Professor Scott Brown

Background: Treatment options for low-risk differentiated thyroid cancer (DTC) include active surveillance (AS), hemithyroidectomy (HT), or total thyroidectomy (TT). Improved understanding of patient values and preferences is required to inform shared decision-making. This study examined factors influencing patient treatment preferences and trade-offs for low-risk DTC. Methods: Adult participants with benign thyroid nodules or low-risk DTC completed an online discrete choice experiment (DCE). Utilizing the scenario of a 50-year-old person with a small solitary DTC, participants chose between three unlabeled treatment options (representing AS, HT, and TT). Risk profiles varied across 5 domains: voice change, thyroid hormone supplementation, hypocalcaemia, chance of future thyroid surgery, and 10-year risk of cancer recurrence. Participants self-reported demographics, disease factors, and answered a decisional regret scale. A conditional logit model was utilized. Results: The DCE was completed by 143 patients across three sites. The conditional logit model demonstrated that participants preferred AS (49%) over TT (29%) or HT (22%). All five domains influenced choices (all p¿<¿0.001), but perceived risk of cancer recurrence exerted most influence. Cancer survivors chose AS less often than those with benign disease (46% vs. 57%), driven by perceived risks of further surgery and cancer recurrence. As the perceived risk of cancer recurrence increased, more participants preferred HT over AS. Conclusion: This study demonstrates that when blinded to the actual treatment, patients prefer the trade-offs associated with AS rather than TT or HT. Perceived risk of cancer recurrence exerted the greatest influence. Accurate risk stratification for cancer recurrence is critical to shared decision-making.

Purpose: Post-event information (PEI) may distort eyewitness memory and lead to erroneous eyewitness testimonies. This paper aims to explore whether factors such as volitional engagement with PEI (e.g. choice to engage with a co-witness) and memory distrust influence misinformation acceptance and the perceived credibility of a co-witness. Design/methodology/approach: Participants (n = 223) completed the Memory Distrust Scale and then watched a short mock crime video. Thereafter, two-thirds of the participants were asked whether they would prefer or not to listen to a co-witness' account of the witnessed event (choice condition), and one-third of the participants did not have the choice (control condition). Every participant listened to the co-witness account (which contained items of misinformation); thus, those who preferred to listen to the testimony were in the choice-yes (i.e., volition) condition and those who preferred not to listen were in the choice-no (i.e., non-volition) condition. Finally, participants completed a cued recall task assessing their memory of the video and acceptance of misinformation. They also provided ratings to establish the perceived credibility of the co-witness. Findings: The results indicated that neither volition nor memory distrust influenced misinformation acceptance. However, those who preferred to listen to the testimony (i.e., the choice-yes condition) perceived the co-witness as more credible than those in the choice-no or control conditions. Practical implications: The findings suggest that witnesses are susceptible to misinformation regardless of their willingness to engage with or avoid PEI. Further implications and future research directions are discussed. Originality/value: To the best of the authors' knowledge, this study is the first to investigate the role of volition and memory distrust as a trait in eyewitnesses tendency to engage with or avoid post-event information. The research explores whether these mechanisms impact upon memory conformity and perceived co-witness credibility.

Whenever someone in a team tries to help others, it is crucial that they have some understanding of other team members' goals. In modern teams, this applies equally to human and artificial ("bot") assistants. Understanding when one does not know something is crucial for stopping the execution of inappropriate behavior and, ideally, attempting to learn more appropriate actions. From a statistical point of view, this can be translated to assessing whether none of the hypotheses in a considered set is correct. Here we investigate a novel approach for making this assessment based on monitoring the maximum a posteriori probability (MAP) of a set of candidate hypotheses as new observations arrive. Simulation studies suggest that this is a promising approach, however, we also caution that there may be cases where this is more challenging. The problem we study and the solution we propose are general, with applications well beyond human¿bot teaming, including for example the scientific process of theory development.

In the modern world, many important tasks have become too complex for a single unaided individual to manage. Teams conduct some safety-critical tasks to improve task performance and minimize the risk of error. These teams have traditionally consisted of human operators, yet, nowadays, artificial intelligence and machine systems are incorporated into team environments to improve performance and capacity. We used a computerized task modeled after a classic arcade game to investigate the performance of human¿machine and human¿human teams. We manipulated the group conditions between team members; sometimes, they were instructed to collaborate, compete, or work separately. We evaluated players' performance in the main task (gameplay) and, in post hoc analyses, participant behavioral patterns to inform group strategies. We compared game performance between team types (human¿human vs. human¿machine) and group conditions (competitive, collaborative, independent). Adapting workload capacity analysis to human¿machine teams, we found performance under both team types and all group conditions suffered a performance efficiency cost. However, we observed a reduced cost in collaborative over competitive teams within human¿human pairings, but this effect was diminished when playing with a machine partner. The implications of workload capacity analysis as a powerful tool for human¿machine team performance measurement are¿discussed.

