational Science
Behind the paper: looking back on thinking ahead
Commentary: Gobet F and Waters AJ, Searching for answers: expert pattern recognition and planning, Trends in Cognitive Sciences
When I had just started my NYU job in 2013, I pitched a project to incoming Ph.D. students. It was a wild new direction for me: using board games to understand how people think ahead. Bas van Opheusden took on the challenge. Have you ever wondered what makes chess grandmasters so good at the game? The science of expertise in chess is almost 75 years old, starting with Adriaan de Groot's beautiful PhD thesis. De Groot, and later Simon & Chase, linked chess expertise to pattern recognition. But what about thinking ahead? Could it be that experts are better because they plan further into the future? It's complicated. Some studies support deeper planning by chess experts, some don't. We believe the problem is the complexity of chess. Building computational models for human chess play is really hard, so studies mostly use verbal reports. We tried to find a game that's simpler than chess, but complex enough that planning 5-6 steps into the future is sometimes necessary. Our answer: four-in-a-row, a variant of tic-tac-toe and Go-Moku (五子棋). Four-in-a-row turned out to be the perfect paradigm. The rules are easy but it's remarkably fun, and calculating multiple moves ahead pays off. Play against the computer here. We collected data from people playing 4-in-a-row against each other or against custom-designed computer opponents. We had to develop new statistical methods to fit our computational models, and run the computer cluster for an insane amount of time. We found a robust increase in planning depth with expertise. Also in the paper: eye tracking, a time pressure manipulation, and replication in large-scale naturalistic data. Thanks to Bas, we have also started lots of collaborations based on this paradigm.

Lee JL, Denison R, Ma WJ (2023)
Re-thinking the fixed-criterion model of perceptual decision-making
Neuroscience of Consciousness Vol. 2023, Issue. 1, DOI: 10.1093/nc/niad010
PDF | Code
Perceptual decision-making is often conceptualized as the process of comparing an internal decision variable to a criterion. Consciousness researchers investigating a phenomenon called "subjective inflation" have proposed that the criterion governing subjective visibility is fixed. That is, it does not adjust to changes in sensory uncertainty. Decision-making researchers, on the other hand, have concluded that the criterion does adjust to account for sensory uncertainty, including under inattention. We try to reconcile these perspectives.

Huang J, Velarde I, Ma WJ, Baldassano C (2023)
Schema-based predictive eye movements support sequential memory encoding
eLife 12:e82599. DOI: 10.7554/eLife.82599
PDF | Data and Code (Open Science Framework)

van den Berg R, Zou Q, Li Y, Ma WJ (2023)
No effect of monetary reward in a visual working memory task
PLoS ONE 18(1), e0280257. DOI: doi.org/10.1371/journal.pone.0280257
PDF | Data and Code (Open Science Framework)
Previous work has shown that humans distribute their visual working memory (VWM) resources flexibly across items: the higher the importance of an item, the better it is remembered. A related, but much less studied question is whether people also have control over the total amount of VWM resource allocated to a task. Here, we address this question by testing whether increasing monetary incentives results in better overall VWM performance.

Schütt H, Yoo A, Calder-Travis J, Ma WJ (2023)
Point estimate observers: A new class of models for perceptual decision making
Psychological Review, DOI: doi.org/10.1037/rev0000402
PDF | Data and Code (Github)
Bayesian optimal inference is often heralded as a principled, general framework for human perception. However, optimal inference requires integration over all possible world states, which quickly becomes intractable in complex real-world settings. Additionally, deviations from optimal inference have been observed in human decisions. As a candidate alternative framework to address these issues, we propose point estimate observers, which evaluate only a single best estimate of the world state per response category. We compare the predicted behavior of these model observers to human decisions in five perceptual categorisation tasks. Compared to the Bayesian observer, the point estimate observer loses decisively in one task, ties in two and wins in two tasks. Thus, the point estimate observer is competitive with the Bayesian observer and should be considered in future model development and experimental studies.

2022

Chancel M, Ehrsson H, Ma WJ (2022)
Uncertainty-based inference of a common cause for body ownership
eLife 11: e77221. DOI: 10.7554/eLife.77221
PDF | Data
Body ownership in the rubber hand illusion is quantitatively predicted by uncertainty-based inference of a common cause: our causal inference model fitted body ownership data well at the individual-subject level (not just correlations with questionnaire ratings). We did not use a proxy task such as proprioceptive drift or reaching error, but direct ``detection-like" judgment of illusion elicitation. By manipulating visual noise, we showed that participants are taking trial-to-trial sensory uncertainty into account to make their judgment about body ownership. We compared different versions of the model for body ownership and synchrony judgment tasks and found interesting differences and similarities across the two types of tasks.

planning
value-based decision-making
neural networks/AI
four-in-a-row

Zheng Z, Lin XL, Topping J, Ma WJ (2022)
Comparing Machine and Human Learning in a Planning Task of Intermediate Complexity
Proceedings of the 44th Annual Meeting of the Cognitive Science Society, Pages 3601-3607.
PDF

planning
value-based decision-making
learning
four-in-a-row
big data

Kuperwajs I, Ma WJ (2022)
A joint analysis of dropout and learning functions in human decision-making with massive online data.
Proceedings of the 44th Annual Meeting of the Cognitive Science Society, Pages 1197-1203.
PDF

planning
value-based decision-making
neural networks/AI
four-in-a-row
big data

Kuperwajs I, Schuett HH, Ma WJ (2022)
Improving a model of human planning via large-scale data and deep neural networks
Proceedings of the 44th Annual Meeting of the Cognitive Science Society, Pages 1190-1196.
PDF

planning
value-based decision-making
learning
four-in-a-row

Ma I, Phaneuf CV, Van Opheusden B, Ma WJ, Hartley C (2022)
Distinct Developmental Trajectories In The Cognitive Components Of Complex Planning
Proceedings of the 44th Annual Meeting of the Cognitive Science Society, Pages 1131-1137.
PDF

