The paper presents a novel formal model of the active buffer of working memory. The model uses sy... more The paper presents a novel formal model of the active buffer of working memory. The model uses synchronic oscillations in order to bind an item and its corresponding context into one representation, while asynchronic oscillations allow the model to maintain several separate representations. Due to bindings, the model exerts proper control over the buffer's contents, as demonstrated by effective rejection of distractors. Most importantly, the model predicts an inherent limitation in WM capacity that arises from the trade-off between the number versus the stability of representations bound by oscillations, which depends on the strength of lateral inhibition present among oscillating items. The systematic variation in inhibition leads to exact replication of the capacity distribution observed in a large sample of participants, as well as to prediction of a few novel, capacity-related experimental effects.
Using a biologically plausible computational model of working memory, we investigated the neuroco... more Using a biologically plausible computational model of working memory, we investigated the neurocognitive mechanisms of finding a simple relation, namely whether syllables located in one row, column, or diagonal line of the matrix of stimuli end with the identical vowel. Though simple, such a task is however a very important tool for understanding of how people cope with various reasoning tasks, because it predicts individual differences in human reasoning in an excellent way. The common mechanism for both finding relations and reasoning may consist of integration of relations due to processing of multiple asynchronous oscillations. It was shown that the distribution of individual ability to integrate relations depends on the level of control over model's internal inhibition, which determines the number of oscillations the model uses to encode complex relational structures.
This study tested the dual-component model of working memory (WM) against its unitary alternative... more This study tested the dual-component model of working memory (WM) against its unitary alternative. The former account predicts that WM consists of two functionally distinct mechanisms: a very accessible but capacity-limited primary memory (PM) and a less accessible secondary memory (SM). The latter account assumes only one long-term memory component. We used a novel version of the Sternberg serial recognition paradigm, which selectively impedes access to either early or late items, by asking participants about the location of a probe in relation to either the end or the start of encoded memory set, respectively. When locations matched probes, our manipulation harmed recognition of early items, while it left late items intact, in the case of both latency and accuracy. However, in trials in which locations did not match probes, such an effect regarded only latency but not accuracy. This result suggests that a way of access to WM may depend on the level of conflict among accessed memor...
Previous research on the effect of response set size on the interference resulting from processin... more Previous research on the effect of response set size on the interference resulting from processing in the Stroop task – a paradigmatic test of executive control – brought equivocal conclusions. In the present paper, we analyze what predictions regarding this effect should be drawn from the most influential computational models of the Stroop as well as from our own new model. Then we test these predictions in an experiment, by manipulating response set size as well as the stimuli/response set sizes proportion, finding both evidence in favor of our model and data which is not explained by any Stroop model.
Existing computational models differ in predictions concerning the set size effect in the Stroop ... more Existing computational models differ in predictions concerning the set size effect in the Stroop task, which is the relation between a number of stimuli/responses and the magnitude of interference. In some studies, no set size effects were observed, while in other substantial set size effects were present. We administered two experiments in order to resolve this discrepancy in the case of manual Stroop task. Experiment 1 compared conditions including four, six, and eight stimuli/response mappings in the picture-word Stroop. No reliable set size effects have been found, apart from a weak effect observed when working memory load imposed by the task was deliberately decreased. Experiment 2 tested conditions consisting of four versus eight mappings in the color-word Stroop, and it has replicated results of Experiment 1. As both experiments had sufficient power to detect set size effects if they existed, our data are inconsistent with models which predict such effects.
This paper presents a preliminary work on a new architecture of executive control (DUCCA), aimed ... more This paper presents a preliminary work on a new architecture of executive control (DUCCA), aimed at integration and extension of some leading approaches to executive control. We present DUCCA assumptions and operation and use the architecture to simulate a few effects observed in Stroop-like task, a hallmark test of how control deals with interference. The focus of DUCCA is on how strategical use of general executive mechanisms contributes to Stroop effect. We explain also what is usually neglected in Stroop modeling: the significant individual differences in task performance.
In this commentary, I will focus on a few key issues and priorities for future research on indivi... more In this commentary, I will focus on a few key issues and priorities for future research on individual differences in attention. More specifically, I will consider the following: (a) the kinds of tasks that are most relevant for assessing attentional and executive processes; (b) attentional biases and cognitive performance; and (c) the most relevant dimensions of individual differences.