Purpose: To explore in a sample of adult cancer patients: (1) the relative influence of initiation source, information format and consultation format on preferred approach to life expectancy disclosure using a discrete choice experiment (DCE); and (2) whether patient age, cancer type and perceived prognosis were associated with preferences within the three attributes. Methods: A DCE survey of adult solid tumour and haematological cancer patients. Participants chose between three hypothetical scenarios about life expectancy disclosure consisting of three attributes: initiation source (i.e. doctor versus patient-initiated discussion), information content (i.e. estimate presented as best-worst-typical length of life case scenario versus median survival time) and consultation format (i.e. two 20-min versus one 40-min consultation). Respondents selected their most preferred scenario within each question. Results: Three hundred and two patients completed the DCE (78% consent rate). Initiation source was the most influential predictor of patient choice. More preferred a doctor deliver life expectancy information as soon as it is available rather than waiting for the patient to ask (59% vs 41% z = - 7.396, p < 0.01). More patients preferred the two 20-min rather than the one 40-min consultation format (55% vs 45%, z = 4.284, p < 0.01). Information content did not influence choice. Age, cancer type, and patient-perceived prognosis were not associated with preferences. Conclusion: Healthcare professionals should assess cancer patients' preferences for engaging in life expectancy discussions as soon as they have this information, and ensure patients have adequate time to consider the information they receive, seek additional information and involve others if they wish.

BackTrack is a multi-component, community-based intervention designed to build capacity amongst 14¿17-year-old high risk young people. The aim of the current study seeks to explore community value and preferences for reducing youth crime and improving community safety using BackTrack in a rural setting in Armidale, New South Wales, Australia. The study design used discrete choice experiments (DCEs), designed in accordance with the 10-item checklist outlined by the International Society for Pharmacoeconomics and Outcomes Research. The DCE was pilot tested on 43 participants to test feasibility and comprehension. A revised version of the survey was subsequently completed by 282 people over a 12-day period between 30 May 2016 and 10 June 2016, representing a survey response rate of 35%. Ninety per cent of respondents were residents of Armidale, the local rural town where BackTrack was implemented. The DCE generated results that consistently demonstrated a preference for social programs to address youth crime and community safety in the Armidale area. Respondents chose BackTrack over Greater Police Presence 75% of the time with an annual benefit of Australian dollars (AUD) 150 per household, equivalent to a community benefit of AUD 2.04 million. This study estimates a strong community preference for BackTrack relative to more policing (a community willing to pay equivalent to AUD 2.04 million) highlighting the clear value of including community preferences when evaluating community-based programs for high-risk young people.

Few fathers enrol in web-based preventive parenting programs for adolescent mental health, despite the evidence of the benefits associated with their participation. To inform the development of father-inclusive programs, this study used a discrete choice experiment (DCE) design to determine (a) the relative influence of number of sessions, program benefits, program participants, and user control over program content on fathers' preferences for web-based preventive parenting programs; and (b) whether selected father characteristics were associated with their preferences. One hundred and seventy-one fathers completed the DCE survey, which comprised 25 choices between hypothetical programs. Programs that included the participant's adolescent child (z = 10.06, p < 0.0001), or parenting partner (z = 7.30, p < 0.001) were preferred over those designed for fathers only. Participants also preferred program content that was recommended for them by experts (z = -4.31, p < 0.0001) and programs with fewer sessions (z = -2.94, p < 0.01). Program benefits did not predict fathers' choice of program. Prior use of a parenting program, level of education, perceived role of parenting for adolescent mental health, and being part of a dual-working family were associated with preferences. Application of these findings may improve paternal enrolment in web-based preventive parenting programs.

It is commonly assumed that a specific testing occasion (task, design, procedure, etc.) provides insights that generalize beyond that occasion. This assumption is infrequently carefully tested in data. We develop a statistically principled method to directly estimate the correlation between latent components of cognitive processing across tasks, contexts, and time. This method simultaneously estimates individual-participant parameters of a cognitive model at each testing occasion, group-level parameters representing across-participant parameter averages and variances, and across-task correlations. The approach provides a natural way to "borrow" strength across testing occasions, which can increase the precision of parameter estimates across all testing occasions. Two example applications demonstrate that the method is practical in standard designs. The examples, and a simulation study, also provide evidence about the reliability and validity of parameter estimates from the linear ballistic accumulator model. We conclude by highlighting the potential of the parameter-correlation method to provide an "assumption-light" tool for estimating the relatedness of cognitive processes across tasks, contexts, and time.

