Incremental implicit learning of bundles of statistical patterns

Using position rather than color at the traffic light - Covariation learning-based deviation from instructions in attention deficit/hyperactivity disorder

Based on instructions people can form task representations that shield relevant from seemingly irrelevant information. It has been documented that instructions can tie people to a particular way of performing a task despite that in principle a more efficient way could be learned and used. Since task shielding can lead to persistence of inefficient variants of task performance, it is relevant to test whether individuals with attention deficit/hyperactivity disorder (ADHD) - characterized by less task shielding - are more likely and quicker to escape a suboptimal instructed variant of performing a task. The paradigm used in this online study builds on the observation that in many environments different covarying features could be used to determine the appropriate response. For instance, as they approach a traffic light, drivers and pedestrians monitor the color (instructed stimulus feature) and/or the position of the signal (covarying stimulus feature, more efficient in case of reduced color sight). Similarly, we instructed participants to respond to the color of a stimulus without mentioning that color covaried with the position of the stimulus. In order to assess whether with practice participants would use the non-instructed feature position to an increasing extent, we compared reaction times and error rates for standard trials to trials in which color was either ambiguous or did not match the usual covariation. Results showed that the covariation learning task can be administered online to adult participants with and without ADHD. Performance differences suggested that with practice ADHD participants (n = 43 out of a total N = 245) might increase attention to non-instructed stimulus features. Yet, they used the non-instructed covarying stimulus feature to a similar extent as other participants. Together the results suggest that participants with ADHD do not lag behind in abandoning instructed task processing in favor of a learned alternative strategy.

Using Rational Models to Interpret the Results of Experiments on Accent Adaptation

Exposure to unfamiliar non-native speech tends to improve comprehension. One hypothesis holds that listeners adapt to non-native-accented speech through distributional learning—by inferring the statistics of the talker's phonetic cues. Models based on this hypothesis provide a good fit to incremental changes after exposure to atypical native speech. These models have, however, not previously been applied to non-native accents, which typically differ from native speech in many dimensions. Motivated by a seeming failure to replicate a well-replicated finding from accent adaptation, we use ideal observers to test whether our results can be understood solely based on the statistics of the relevant cue distributions in the native- and non-native-accented speech. The simple computational model we use for this purpose can be used predictively by other researchers working on similar questions. All code and data are shared.

Children's Learning of Non-adjacent Dependencies Using a Web-Based Computer Game Setting

Infants show impressive speech decoding abilities and detect acoustic regularities that highlight the syntactic relations of a language, often coded via non-adjacent dependencies (NADs, e.g., is singing). It has been claimed that infants learn NADs implicitly and associatively through passive listening and that there is a shift from effortless associative learning to a more controlled learning of NADs after the age of 2 years, potentially driven by the maturation of the prefrontal cortex. To investigate if older children are able to learn NADs, Lammertink et al. (2019) recently developed a word-monitoring serial reaction time (SRT) task and could show that 6-11-year-old children learned the NADs, as their reaction times (RTs) increased then they were presented with violated NADs. In the current study we adapted their experimental paradigm and tested NAD learning in a younger group of 52 children between the age of 4-8 years in a remote, web-based, game-like setting (whack-a-mole). Children were exposed to Italian phrases containing NADs and had to monitor the occurrence of a target syllable, which was the second element of the NAD. After exposure, children did a "Stem Completion" task in which they were presented with the first element of the NAD and had to choose the second element of the NAD to complete the stimuli. Our findings show that, despite large variability in the data, children aged 4-8 years are sensitive to NADs; they show the expected differences in r RTs in the SRT task and could transfer the NAD-rule in the Stem Completion task. We discuss these results with respect to the development of NAD dependency learning in childhood and the practical impact and limitations of collecting these data in a web-based setting.

Statistical learning of multiple speech streams: A challenge for monolingual infants

Language acquisition depends on the ability to detect and track the distributional properties of speech. Successful acquisition also necessitates detecting changes in those properties, which can occur when the learner encounters different speakers, topics, dialects, or languages. When encountering multiple speech streams with different underlying statistics but overlapping features, how do infants keep track of the properties of each speech stream separately? In four experiments, we tested whether 8-month-old monolingual infants (N = 144) can track the underlying statistics of two artificial speech streams that share a portion of their syllables. We first presented each stream individually. We then presented the two speech streams in sequence, without contextual cues signaling the different speech streams, and subsequently added pitch and accent cues to help learners track each stream separately. The results reveal that monolingual infants experience difficulty tracking the statistical regularities in two speech streams presented sequentially, even when provided with contextual cues intended to facilitate separation of the speech streams. We discuss the implications of our findings for understanding how infants learn and separate the input when confronted with multiple statistical structures.

