Continuous dynamics in behavior reveal perceptual nonlinearities aid speech categorization in noise

Spoken language comprehension requires listeners map continuous features of the speech signal to discrete category labels. Categories are however malleable to surrounding context; listeners’ percept can dynamically shift depending on the sequencing of adjacent stimuli resulting in a warping of the heard phonetic category (i.e., hysteresis). Here, we investigated whether such perceptual nonlinearities—which amplify categorical hearing—might further aid speech processing in noise-degraded listening scenarios. We measured continuous dynamics in perception and category judgments of an acoustic-phonetic vowel gradient via mouse tracking. Tokens were presented in serial vs. random orders to induce more/less perceptual warping while listeners categorized continua in clean and noise conditions. Listeners’ responses were faster and their mouse trajectories closer to the ultimate behavioral selection (marked visually on the screen) in serial vs. random order, suggesting increased perceptual attraction to category exemplars. Interestingly, order effects emerged earlier and persisted later in the trial time course when categorizing speech in noise. These data describe a new functional benefit of perceptual nonlinearities to speech perception yet undocumented: warping strengthens the behavioral attraction to relevant speech categories while simultaneously assisting perception in degraded acoustic environments.

The ubiquity of selective attention in the processing of feedback during category learning

Feedback is essential for many kinds of learning, but the cognitive processes involved in learning from feedback are unclear. Models of category learning incorporate selective attention to stimulus features while generating a response, but during the feedback phase of an experiment, it is assumed that participants receive complete information about stimulus features as well as the correct category. The present work looks at eye tracking data from six category learning datasets covering a variety of category complexities and types. We find that selective attention to task-relevant information is pervasive throughout feedback processing, suggesting a role for selective attention in memory encoding of category exemplars. We also find that error trials elicit additional stimulus processing during the feedback phase. Finally, our data reveal that participants increasingly skip the processing of feedback altogether. At the broadest level, these three findings reveal that selective attention is ubiquitous throughout the entire category learning task, functioning to emphasize the importance of certain stimulus features, the helpfulness of extra stimulus encoding during times of uncertainty, and the superfluousness of feedback once one has learned the task. We discuss the implications of our findings for modelling efforts in category learning from the perspective of researchers trying to capture the full dynamic interaction of selective attention and learning, as well as for researchers focused on other issues, such as category representation, whose work only requires simplifications that do a reasonable job of capturing learning.

role of f0 alignment in distinguishing intonation categories: evidence from American english

Under the autosegmental-metrical (AM) theory of intonation, the temporal alignment of fundamental frequency (F0) patterns with respect to syllables has been claimed to distinguish pitch accent categories. Several experiments test whether differences in F0 peak or valley alignment in American English phrases would produce evidence consistent with a change from (1) a H* to a H+L* pitch accent, and (2) a L* to a L+H* pitch accent. Four stimulus series were constructed in which F0 peak or valley alignment was shifted across portions of short phrases with varying stress. In Experiment 1, participants discriminated pairs of stimuli in an AX task. In Experiment 2, participants classified stimuli as category exemplars using an AXB task. In Experiment 3, participants imitated stimuli; the alignment of F0 peaks and valleys in their productions was measured. Finally, in Experiment 4, participants judged the relative prominence of initial and final syllables in stimuli to determine whether alignment differences generated a stress shift. The results support the distinctions between H* and H+L* and between L+H* and L*. Moreover, evidence consistent with an additional category not currently predicted by most AM theories was obtained, which is proposed here to be H*+H. The results have implications for understanding phonological contrasts, phonetic interpolation in English intonation, and the transcription of prosodic contrasts in corpus-based analysis.

A neural network model of the effect of prior experience with regularities on subsequent category learning

Categories are often structured by the similarities of instances within the category. A popular dual systems theory of category learning argues that the structure of exemplars forming categories determines the mechanisms that drive learning. Category distributions are necessarily defined by dimensions or features. Researchers typically assume that there is a direct, linear relationship between the physical input dimensions across which category exemplars are defined and the psychological representation of these dimensions, but this assumption is not always warranted. Through a set of simulations, we demonstrate that the psychological representations of input dimensions developed through prior experience can place drastic constraints on category learning. We compare the model’s behavior to auditory, visual, and cross-modal human category learning and make conclusions regarding the nature of the psychological representations of the dimensions in those studies. These simulations support the conclusion that the nature of psychological representations is a critical aspect to understanding the mechanisms underlying category learning.