Objective: To explore the preferences of people with cancer regarding the timing and format of information provision about emotional concerns that may occur when undergoing medical procedures. Methods: Eligible cancer survivors were mailed a survey containing discrete choice scenarios examining their timing and format preferences for information about potential emotional concerns associated with an upcoming hypothetical medical procedure. Results: Of 356 eligible patients, 271 (76 %) completed the survey. Both face-to-face discussion and written materials were preferred as the mode of information delivery over access to a website. In order of descending preference, participants preferred to receive the information 1 week, 3 days and the day of the procedure. There were no differences in preferences for timing or format between subgroups based on age, gender, education and cancer type. Conclusion: This study has demonstrated that cancer patients prefer receiving information about emotional concerns that might be experienced as part of a medical procedure in either written or via face-to-face format, and one week before the procedure. Practice Implications: In order to provide patient-centred care, clinicians and the healthcare system more broadly should consider patient preferences for information delivery about upcoming medical procedures. Information: preparation for medical procedures; discrete choice; oncology; patient preference; emotional response.

With the advancement of technologies like in-car navigation and smartphones, concerns around how cognitive functioning is influenced by "workload" are increasingly prevalent. Research shows that spreading effort across multiple tasks can impair cognitive abilities through an overuse of resources, and that similar overload effects arise in difficult single-task paradigms. We developed a novel lab-based extension of the Detection Response Task, which measures workload, and paired it with a Multiple Object Tracking Task to manipulate cognitive load. Load was manipulated either by changing within-task difficulty or by the addition of an extra task. Using quantitative cognitive modelling we showed that these manipulations cause similar cognitive impairments through diminished processing rates, but that the introduction of a second task tends to invoke more cautious response strategies that do not occur when only difficulty changes. We conclude that more prudence should be exercised when directly comparing multi-tasking and difficulty-based workload impairments, particularly when relying on measures of central tendency.

Most data analyses rely on models. To complement statistical models, psychologists have developed cognitive models, which translate observed variables into psychologically interesting constructs. Response time models, in particular, assume that response time and accuracy are the observed expression of latent variables including 1) ease of processing, 2) response caution, 3) response bias, and 4) non-decision time. Inferences about these psychological factors, hinge upon the validity of the models' parameters. Here, we use a blinded, collaborative approach to assess the validity of such model-based inferences. Seventeen teams of researchers analyzed the same 14 data sets. In each of these two-condition data sets, we manipulated properties of participants' behavior in a two-alternative forced choice task. The contributing teams were blind to the manipulations, and had to infer what aspect of behavior was changed using their method of choice. The contributors chose to employ a variety of models, estimation methods, and inference procedures. Our results show that, although conclusions were similar across different methods, these "modeler's degrees of freedom" did affect their inferences. Interestingly, many of the simpler approaches yielded as robust and accurate inferences as the more complex methods. We recommend that, in general, cognitive models become a typical analysis tool for response time data. In particular, we argue that the simpler models and procedures are sufficient for standard experimental designs. We finish by outlining situations in which more complicated models and methods may be necessary, and discuss potential pitfalls when interpreting the output from response time models.

Decision-making involves a tradeoff between pressures for caution and urgency. Previous research has investigated how well humans optimize this tradeoff, and mostly concluded that people adopt a sub-optimal strategy that over-emphasizes caution. This emphasis reduces how many decisions can be made in a fixed time, which reduces the "reward rate". However, the strategy that is optimal depends critically on the timing properties of the experiment design: the slower the rate of decision opportunities, the more cautious the optimal strategy. Previous studies have almost uniformly adopted very fast designs, which favor very urgent decision-making. This raises the possibility that previous findings¿that humans adopt strategies that are too cautious¿could either be ascribed to human caution, or to the experiments' design. To test this, we used a slowed-down decision-making task in which the optimal strategy was quite cautious. With this task, and in contrast to previous findings, the average strategy adopted across participants was very close to optimal, with about equally many participants adopting too-cautious as too-urgent strategies. Our findings suggest that task design can play a role in inferences about optimality, and that previous conclusions regarding human sub-optimality are conditional on the task settings. This limits claims about human optimality that can be supported by the available evidence.