A Tale of Two Positivities and the N400: Distinct Neural Signatures Are Evoked by Confirmed and Violated Predictions at Different Levels of Representation

It has been proposed that hierarchical prediction is a fundamental computational principle underlying neurocognitive processing. Here, we ask whether the brain engages distinct neurocognitive mechanisms in response to inputs that fulfill versus violate strong predictions at different levels of representation during language comprehension. Participants read three-sentence scenarios in which the third sentence constrained for a broad event structure, for example, {Agent caution animate-Patient}. High constraint contexts additionally constrained for a specific event/lexical item, for example, a two-sentence context about a beach, lifeguards, and sharks constrained for the event, {Lifeguards cautioned Swimmers}, and the specific lexical item swimmers. Low constraint contexts did not constrain for any specific event/lexical item. We measured ERPs on critical nouns that fulfilled and/or violated each of these constraints. We found clear, dissociable effects to fulfilled semantic predictions (a reduced N400), to event/lexical prediction violations (an increased late frontal positivity), and to event structure/animacy prediction violations (an increased late posterior positivity/P600). We argue that the late frontal positivity reflects a large change in activity associated with successfully updating the comprehender's current situation model with new unpredicted information. We suggest that the late posterior positivity/P600 is triggered when the comprehender detects a conflict between the input and her model of the communicator and communicative environment. This leads to an initial failure to incorporate the unpredicted input into the situation model, which may be followed by second-pass attempts to make sense of the discourse through reanalysis, repair, or reinterpretation. Together, these findings provide strong evidence that confirmed and violated predictions at different levels of representation manifest as distinct spatiotemporal neural signatures.

A Tale of Two Positivities and the N400: Distinct Neural Signatures Are Evoked by Confirmed and Violated Predictions at Different Levels of Representation

It has been proposed that hierarchical prediction is a fundamental computational principle underlying neurocognitive processing. Here, we ask whether the brain engages distinct neurocognitive mechanisms in response to inputs that fulfill versus violate strong predictions at different levels of representation during language comprehension. Participants read three-sentence scenarios in which the third sentence constrained for a broad event structure, for example, {Agent caution animate-Patient}. High constraint contexts additionally constrained for a specific event/lexical item, for example, a two-sentence context about a beach, lifeguards, and sharks constrained for the event, {Lifeguards cautioned Swimmers}, and the specific lexical item swimmers. Low constraint contexts did not constrain for any specific event/lexical item. We measured ERPs on critical nouns that fulfilled and/or violated each of these constraints. We found clear, dissociable effects to fulfilled semantic predictions (a reduced N400), to event/lexical prediction violations (an increased late frontal positivity), and to event structure/animacy prediction violations (an increased late posterior positivity/P600). We argue that the late frontal positivity reflects a large change in activity associated with successfully updating the comprehender's current situation model with new unpredicted information. We suggest that the late posterior positivity/P600 is triggered when the comprehender detects a conflict between the input and her model of the communicator and communicative environment. This leads to an initial failure to incorporate the unpredicted input into the situation model, which may be followed by second-pass attempts to make sense of the discourse through reanalysis, repair, or reinterpretation. Together, these findings provide strong evidence that confirmed and violated predictions at different levels of representation manifest as distinct spatiotemporal neural signatures.

Neural evidence for Bayesian trial-by-trial adaptation on the N400 during semantic priming

When semantic information is activated by a context prior to new bottom-up input (i.e. when a word is predicted), semantic processing of that incoming word is typically facilitated, attenuating the amplitude of the N400 event related potential (ERP) - a direct neural measure of semantic processing. N400 modulation is observed even when the context is a single semantically related "prime" word. This so-called "N400 semantic priming effect" is sensitive to the probability of encountering a related prime-target pair within an experimental block, suggesting that participants may be adapting the strength of their predictions to the predictive validity of their broader experimental environment. We formalize this adaptation using a Bayesian learning model that estimates and updates the probability of encountering a related versus an unrelated prime-target pair on each successive trial. We found that our model's trial-by-trial estimates of target word probability accounted for significant variance in trial-by-trial N400 amplitude. These findings suggest that Bayesian principles contribute to how comprehenders adapt their semantic predictions to the statistical structure of their broader environment, with implications for the functional significance of the N400 component and the predictive nature of language processing.