2021

Lee JL, Ma WJ (2021)
Point-estimating observer models for latent cause detection
PLoS Computational Biology 17(10): e1009159. DOI: 10.1371/journal.pcbi.1009159
PDF | Data and Code (Github)
The spatial distribution of visual items allows us to infer the presence of latent causes in the world. For instance, a spatial cluster of ants allows us to infer the presence of a common food source. However, optimal inference requires the integration of a computationally intractable number of world states in real world situations. For example, optimal inference about whether a common cause exists based on N spatially distributed visual items requires marginalizing over both the location of the latent cause and 2N possible affiliation patterns (where each item may be affiliated or non-affiliated with the latent cause). How might the brain approximate this inference?

planning
inference/uncertainty
optimality/rationality
value-based decision-making
four-in-a-row
big data

Kuperwajs I, Ma WJ (2021)
Planning to plan: a Bayesian model for optimizing the depth of decision tree search
Proceedings of the 43rd Annual Meeting of the Cognitive Science Society, Pages 91-97.
PDF

planning

Ma I, Ma WJ, Gureckis TM (2021)
Information sampling for contingency planning
Proceedings of the 43rd Annual Meeting of the Cognitive Science Society, Pages 1000-1006.
PDF

Li HH, Sprague TC, Yoo AH, Ma WJ, Curtis CE (2021)
Joint representation of working memory and uncertainty in human cortex
Neuron Vol. 109, Issue 22, Pages 3699-3712.e6. DOI: 10.1016/j.neuron.2021.08.022
PDF Main | PDF Supp | Data and Code (Open Science Framework) | Press release
Neural noise shows the uncertainty of our memories (Quanta Magazine, Jan 18, 2022)
Neural representations of visual working memory (VWM) are noisy, and thus, decisions based on VWM are inevitably subject to uncertainty. However, the mechanisms by which the brain simultaneously represents the content and uncertainty of memory remain largely unknown. Here, inspired by the theory of probabilistic population codes, we test the hypothesis that the human brain represents an item maintained in VWM as a probability distribution over stimulus feature space, thereby capturing both its content and uncertainty. We used a neural generative model to decode probability distributions over memorized locations from fMRI activation patterns. We found that the mean of the probability distribution decoded from retinotopic cortical areas predicted memory reports on a trial-by-trial basis. Moreover, in several of the same mid-dorsal stream areas the spread of the distribution predicted subjective trial-by-trial uncertainty judgments. These results provide evidence that VWM content and uncertainty are jointly represented by probabilistic neural codes.

Yoo A, Acerbi L, Ma WJ (2021)
Uncertainty is maintained and used in working memory
Journal of Vision 21(8): 13. DOI: 10.1167/jov.21.8.13
PDF | Data and Code (Github)
What are the contents of working memory? In both behavioral and neural computational models, the working memory representation of a stimulus is typically described by a single number, namely a point estimate of that stimulus. Here, we asked if people also maintain the uncertainty associated with a memory, and if people use this uncertainty in subsequent decisions.

Li Z, Ma WJ (2021)
An uncertainty-based model of the effects of fixation on choice
PLoS Computational Biology 17(10): e1009159. DOI: 10.1371/journal.pcbi.1009190
PDF Main | PDF Supp | Data and Code (Open Science Framework)
When people view a consumable item for a longer amount of time, they choose it more frequently; this also seems to be the direction of causality. The leading model of this effect is a drift-diffusion model with a fixation-based attentional bias. While this model accounts for the data, it is not normative, in the sense that it does not provide a rationale for this behavioral tendency. Here, we propose a partially normative account for the same data.

2020

van Opheusden, B, Acerbi L, Ma WJ (2020)
Unbiased and efficient log-likelihood estimation with inverse binomial sampling
PLoS Computational Biology 16 (12): e1008483. DOI: https://doi:10.1371/journal.pcbi.1008483
PDF | Code (GitHub)
Many models (not limited to psychology and neuroscience) don't have a log-likelihood in closed form, but we can easily sample observations from the model. When you fit such a simulated (sampled) model using ML estimation, do you ever encounter log(0), and then replace that by log(some small number) so that your code runs? I used to do that all the time. Bad idea!! A few years ago, Bas van Opheusden explained to me that this inevitably leads to biases, because log(true probability) can go all the way to negative infinity, whereas log(small number) cannot. And this is just one symptom. Even if you don't run into log(0) issues, the log(proportion of samples) estimator ("fixed sampling") is biased no matter what. To fix the issue, Bas reinvented what turned out to be an old method by the statistician De Groot (1959). Basic idea: you sample UNTIL you obtain the observed outcome, and use HOW LONG IT TOOK YOU, plugging that into an equation. This gives a uniformly unbiased estimate of the log likelihood. Bas worked with Luigi on computational improvements to reduce variance, interfacing with parameter optimization, and applications. None of this applies if you have an analytical expression for the log likelihood, but sadly, most interesting computational models are not of that type.

Calder-Travis J, Ma WJ (2020)
Explaining the effects of distractor statistics in visual search
Journal of Vision Vol. 20, article number 11. DOI: https://doi.org/10.1167/jov.20.13.11
PDF | Interactive | Data and Code (Open Science Framework) | Video of virtual journal club on Nov 18, 2020 | Slides
Visual search, the task of detecting or locating target items amongst distractor items in a visual scene, is an important function for animals and humans. Different theoretical accounts make differing predictions for the effects of distractor statistics. Here we use a task in which we parametrically vary distractor items, allowing for a simultaneously fine-grained and comprehensive study of distractor statistics. We found effects of target-distractor similarity, distractor variability, and an interaction between the two, although the effect of the interaction on performance differed from the one expected. To explain these findings, we constructed computational process models that make trial-by-trial predictions for behaviour based on the full set of stimuli in a trial. These models, including a Bayesian observer model, provided excellent accounts of both the qualitative and quantitative effects of distractor statistics, as well as of the effect of changing the statistics of the environment (in the form of distractors being drawn from a different distribution). We conclude with a broader discussion of the role of computational process models in the understanding of visual search.