Objective: Using a vignette-style DCE in a sample of oncology patients, this study explored: (1) the relative influence of the patient's level of concern about their depression on preferences for care, (2) the relative influence of depression severity according to a mental health checklist on preferred treatment-seeking options, and (3) whether patient age and gender were associated with depression care preference. Methods: A discrete choice experiment (DCE) survey of cancer patients was conducted. Hypothetical vignettes to elicit care preferences were created using two attributes: the cancer patient's level of concern about depression (a little or a great deal) and results of a mental health checklist (not depressed or very depressed). Three response options for care preferences were presented, including a self-directed approach, shared care approach, and clinician-directed referral approach. Participants chose their most and least preferred options. Results: A total of 281 cancer patients completed the survey. There was a significant association between level of concern and the most preferred option. Those with a great deal of concern about depression preferred to receive referral from their clinician more than those with a little concern about depression. Males were significantly more likely to select a self-directed approach as their most preferred option. Conclusions: An oncology patient's level of concern about depression may influence the type of care they want to receive from their cancer doctor for depression. This finding has implications for depression screening in clinical practice.

An important feature of human cognition is the ability to flexibly and efficiently adapt behavior in response to continuously changing contextual demands. We leverage a large-scale dataset from Lumosity, an online cognitive-training platform, to investigate how cognitive processes involved in cued switching between tasks are affected by level of task practice across the adult lifespan. We develop a computational account of task switching that specifies the temporal dynamics of activating task-relevant representations and inhibiting task-irrelevant representations and how they vary with extended task practice across a number of age groups. Practice modulates the level of activation of the task-relevant representation and improves the rate at which this information becomes available, but has little effect on the task-irrelevant representation. While longterm practice improves performance across all age groups, it has a greater effect on older adults. Indeed, extensive task practice can make older individuals functionally similar to lesspracticed younger individuals, especially for cognitive measures that focus on the rate at which task-relevant information becomes available.

Only 2¿3% of cancer patients enrol in a trial. We surveyed patients' willingness to change clinician or treating centre, or to travel, to participate in trials, to improve trial recruitment. Of 188 respondents, 79% were willing to participate in a trial in at least one scenario. Increasing travel time, change in oncologist, private health insurance and out of pocket expenses decreased likelihood of joining a trial. Rural and regional patients, and those from lower socio-economic areas, were more willing to travel. To optimise access to trials, clinicians should refer within and between institutions.

The "law of practice"-a simple nonlinear function describing the relationship between mean response time (RT) and practice- has provided a practically and theoretically useful way of quantifying the speed-up that characterizes skill acquisition. Early work favored a power law, but this was shown to be an artifact of biases caused by averaging over participants who are individually better described by an exponential law. However, both power and exponential functions make the strong assumption that the speedup always proceeds at a steadily decreasing rate, even though there are sometimes clear exceptions. We propose a new law that can both accommodate an initial delay resulting in a slower-faster-slower rate of learning, with either power or exponential forms as limiting cases, and which can account for not only mean RT but also the effect of practice on the entire distribution of RT. We evaluate this proposal with data from a broad array of tasks using hierarchical Bayesian modeling, which pools data across participants while minimizing averaging artifacts, and using inference procedures that take into account differences in flexibility among laws. In a clear majority of paradigms our results supported a delayed exponential law.

Evidence accumulation models of decision-making have led to advances in several different areas of psychology. These models provide a way to integrate response time and accuracy data, and to describe performance in terms of latent cognitive processes. Testing important psychological hypotheses using cognitive models requires a method to make inferences about different versions of the models which assume different parameters to cause observed effects. The task of model-based inference using noisy data is difficult, and has proven especially problematic with current model selection methods based on parameter estimation. We provide a method for computing Bayes factors through Monte-Carlo integration for the linear ballistic accumulator (LBA; Brown and Heathcote, 2008), a widely used evidence accumulation model. Bayes factors are used frequently for inference with simpler statistical models, and they do not require parameter estimation. In order to overcome the computational burden of estimating Bayes factors via brute force integration, we exploit general purpose graphical processing units; we provide free code for this. This approach allows estimation of Bayes factors via Monte-Carlo integration within a practical time frame. We demonstrate the method using both simulated and real data. We investigate the stability of the Monte-Carlo approximation, and the LBA's inferential properties, in simulation studies.

