Phillip M. Alday

Towards a Computational Model of Actor-Based Language Comprehension

Neurophysiological data from a range of typologically diverse languages provide evidence for a cross-linguistically valid, actor-based strategy of understanding sentence-level meaning. This strategy seeks to identify the participant primarily responsible for the state of affairs (the actor) as quickly and unambiguously as possible, thus resulting in competition for the actor role when there are multiple candidates. Due to its applicability across languages with vastly different characteristics, we have proposed that the actor strategy may derive from more basic cognitive or neurobiological organizational principles, though it is also shaped by distributional properties of the linguistic input (e.g. the morphosyntactic coding strategies for actors in a given language). Here, we describe an initial computational model of the actor strategy and how it interacts with language-specific properties. Specifically, we contrast two distance metrics derived from the output of the computational model (one weighted and one unweighted) as potential measures of the degree of competition for actorhood by testing how well they predict modulations of electrophysiological activity engendered by language processing. To this end, we present an EEG study on word order processing in German and use linear mixed-effects models to assess the effect of the various distance metrics. Our results show that a weighted metric, which takes into account the weighting of an actor-identifying feature in the language under consideration outperforms an unweighted distance measure. We conclude that actor competition effects cannot be reduced to feature overlap between multiple sentence participants and thereby to the notion of similarity-based interference, which is prominent in current memory-based models of language processing. Finally, we argue that, in addition to illuminating the underlying neurocognitive mechanisms of actor competition, the present model can form the basis for a more comprehensive, neurobiologically plausible computational model of constructing sentence-level meaning.

A common misapplication of statistical inference: Nuisance control with null-hypothesis significance tests.

Experimental research on behavior and cognition frequently rests on stimulus or subject selection where not all characteristics can be fully controlled, even when attempting strict matching. For example, when contrasting patients to controls, variables such as intelligence or socioeconomic status are often correlated with patient status. Similarly, when presenting word stimuli, variables such as word frequency are often correlated with primary variables of interest. One procedure very commonly employed to control for such nuisance effects is conducting inferential tests on confounding stimulus or subject characteristics. For example, if word length is not significantly different for two stimulus sets, they are considered as matched for word length. Such a test has high error rates and is conceptually misguided. It reflects a common misunderstanding of statistical tests: interpreting significance not to refer to inference about a particular population parameter, but about 1. the sample in question, 2. the practical relevance of a sample difference (so that a nonsignificant test is taken to indicate evidence for the absence of relevant differences). We show inferential testing for assessing nuisance effects to be inappropriate both pragmatically and philosophically, present a survey showing its high prevalence, and briefly discuss an alternative in the form of regression including nuisance variables.

The Decision Decoding ToolBOX (DDTBOX) -- A Multivariate Pattern Analysis Toolbox for Event-Related Potentials

In recent years, neuroimaging research in cognitive neuroscience has increasingly used multivariate pattern analysis (MVPA) to investigate higher cognitive functions. Here we present DDTBOX, an open-source MVPA toolbox for electroencephalography (EEG) data. DDTBOX runs under MATLAB and is well integrated with the EEGLAB/ERPLAB and Fieldtrip toolboxes (Delorme and Makeig 2004; Lopez-Calderon and Luck 2014; Oostenveld et al. 2011). It trains support vector machines (SVMs) on patterns of event-related potential (ERP) amplitude data, following or preceding an event of interest, for classification or regression of experimental variables. These amplitude patterns can be extracted across space/electrodes (spatial decoding), time (temporal decoding), or both (spatiotemporal decoding). DDTBOX can also extract SVM feature weights, generate empirical chance distributions based on shuffled-labels decoding for group-level statistical testing, provide estimates of the prevalence of decodable information in the population, and perform a variety of corrections for multiple comparisons. It also includes plotting functions for single subject and group results. DDTBOX complements conventional analyses of ERP components, as subtle multivariate patterns can be detected that would be overlooked in standard analyses. It further allows for a more explorative search for information when no ERP component is known to be specifically linked to a cognitive process of interest. In summary, DDTBOX is an easy-to-use and open-source toolbox that allows for characterising the time-course of information related to various perceptual and cognitive processes. It can be applied to data from a large number of experimental paradigms and could therefore be a valuable tool for the neuroimaging community.