Zhou Y, Acerbi L, Ma WJ (2020)
The role of sensory uncertainty in simple contour integration
PLoS Computational Biology 16 (11): e1006308. DOI: doi.org/10.1371/journal.pcbi.1006308
PDF Main | PDF Supp | Data and Code (Github)
A Bayesian approach to collinearity detection, which you could call a simple form of perceptual organization. We focus on the effects of uncertainty as controlled by retinal eccentricity. Yanli started this work as an undergraduate research assistant in 2015, continued while a Masters student in Data Science, and finished while (now) a PhD student in Data Science. Luigi started it as a first-year postdoc in the lab, did another postdoc since, and is now faculty at the University of Helsinki. This is how long papers can take (the pandemic may have made it longer than normal). But I enjoyed the process, especially because Yanli kept energetically coming back to the project even though revisions were spaced far apart and she had moved on. I would say that the work falls into the category of "Bayesian models of slightly more interesting perceptual computations". The Bayesian poster children of combining a likelihood with a prior, and cue combination are nowadays interesting only for their new applications, not for their complications. Closely related work from the lab: Shaiyan Keshvari's and Aspen Yoo's work on change detection. The parallel is that change detection involves changes across time and collinearity detection involves changes across space (behind an occluder). A difference is that uncertainty was controlled through shape there and through eccentricity here (shape wouldn't make much sense for collinearity detection). Also related: Ronald van den Berg's work on sameness judgment, but that focused on set size and not uncertainty.

Li HH, Ma WJ (2020)
Confidence reports in decision-making with multiple alternatives violate the Bayesian confidence hypothesis
Nature Communications 11, article number 2004. DOI: 10.1038/s41467-020-15581-6
PDF Main | PDF Supp | Data and Code (Github) | Interactive | Video of virtual journal club on May 4, 2020 | Slides

Honig M, Ma WJ, Fougnie D (2020)
Humans incorporate trial-to-trial working memory uncertainty into rewarded decisions
Proceedings of the National Academies of Science 117 (15), 8391-8397. DOI: 10.1073/pnas.1918143117
PDF Main | PDF Supp | Data and Code (Open Science Framework) | Video of virtual journal club on April 3, 2020

Walker EY, Cotton RJ, Ma WJ, Tolias AS (2020)
A neural code for probabilistic computation in visual cortex
Nature Neuroscience 23, 122-129. DOI: 10.1038/s41593-019-0554-5
PDF Main | PDF Supp |Github | Video of virtual journal club on July 6, 2020 | Slides

2019

Ma WJ (2019),
Bayesian decision models: a primer
Neuron 104 (1): 164-175. DOI: 10.1016/j.neuron.2019.09.037
PDF Main | PDF Supp | PDF Main+Supp

van Opheusden B, Ma WJ(2019),
Tasks for aligning human and machine planning
Current Opinion in Behavioral Sciences 29: 127-133. DOI: 10.1016/j.cobeha.2019.07.002
PDF
Perspective paper stemming from our work on human tree search in four-in-a-row. Any good ideas in this paper came from Bas.

Norton EH, Acerbi L, Ma WJ, Landy MS (2019)
Human online adaptation to changes in prior probability
PLoS Computational Behavior 15 (7): e1006681. DOI: 10.1371/journal.pcbi.1006681
PDF Main | PDF Supp | Code (Github)
I had a very minor role in this paper, but I am very happy how it turned out, especially in terms of the wide range of models considered and the model fitting techniques used. Also an interesting tidbit: one of the tasks (binary categorization under random changes in prior probability) is essentially the same as the task used by the International Brain Lab.

Song M, Bnaya Z, Ma WJ (2019),
Sources of suboptimality in a minimalistic explore-exploit task
Nature Human Behaviour 3, 361-368. DOI: 10.1038/s41562-018-0526-x
PDF Main | PDF Supp | Code (Github)
The lab's latest foray into sequential decision-making! This paper started with my visit to Beijing in January 2015. I knew I would be in the city for just 5 days, yet went twice to the same breakfast place (in the neighborhood Hepingli). Why would I do that? I started wondering about the trade-offs between exploration and exploitation. Coming from perception, my first question was about (deviations from) optimality, but there did not seem to be a whole lot on that. Mingyu started on this project in April 2015, and Zahy joined in October 2015. We tried to do a clean experiment, without too many complicating factors. Shortly after, we heard about a 2011 CogSci paper by Ke Sang, Peter Todd, and Robert Goldstone, but fortunately there were enough interesting questions for two groups to have very similar approaches. Please also check out their latest preprint here. As to Mingyu and Zahy's paper, it took 3 years and 10 months from the start of the work to publication; an interesting fact is that Mingyu was a primary lab member only during the summer of 2015, which might have made the timeline a little longer than average. Mingyu is currently doing great as a PhD student in the lab of Yael Niv at Princeton.

A Emin Orhan, Ma WJ (2019),
A diverse range of factors affect the nature of neural representations underlying short-term memory
Nature Neuroscience 22, 275-283. DOI: 10.1038/s41593-018-0314-y
PDF Main | PDF Supp | Code (Github)
How a working memory is maintained in a neural population might depend on the nature of the task, in particular its temporal complexity. Emin tested 5 tasks, including delayed estimation, change detection, and comparison. This paper started out as a project about the maintenance of uncertainty in working memory, but Emin drew it more broadly and made it about distinguishing two fundamentally different neural mechanisms of maintenance. The paper is almost entirely his work, with me only giving high-level comments from the sidelines. (This should not be surprising to those following my work - I specialize much more in modeling behavior than in modeling neural mechanisms.) Emin is one of the most original minds I know in computational neuroscience. In case you would like to consider him for a faculty job, here is his Google Scholars profile, and here is his blog about computational neuroscience and machine learning.