Neuroscientific theories of attention-deficit/hyperactivity disorder (ADHD) alternately posit that cognitive aberrations in the disorder are due to acute attentional lapses, slowed neural processing, or reduced signal-to-noise ratios. However, they make similar predictions about behavioral summary statistics (response times [RTs] and accuracy), hindering the field's ability to produce strong and specific tests of these theories. The current study uses the linear ballistic accumulator (LBA; Brown & Heathcote, 2008), a mathematical model of choice RT tasks, to distinguish between competing theory predictions. Children with ADHD (n = 80) and age-matched controls (n = 32) completed a numerosity discrimination paradigm at 2 levels of difficulty, and RT data were fit to the LBA model to test theoretical predictions. Individuals with ADHD displayed slowed processing of evidence for correct responses (signal) relative to their peers but comparable processing of evidence for error responses (noise) and between-trial variability in processing (performance lapses). The findings are inconsistent with accounts that posit an increased incidence of attentional lapses in the disorder and provide partial support for those that posit slowed neural processing and lower signal-to-noise ratios. Results also highlight the utility of well-developed cognitive models for distinguishing between the predictions of etiological theories of psychopathology.

Objectives: To explore in a sample of medical oncology outpatients and their nominated support persons (SPs): (1) the relative influence of pain, consciousness and life extension on end-of-life choices using a discrete choice experiment (DCE); (2) the extent to which SPs can predict the choices of index patients and (3) whether having a previous end-of-life discussion was associated with dyad agreement. Methods: Adult medical oncology patients and their SPs were approached for consent to complete a survey containing a DCE. Participants chose between three unlabelled care scenarios characterised by three attributes: pain (mild, moderate or severe), consciousness (some, half or most of time) and extension of life (1, 2 or 3¿weeks). Respondents selected (1) most-preferred and (2) least-preferred scenarios within each question. SPs answered the same questions but from patient's perspective. Results: A total of 110 patients and 64 SPs responded overall (42 matched patient-SP dyads). For patients, pain was the most influential predictor of most- and least-preferred scenarios (z = 12.5 and z = 12.9). For SPs, pain was the only significant predictor of most and least-preferred scenarios (z = 9.7 and z = 11.5). Dyad agreement was greater for choices about least- (69%) compared to most-preferred scenarios (55%). Agreement was slightly higher for dyads reporting a previous EOL discussion (68 versus 48%; p = 0.065). Conclusion: Patients and SPs place significant value on avoiding severe pain when making end-of-life choices, over and above level of consciousness or life extension. People's views about end-of-life scenarios they most as well as least prefer should be sought.

Understanding individual differences in cognitive performance is an important part of understanding how variations in underlying cognitive processes can result in variations in task performance. However, the exploration of individual differences in the components of the decision process¿such as cognitive processing speed, response caution, and motor execution speed¿in previous research has been limited. Here, we assess the heritability of the components of the decision process, with heritability having been a common aspect of individual differences research within other areas of cognition. Importantly, a limitation of previous work on cognitive heritability is the underlying assumption that variability in response times solely reflects variability in the speed of cognitive processing. This assumption has been problematic in other domains, due to the confounding effects of caution and motor execution speed on observed response times. We extend a cognitive model of decision-making to account for relatedness structure in a twin study paradigm. This approach can separately quantify different contributions to the heritability of response time. Using data from the Human Connectome Project, we find strong evidence for the heritability of response caution, and more ambiguous evidence for the heritability of cognitive processing speed and motor execution speed. Our study suggests that the assumption made in previous studies¿that the heritability of cognitive ability is based on cognitive processing speed¿may be incorrect. More generally, our methodology provides a useful avenue for future research in complex data that aims to analyze cognitive traits across different sources of related data, whether the relation is between people, tasks, experimental phases, or methods of measurement.

Recently, Veksler, Myers, and Gluck (2015) proposed model flexibility analysis as a method that "aids model evaluation by providing a metric for gauging the persuasiveness of a given fit" (p. 755) Model flexibility analysis measures the complexity of a model in terms of the proportion of all possible data patterns it can predict. We show that this measure does not provide a reliable way to gauge complexity, which prevents model flexibility analysis from fulfilling either of the 2 aims outlined by Veksler et al. (2015): absolute and relative model evaluation. We also show that model flexibility analysis can even fail to correctly quantify complexity in the most clear cut case, with nested models. We advocate for the use of well-established techniques with these characteristics, such as Bayes factors, normalized maximum likelihood, or cross-validation, and against the use of model flexibility analysis. In the discussion, we explore 2 issues relevant to the area of model evaluation: the completeness of current model selection methods and the philosophical debate of absolute versus relative model evaluation.

