My current research explores the growth of immediate memory capacity across age, and capacity in young adults. I’m interested in testing immediate memory performance in participants at different ages using different classes of stimuli to show that memory spans for different stimulus classes improve similarly. My current work suggests that chunking abilities do not improve with age, and that improvements in immediate memory instead reflect a true increase in the number of chunks available for encoding materials. Current findings also suggest that chunking is a core ability in children, and that differences in immediate memory span for materials from different stimulus classes are robust across development. If you want to join the team, some opportunities are listed occasionally in the Research Interests menu under the Research grants section. Feel free to contact me !
The ability to learn sequences depends on different factors governing sequence structure, such as transitional probability (TP, probability of a stimulus given a previous stimulus), adjacent or nonadjacent dependency, and frequency. Current evidence indicates that adjacent and nonadjacent pairs are not equally learnable; the same applies to second-order and first-order TPs and to the frequency of the sequences. However, the relative importance of these factors and interactive effects on learning remain poorly understood. The first experiment tested the effects of TPs and dependency separately on the learning of nonlinguistic visual sequences, and the second experiment used the factors of the first experiment and added a frequency factor to test their interactive effects with verbal sequences of stimuli (pseudo-words). The results of both experiments showed higher performance during online learning for first-order TPs in adjacent pairs. Moreover, Experiment 2 indicated poorer performance during offline recall for nonadjacent dependencies and low-frequency sequences. We discuss the results that different factors are not used equally in prediction and memorization.
This study simultaneously manipulates within-category (rule-based vs. similarity-based), between-category (blocked vs. interleaved), and across-blocks (constant vs. variable) orders to investigate how different types of presentation order interact with one another. With regard to within-category orders, stimuli were presented either in a “rule plus exceptions” fashion (in the rule-based order) or by maximizing the similarity between contiguous examples (in the similarity-based order). As for the between-category manipulation, categories were either blocked (in the blocked order) or alternated (in the interleaved order). Finally, the sequence of stimuli was either repeated (in the constant order) or varied (in the variable order) across blocks. This research offers a novel approach through both an individual and concurrent analysis of the studied factors, with the investigation of across-blocks manipulations being unprecedented. We found a significant interaction between within-category and across-blocks orders, as well as between between-category and across-blocks orders. In particular, the combination similarity-based + variable orders was the most detrimental, whereas the combination blocked + constant was the most beneficial. We also found a main effect of across-blocks manipulation, with faster learning in the constant order as compared to the variable one. With regard to the classification of novel stimuli, learners in the rule-based and interleaved orders showed generalization patterns that were more consistent with a specific rule-based strategy, as compared to learners in the similarity-based and blocked orders, respectively. This study shows that different types of order can interact in a subtle fashion and thus should not be considered in isolation.
In this article, we develop a new general inference method for selecting learning models. The method relies upon a specific hold-out cross-validation, which takes into account the dependency within the data. This allows us to retrieve the model that best fits the learning strategy of a single individual. The novelty of our approach lies on the choice of the testing set, both in the experimental design and in the data analysis. This individual approach is then applied to two category learning models (ALCOVE and Component-cue) on data-sets manipulating presentation order, after verification of the reliability of our method. We found that both models performed equally well during transfer, but Component-cue best fits the majority of participants during learning. To further analyze these models, we also investigated a potential relation between the underlying mechanisms of the models and the actual types of presentation order assigned to participants.
International Conference on Interdisciplinary Advances in Statistical Learning, Jun 2022, San Sebastian, Spain
annee_publi
2022
resume
The ability to learn transitional probabilities (TPs) and frequency is central to language processing. Current evidence indicates that both frequency and transitional probability are involved in the memorization of sequences, but the questions of which prevails and why it would prevail in statistical learning remain unclear. The present study investigated the respective roles of transitional probability and frequency in statistical learning of pseudowords in two different tasks that focused on either prediction or recognition. The learning phase consisted of the repeated presentation of sixteen three-syllable pseudowords for which participants were asked to perform a target detection task on vowels (fully predictable based on TPs). The evolution of the rate of correct answer and response times during the learning phase was recorded. After the detection task, a two-alternative forced-choice task (2AFC) required participants to choose between a pseudoword and a lure. Results indicated a prevalence of TPs during the detection task, but a prevalence of frequency during the 2AFC task. Our findings suggests that TPs and frequency can be used flexibly depending on which process (learning or recognition) is more adapted to the task.