The concept of "intelligence" has evolved in order to account for two facts, namely, intraindivid... more The concept of "intelligence" has evolved in order to account for two facts, namely, intraindividual stability and interindividual variability of human intellectual performance. On one hand, people who outperform others in one class of tasks that involve reasoning, abstracting, or learning, will most probably excel in any other class of such tasks. On the other hand, within any class of cognitive tasks, one can find people who perform in an outstanding way as well as ones who fail. Early studies on the structure of intelligence examined if there is one general ability factor that manifests itself in all cognitive activities (g factor; Spearman, 1927) or maybe more domain-specific factors (e.g., linguistic, mathematical, etc., Thurstone, 1938) exist. These studies have converged to the widely accepted proposal of the three-layer hierarchy of factors, with general ability (g) on its highest level, several subordinate group factors (loading groups of tasks like mnemonic or perceptual tasks) on the middle level, and many low-level specific factors for particular tasks (e.g., one for perceptual speed). The most important middle-level factor is general fluid intelligence (Gf; Cattell, 1971), which represents human ability to adapt to the novelty and complexity by means of discovery of abstract relations in the environment and by their efficient goal-oriented application. In this respect, Gf differs from crystallized intelligence (Gc factor; ibidem), which consists in the application of one's existing knowledge to the requirements of a situation.
We present a novel computational model of the active buffer of working memory (WM). The model use... more We present a novel computational model of the active buffer of working memory (WM). The model uses synchronous oscillations in order to bind an item and its corresponding context into one representation, while asynchronous oscillations are used to separate the representations. Due to the bindings, the model can ascribe proper meanings to items, as demonstrated by the replication of the effective rejection of distractors. The model predicts the inherent limitation of WM capacity in range of 1 to around six items that arises from the trade-off between the number and stability of separate oscillations. This trade-off depends on the strength of lateral inhibition exerted. The systematic variation in inhibition led to the exact replication of capacity distribution observed in a large sample, as well as to the prediction of a few novel capacity-related experimental effects. Finally, we showed that the differences in capacity can underlie the differences in a more complex ability of detecting relations governing a pattern of stimuli, called relational integration, which is known to be strongly related to the effectiveness of higher-order cognitive processing.
Modeling of individual or group differences, believed to be a powerful test for computational mod... more Modeling of individual or group differences, believed to be a powerful test for computational models, is still rare in current cognitive science. In this paper, we discuss alternative approaches to the computational modeling of both qualitative and quantitative differences among individuals as well as groups of individuals. Then, an example is presented of how accounting for individual differences in short term memory (STM) search can bring us insight into cognitive processes underlying this phenomenon, insight that otherways would be impossible. The two-phase computational model of memory search implements the idea of working memory (WM) focus of attention (FA): due to updating process a few items may be actively kept and easily accessed in ACT-R goal buffer. FA is being scanned serially first, and if the scan result is negative, a parallel chunk retrieval from active part of declarative memory outside the FA may run with certain probability. The model aptly simulates steep decrease in accuracy as well as steep increase in latency for responses to five most recent stimuli. The model also predicts the observed effect of faster negative responses than positive responses to less recent stimuli. Most important, with manipulation to only one of its parameters (i.e., the capacity of FA) our model is able to predict 94% of variance for two groups of participants that differed in latency patterns (i.e., 'serial-like' vs. 'parallel-like' groups) of the search process.
We examined whether fluid intelligence (Gf) is better predicted by the storage capacity of active... more We examined whether fluid intelligence (Gf) is better predicted by the storage capacity of active memory or by the effectiveness of executive control. In two psychometric studies, we measured storage capacity with three kinds of task which required the maintenance of a visual array, the monitoring of simple relations among perceptually available stimuli, or the quick update of information. Executive control was measured with tasks reflecting three executive functions, namely attention control, interference resolution, and response inhibition. Using structural equation modeling, we found that all storage tasks loaded on one latent variable, which predicted on average 70% of variance in Gf . On the contrary, neither interference resolution nor response inhibition was substantially related to Gf or to any other variable (Study 1). Although attention control predicted on average 25% of Gf variance (Studies 1 and 2), when storage capacity was statistically controlled for, attention control no longer significantly explained Gf.