Organisms making repeated simple decisions are faced with a tradeoff between urgent and cautious strategies. While animals can adopt a statistically optimal policy for this tradeoff, findings about human decision-makers have been mixed. Some studies have shown that people can optimize this "speed¿accuracy tradeoff", while others have identified a systematic bias towards excessive caution. These issues have driven theoretical development and spurred debate about the nature of human decision-making. We investigated a potential resolution to the debate, based on two factors that routinely differ between human and animal studies of decision-making: the effects of practice, and of longer-term feedback. Our study replicated the finding that most people, by default, are overly cautious. When given both practice and detailed feedback, people moved rapidly towards the optimal policy, with many participants reaching optimality with less than 1 h of practice. Our findings have theoretical implications for cognitive and neural models of simple decision-making, as well as methodological implications.

Constant decision-making underpins much of daily life, from simple perceptual decisions about navigation through to more complex decisions about important life events. At many scales, a fundamental task of the decision-maker is to balance competing needs for caution and urgency: fast decisions can be more efficient, but also more often wrong. We show how a single mathematical framework for decision-making explains the urgency/caution balance across decision-making at two very different scales. This explanation has been applied at the level of neuronal circuits (on a time scale of hundreds of milliseconds) through to the level of stable personality traits (time scale of years).

Recent decades have witnessed amazing advances in both mathematical models of cognition and in the field of cognitive neuroscience. These developments were initially independent of one another, but recently the fields have started to become interested in joining forces. The resulting joint modeling of behavioral and neural data can be difficult, but has proved fruitful. We briefly review different approaches used in decision-making research for linking behavioral and neural data, and also provide an example. Our example provides a tight link between behavioral data and evoked scalp potentials measured during mental rotation. The example model illustrates a powerful hypothesis-driven way of linking such data sets. We demonstrate the use of such a model, provide a model comparison against interesting alternatives, and discuss the conclusions that follow from applying such a joint model.

Listeners rely on multiple acoustic cues to recognize any phoneme. The relative contribution of these cues to listeners¿ perception is typically inferred from listeners¿ categorization of sounds in a two-alternative forced-choice task. Here we advocate the use of an evidence accumulation model to analyze categorization as well as response time data from such cue weighting paradigms in terms of the processes that underlie the listeners¿ categorization. We tested 30 Dutch listeners on their categorization of speech sounds that varied between typical /¿/ and /a¿/ in vowel quality (F1 and F2) and duration. Using the linear ballistic accumulator model, we found that the changes in spectral quality and duration lead to changes in the speed of information processing, and the effects were larger for spectral quality. In addition, for stimuli with atypical spectral information, listeners accumulate evidence faster for /¿/ compared to /a¿/. Finally, longer durations of sounds did not produce longer estimates of perceptual encoding time. Our results demonstrate the utility of evidence accumulation models for learning about the latent processes that underlie phoneme categorization. The implications for current theory in speech perception as well as future directions for evidence accumulation models are discussed.

Reasoning and inference are well-studied aspects of basic cognition that have been explained as statistically optimal Bayesian inference. Using a simplified experimental design, we conducted quantitative comparisons between Bayesian inference and human inference at the level of individuals. In 3 experiments, with more than 13,000 participants, we asked people for prior and posterior inferences about the probability that 1 of 2 coins would generate certain outcomes. Most participants' inferences were inconsistent with Bayes' rule. Only in the simplest version of the task did the majority of participants adhere to Bayes' rule, but even in that case, there was a significant proportion that failed to do so. The current results highlight the importance of close quantitative comparisons between Bayesian inference and human data at the individual-subject level when evaluating models of cognition.

Cognitive neuroscientists sometimes apply formal models to investigate how the brain implements cognitive processes. These models describe behavioral data in terms of underlying, latent variables linked to hypothesized cognitive processes. A goal of model-based cognitive neuroscience is to link these variables to brain measurements, which can advance progress in both cognitive and neuroscientific research. However, the details and the philosophical approach for this linking problem can vary greatly. We propose a continuum of approaches that differ in the degree of tight, quantitative, and explicit hypothesizing. We describe this continuum using four points along it, which we dub qualitative structural, qualitative predictive, quantitative predictive, and single model linking approaches. We further illustrate by providing examples from three research fields (decision making, reinforcement learning, and symbolic reasoning) for the different linking approaches.