Probability Learning in an Uncertain World: How Children Adjust to Changing Contingencies

We regularly make predictions about future events, even in a world where events occur probabilistically rather than deterministically. Our environment may even be non-stationary such that the probability of an event may change suddenly or from one context to another. 4-6 year olds and adults viewed 3 boxes and guessed the location of a hidden toy. After 80 trials with one set of probabilities assigned to the 3 boxes, the spatial distribution of these probabilities was altered. Adults easily responded to this change, with participants who maximized in the first half (by choosing the most common location at a higher rate than it was presented) being the fastest at making this shift. Only the older children successfully switched to the new location, with younger children either partially switching, perseverating on their original strategy, or failing to learn the first distribution, suggesting a fundamental development in children's response to changing probabilities.

Individual differences in learning social and nonsocial network structures

How do people acquire knowledge about which individuals belong to different cliques or communities? And to what extent does this learning process differ from the process of learning higher-order information about complex associations between nonsocial bits of information? Here, the authors use a paradigm in which the order of stimulus presentation forms temporal associations between the stimuli, collectively constituting a complex network. They examined individual differences in the ability to learn community structure of networks composed of social versus nonsocial stimuli. Although participants were able to learn community structure of both social and nonsocial networks, their performance in social network learning was uncorrelated with their performance in nonsocial network learning. In addition, social traits, including social orientation and perspective-taking, uniquely predicted the learning of social community structure but not the learning of nonsocial community structure. Taken together, the results suggest that the process of learning higher-order community structure in social networks is partially distinct from the process of learning higher-order community structure in nonsocial networks. The study design provides a promising approach to identify neurophysiological drivers of social network versus nonsocial network learning, extending knowledge about the impact of individual differences on these learning processes. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

Incidental covariation learning leading to strategy change

As they approach a traffic light, drivers and pedestrians monitor the color (instructed stimulus feature) and/or the position of the signal (covarying stimulus feature) for response selection. Many studies have pointed out that instructions can effectively determine the stimulus features used for response selection in a task. This leaves open whether and how practice with a correlating alternative stimulus feature can lead to a strategy change from an instructed to a learned variant of performing the task. To address this question, we instructed participants to respond to the position of a stimulus within a reference frame, at the same time, during task performance, an unmentioned second stimulus feature, the color, covaried with stimulus position and allowed the use of an alternative response strategy. To assess the impact of the non-instructed stimulus feature of color on response selection throughout practice, the spatial position of the stimulus was ambiguous on some trials. Group average increases in color usage were based on a mixture of (1) participants who, despite extended practice on the covariation, exclusively relied on the instructed stimulus feature and (2) those who abruptly started to rely heavily on stimulus color to select responses in ambiguous trials. When the instructed and uninstructed feature predicted different actions, choices were still biased by the uninstructed color feature, albeit more weakly. A second experiment showed that the influence of color generalized across frequently and infrequently presented combinations of position and color. Strategy changes were accompanied by awareness in both experiments. The results suggest that incidental covariation learning can trigger spontaneous voluntary strategy change involving a re-configuration of the instructed task set.

What Mechanisms Underlie Implicit Statistical Learning? Transitional Probabilities Versus Chunks in Language Learning

In a prior review, Perrruchet and Pacton (2006) noted that the literature on implicit learning and the more recent studies on statistical learning focused on the same phenomena, namely the domain-general learning mechanisms acting in incidental, unsupervised learning situations. However, they also noted that implicit learning and statistical learning research favored different interpretations, focusing on the selection of chunks and the computation of transitional probabilities aimed at discovering chunk boundaries, respectively. This paper examines the state of the debate 12 years later. The link between contrasting theories and their historical roots has disappeared, but a number of studies were aimed at contrasting the predictions of these two approaches. Overall, these studies strongly question the still prevalent account based on the statistical computation of pairwise associations. Various chunk-based models provide much better predictions in a number of experimental situations. However, these models rely on very different conceptual frameworks, as illustrated by a comparison between Bayesian models of word segmentation, PARSER, and a connectionist model (TRACX).