Ma WJ (2019),
Identifying suboptimalities with factorial model comparison
Behavioral and Brain Sciences 41, e234. DOI: 10.1017/S0140525X18001541
PDF
Response to Rahnev and Denison (2019), Suboptimality in perceptual decision making, Behavioral and Brain Sciences 41, e223.

visual decision-making
encoding/ representation
decision rules
set size effects

Shen S, Ma WJ (2019),
Variable precision in visual perception
Psychological Review 126 (1), 89-132. DOI: 10.1037/rev0000128.
PDF | Data and Code (Github)
The paper from hell! This took Shan and me 7 years from conception to publication, and it was not even her thesis work!! 11 experiments, up to 308 models per experiment (Fig. B12). In the end, 43 pages and a giant negative result, but new methods for model comparison. We set out to detect one of my babies (variable precision) in perception (rather than in visual working memory). But we wanted to do thorough model comparison, in the style of Luigi Acerbi, Ronald van den Berg, Jan Drugowitsch, Valentin Wyart etc. But... the more factors we included (lapses, decision noise, oblique effect), the less evidence remained for "pure" variable precision (e.g. due to attentional fluctuations). But I'm actually quite happy, as it means that as long you include those factors in your model, you might not need variable precision. Open question: is the situation the same in visual working memory (as suggested by Mike Pratte and Frank Tong in 2017)? Take-home message: think twice before you start working with me!

2018

visual decision-making
inference/uncertainty
comparative cognition
learning

Ma WJ, Higham JP (2018),
The role of familiarity in signaler-receiver interactions
Journal of the Royal Society Interface 15: 20180568. DOI: 10.1098/rsif.2018.056
PDF | Supp |Code (Dryad) | Code (GitHub)
Do you ever wonder whether your professor can still do real work? This is my first paper in ecology / animal communication, a field far from most of my lab's work. In this project, primatologist James P. Higham acted as the PI and I as the PhD student. It was absolutely exhilarating! James taught me a lot about monkey mating behavior and provided the big picture of the project. I did the math and the coding, and of course that came with familiar feelings of dread about bugs being left in my code. The only reason I could even contribute was because at a computational level, inferring ovulation time from observations of facial color is not that different from less naturalistic forms of perception that are my lab's bread and butter. So yes, it is possible to have your own side projects as a PI, and those side projects often make life interesting, especially if they are forays into neighboring fields. And academia is perfect for continuing to learn actively throughout your career.

visual decision-making
decision rules
inference/uncertainty
confidence

Adler WT, Ma WJ (2018),
Comparing Bayesian and non-Bayesian accounts of human confidence reports
PLoS Computational Biology 14 (11): e1006572. DOI: 10.1371/journal.pcbi.1006572
PDF | Data and Code (Github)
The message of this paper is intentionally provocative. Although we love the Bayesian confidence hypothesis and it does a good approximate job of accounting for our human data, it fails in the details. We don't really believe that the better-fitting model can compete with the Bayesian framework in generality or principledness (although we do a large number of experimental and analysis controls), but we pose the question: for a given task, how much failure can you tolerate in a Bayesian model before you reject it?

Yoo AH, Klyszejko Z, Curtis CE, Ma WJ (2018),
Strategic allocation of working memory resource
Scientific Reports 8: 16162. DOI: 10.1038/s41598-018-34282-1.
PDF | PDF Supp | Data and Code (Github)
First joint paper with the lab of Clay Curtis. Main findings: (1) Knowing in advance the relevance of different objects helps to allocate the appropriate amounts of working memory resource. This is related to work by Stephen Emrich). (2) Some indication that this is done near-optimally. (3) Behavioral reports of confidence (post-decision wager) track memory error, not just across but also within relevance levels. This is related to work by Rosanne Rademaker and Frank Tong, and to our own work with Maija Honig and Daryl Fougnie. We are now using this paradigm for a neuroimaging study, with Tommy Sprague and Masih Rahmati.

visual decision-making
inference/uncertainty
confidence

Adler WT, Ma WJ (2018),
Limitations of proposed signatures of Bayesian confidence
Neural Computation 30 (12), 3327-3354. DOI: 10.1162/neco_a_01141.
PDF | Code (Github)
Everybody knows that confidence ratings are Bayesian. Except that nobody really knows, because very few papers compare against alternatives! Wouldn't it be nice to have qualitative signatures of Bayesian confidence in behavioral data? This was the bold idea of Balazs Hangya and Joshua Sanders in Adam Kepecs' lab (Neural Computation and Neuron, 2016). Will was intrigued and dug into the details. For two signatures, we identified conditions that seem rather constraining, and we try to dissect those here. Moreover, last year, Joaquin Navajas working with Bahador Bahrami proposed a separate "X-pattern" or "divergence" signature. It looks the same as one of Hangya et al., but it turns out not to be, and it doesn't seem either necessary or sufficient for confidence to be Bayesian :-( and I conclude here that, tragically but not unexpectedly, there might be no shortcut to quantitative model comparison (of course on top of clever experiments). The jury is still out on whether confidence ratings are Bayesian. In the meantime, we want to thank Hangya, Sanders, Kepecs, and Navajas for being very open to discussing the pros and cons of their measures. This was not a given, since they knew we were going to write this critical paper.

visual decision-making
inference/uncertainty
confidence
attention

Denison R, Adler WT, Carrasco M, Ma WJ (2018),
Humans incorporate attention-dependent uncertainty into perceptual decisions and confidence
Proceedings of the National Academy of Sciences 115 (43), 11090-11095. DOI: 10.1073/pnas.1717720115
PDF | PDF Supp | Data and Code (Github)
First joint paper with the lab of Marisa Carrasco. Will and Rachel pretty much ran this study without their PIs. They found that when variations in attention cause variations in uncertainty, the brain takes the latter into account in a subsequent categorical decision.

visual decision-making
attention
disorders

Mihali A, Young AG, Adler LA, Halassa MM, Ma WJ (2018),
A low-level perceptual correlate of behavioral and clinical deficits in ADHD
Computational Psychiatry 2: 141-163. DOI: 10.1162/cpsy_a_00018
PDF | PDF Supp | Data and Code (Github)
This is my first special populations paper (ADHD), and I might never have done one were it not for my coauthors. We introduce a task that is at its core perceptual categorization of a 1d stimulus, but with both a spatial switching and a feature switching component. In addition, we aimed to bring out executive control deficits by allowing participants to press one of 8 buttons. All 8 are used in the task but on a given trial, only 2 are relevant. ADHD participants had both higher perceptual variability and higher "task-irrelevant motor output". In addition, those were correlated with each other. Based on the perceptual variability parameter alone, we could already classify participants as ADHD vs control with 77% accuracy. ADHD is not my field but I found this remarkably high. The high cognitive load might be necessary to bring out the perceptual deficits.