The paper presents a novel formal model of the active buffer of working memory. The model uses sy... more The paper presents a novel formal model of the active buffer of working memory. The model uses synchronic oscillations in order to bind an item and its corresponding context into one representation, while asynchronic oscillations allow the model to maintain several separate representations. Due to bindings, the model exerts proper control over the buffer's contents, as demonstrated by effective rejection of distractors. Most importantly, the model predicts an inherent limitation in WM capacity that arises from the trade-off between the number versus the stability of representations bound by oscillations, which depends on the strength of lateral inhibition present among oscillating items. The systematic variation in inhibition leads to exact replication of the capacity distribution observed in a large sample of participants, as well as to prediction of a few novel, capacity-related experimental effects.
Using a biologically plausible computational model of working memory, we investigated the neuroco... more Using a biologically plausible computational model of working memory, we investigated the neurocognitive mechanisms of finding a simple relation, namely whether syllables located in one row, column, or diagonal line of the matrix of stimuli end with the identical vowel. Though simple, such a task is however a very important tool for understanding of how people cope with various reasoning tasks, because it predicts individual differences in human reasoning in an excellent way. The common mechanism for both finding relations and reasoning may consist of integration of relations due to processing of multiple asynchronous oscillations. It was shown that the distribution of individual ability to integrate relations depends on the level of control over model's internal inhibition, which determines the number of oscillations the model uses to encode complex relational structures.
This study tested the dual-component model of working memory (WM) against its unitary alternative... more This study tested the dual-component model of working memory (WM) against its unitary alternative. The former account predicts that WM consists of two functionally distinct mechanisms: a very accessible but capacity-limited primary memory (PM) and a less accessible secondary memory (SM). The latter account assumes only one long-term memory component. We used a novel version of the Sternberg serial recognition paradigm, which selectively impedes access to either early or late items, by asking participants about the location of a probe in relation to either the end or the start of encoded memory set, respectively. When locations matched probes, our manipulation harmed recognition of early items, while it left late items intact, in the case of both latency and accuracy. However, in trials in which locations did not match probes, such an effect regarded only latency but not accuracy. This result suggests that a way of access to WM may depend on the level of conflict among accessed memor...
Previous research on the effect of response set size on the interference resulting from processin... more Previous research on the effect of response set size on the interference resulting from processing in the Stroop task – a paradigmatic test of executive control – brought equivocal conclusions. In the present paper, we analyze what predictions regarding this effect should be drawn from the most influential computational models of the Stroop as well as from our own new model. Then we test these predictions in an experiment, by manipulating response set size as well as the stimuli/response set sizes proportion, finding both evidence in favor of our model and data which is not explained by any Stroop model.
Existing computational models differ in predictions concerning the set size effect in the Stroop ... more Existing computational models differ in predictions concerning the set size effect in the Stroop task, which is the relation between a number of stimuli/responses and the magnitude of interference. In some studies, no set size effects were observed, while in other substantial set size effects were present. We administered two experiments in order to resolve this discrepancy in the case of manual Stroop task. Experiment 1 compared conditions including four, six, and eight stimuli/response mappings in the picture-word Stroop. No reliable set size effects have been found, apart from a weak effect observed when working memory load imposed by the task was deliberately decreased. Experiment 2 tested conditions consisting of four versus eight mappings in the color-word Stroop, and it has replicated results of Experiment 1. As both experiments had sufficient power to detect set size effects if they existed, our data are inconsistent with models which predict such effects.
This paper presents a preliminary work on a new architecture of executive control (DUCCA), aimed ... more This paper presents a preliminary work on a new architecture of executive control (DUCCA), aimed at integration and extension of some leading approaches to executive control. We present DUCCA assumptions and operation and use the architecture to simulate a few effects observed in Stroop-like task, a hallmark test of how control deals with interference. The focus of DUCCA is on how strategical use of general executive mechanisms contributes to Stroop effect. We explain also what is usually neglected in Stroop modeling: the significant individual differences in task performance.
In this commentary, I will focus on a few key issues and priorities for future research on indivi... more In this commentary, I will focus on a few key issues and priorities for future research on individual differences in attention. More specifically, I will consider the following: (a) the kinds of tasks that are most relevant for assessing attentional and executive processes; (b) attentional biases and cognitive performance; and (c) the most relevant dimensions of individual differences.