The extraction of cooccurrences between two events, A and B, is a central learning mechanism shared by all species capable of associative learning. Formally, the cooccurrence of events A and B appearing in a sequence is measured by the transitional probability (TP) between these events, and it corresponds to the probability of the second stimulus given the first (i.e., p(B|A)). In the present study, nonhuman primates (Guinea baboons, Papio papio) were exposed to a serial version of the XOR (i.e., exclusive-OR), in which they had to process sequences of three stimuli: A, B, and C. In this manipulation, first-order TPs (i.e., AB and BC) were uninformative due to their transitional probabilities being equal to .5 (i.e., p(B|A) = p(C|B) = .5), while secondorder TPs were fully predictive of the upcoming stimulus (i.e., p(C|AB) = 1). In Experiment 1, we found that baboons were able to learn second-order TPs, while no learning occurred on first-order TPs. In Experiment 2, this pattern of results was replicated, and a final test ruled out an alternative interpretation in terms of proximity to the reward. These results indicate that a non-human primate species can learn a nonlinearly separable problem such as the XOR. They also provide fine-grained empirical data to test models of statistical learning on the interaction between the learning of different orders of TPs. Recent bioinspired models of associative learning are also introduced as promising alternatives to the modeling of statistical learning mechanisms.
Most categorization models are insensitive to the order in which stimuli are presented. However, a vast array of studies have shown that the sequence received during learning can influence how categories are formed. In this paper, the objective was to better account for effects of serial order. We developed a model called Ordinal General Context Model (OGCM) based on the Generalized Context Model (GCM), which we modified to incorporate ordinal information. OGCM incorporates serial order as a feature along ordinary physical features, allowing it to account for the effect of sequential order as a form of distortion of the feature space. The comparison between the models showed that integrating serial order during learning in the OGCM provided the best account of classification of the stimuli in our data-sets.
62nd Virtual annual meeting of the Psychonomic Society, Nov 2021, Virtuel, United States
annee_publi
2021
resume
The order of stimuli within sequences and the transitional probabilities (TPs) these orders generate are central information in language acquisition, but less is known about how this type of information is extracted by general learning mechanisms. The present study focused on the statistical learning of second-order TPs (i.e., only the combination of two stimuli allowing to predict the third) of visual sequences. Eight three-item sequences exclusively governed by second-order TPs were presented. The response times were measured with oculometry. The task included a learning phase and a switch phase which reset the second-order TPs (e.g., the sequences ABC and BAF became respectively ABF and BAC). Results indicated a sole decrease of RTs between the second and the third stimulus and an increase of RTs during the switch phase that suggested the learnability of second-order TPs. We discuss this result in light of language acquisition.
Data compression in memory is a cognitive process allowing participants to cope with complexity to reduce information load. However, previous studies have not yet considered the hypothesis that this process could also lead to over-simplifying information due to haphazard amplification of the compression process itself. For instance, we could expect that the over-regularized features of a visual scene could produce false recognition of patterns, not because of storage capacity limits but because of an errant compression process. To prompt memory compression in our participants, we used multielement visual displays for which the underlying information varied in compressibility. The compressibility of our material could vary depending on the number of common features between the multi-dimensional objects in the displays. We measured both accuracy and response times by probing memory representations with probes that we hypothesized could modify the participants’ representations. We confirm that more compressible information facilitates performance, but a more novel finding is that compression can produce both typical memory errors and lengthened response times. Our findings provide clearer evidence of the forms of compression that participants carry out.