working memory
attention
optimality/rationality
set size effects

Van den Berg R, Ma WJ (2018),
A resource-rational theory of set size effects in visual working memory
eLife 7: e34963. DOI: 10.7554/eLife.34963
PDF | Data and Code (Dryad) | Data and Code (GitHub)
This paper took no less than 8 years from conception to publication, but we both feel that the review process - including multiple rejections - made the paper stronger, and we are pretty proud of the result. Resource rationality has been described extensively by Tom Griffiths and others, but to our knowledge it is new in the realm of working memory. Our cost term besides behavioral errors is a neural cost inspired by Lennie 2003.

decision rules
inference/uncertainty
multisensory perception

Acerbi L*, Dokka K*, Angelaki DA, Ma WJ (2018),
Bayesian comparison of explicit and implicit causal inference strategies in multisensory heading perception
PLoS Computational Biology 14 (7): e1006110. DOI: 10.1371/journal.pcbi.1006110
PDF | PDF Supp | Data and Code (Github)
First joint paper with the lab of Dora Angelaki. We originally thought that the data would be able to strongly distinguish models of multisensory perception. However, behavioral models have many plausible tweaks (e.g. observer assumptions, heteroskedasticity, decision noise), and when we allowed for those, the models became less distinguishable. Thus, the story became more methodological: how to comprehensively compare models of multisensory perception. Then, two remarkable things happened: (1) @AcerbiLuigi and Kalpana stubbornly kept working on it, even though the message was now less sexy; (2) The PLoS Computational Biology reviewers and editor saw the merits of such a methodological contribution. Remarkable because in my experience, methodological rigor often loses out to sexy messaging. For example, authors present only one favored model, without any comparison, or they consider only a bare-bones version of each model. Comparing more models rarely makes a conclusion crisper, but almost always makes it more appropriately nuanced.

working memory

Oberauer K, Lewandowsky S, Awh E, Brown GDA, Conway A, Cowan N, Donkin C, Farrell S, Hitch GJ, Hurlstone M, Ma WJ, Morey CC, Nee DE, Schweppe J, Vergauwe E, Ward G (2018),
Benchmarks for models of short-term and working memory
Psychological Bulletin 144 (9), 885–958. DOI: 10.1037/bul0000153
PDF | Data and Code (GitHub) | Data and Code (Open Science Framework)

methods

Acerbi L, Ma WJ (2017),
Practical Bayesian optimization for model fitting with Bayesian Adaptive Direct Search
Advances in Neural Information Processing Systems 20, 1834-1844.
Publisher: Curran Associates, Inc. Editors: Guyon I, Luxburg UV, Bengio S, Wallach H, Fergus R, Vishwanathan S, Garnett R.
PDF | Code (MatLab) | Code (Python) | Documentation
Computational models in fields such as computational neuroscience are often evaluated via stochastic simulation or numerical approximation. Fitting these models implies a difficult optimization problem over complex, possibly noisy parameter landscapes. Bayesian optimization (BO) has been successfully applied to solving expensive black-box problems in engineering and machine learning. Here we explore whether BO can be applied as a general tool for model fitting. First, we present a novel hybrid BO algorithm, Bayesian adaptive direct search (BADS), that achieves competitive performance with an affordable computational overhead for the running time of typical models. We then perform an extensive benchmark of BADS vs. many common and state-of-the-art nonconvex, derivative- free optimizers, on a set of model-fitting problems with real data and models from six studies in behavioral, cognitive, and computational neuroscience. With default settings, BADS consistently finds comparable or better solutions than other methods, including ‘vanilla’ BO, showing great promise for advanced BO techniques, and BADS in particular, as a general model-fitting tool.

planning

Van Opheusden B, Galbiati G, Bnaya Z, Li Y, Ma WJ (2017),
A computational model for decision tree search
Proceedings of the 39th Annual Meeting of the Cognitive Science Society, 1254-1259.
PDF | Website to try out the experiments and explore the data

working memory
encoding/ representation
set size effects

Shin H, Zou Q, Ma WJ (2017),
The effects of delay duration on visual working memory for orientation
Journal of Vision 17 (14): 10. DOI: 10.1167/17.14.10
PDF | Data and Code (Github)

working memory
decision rules
inference/uncertainty
comparative cognition

Devkar D, Wright AA, Ma WJ (2017),
Monkeys and humans take local uncertainty into account when localizing a change
Journal of Vision 17 (11): 4. DOI: 10.1167/17.11.4
PDF

multisensory perception
inference/uncertainty

Dobs K, Ma WJ, Reddy L (2017),
Near-optimal integration of facial form and motion
Scientific Reports 7: 11002. DOI: 10.1038/s41598-017-10885-y
PDF | PDF Supp | Data and Code (GitHub)

academia

Ma WJ (2017),
The stories behind a CV
Science 357 (6354): 942. DOI: 10.1126/science.357.6354.942
PDF

working memory
encoding/ representation
decision rules

Shin H, Ma WJ (2017),
Visual short-term memory for oriented, colored objects
Journal of Vision 17 (9): 12. DOI: 10.1167/17.9.12
PDF

neural networks/AI
inference/uncertainty
learning

Orhan AE, Ma WJ (2017),
Efficient probabilistic inference in generic neural networks trained with non-probabilistic feedback
Nature Communications 8, 138. DOI: 10.1038/s41467-017-00181-8
PDF | PDF Supp | Code (GitHub)

visual decision-making
inference/uncertainty
illusions

Zhu JE, Ma WJ (2017),
Orientation-dependent biases in length judgments of isolated stimuli
Journal of Vision, 17 (2): 20. DOI: 10.1167/17.2.20
PDF | Data and Code (GitHub)

working memory
encoding/ representation
confidence

Van den Berg R, Yoo AH, Ma WJ (2017),
Fechner’s Law in metacognition: a quantitative model of visual working memory confidence
Psychological Review, 124 (2), 197-214. DOI: 10.1037/rev0000060
PDF | Data and Code (GitHub)

methods
inference/uncertainty

Mihali A, Van Opheusden B, Ma WJ (2017),
Bayesian microsaccade detection
Journal of Vision, 17 (1): 13. DOI: 10.1167/17.1.13
PDF | Data and Code (Github)