Synesthetes perseverate in implicit learning: Evidence from a non-stationary statistical learning task

Synesthetes automatically and consistently experience additional sensory or cognitive perceptions in response to particular environmental stimuli. Recent evidence suggests that the propensity to develop synesthesia is genetic while the particular associations experienced by a given synesthete are influenced by learning. Despite the potential role of implicit learning in the formation of synesthetic associations, there has been minimal investigation of synesthetes' implicit learning abilities. In this study, we examine linguistic-colour synesthetes' ability to implicitly learn from and adjust to non-stationary statistics in a domain unrelated to their particular form of synesthesia. Engaging participants in a computer game Whack-the-mole, we utilise the online measure of reaction time to assess the time course of learning. Participants are exposed to "worlds" of probabilities that, unbeknownst to them, undergo unannounced changes, creating unpredictable statistical shifts devoid of accompanying cues. The same small set of probability worlds are repeated throughout the experiment to investigate participants' ability to retain and learn from this repetitive probabilistic information. The reaction time data provide evidence that synesthetes require more information than nonsynesthetes to benefit from the non-stationary probability distributions. These findings demonstrate that linguistic-colour synesthetes' implicit learning abilities-in a domain far from their synesthetic experiences-differ from those of nonsynesthetes.

Sociophonetics: The Role of Words, the Role of Context, and the Role of Words in Context

This paper synthesizes a wide range of literature from sociolinguistics and cognitive psychology, to argue for a central role for the "word" as a vehicle of language variation and change. Three crucially interlinked strands of research are reviewed-the role of context in associative learning, the word-level storage of phonetic and contextual detail, and the phonetic consequences of skewed distributions of words across different contexts. I argue that the human capacity for associative learning, combined with attention to fine-phonetic detail at the level of the word, plays a significant role in predicting a range of subtle but systematically robust observed socioindexical patterns in speech production and perception.

Re-examining selective adaptation: Fatiguing feature detectors, or distributional learning?

When a listener hears many good examples of a /b/ in a row, they are less likely to classify other sounds on, e.g., a /b/-to-/d/ continuum as /b/. This phenomenon is known as selective adaptation and is a well-studied property of speech perception. Traditionally, selective adaptation is seen as a mechanistic property of the speech perception system, and attributed to fatigue in acoustic-phonetic feature detectors. However, recent developments in our understanding of non-linguistic sensory adaptation and higher-level adaptive plasticity in speech perception and language comprehension suggest that it is time to re-visit the phenomenon of selective adaptation. We argue that selective adaptation is better thought of as a computational property of the speech perception system. Drawing on a common thread in recent work on both non-linguistic sensory adaptation and plasticity in language comprehension, we furthermore propose that selective adaptation can be seen as a consequence of distributional learning across multiple levels of representation. This proposal opens up new questions for research on selective adaptation itself, and also suggests that selective adaptation can be an important bridge between work on adaptation in low-level sensory systems and the complicated plasticity of the adult language comprehension system.

Robust speech perception: recognize the familiar, generalize to the similar, and adapt to the novel

Successful speech perception requires that listeners map the acoustic signal to linguistic categories. These mappings are not only probabilistic, but change depending on the situation. For example, one talker's /p/ might be physically indistinguishable from another talker's /b/ (cf. lack of invariance). We characterize the computational problem posed by such a subjectively nonstationary world and propose that the speech perception system overcomes this challenge by (a) recognizing previously encountered situations, (b) generalizing to other situations based on previous similar experience, and (c) adapting to novel situations. We formalize this proposal in the ideal adapter framework: (a) to (c) can be understood as inference under uncertainty about the appropriate generative model for the current talker, thereby facilitating robust speech perception despite the lack of invariance. We focus on 2 critical aspects of the ideal adapter. First, in situations that clearly deviate from previous experience, listeners need to adapt. We develop a distributional (belief-updating) learning model of incremental adaptation. The model provides a good fit against known and novel phonetic adaptation data, including perceptual recalibration and selective adaptation. Second, robust speech recognition requires that listeners learn to represent the structured component of cross-situation variability in the speech signal. We discuss how these 2 aspects of the ideal adapter provide a unifying explanation for adaptation, talker-specificity, and generalization across talkers and groups of talkers (e.g., accents and dialects). The ideal adapter provides a guiding framework for future investigations into speech perception and adaptation, and more broadly language comprehension.