Using models to correct data: paleodiversity and the fossil record

Despite an enormous philosophical literature on models in science, surprisingly little has been written about data models and how they are constructed. In this paper, I examine the case of how paleodiversity data models are constructed from the fossil data. In particular, I show how paleontologists are using various model-based techniques to correct the data. Drawing on this research, I argue for the following related theses: first, the ‘purity’ of a data model is not a measure of its epistemic reliability. Instead it is the fidelity of the data that matters. Second, the fidelity of a data model in capturing the signal of interest is a matter of degree. Third, the fidelity of a data model can be improved ‘vicariously’, such as through the use of post hoc model-based correction techniques. And, fourth, data models, like theoretical models, should be assessed as adequate (or inadequate) for particular purposes.

     

Creativity and modelling the measurement process of the Higgs self-coupling at the LHC and HL-LHC

This paper provides an account of the nature of creativity in high-energy physics experiments through an integrated historical and philosophical study of the current and planned attempts to measure the self-coupling of the Higgs boson by two experimental collaborations (ATLAS and CMS) at the Large Hadron Collider (LHC) and the planned High Luminosity Large Hadron Collider (HL-LHC). A notion of creativity is first identified broadly as an increase in the epistemic value of a measurement outcome from an unexpected transformation, and narrowly as a condition for knowledge of the measurement of the self-coupling of the Higgs. Drawing upon Tal’s model-based epistemology of measurement (2012) this paper shows how without change to ‘readings’ (or ‘instrument indicators’) a transformation to the model of the measurement process can increase the epistemic value of the measurement outcome. Such transformations are attributed to the creativity of the experimental collaboration. Creativity, in this context, is both a product, a creative and improved model, and the distributed collaborative process of transformation to the model of the measurement process. For the case of the planned measurements at the HL-LHC, where models of the measurement process perform the epistemic function of prediction, creativity is included in the models of the measurement process, both as projected quantified creativity and as an assumed property of the future collaborations.

     

Observation Versus Experiment: An Adequate Framework for Analysing Scientific Experimentation?

Observation and experiment as categories for analysing scientific practice have a long pedigree in writings on science. There has, however, been little attempt to delineate observation and experiment with respect to analysing scientific practice; in particular, scientific experimentation, in a systematic manner. Someone who has presented a systematic account of observation and experiment as categories for analysing scientific experimentation is Ian Hacking. In this paper, I present a detailed analysis of Hacking’s observation versus experiment account. Using a range of cases from various fields of scientific enquiry, I argue that the observation versus experiment account is not an adequate framework for delineating scientific experimentation in a systematic manner.     

Probabilistic models of language processing and acquisition

Probabilistic methods are providing new explanatory approaches to fundamental cognitive science questions of how humans structure, process and acquire language. This review examines probabilistic models defined over traditional symbolic structures. Language comprehension and production involve probabilistic inference in such models; and acquisition involves choosing the best model, given innate constraints and linguistic and other input. Probabilistic models can account for the learning and processing of language, while maintaining the sophistication of symbolic models. A recent burgeoning of theoretical developments and online corpus creation has enabled large models to be tested, revealing probabilistic constraints in processing, undermining acquisition arguments based on a perceived poverty of the stimulus, and suggesting fruitful links with probabilistic theories of categorization and ambiguity resolution in perception.     

对整合模型和占优启发式模型的检验：基于信息加工过程的眼动研究证据

基于信息加工过程视角, 本研究采用眼动技术检验风险决策整合模型和占优启发式模型。结果表明, 自主决策任务条件下决策过程反应时及信息搜索模式均不同于期望价值(EV)迫选任务条件下的决策过程反应时及信息搜索模式; 自主决策任务条件下决策过程反应时并未随着选项间整体值(CPT值)差值的变大而变快, 且基于特征(attribute-based)的信息搜索多于基于选项(option-based)的信息搜索, 不符合整合模型预期。此外, 决策者亦未按照占优启发式模型所假定的决策步骤进行决策。基于信息加工过程的检验结果既不利于以累积预期理论为代表的整合模型, 亦不利于占优启发式模型。文章建议从决策过程视角检验已有决策模型及建立新的启发式决策过程模型(process model)。     

Models and the locus of their truth

If models can be true, where is their truth located? Giere (Explaining science, University of Chicago Press, Chicago, 1998) has suggested an account of theoretical models on which models themselves are not truth-valued. The paper suggests modifying Giere’s account without going all the way to purely pragmatic conceptions of truth—while giving pragmatics a prominent role in modeling and truth-acquisition. The strategy of the paper is to ask: if I want to relocate truth inside models, how do I get it, what else do I need to accept and reject? In particular, what ideas about model and truth do I need? The case used as an illustration is the world’s first economic model, that of von Thünen (1826/1842) on agricultural land use in the highly idealized Isolated State.     