2016

planning

Van Opheusden B, Galbiati G, Bnaya Z, Ma WJ (2016),
Do people think like computers?
Computers and Games 2016, Leiden, The Netherlands, Series Volume 10068. Eds. Plaat A, Kosters W, Van den Herik J 212-224. DOI: 10.1007/978-3-319-50935-8
PDF | Website to try out the experiments and explore the data

multisensory perception
inference/uncertainty
illusions

Peters M, Ma WJ, Shams L (2016),
The size-weight illusion is not anti-Bayesian after all: A unifying Bayesian account
PeerJ 4: e2124. DOI: 10.7717/peerj.2124
PDF | Supp

visual search
decision rules
optimality/rationality
visual decision-making

Shen S, Ma WJ (2016),
A detailed comparison of optimality and simplicity in perceptual decision-making
Psychological Review, 123 (4), 452-480. DOI: 10.1037/rev0000028
PDF | Data and Code (GitHub)

working memory
encoding/ representation

Shin H, Ma WJ (2016),
Crowdsourced single-trial probes of visual working memory for irrelevant features
Journal of Vision 16 (5): 10. DOI: 10.1167/16.5.10
PDF | Data and Code (GitHub)

working memory
attention

Cardoso-Leite P, Kludt R, Vignola G, Ma WJ, Green CS, Bavelier D (2016),
Technology consumption and cognitive control: contrasting action video game experience with media multitasking
Attention, Perception, and Psychophysics 78 (1), 218-241. DOI: 10.3758/s13414-015-0988-0
PDF

visual search
visual decision-making
decision rules
optimality/rationality

Bhardwaj M, Van den Berg R, Ma WJ, Josic K (2016),
Do people take stimulus correlations into account in visual search?
PLoS ONE 11 (3): e0149402. DOI: 10.1371/journal.pone.0149402
PDF

2015

working memory
encoding/ representation
set size effects
comparative cognition

Devkar D, Wright AA, Ma WJ (2015),
The same type of visual working memory limitations in humans and monkeys
Journal of Vision 15 (16): 13. DOI: 10.1167/15.16.13
PDF

visual decision-making
inference/uncertainty
neural coding

Van Bergen RS, Ma WJ, Pratte MS, Jehee JFM (2015),
Sensory uncertainty decoded from visual cortex predicts behavior
Nature Neuroscience 18, 1728-1730. DOI: 10.1038/nn.4150
PDF | PDF Supp

visual search
decision rules
encoding/ representation
visual decision-making

Bhardwaj M, Carroll S, Ma WJ, Josic K (2015),
Visual decisions in the presence of measurement and stimulus correlations
Neural Computation 27, 2318-2353. DOI: 10.1162/NECO_a_00778
PDF

time
illusions

Cai MB, Eagleman DE, Ma WJ (2015),
Perceived duration is reduced by repetition, but not by high-level expectation
Journal of Vision 15 (13): 19. DOI: 10.1167/15.13.19
PDF

neural coding
optimality/rationality
encoding/ representation
set size effects

Orhan AE, Ma WJ (2015),
Neural population coding of multiple stimuli
Journal of Neuroscience 35 (9), 3825-41. DOI: 10.1523/JNEUROSCI.4097-14.2015
PDF

review/tutorial
visual search
visual decision-making
decision rules

Ma WJ, Shen S, Dziugaite G, Van den Berg R (2015),
Requiem for the max rule?
Vision Research 116, 179-193. DOI: 10.1016/j.visres.2014.12.019
PDF

motivation
neural coding

Marsden KE*, Ma WJ*, Deci EL, Ryan RM, Chiu PH (2015),
Diminished neural responses predict enhanced intrinsic motivation and sensitivity to external incentive
Cognitive, Affective, and Behavioral Neuroscience 15 (2), 276-286. DOI: 10.3758/s13415-014-0324-5
PDF | PDF Supp

2014

methods
visual decision-making
inference/uncertainty

Acerbi L, Ma, WJ, Vijayakumar S (2014),
A framework for testing identifiability of Bayesian models of perception
Advances in Neural Information Processing Systems 27 (NIPS ’14)
PDF

review/tutorial
neural coding
inference/uncertainty

Ma WJ, Jazayeri M. (2014)
Neural coding of uncertainty and probability
Annual Review of Neuroscience 37, 205-20. DOI: 10.1146/annurev-neuro-071013-014017
PDF

methods
working memory
encoding/ representation

Van den Berg R, Ma WJ (2014)
"Plateau"-related summary statistics are uninformative for comparing working memory models
Attention, Perception, and Psychophysics 76 (7), 2117-35. DOI: 10.3758/s13414-013-0618-7
PDF

review/tutorial
working memory
encoding/ representation
set size effects

Ma WJ, Husain M, Bays P (2014),
Changing concepts of working memory
Nature Neuroscience 17, 347-56. DOI: 10.1038/nn.3655
PDF

working memory
encoding/ representation
set size effects

Van den Berg R, Awh E, Ma WJ (2014),
Factorial comparison of working memory models
Psychological Review 121 (1), 124-49. DOI: 10.1037/a0035234
PDF | Data and Code (GitHub)

2013

visual decision-making
decision rules
inference/uncertainty

Qamar AT*, Cotton RJ*, George R*, Beck JM, Prezhdo E, Laudano A, Tolias A, Ma WJ (2013),
Trial-to-trial, uncertainty-based adjustment of decision boundaries in visual categorization
Proceedings of the National Academy of Sciences 110 (50), 20332-7. DOI: 10.1073/pnas.1219756110
PDF

multisensory perception
encoding/ representation
adaptation

Zaidel A, Ma WJ, Angelaki D (2013),
Supervised calibration relies on the multisensory percept
Neuron 80 (6), 1544-57. DOI: 10.1016/j.neuron.2013.09.026
PDF | PDF Supp

multisensory perception
inference/uncertainty
decision rules

Magnotti JF, Ma WJ, Beauchamp MS (2013),
Causal inference of asynchronous audiovisual speech
Frontiers in Psychology 4: 798. DOI: 10.3389/fpsyg.2013.00798
PDF