The concept of "intelligence" has evolved in order to account for two facts, namely, intraindivid... more The concept of "intelligence" has evolved in order to account for two facts, namely, intraindividual stability and interindividual variability of human intellectual performance. On one hand, people who outperform others in one class of tasks that involve reasoning, abstracting, or learning, will most probably excel in any other class of such tasks. On the other hand, within any class of cognitive tasks, one can find people who perform in an outstanding way as well as ones who fail. Early studies on the structure of intelligence examined if there is one general ability factor that manifests itself in all cognitive activities (g factor; Spearman, 1927) or maybe more domain-specific factors (e.g., linguistic, mathematical, etc., Thurstone, 1938) exist. These studies have converged to the widely accepted proposal of the three-layer hierarchy of factors, with general ability (g) on its highest level, several subordinate group factors (loading groups of tasks like mnemonic or perceptual tasks) on the middle level, and many low-level specific factors for particular tasks (e.g., one for perceptual speed). The most important middle-level factor is general fluid intelligence (Gf; Cattell, 1971), which represents human ability to adapt to the novelty and complexity by means of discovery of abstract relations in the environment and by their efficient goal-oriented application. In this respect, Gf differs from crystallized intelligence (Gc factor; ibidem), which consists in the application of one's existing knowledge to the requirements of a situation.
We present a novel computational model of the active buffer of working memory (WM). The model use... more We present a novel computational model of the active buffer of working memory (WM). The model uses synchronous oscillations in order to bind an item and its corresponding context into one representation, while asynchronous oscillations are used to separate the representations. Due to the bindings, the model can ascribe proper meanings to items, as demonstrated by the replication of the effective rejection of distractors. The model predicts the inherent limitation of WM capacity in range of 1 to around six items that arises from the trade-off between the number and stability of separate oscillations. This trade-off depends on the strength of lateral inhibition exerted. The systematic variation in inhibition led to the exact replication of capacity distribution observed in a large sample, as well as to the prediction of a few novel capacity-related experimental effects. Finally, we showed that the differences in capacity can underlie the differences in a more complex ability of detecting relations governing a pattern of stimuli, called relational integration, which is known to be strongly related to the effectiveness of higher-order cognitive processing.
Modeling of individual or group differences, believed to be a powerful test for computational mod... more Modeling of individual or group differences, believed to be a powerful test for computational models, is still rare in current cognitive science. In this paper, we discuss alternative approaches to the computational modeling of both qualitative and quantitative differences among individuals as well as groups of individuals. Then, an example is presented of how accounting for individual differences in short term memory (STM) search can bring us insight into cognitive processes underlying this phenomenon, insight that otherways would be impossible. The two-phase computational model of memory search implements the idea of working memory (WM) focus of attention (FA): due to updating process a few items may be actively kept and easily accessed in ACT-R goal buffer. FA is being scanned serially first, and if the scan result is negative, a parallel chunk retrieval from active part of declarative memory outside the FA may run with certain probability. The model aptly simulates steep decrease in accuracy as well as steep increase in latency for responses to five most recent stimuli. The model also predicts the observed effect of faster negative responses than positive responses to less recent stimuli. Most important, with manipulation to only one of its parameters (i.e., the capacity of FA) our model is able to predict 94% of variance for two groups of participants that differed in latency patterns (i.e., 'serial-like' vs. 'parallel-like' groups) of the search process.
We examined whether fluid intelligence (Gf) is better predicted by the storage capacity of active... more We examined whether fluid intelligence (Gf) is better predicted by the storage capacity of active memory or by the effectiveness of executive control. In two psychometric studies, we measured storage capacity with three kinds of task which required the maintenance of a visual array, the monitoring of simple relations among perceptually available stimuli, or the quick update of information. Executive control was measured with tasks reflecting three executive functions, namely attention control, interference resolution, and response inhibition. Using structural equation modeling, we found that all storage tasks loaded on one latent variable, which predicted on average 70% of variance in Gf . On the contrary, neither interference resolution nor response inhibition was substantially related to Gf or to any other variable (Study 1). Although attention control predicted on average 25% of Gf variance (Studies 1 and 2), when storage capacity was statistically controlled for, attention control no longer significantly explained Gf.
Uploads
Papers by Adam Chuderski