Age-Related Changes in Predictive Capacity Versus Internal Model Adaptability: Electrophysiological Evidence that Individual Differences Outweigh Effects of Age

Hierarchical predictive coding has been identified as a possible unifying principle of brain function, and recent work in cognitive neuroscience has examined how it may be affected by age–related changes. Using language comprehension as a test case, the present study aimed to dissociate age-related changes in prediction generation versus internal model adaptation following a prediction error. Event-related brain potentials (ERPs) were measured in a group of older adults (60–81 years; n = 40) as they read sentences of the form “The opposite of black is white/yellow/nice.” Replicating previous work in young adults, results showed a target-related P300 for the expected antonym (“white”; an effect assumed to reflect a prediction match), and a graded N400 effect for the two incongruous conditions (i.e. a larger N400 amplitude for the incongruous continuation not related to the expected antonym, “nice,” versus the incongruous associated condition, “yellow”). These effects were followed by a late positivity, again with a larger amplitude in the incongruous non-associated versus incongruous associated condition. Analyses using linear mixed-effects models showed that the target-related P300 effect and the N400 effect for the incongruous non-associated condition were both modulated by age, thus suggesting that age-related changes affect both prediction generation and model adaptation. However, effects of age were outweighed by the interindividual variability of ERP responses, as reflected in the high proportion of variance captured by the inclusion of by-condition random slopes for participants and items. We thus argue that – at both a neurophysiological and a functional level – the notion of general differences between language processing in young and older adults may only be of limited use, and that future research should seek to better understand the causes of interindividual variability in the ERP responses of older adults and its relation to cognitive performance.

Electrophysiology Reveals the Neural Dynamics of Naturalistic Auditory Language Processing: Event-Related Potentials Reflect Continuous Model Updates

Abstract The recent trend away from ANOVA-based analyses places experimental investigations into the neurobiology of cognition in more naturalistic and ecologically valid designs within reach. Using mixed-effects models for epoch-based regression, we demonstrate the feasibility of examining event-related potentials (ERPs), and in particular the N400, to study the neural dynamics of human auditory language processing in a naturalistic setting. Despite the large variability between trials during naturalistic stimulation, we replicated previous findings from the literature: the effects of frequency, animacy, and word order and find previously unexplored interaction effects. This suggests a new perspective on ERPs, namely, as a continuous modulation reflecting continuous stimulation instead of a series of discrete and essentially sequential processes locked to discrete events.

A modality-independent, neurobiological grounding for the combinatory capacity of the language-ready brain: Comment on “Towards a Computational Comparative Neuroprimatology: Framing the language-ready brain” by Michael A. Arbib

In this comprehensive review of his past and current work on language evolution, Arbib [1] argues that “the capability for protosign – rather than elaborations intrinsic to the core vocalization systems – may [...] have provided the essential scaffolding for protospeech and evolution of the human language-ready brain” (p. 25). He hypothesises that this evolutionary trajectory is based on the mirror system and mechanisms of complex imitation that developed by drawing on systems “beyond the mirror”. As Arbib himself discusses in detail, the claim that gestural combinatorics of increasing complexity and symbolisation formed a prerequisite for the evolution of auditory speech and language is rather controversial. Though, in our own previous work, we have emphasised the importance of the computational properties of the auditory system in defining the language-ready brain [2], we would like to focus on a somewhat different, and perhaps even more foundational issue for the purposes of this commentary: are there basic neurobiological mechanisms that underlie combinatory processing irrespective of modality? The answer to this question may be surprisingly simple. Internal (forward and inverse) models, currently one of the “hot topics” in cognitive neuroscience (e.g. [3]), indeed provide a basic, biologically plausible and – perhaps even more importantly – unifying mechanism for combinatory processing. Originally proposed for the domain of motor control and sensorimotor integration [4], internal models can serve as models of the body and its surrounding environment [5], and they have been used to account for non-motor processing in domains such as one’s own sense of agency [6], social interaction [7], theory of mind [8] and language [2,9,10]. The common denominator between forward models in all of these areas is that they serve to encode predictions about the environment that can be compared against sensory information, thereby either confirming the model or leading to its adaptation. This type of mechanism provides a natural foundation for combinatory processing, as it (a) implements the (predictive) processing of previously encountered sequences; and (b) allows for the interpretation of novel sequences, by enabling an input item

