Limits...
Using data-driven model-brain mappings to constrain formal models of cognition.

Borst JP, Nijboer M, Taatgen NA, van Rijn H, Anderson JR - PLoS ONE (2015)

Bottom Line: Although such mappings can be based on the experience of the modeler or on a reading of the literature, a formal method is preferred to prevent researcher-based biases.We then validated this mapping by applying it to two new datasets with associated models.The new mapping was at least as powerful as an existing mapping that was based on the literature, and indicated where the models were supported by the data and where they have to be improved.

View Article: PubMed Central - PubMed

Affiliation: Carnegie Mellon University, Dept. of Psychology, Pittsburgh, United States of America; University of Groningen, Dept. of Artificial Intelligence, Groningen, the Netherlands.

ABSTRACT
In this paper we propose a method to create data-driven mappings from components of cognitive models to brain regions. Cognitive models are notoriously hard to evaluate, especially based on behavioral measures alone. Neuroimaging data can provide additional constraints, but this requires a mapping from model components to brain regions. Although such mappings can be based on the experience of the modeler or on a reading of the literature, a formal method is preferred to prevent researcher-based biases. In this paper we used model-based fMRI analysis to create a data-driven model-brain mapping for five modules of the ACT-R cognitive architecture. We then validated this mapping by applying it to two new datasets with associated models. The new mapping was at least as powerful as an existing mapping that was based on the literature, and indicated where the models were supported by the data and where they have to be improved. We conclude that data-driven model-brain mappings can provide strong constraints on cognitive models, and that model-based fMRI is a suitable way to create such mappings.

Show MeSH
Convolving module activity with a hemodynamic response function.(a) HRF, (b) convolving short periods of module activity with the HRF, and (c) example of the problem state module and manual module for four conditions of a multitasking experiment [8]. For illustrative purposes only, similar to Fig. 3 in [72].
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4352055&req=5

pone.0119673.g003: Convolving module activity with a hemodynamic response function.(a) HRF, (b) convolving short periods of module activity with the HRF, and (c) example of the problem state module and manual module for four conditions of a multitasking experiment [8]. For illustrative purposes only, similar to Fig. 3 in [72].

Mentions: Fig. 3 illustrates this process (for more details, including example model code in Lisp and convolution code in Matlab, see [20]). Panel A shows the HRF in response to neural activity at time 0. The HRF increases slowly, and only reaches its peak around 5 seconds after the neural activity. Fig. 3B shows the result of convolving a demand function (gray) with the HRF. The predicted BOLD response depends on the amount and duration of periods of module activity. In essence, for each period of activity an HRF is assumed, which are summed to predict the final response [9]. Fig. 3C shows the results of this process when applied to two modules of a complex model (cf. Fig. 1; [8]). The figure depicts four conditions from left to right, which increase in difficulty. Whereas the manual module predicts a lower BOLD response for the more difficult conditions (because the same amount of key-presses are spaced out over more time), the problem state module predicts a strong increase in activity with task difficulty. These predictions can be compared to fMRI data, and can thus be used to evaluate ACT-R models (we will discuss two examples in this paper). We will refer to this type of analysis as a region-of-interest (ROI) analysis.


Using data-driven model-brain mappings to constrain formal models of cognition.

Borst JP, Nijboer M, Taatgen NA, van Rijn H, Anderson JR - PLoS ONE (2015)

Convolving module activity with a hemodynamic response function.(a) HRF, (b) convolving short periods of module activity with the HRF, and (c) example of the problem state module and manual module for four conditions of a multitasking experiment [8]. For illustrative purposes only, similar to Fig. 3 in [72].
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4352055&req=5

pone.0119673.g003: Convolving module activity with a hemodynamic response function.(a) HRF, (b) convolving short periods of module activity with the HRF, and (c) example of the problem state module and manual module for four conditions of a multitasking experiment [8]. For illustrative purposes only, similar to Fig. 3 in [72].
Mentions: Fig. 3 illustrates this process (for more details, including example model code in Lisp and convolution code in Matlab, see [20]). Panel A shows the HRF in response to neural activity at time 0. The HRF increases slowly, and only reaches its peak around 5 seconds after the neural activity. Fig. 3B shows the result of convolving a demand function (gray) with the HRF. The predicted BOLD response depends on the amount and duration of periods of module activity. In essence, for each period of activity an HRF is assumed, which are summed to predict the final response [9]. Fig. 3C shows the results of this process when applied to two modules of a complex model (cf. Fig. 1; [8]). The figure depicts four conditions from left to right, which increase in difficulty. Whereas the manual module predicts a lower BOLD response for the more difficult conditions (because the same amount of key-presses are spaced out over more time), the problem state module predicts a strong increase in activity with task difficulty. These predictions can be compared to fMRI data, and can thus be used to evaluate ACT-R models (we will discuss two examples in this paper). We will refer to this type of analysis as a region-of-interest (ROI) analysis.

Bottom Line: Although such mappings can be based on the experience of the modeler or on a reading of the literature, a formal method is preferred to prevent researcher-based biases.We then validated this mapping by applying it to two new datasets with associated models.The new mapping was at least as powerful as an existing mapping that was based on the literature, and indicated where the models were supported by the data and where they have to be improved.

View Article: PubMed Central - PubMed

Affiliation: Carnegie Mellon University, Dept. of Psychology, Pittsburgh, United States of America; University of Groningen, Dept. of Artificial Intelligence, Groningen, the Netherlands.

ABSTRACT
In this paper we propose a method to create data-driven mappings from components of cognitive models to brain regions. Cognitive models are notoriously hard to evaluate, especially based on behavioral measures alone. Neuroimaging data can provide additional constraints, but this requires a mapping from model components to brain regions. Although such mappings can be based on the experience of the modeler or on a reading of the literature, a formal method is preferred to prevent researcher-based biases. In this paper we used model-based fMRI analysis to create a data-driven model-brain mapping for five modules of the ACT-R cognitive architecture. We then validated this mapping by applying it to two new datasets with associated models. The new mapping was at least as powerful as an existing mapping that was based on the literature, and indicated where the models were supported by the data and where they have to be improved. We conclude that data-driven model-brain mappings can provide strong constraints on cognitive models, and that model-based fMRI is a suitable way to create such mappings.

Show MeSH