Associative isolation: Unifying associative and list memory

Rather than treating paired associate and serial learning as involving the acquisition of distinct types of information [e.g. Murdock (1974). Human memory: Theory and data. New York: Wiley] I propose an Isolation Principle which treats the two as ends of a continuum. According to this principle, consecutive pairs of items are relatively isolated from other studied items in paired associates learning, but not isolated in serial list learning. The consequence is that variability that dominates forward and backward probed recall is highly correlated in pairs but less so, due to differential interference, in lists. This can explain an important dissociation: whereas forward and backward probes of pairs are nearly perfectly correlated, the correlation is only moderate for serial lists. I demonstrate this in a chaining model by varying item-to-item associative strengths and in a positional coding model by varying the similarity structure of item positions. This enables a range of models to account for data on pairs and lists, as well as potential intermediate or hybrid paradigms, within a single theoretical framework.     

Mathematical models in Newton's Principia: A new view of the ‘Newtonian Style’

In this essay I argue against I. Bernard Cohen's influential account of Newton's methodology in the Principia: the ‘Newtonian Style’. The crux of Cohen's account is the successive adaptation of ‘mental constructs’ through comparisons with nature. In Cohen's view there is a direct dynamic between the mental constructs and physical systems. I argue that his account is essentially hypothetical‐deductive, which is at odds with Newton's rejection of the hypothetical‐deductive method. An adequate account of Newton's methodology needs to show how Newton's method proceeds differently from the hypothetical‐deductive method. In the constructive part I argue for my own account, which is model based: it focuses on how Newton constructed his models in Book I of the Principia. I will show that Newton understood Book I as an exercise in determining the mathematical consequences of certain force functions. The growing complexity of Newton's models is a result of exploring increasingly complex force functions (intra‐theoretical dynamics) rather than a successive comparison with nature (extra‐theoretical dynamics). Nature did not enter the scene here. This intra‐theoretical dynamics is related to the ‘autonomy of the models’.     

A simple computational algorithm of model-based choice preference

A broadly used computational framework posits that two learning systems operate in parallel during the learning of choice preferences—namely, the model-free and model-based reinforcement-learning systems. In this study, we examined another possibility, through which model-free learning is the basic system and model-based information is its modulator. Accordingly, we proposed several modified versions of a temporal-difference learning model to explain the choice-learning process. Using the two-stage decision task developed by Daw, Gershman, Seymour, Dayan, and Dolan (2011), we compared their original computational model, which assumes a parallel learning process, and our proposed models, which assume a sequential learning process. Choice data from 23 participants showed a better fit with the proposed models. More specifically, the proposed eligibility adjustment model, which assumes that the environmental model can weight the degree of the eligibility trace, can explain choices better under both model-free and model-based controls and has a simpler computational algorithm than the original model. In addition, the forgetting learning model and its variation, which assume changes in the values of unchosen actions, substantially improved the fits to the data. Overall, we show that a hybrid computational model best fits the data. The parameters used in this model succeed in capturing individual tendencies with respect to both model use in learning and exploration behavior. This computational model provides novel insights into learning with interacting model-free and model-based components.     

An experimental investigation of the interaction between displays and models in multicriteria decision support

An experiment in multicriteria decision support was conducted that examined two types of information displays, two different multiattribute models and two levels of task complexity in a laboratory setting with students as subjects. The findings focused on the ‘process’ behaviour and the ‘design’ elements of the human-machine interface. Findings concerning the process behaviour suggested that the subjects' actions on data acquisition were guided by their cognitive preference for the attributes, but that the subjects had no discernible search pattern for evaluation. Findings for design included a significant interaction between display and model in weight estimation and assessment. The practical implications for software design are also discussed.     

Hierarchical single- and dual-process models of recognition memory

Recognition memory is commonly modeled as either a single, continuous process within the theory of signal detection, or with two-process models such as Yonelinas’ dual-process model. Previous attempts to determine which model provides a better account of the data have relied on fitting the models to data that are averaged over items. Because such averaging distorts conclusions, we develop and compare hierarchical versions of competing single and dual-process models that account for item variability. The dual-process model provides a superior account of a typical data set when models are compared with the deviance information criterion. Parameters of the dual-process model are highly correlated, however, suggesting that a single-process model may exist that can provide a better account of the data.     

The Psychoanalysis of Narcissism and the Inevitability of “Postmodern” Psychoanalysis

In my exploration of postmodern psychoanalysis, I take as my starting point the complex issue of narcissism as developed by Freud. After pointing out some of the theoretical and clinical difficulties and complexities that arise in the analysis of narcissism, I discuss the diversity of the models and interpretations that then become possible; plurality in psychoanalysis is an inevitable consequence of this. The analysis of narcissism thus opens up many questions that lead me to explore some of the answers that both our forebears and present-day psychoanalysts have suggested.     