Functional neuroimaging data indicate the dorsal striatum is engaged when people are required to vary the cautiousness of their decisions, by emphasizing the speed or accuracy of responding in laboratory-based decision tasks. However, the functional contribution of the striatum to decision making is unknown. In the current study we tested patients with focal ischemic lesions of the dorsal striatum and matched non-lesion control participants on a speed-accuracy tradeoff (SAT) task. Analysis using a computational model of response selection in a competitive and time-pressured context indicated that the decisions of patients with striatal lesions were less cautious than those of matched controls. This deficit was most prominent when the accuracy of decisions was emphasized. The results are consistent with the hypothesis that the striatum plays an important role in strategically setting response caution, an essential function for flexible behavior.

Theory development in both psychology and neuroscience can benefit by consideration of both behavioral and neural data sets. However, the development of appropriate methods for linking these data sets is a difficult statistical and conceptual problem. Over the past decades, different linking approaches have been employed in the study of perceptual decision-making, beginning with rudimentary linking of the data sets at a qualitative, structural level, culminating in sophisticated statistical approaches with quantitative links. We outline a new approach, in which a single model is developed that jointly addresses neural and behavioral data. This approach allows for specification and testing of quantitative links between neural and behavioral aspects of the model. Estimating the model in a Bayesian framework allows both data sets to equally inform the estimation of all model parameters. The use of a hierarchical model architecture allows for a model, which accounts for and measures the variability between neurons. We demonstrate the approach by re-analysis of a classic data set containing behavioral recordings of decision-making with accompanying single-cell neural recordings. The joint model is able to capture most aspects of both data sets, and also supports the analysis of interesting questions about prediction, including predicting the times at which responses are made, and the corresponding neural firing rates.

For decades sequential sampling models have successfully accounted for human and monkey decision-making, relying on the standard assumption that decision makers maintain a pre-set decision standard throughout the decision process. Based on the theoretical argument of reward rate maximization, some authors have recently suggested that decision makers become increasingly impatient as time passes and therefore lower their decision standard. Indeed, a number of studies show that computational models with an impatience component provide a good fit to human and monkey decision behavior. However, many of these studies lack quantitative model comparisons and systematic manipulations of rewards. Moreover, the often-cited evidence from single-cell recordings is not unequivocal and complimentary data from human subjects is largely missing. We conclude that, despite some enthusiastic calls for the abandonment of the standard model, the idea of an impatience component has yet to be fully established; we suggest a number of recently developed tools that will help bring the debate to a conclusive settlement.

There is growing interest in diffusion models to represent the cognitive and neural processes of speeded decision making. Sequential-sampling models like the diffusion model have a long history in psychology. They view decision making as a process of noisy accumulation of evidence from a stimulus. The standard model assumes that evidence accumulates at a constant rate during the second or two it takes to make a decision. This process can be linked to the behaviors of populations of neurons and to theories of optimality. Diffusion models have been used successfully in a range of cognitive tasks and as psychometric tools in clinical research to examine individual differences. In this review, we relate the models to both earlier and more recent research in psychology.

For nearly 50 years, the dominant account of decision-making holds that noisy information is accumulated until a fixed threshold is crossed. This account has been tested extensively against behavioral and neurophysiological data for decisions about consumer goods, perceptual stimuli, eyewitness testimony, memories, and dozens of other paradigms, with no systematic misfit between model and data. Recently, the standard model has been challenged by alternative accounts that assume that less evidence is required to trigger a decision as time passes. Such "collapsing boundaries" or "urgency signals" have gained popularity in some theoretical accounts of neurophysiology. Nevertheless, evidence in favor of these models is mixed, with support coming from only a narrow range of decision paradigms compared with a long history of support from dozens of paradigms for the standard theory. We conducted the first large-scale analysis of data from humans and nonhuman primates across three distinct paradigms using powerful model-selection methods to compare evidence for fixed versus collapsing bounds. Overall, we identified evidence in favor of the standard model with fixed decision boundaries. We further found that evidence for static or dynamic response boundaries may depend on specific paradigms or procedures, such as the extent of task practice. We conclude that the difficulty of selecting between collapsing and fixed bounds models has received insufficient attention in previous research, calling into question some previous results.

Trueblood, Brown, and Heathcote (2014) developed a new model, called the multiattribute linear ballistic accumulator (MLBA), to explain contextual preference reversals in multialternative choice. MLBA was shown to provide good accounts of human behavior through both qualitative analyses and quantitative fitting of choice data. Tsetsos, Chater, and Usher (2015) investigated the ability of MLBA to simultaneously capture 3 prominent context effects (attraction, compromise, and similarity). They concluded that MLBA must set a "fine balance" of competing forces to account for all 3 effects simultaneously and that its predictions are sensitive to the position of the stimuli in the attribute space. Through a new experiment, we show that the 3 effects are very fragile and that only a small subset of people shows all 3 simultaneously. Thus, the predictions that Tsetsos et al. generated from the MLBA model turn out to match closely real data in a new experiment. Support for these predictions provides strong evidence for the MLBA. A corollary is that a model that can "robustly" capture all 3 effects simultaneously is not necessarily a good model. Rather, a good model captures patterns found in human data, but cannot accommodate patterns that are not found.