review/tutorial
neural coding
inference/uncertainty

Pouget A, Beck JM, Ma WJ, Latham PE (2013),
Probabilistic brains: knowns and unknowns
Nature Neuroscience 16 (9), 1170-8. DOI: 10.1038/nn.3495
PDF

visual search
visual decision-making
encoding/ representation
set size effects

Mazyar H*, Van den Berg R*, Seilheimer R*, Ma WJ (2013),
Independence is elusive: set size effects on encoding precision in visual search
Journal of Vision 13 (5), 8. DOI: 10.1167/13.5.8
PDF

working memory
encoding/ representation
set size effects

Keshvari S, Van den Berg R, Ma WJ (2013),
No evidence for an item limit in change detection
PLoS Computational Biology 9(2): e1002927. DOI: 10.1371/journal.pcbi.1002927
PDF | Data and Code (GitHub)

multisensory perception
neural networks/AI
inference/uncertainty

Ma WJ and Rahmati M (2013),
Towards a neural implementation of causal inference in cue combination
Multisensory Research 26, 159-76. DOI: 10.1163/22134808-00002407
PDF

2012

inference/uncertainty
review/tutorial

Ma WJ (2012),
Organizing probabilistic models of perception
Trends in Cognitive Sciences 16 (10), 511-8. DOI: 10.1016/j.tics.2012.08.010
PDF

encoding/ representation
inference/uncertainty
visual decision-making
working memory

Keshvari S, Van den Berg R, Ma WJ (2012),
Probabilistic computation in human perception under variability in encoding precision
PLoS ONE 7(6): e40216. DOI: 10.1371/journal.pone.0040216
PDF | Data and Code

visual search
visual decision-making
encoding/ representation
inference/uncertainty
optimality/rationality

Mazyar H*, Van den Berg R*, Ma WJ (2012),
Does precision decrease with set size?
Journal of Vision 12 (6), 10. DOI: 10.1167/12.6.10
PDF

encoding/ representation
neural coding

Berens P, Ecker AS, Cotton RJ, Ma WJ, Bethge M, Tolias AS (2012),
A fast and simple population code for orientation in primate V1
Journal of Neuroscience 32 (31), 10618-26. DOI: 10.1523/jneurosci.1335-12.2012
PDF

working memory
attention
inference/uncertainty

Van den Berg R*, Shin H*, Chou WC, George R, Ma WJ (2012),
Variability in encoding precision accounts for visual short-term memory limitations
Proceedings of the National Academy of Sciences 109 (22), 8780-5. DOI: 10.1073/pnas.1117465109
PDF | PDF Supp | Code Download | Code (GitHub)
Press release "Short-term memory not all-or-nothing" by BCM
Press release "Kortetermijngeheugen is flexibeler dan gedacht" by the Netherlands Organisation for Scientific Research (NWO)

inference/uncertainty
review/tutorial

Beck JM*, Ma WJ*, Pitkow X, Latham PE, Pouget A (2012),
Not noisy, just wrong: the role of suboptimal inference in behavioral variability
Neuron 74 (1), 30-39. DOI: 10.1016/j.neuron.2012.03.016
PDF

working memory
time

Wright AA, Katz JS, Ma WJ (2012),
How to be proactive about interference: lessons from animal memory
Psychological Science 23 (5), 453-8. DOI: 10.1177/0956797611430096
PDF | PDF Supp

optimality/rationality
visual decision-making

Van den Berg R, Ma WJ (2012),
Robust averaging during perceptual judgment is not optimal
Proceedings of the National Academy of Sciences 2012. DOI: 10.1073/pnas.1119078109
PDF
Reply by De Gardelle and Summerfield
Our response to their reply

inference/uncertainty
optimality/rationality
perceptual organization
visual decision-making

Van den Berg R, Vogel M, Josic K, Ma WJ (2012),
Optimal inference of sameness
Proceedings of the National Academy of Sciences 109 (8), 3178-83. DOI: 10.1073/pnas.1108790109
PDF | PDF Supp | Code Download | Code (GitHub)

2011

comparative cognition
working memory

Elmore LC, Ma WJ, Magnotti JF, Leising KJ, Passaro AD, Katz JS, Wright AA (2011),
Visual short-term memory compared in rhesus monkeys and humans
Current Biology 21 (11), 975-9. DOI: 10.1016/j.cub.2011.04.031
PDF | PDF Supp
Commentary by Jonathon Crystal: Comparative cognition: comparing human and monkey memory

decision rules
inference/uncertainty
optimality/rationality
visual decision-making
visual search

Ma WJ*, Navalpakkam V*, Beck JM*, Van den Berg R*, Pouget A (2011),
Behavior and neural basis of near-optimal visual search
Nature Neuroscience 14 (6), 783-90. DOI: 10.1038/nn.2814
PDF | PDF Supp | Code Download | Code (GitHub)
Press release "Brain Performs Near Optimal Visual Search" by BCM
Press release "Bekenden zien is een kwestie van hersenstrategie" by Volkskrant (Dutch)
Press release "Het brein: snelle en goede detective" by Parool (Dutch)
Press release "Mensenhersens Model voor Zoekmachine" by VPRO's Labyrint (Dutch) (Radio)

neural networks/AI
neural coding
multisensory perception

Ma WJ, Beck JM, Pouget A (2011),
A neural implementation of optimal cue integration
In Sensory Cue Integration. Trommershäuser J, Körding K, Landy MS (eds.), Oxford University Press, New York, NY.
PDF

2010

inference/uncertainty
neural coding
review/tutorial

Ma WJ (2010),
Signal detection theory, uncertainty, and Poisson-like population codes
Vision Research 50, 2308-19. DOI: 10.1016/j.visres.2010.08.035
PDF

