MSMaarten Speekenbrink

Maarten Speekenbrink

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Maarten Speekenbrink has a total of 614 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Photoentrainment as Bayesian Inference: How the central circadian clock weights and integrate the intensity, wavelength and rate of change of natural light to estimate astronomical time

    by Danny M. Ball, Vincent Walsh, Russell G. Foster, Maarten Speekenbrink

    Synchronising with the Earth’s 24-hour rotation using natural light as a timing cue presents a problem for the mammalian brain’s central circadian clock. Natural light is highly variable and can take a range of values at any given time of day, for example, fluctuating due to differences in weather conditions and across seasons. Despite this variability, the central clock maintains close alignment with astronomical time, and here we describe how this may be achieved computationally. We present a Bayesian model in which the clock’s endogenous oscillation is used to predict the probability of observing the features of light, intensity, spectral composition, and their co-occurrence given the time of day, drawn from a distribution of values typically experienced. When observed light falls outside the predicted range, a prediction error provides evidence that the clock may be misaligned with astronomical time. How much the clock then adjusts its phase depends on the relative uncertainty associated with its internal prediction and observed light. These are combined as a weighted average according to their relative reliability, drawing the clock’s updated phase towards whichever provides the more reliable estimate, either the clock’s prediction or the observation, of astronomical time. Using natural light measurements collected over one week from 59 human observers and 100 days of environmental spectral measurements, we characterise how the variability of natural light systematically changes across the day and show how the clock may exploit this structure for circadian timekeeping. The model explains how the canonical shape of the phase-response curve emerges from the relationship between light distributions and time of day, and how properties of light, including intensity and spectral composition, can be integrated according to their reliability. The central clock’s synchrony to light thus represents a specific instance of a general computational strategy employed by the mammalian nervous system, whereby sensory information is interpreted relative to an internally generated prediction.

  • Top Cited Papers

    Investigating the analytical robustness of the social and behavioural sciences

    by Balazs Aczel, Barnabas Szaszi, Harry T. Clelland, Marton Kovacs, Felix Holzmeister, Don van Ravenzwaaij, Hannah Schulz-Kümpel, Sabine Hoffmann, Gustav Nilsonne, Livia Kosa, et al.

    The same dataset can be analysed in different justifiable ways to answer the same research question, potentially challenging the robustness of empirical science1-3. In this crowd initiative, we investigated the degree to which research findings in the social and behavioural sciences are contingent on analysts' choices. We examined a stratified random sample of 100 studies published between 2009 and 2018, in which, for one claim per study, at least five reanalysts independently reanalysed the original data. The statistical appropriateness of the reanalyses was assessed in peer evaluations, and the robustness indicators were inspected along a range of research characteristics and study designs. We found that 34% of the independent reanalyses yielded the same result (within a tolerance region of ±0.05 Cohen's d) as the original report; with a four times broader tolerance region, this indicator increased to 57%. Of the reanalyses conducted, 74% reached the same conclusion as the original investigation, 24% yielded no effects or inconclusive results and 2% reported the opposite effect. This exploratory study indicates that the common single-path analyses in social and behavioural research should not be simply assumed to be robust to alternative analyses4. Therefore, we recommend the development and use of practices to explore and communicate this neglected source of uncertainty.


  • Open Access

    (Re)building Cooperation: Effects of a Cognitive Intervention on Cooperative Behaviour in Games

    by Ismail Guennouni, Samuel Dupret, Quentin J. M. Huys, Maarten Speekenbrink

    Cooperation often breaks down when one party appears to withdraw trust, yet most research has focused on initial trust formation rather than how the aggrieved party can resist the impulse to retaliate. In a randomised controlled online trial (N = 318), we tested whether a brief cognitive intervention inspired by Dialectical Behaviour Therapy principles could promote resilient cooperation in the face of trust withdrawals. Participants played a Repeated Trust Game as trustees against adaptive computerised investors driven by hidden Markov models (HMMs) trained on human data, providing behavioural realism under full experimental control. Following the intervention, participants returned a significantly higher proportion of the amount received than an active control group across the post-manipulation game. The groups diverged most sharply at the pre-programmed low investment, where the control group retaliated by reducing returns while the intervention group maintained the same level of returns. This shift occurred without a change in self-reported emotional reactivity, dissociating the preserved emotional reaction from the retaliatory behaviour it drives. HMM analysis of participants’ own return patterns revealed that participants in the intervention condition were less likely to become stuck in low-return behavioural states following the pre-programmed low investment, though a notable minority did not respond to the intervention. The intervention effect did not transfer to a Prisoner’s Dilemma game, consistent with context-specific effects. These findings provide proof-of-concept evidence that targeted cognitive strategies can support cooperation after trust violations, and show that HMM-based agents are a useful tool for studying social behaviour under controlled conditions.


  • Open Access

    Less is more: Local focus in continuous time causal learning.

    by Victor Btesh, Neil R. Bramley, Maarten Speekenbrink, David A. Lagnado

    In this study, we investigated human causal learning in a continuous time and space setting. We find participants to be capable active causal structure learners, and with the help of computational modeling explore how they mitigate the complexity of continuous dynamics data to achieve this. We propose that participants combine systematic interventions with a narrowed focus on causal dynamics that occur during and directly downstream of their interventions. This task decomposition approach achieves comparable accuracy to attending to all the dynamics, while discarding almost half of the data. We argue this strategy makes sense from a resource rationality perspective: Ignoring dynamics outside of interventions saves computational cost while the interventions naturally decompose the global learning problem into a series of more manageable subproblems. We also find that when the causal relata are given real-world labels, participants will use their domain-specific priors to guide their structure inferences. In particular, individuals with accurate prior expectations were less likely to make the common local computations error of mistaking an indirect for a direct relationship. Overall, our experiments reinforce the idea that humans are frugal and intuitive active learners who combine actions and inference to optimize learning while minimizing effort. (PsycInfo Database Record (c) 2025 APA, all rights reserved).