Trait anxiety is associated with reduced typicality asymmetry in fear generalization

In fear conditioning, training with typical category exemplars has been shown to promote fear generalization to novel exemplars of the same category, whereas training with atypical category exemplars supports limited if any generalization to other category members, amounting to a typicality asymmetry in fear generalization. The present study sought to examine how trait anxiety bears on typicality asymmetry in fear generalization. Participants in one condition were presented with typical exemplars during fear acquisition and atypical exemplars of the same category in the subsequent generalization test (typical condition), whereas in the other group, atypical and typical exemplars were presented during fear acquisition and generalization test, respectively (atypical condition). We observed a typicality asymmetry in fear generalization in self-reported expectancy ratings in low trait anxious individuals only. High trait anxious individuals showed a similar degree of fear generalization in both conditions. The current results help illuminate why some individuals are at risk for exhibiting broad fear generalization after exposure to an aversive event.

Exemplar learning reveals the representational origins of expert category perception

Irrespective of whether one has substantial perceptual expertise for a class of stimuli, an observer invariably encounters novel exemplars from this class. To understand how novel exemplars are represented, we examined the extent to which previous experience with a category constrains the acquisition and nature of representation of subsequent exemplars from that category. Participants completed a perceptual training paradigm with either novel other-race faces (category of experience) or novel computer-generated objects (YUFOs) that included pairwise similarity ratings at the beginning, middle, and end of training, and a 20-d visual search training task on a subset of category exemplars. Analyses of pairwise similarity ratings revealed multiple dissociations between the representational spaces for those learning faces and those learning YUFOs. First, representational distance changes were more selective for faces than YUFOs; trained faces exhibited greater magnitude in representational distance change relative to untrained faces, whereas this trained–untrained distance change was much smaller for YUFOs. Second, there was a difference in where the representational distance changes were observed; for faces, representations that were closer together before training exhibited a greater distance change relative to those that were farther apart before training. For YUFOs, however, the distance changes occurred more uniformly across representational space. Last, there was a decrease in dimensionality of the representational space after training on YUFOs, but not after training on faces. Together, these findings demonstrate how previous category experience governs representational patterns of exemplar learning as well as the underlying dimensionality of the representational space.

Category norms with a cross-sectional sample of adults in the United States: Consideration of cohort, age, and historical effects on semantic categories

This paper describes normative data for newly collected exemplar responses to 70 semantic categories described in previous norming studies (Battig & Montague, 1969; Van Overschelde, Rawson, & Dunlosky, 2004; Yoon et al., 2004). These categories were presented to 246 Young (18 – 39 years), Middle (40 – 59 years), and Older (60 years and older) English-speaking adults living in the United States who were asked to generate as many category exemplars as possible for each of the 70 categories. In order to understand differences in normative responses, we analyzed these responses a) between age groups within the current sample and b) in comparison to three previously-published sets of norms. Experimental studies using such norms typically assume invariance of normative likelihoods across age and historical time. We replicate previous findings such that exemplar frequency correlations suggest moderate stability in generated category members between age groups and cohorts for many, but not all, categories. Further, analyses of rank order correlations highlight that the traditional measure of typicality may not capture all aspects of typicality, namely that for some categories there is high consistency in the frequency of exemplars across age groups and/or norms, but the ordering of those exemplars differs significantly. We include a cluster analysis to aid in grouping categories based on relative stability across time, cohort, and age groups. These results emphasize the importance of maintaining and updating age-differentiated category norms.

Individual differences among deep neural network models

Deep neural networks (DNNs) excel at visual recognition tasks and are increasingly used as a modelling framework for neural computations in the primate brain. However, each DNN instance, just like each individual brain, has a unique connectivity and representational profile. Here, we investigate individual differences among DNN instances that arise from varying only the random initialization of the network weights. Using representational similarity analysis, we demonstrate that this minimal change in initial conditions prior to training leads to substantial differences in intermediate and higher-level network representations, despite achieving indistinguishable network-level classification performance. We locate the origins of the effects in an under-constrained alignment of category exemplars, rather than a misalignment of category centroids. Furthermore, while network regularization can increase the consistency of learned representations, considerable differences remain. These results suggest that computational neuroscientists working with DNNs should base their inferences on multiple networks instances instead of single off-the-shelf networks.