2009

optimality/rationality
inference/uncertainty
attention
visual decision-making

Ma WJ, Huang W (2009),
No capacity limit in attentional tracking: evidence for probabilistic inference under a resource constraint
Journal of Vision, 9 (11): 3, 1-30. DOI: 10.1167/9.11.3
PDF | Code

inference/uncertainty
multisensory perception

Ma WJ*, Zhou X*, Ross LA, Foxe JJ, Parra LC (2009),
Lip-reading aids word recognition most in moderate noise: a Bayesian explanation using high-dimensional feature space
PLoS ONE 4 (3): e4638. DOI: 10.1371/journal.pone.0004638
PDF | PDF Supp | Sample Stimuli
Press release "What you see affects what you hear" by BCM
Press release "Visual Cues Help People Understand Spoken Words" by Science Daily
Press release "Why your brain can't always make good decisions" by CNN
Press release "Blinded by the lyric? Study reveals why we get the words wrong" by MSNBC

motivation

Zhou MM, Ma WJ, Deci E (2009),
The importance of autonomy for rural Chinese children’s motivation for learning
Learning and Individual Differences, 19, 492-8. DOI:10.1016/j.lindif.2009.05.003
PDF

inference/uncertainty
optimality/rationality
review/tutorial

Ma WJ (2009),
Bayesian approach to perception
In SAGE Encyclopedia of Perception, Goldstein EB ed., 201-5.
PDF

neural coding
review/tutorial

Ma WJ, Pouget A (2009),
Population codes: theoretical aspects
In Encyclopedia of Neuroscience, Squire LR ed., 7, 749-55. Academic Press, Oxford.
PDF

2008

inference/uncertainty
neural coding
visual decision-making

Beck JM*, Ma WJ*, Kiani R, Hanks TD, Churchland AK, Roitman JD, Shadlen MN, Latham, PE, and Pouget A (2008),
Probabilistic Population Codes for Bayesian Decision Making
Neuron 60 (6), 1142-5. DOI: 10.1016/j.neuron.2008.09.021
PDF | PDF Supp | Commentary
Press release "Brain smarter than we are" by Scientific American
Press release "Why your brain can't always make good decisions" by CNN

encoding/ representation
multisensory perception
neural coding

Ma WJ, Pouget A (2008),
Linking neurons to behavior in multisensory perception: a computational review
Brain Research 1242, 4-12. DOI: 10.1016/j.brainres.2008.04.082
PDF

inference/uncertainty
neural coding
review/tutorial

Ma WJ, Beck JM, Pouget A (2008),
Spiking networks for Bayesian inference and choice
Current Opinion in Neurobiology 18, 217-22. DOI: 10.1016/j.conb.2008.07.004
PDF

methods
multisensory perception
visual decision-making

Beierholm U, Kording K, Shams L, Ma WJ (2008),
Comparing Bayesian models for multisensory cue combination without mandatory integration
Advances in Neural Information Processing Systems 20, 81-88. MIT Press, Cambridge, MA.
PDF

2007

encoding/ representation
inference/uncertainty
neural coding

Beck JM, Ma WJ, Latham PE, Pouget A (2007),
Probabilistic population codes and the exponential family of distributions
Progress in Brain Research 165, 509-19. DOI: 10.1016/S0079-6123(06)65032-2
PDF

decision rules
inference/uncertainty
multisensory perception
optimality/rationality

Kording K*, Beierholm U*, Ma WJ*, Quartz S, Tenenbaum JB, Shams L (2007),
Causal inference in multisensory perception
PLoS ONE 2 (9), e943. DOI: 10.1371/journal.pone.0000943
PDF | PDF Supp | Figure 1 | Figure 2

2006

inference/uncertainty
multisensory perception
optimality/rationality
neural coding

Ma WJ*, Beck JM*, Latham PE, Pouget A (2006),
Bayesian inference with probabilistic population codes
Nature Neuroscience 9 (11), 1432-8 165, 509-19. DOI: 10.1038/nn1790
PDF | PDF Supp | Commentary

illusions
neural networks/AI
time

Ma WJ, Hamker F, Koch C (2006),
Neural mechanisms underlying temporal aspects of conscious visual perception
In The first half second: the microgenesis and temporal dynamics of unconscious and conscious visual processes, Ögmen H, Breitmeyer BG eds., 275-94, MIT Press, Cambridge MA.
PDF | References

2005

inference/uncertainty
multisensory perception

Shams L, Ma WJ, Beierholm U (2005),
Sound-induced flash illusion as an optimal percept
NeuroReport 16 (17), 1923-7. DOI: 10.1097/01.wnr.0000187634.68504.bb
PDF

2004

decision rules
encoding/ representation
working memory

Wilken P, Ma WJ (2004),
A detection theory account of change detection
Journal of Vision 4 (12): 11, 1120-35. DOI: 10.1167/4.12.11
PDF

planning
four-in-a-row
neural networks/AI
inference/uncertainty
value-based decision-making
optimality/rationality
learning
big data

Ionatan Kuperwajs (2023),
Cognitive mechanisms of complex planning
PDF

visual decision-making
working memory
decision rules
confidence
inference/uncertainty

Aspen Yoo (2019),
The role of uncertainty and priority in visual working memory
PDF

decision-making
planning
methods
four-in-a-row

Sebastiaan van Opheusden (2019),
Complex decision-making
PDF

visual search
inference/uncertainty
visual decision-making
attention
methods
disorders
optimality/rationality

Andra Mihali (2018),
Measures and models of covert visual attention in neurotypical function and ADHD
PDF

confidence
decision rules
inference/uncertainty
visual decision-making
attention
optimality/rationality

Will Adler (2018),
Computational mechanisms underlying human confidence reports
PDF

neural coding
encoding/representation
visual decision-making
set size effects

Stuart Jackson (2016),
Encoding-decoding models of luminance contrast processing
PDF

encoding/representation
visual decision-making
working memory
set size effects
decision rules

Hongsup Shin (2015),
Measuring and modeling of human visual short-term memory
PDF

encoding/representation
visual decision-making
working memory
set size effects
decision rules
inference/uncertainty
comparative cognition

Deepna Devkar (014),
Main advisor: Anthony Wright; co-advised by Wei Ji,
Change detection in rhesus monkeys and humans
PDF

visual decision-making
encoding/ representation
visual search
set size effects
decision rules
optimality/rationality

Manisha Bhardwaj (2013),
Main advisor: Kresimir Josic; co-advised by Wei Ji,
Visual decision-making in the presence of stimulus and measurement correlations
PDF

Finding the lab's output