IGT-Open: An open-source,computerized version of the Iowa Gambling Task

The Iowa Gambling Task (IGT) is commonly used to understand the processes involved in decision-making. Though the task was originally run without a computer, using a computerized version of the task has become typical. These computerized versions of the IGT are useful, because they can make the task more standardized across studies and allow for the task to be used in environments where a physical version of the task may be difficult or impossible to use (e.g., while collecting brain imaging data). Though these computerized versions of the IGT have been useful for experimentation, having multiple software implementations of the task could present reliability issues. We present an open-source software version of the Iowa Gambling Task (called IGT-Open) that allows for millisecond visual presentation accuracy and is freely available to be used and modified. This software has been used to collect data from human subjects and also has been used to run model-based simulations with computational process models developed to run in the ACT-R architecture.     

Signal-Detection, Threshold, and Dual-Process Models of Recognition Memory: ROCs and Conscious Recollection

Threshold- and signal-detection-based models have dominated theorizing about recognition memory. Building upon these theoretical frameworks, we have argued for a dual-process model in which conscious recollection (a threshold process) and familiarity (a signal-detection process) contribute to memory performance. In the current paper we assessed several memory models by examining the effects of levels of processing and the number of presentations on recognition memory receiver operating characteristics (ROCs). In general, when the ROCs were plotted in probability space they exhibited an inverted U shape; however, when they were plotted inzspace they exhibited a U shape. An examination of the ROCs showed that the dual-process model could account for the observed ROCs, but that models based solely on either threshold or signal-detection processes failed to provide a sufficient account of the data. Furthermore, an examination of subjects' introspective reports using the remember/know procedure showed that subjects were aware of recollection and familiarity and were able to consistently report on their occurrence. The remember/know data were used to accurately predict the shapes of the ROCs, and estimates of recollection and familiarity derived from the ROC data mirrored the subjective reports of these processes.     

Learning the Natural Numbers as a Child

How do we get out knowledge of the natural numbers? Various philosophical accounts exist, but there has been comparatively little attention to psychological data on how the learning process actually takes place. I work through the psychological literature on number acquisition with the aim of characterising the acquisition stages in formal terms. In doing so, I argue that we need a combination of current neologicist accounts and accounts such as that of Parsons. In particular, I argue that we learn the initial segment of the natural numbers on the basis of the Fregean definitions, but do not learn the natural number structure as a whole on the basis of Hume's principle. Therefore, we need to account for some of the consistency of our number concepts with the Dedekind‐Peano axioms in other terms.     

Observing evidence accumulation during multi-alternative decisions

Most decision-making research has focused on choices between two alternatives. For choices between many alternatives, the primary result is Hick’s Law—that mean response time increases logarithmically with the number of alternatives. Various models for this result exist within specific paradigms, and there are some more general theoretical results, but none of those have been tested stringently against data. We present an experimental paradigm that supports detailed examination of multi-choice data, and analyze predictions from a Bayesian ideal observer model for this paradigm. Data from the experiment deviate from the predictions of the Bayesian model in interesting ways. A simple heuristic model based on evidence accumulation provides a good account for the data, and has attractive properties as a limit case of the Bayesian model.     

A test of Bayesian observer models of processing in the Eriksen flanker task

Two Bayesian observer models were recently proposed to account for data from the Eriksen flanker task, in which flanking items interfere with processing of a central target. One model assumes that interference stems from a perceptual bias to process nearby items as if they are compatible, and the other assumes that the interference is due to spatial uncertainty in the visual system (Yu, Dayan, & Cohen, 2009). Both models were shown to produce one aspect of the empirical data, the below-chance dip in accuracy for fast responses to incongruent trials. However, the models had not been fit to the full set of behavioral data from the flanker task, nor had they been contrasted with other models. The present study demonstrates that neither model can account for the behavioral data as well as a comparison spotlight-diffusion model. Both observer models missed key aspects of the data, challenging the validity of their underlying mechanisms. Analysis of a new hybrid model showed that the shortcomings of the observer models stem from their assumptions about visual processing, not the use of a Bayesian decision process.     

Changes in smoking behaviors from late childhood to adolescence: insights from the Canadian National Longitudinal Survey of Children and Youth.

OBJECTIVE: To examine smoking behaviors in Canadian youth from late childhood to adolescence. By following participants from as young as 10 and 11 years, the authors proposed to identify distinct developmental pathways of smoking acquisition. DESIGN: Growth mixture modeling was used to identify developmental trajectories of smoking among 10- to 17-year-old participants of the Canadian National Longitudinal Survey of Children and Youth. MAIN OUTCOME MEASURES: Developmental trajectories of trying smoking, smoking frequency, and smoking intensity. Results: Five developmental trajectories related to smoking frequency were identified, of which 2 were acquisition patterns that led to daily smoking at age 16-17, and 3 were experimentation patterns that led to nonsmoking at age 16-17. The largest variability in changes in smoking behavior over time was the reported level of smoking frequency. CONCLUSION: Analysis showed that there is more than 1 way in which Canadian children and adolescents acquire smoking behaviors over time. The authors were able to differentiate patterns of experimentation from patterns of acquisition. Whereas experimentation has been generally considered as 1 of the stages in the smoking acquisition process leading to regular smoking, these results indicate that experimentation can be described as a distinct process in itself.