  • Open Access

    State-Dependent Missingness in Hidden Markov Models, with an Application to Drop-Out in a Clinical Trial

    by Maarten Speekenbrink, Ingmar Visser

    Establishing the effectiveness of treatments for psychopathology requires accurate models of its progression over time and the factors that impact it. Longitudinal data is however fraught with missingness, hindering accurate modeling. We re-analyse data on schizophrenia severity in a clinical trial using hidden Markov models (HMMs). We consider missing data in HMMs with a focus on situations where data is missing not at random (MNAR) and missingness depends on the latent states, allowing symptom severity to indirectly impact probability of missingness. In simulations, we show that including a submodel for state-dependent missingness reduces bias when data is MNAR and state-dependent, whilst not reducing accuracy when data is missing-at-random (MAR). When missingness depends on time, a model that allows missingness to be both state- and time-dependent is unbiased. Overall, these results show that modelling missingness as state-dependent and including relevant covariates is a useful strategy in applications of HMMs to time-series with missing data. Applying the model to data from a clinical trial, we find that drop-out is more likely for patients with less severe symptoms, which may lead to a biased assessment of treatment effectiveness.


  • Follow my example, for better and for worse: The influence of behavioral traces on recycling decisions.

    by Sabine Topf, Maarten Speekenbrink

    Recycling behavior can recover valuable materials and mitigate green house gas emissions from landfills and incinerators. The potential positive impact of individuals' recycling behavior depends on others also making an effort, for instance, avoiding contamination. Knowing what other people have done may therefore influence recycling behavior. Behavioral traces are evidence of other people's behavior in a shared environment. Here, they relate to waste items already placed in one of two bins, a mixed recycling bin and a nonrecyclable waste bin. In two online experiments and one real-life intervention study, we investigate the role of behavioral traces on the willingness to recycle as well as the correctness of recycling. We find that seeing behavioral traces of previous recycling behavior makes recycling generally more likely, and people tend to copy item placement. This in turn increases correctness in groups where the average individual has good knowledge of recycling. Introducing correct items at the start of the day in the intervention study did not increase correctness, possibly because the correct items were soon buried by other items. Implications and future directions are discussed. (PsycInfo Database Record (c) 2023 APA, all rights reserved).


  • Human optional stopping in a heteroscedastic world.

    by Hannah Tickle, Konstantinos Tsetsos, Maarten Speekenbrink, Christopher Summerfield

    When making decisions, animals must trade off the benefits of information harvesting against the opportunity cost of prolonged deliberation. Deciding when to stop accumulating information and commit to a choice is challenging in natural environments, where the reliability of decision-relevant information may itself vary unpredictably over time (variable variance or "heteroscedasticity"). We asked humans to perform a categorization task in which discrete, continuously valued samples (oriented gratings) arrived in series until the observer made a choice. Human behavior was best described by a model that adaptively weighted sensory signals by their inverse prediction error and integrated the resulting quantities with a linear urgency signal to a decision threshold. This model approximated the output of a Bayesian model that computed the full posterior probability of a correct response, and successfully predicted adaptive weighting of decision information in neural signals. Adaptive weighting of decision information may have evolved to promote optional stopping in heteroscedastic natural environments. (PsycInfo Database Record (c) 2023 APA, all rights reserved).


  • Open Access

    Chasing Unknown Bandits: Uncertainty Guidance in Learning and Decision Making

    by Maarten Speekenbrink

    In repeated decision problems for which it is possible to learn from experience, people should actively seek out uncertain options, rather than avoid ambiguity or uncertainty, in order to learn and improve future decisions. Research on human behavior in a variety of multiarmed-bandit tasks supports this prediction. Multiarmed-bandit tasks involve repeated decisions between options with initially unknown reward distributions and require a careful balance between learning about relatively unknown options (exploration) and obtaining high immediate rewards (exploitation). Resolving this exploration-exploitation dilemma optimally requires considering not only the estimated value of each option, but also the uncertainty in these estimations. Bayesian learning naturally quantifies uncertainty and hence provides a principled framework to study how humans resolve this dilemma. On the basis of computational modeling and behavioral results in bandit tasks, I argue that human learning, attention, and exploration are guided by uncertainty. These results support Bayesian theories of cognition and underpin the fundamental role of subjective uncertainty in both learning and decision making.


  • Open Access

    Transfer of Learned Opponent Models in Zero Sum Games

    by Ismail Guennouni, Maarten Speekenbrink

    Human learning transfer abilities take advantage of important cognitive building blocks such as an abstract representation of concepts underlying tasks and causal models of the environment. One way to build abstract representations of the environment when the task involves interactions with others is to build a model of the opponent that may inform what actions they are likely to take next. In this study, we explore opponent modelling and its transfer in games where human agents play against computer agents with human-like limited degrees of iterated reasoning. In two experiments, we find that participants deviate from Nash equilibrium play and learn to adapt to their opponent’s strategy to exploit it. Moreover, we show that participants transfer their learning to new games. Computational modelling shows that players start each game with a model-based learning strategy that facilitates between-game transfer of their opponent’s strategy, but then switch to behaviour that is consistent with a model-free learning strategy in the latter stages of the interaction.