Be Careful When Assuming the Obvious: Commentary on “The Placement of the Head that Minimizes Online Memory: A Complex Systems Approach”

Ferrer-i-Cancho (2015) presents a mathematical model of both the synchronic and diachronic nature of word order based on the assumption that memory costs are a never decreasing function of distance and a few very general linguistic assumptions. However, even these minimal and seemingly obvious assumptions are not as safe as they appear in light of recent typological and psycholinguistic evidence. The interaction of word order and memory has further depths to be explored.

The role of phonotactic principles in language processing

ABSTRACT The paper reports the results of a learnability experiment with German speakers, investigating the role of universal phonotactic constraints and language use in language processing. Making use of an artificial language paradigm, participants learned nonce words with existent and non-existent German final consonant clusters adhering to or violating sonority sequencing principles postulated for consonant clusters. Behavioural data and event-related brain potentials in response to these cluster properties were obtained twice, before and after learning word-picture-pairs. The results show (1) that learning and processing of final consonant clusters is facilitated by adherence to the sonority hierarchy, and (2) that actual existence of well- and ill-formed consonant clusters aids processing mechanisms. Thus, both implicit knowledge of universal phonotactic principles and frequency-based factors are demonstrated to play a role in the online-processing of words.

Toward a reliable, automated method of individual alpha frequency (IAF) quantification

Individual alpha frequency (IAF) is a promising electrophysiological marker of interindividual differences in cognitive function. IAF has been linked with trait-like differences in information processing and general intelligence, and provides an empirical basis for the definition of individualized frequency bands. Despite its widespread application, however, there is little consensus on the optimal method for estimating IAF, and many common approaches are prone to bias and inconsistency. Here, we describe an automated strategy for deriving two of the most prevalent IAF estimators in the literature: peak alpha frequency (PAF) and center of gravity (CoG). These indices are calculated from resting-state power spectra that have been smoothed using a Savitzky-Golay filter (SGF). We evaluate the performance characteristics of this analysis procedure in both empirical and simulated EEG data sets. Applying the SGF technique to resting-state data from n = 63 healthy adults furnished 61 PAF and 62 CoG estimates. The statistical properties of these estimates were consistent with previous reports. Simulation analyses revealed that the SGF routine was able to reliably extract target alpha components, even under relatively noisy spectral conditions. The routine consistently outperformed a simpler method of automated peak detection that did not involve spectral smoothing. The SGF technique is fast, open source, and available in two popular programming languages (MATLAB, Python), and thus can easily be integrated within the most popular M/EEG toolsets (EEGLAB, FieldTrip, MNE-Python). As such, it affords a convenient tool for improving the reliability and replicability of future IAF-related research.

Commentary on Sanborn and Chater: Posterior Modes Are Attractor Basins

Sanborn and Chater [1] propose an interesting theory of cognitive and brain function based on Bayesian sampling instead of asymptotic Bayesian inference. Their proposal unifies many current observations and models and, despite focusing primarily on cognitive phenomena, their work provides a springboard for unifying several proposed theories of brain function. It has the potential to serve as a bridge between three influential overarching current theories of cognitive and brain function: Bayesian models, Friston’s [2–4] theory of cortical responses based on the free-energy principle, and attractor-basin dynamics [5,6].