The linear ballistic accumulator model is a theory of decision-making that has been used to analyse data from human and animal experiments. It represents decisions as a race between independent evidence accumulators, and has proven successful in a form assuming a normal distribution for accumulation ("drift") rates. However, this assumption has some limitations, including the corollary that some decision times are negative or undefined. We show that various drift rate distributions with strictly positive support can be substituted for the normal distribution without loss of analytic tractability, provided the candidate distribution has a closed-form expression for its mean when truncated to a closed interval. We illustrate the approach by developing three new linear ballistic accumulation variants, in which the normal distribution for drift rates is replaced by either the lognormal, Fréchet, or gamma distribution. We compare some properties of these new variants to the original normal-rate model.

Feature-based attention has been shown to enhance the responses of neurons tuned to an attended feature while simultaneously suppressing responses of neurons tuned to unattended features. However, the influence of these suppressive neuronal-level modulations on perception is not well understood. Here, we investigated the perceptual consequences of feature-based attention by having subjects judge which of four random dot patterns (RDPs) contained a motion signal (Experiment 1) or which of four RDPs contained the most salient nonrandom motion signal (Experiment 2). Subjects viewed pre-cues which validly, invalidly, or neutrally cued the direction of the target RDP. Behavioral data were fit using the linear ballistic accumulator (LBA) model; the model design that best described the data revealed that the rate of sensory evidence accumulation (drift rate) was highest on valid trials and systematically decreased until the cued direction and the target direction were orthogonal. These results demonstrate behavioral correlates of both feature-based attentional enhancement and suppression.

Rational: Psychostimulant injections in rats have been shown to alter future performance in natural reward conditioning. These effects may represent a persistent impact of drugs on neurocircuits that interface cognitive and motivational processes, which may be further altered in neuropsychiatric conditions that entail increased addiction vulnerability. Objective: This study investigated whether a rat model of schizophrenia with cocaine addiction vulnerability shows altered natural reward conditioning with or without prior cocaine exposure. Methods: Adult rats with SHAM or neonatal ventral hippocampal lesions were given cocaine (15 mg/kg per day for 5 days) or saline injections, followed 7 days later by natural reward-conditioned learning. Over ten daily sessions, water-restricted rats were assessed for durations of head entries into a magazine during random water presentations, a conditioning stimulus phase predictive of the water reward, and an "inappropriate" phase when conditioning stimuli were absent and reward presentation would be delayed. Results: Over repeated sessions, lesioned and SHAM rats showed similar reductions in total magazine entry durations, with similar increases in the allocations of entry times during the water presentation. However, lesioned rats, especially those exposed to cocaine, demonstrated reduced allocations of magazine entry times during the conditioning stimulus phase, and increased allocations during the inappropriate phase. Conclusions: Intact natural reward motivation accompanied by deficient learning of complex contingencies to guide efficient reward approach may represent a form of impulsivity as an addiction vulnerability trait marker in an animal model of schizophrenia. © Springer-Verlag 2004.

Heathcote, Brown, and Mewhort (2002) have introduced a new, robust method of estimating response time distributions. Their method may have practical advantages over conventional maximum likelihood estimation. The basic idea is that the likelihood of parameters is maximized given a few quantiles from the data. We show that Heathcote et al.'s likelihood function is not correct and provide the appropriate correction. However, although our correction stands on firmer theoretical ground than Heathcote et al.'s, it appears to yield worse parameter estimates. This result further indicates that, at least for some distributions and situations, quantile maximum likelihood estimation may have better nonasymptotic properties than a more theoretically justified approach.

We measure the ability of human observers to predict the next datum in a sequence that is generated by a simple statistical process undergoing change at random points in time. Accurate performance in this task requires the identification of change points. We assess individual differences between observers both empirically, and using two kinds of models: a Bayesian approach for change detection and a family of cognitively plausible fast and frugal models. Some individuals detect too many changes and hence perform sub-optimally due to excess variability. Other individuals do not detect enough changes, and perform sub-optimally because they fail to notice short-term temporal trends.