Limits...
Modeling place field activity with hierarchical slow feature analysis.

Schönfeld F, Wiskott L - Front Comput Neurosci (2015)

Bottom Line: Simulations are based on a hierarchical Slow Feature Analysis (SFA) network topped by a principal component analysis (ICA) output layer.The slowness principle is shown to account for the main findings of the presented experimental studies.Future iterations of the model will thus have to incorporate additional information, such as path integration and grid cell activity, in order to be able to also replicate studies that take place during darkness.

View Article: PubMed Central - PubMed

Affiliation: Theory of Neural Systems Group, Institut für Neuroinformatik, Ruhr Universität Bochum Bochum, Germany.

ABSTRACT
What are the computational laws of hippocampal activity? In this paper we argue for the slowness principle as a fundamental processing paradigm behind hippocampal place cell firing. We present six different studies from the experimental literature, performed with real-life rats, that we replicated in computer simulations. Each of the chosen studies allows rodents to develop stable place fields and then examines a distinct property of the established spatial encoding: adaptation to cue relocation and removal; directional dependent firing in the linear track and open field; and morphing and scaling the environment itself. Simulations are based on a hierarchical Slow Feature Analysis (SFA) network topped by a principal component analysis (ICA) output layer. The slowness principle is shown to account for the main findings of the presented experimental studies. The SFA network generates its responses using raw visual input only, which adds to its biological plausibility but requires experiments performed in light conditions. Future iterations of the model will thus have to incorporate additional information, such as path integration and grid cell activity, in order to be able to also replicate studies that take place during darkness.

No MeSH data available.


Related in: MedlinePlus

Left: directional consistency of the artificial cell population during cue rotation. Right: spatial consistency of the population.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4441153&req=5

Figure 7: Left: directional consistency of the artificial cell population during cue rotation. Right: spatial consistency of the population.

Mentions: Figure 6 presents the activity of two representative cells before and after the rotation of a white cue card in an environment that lacks any other distinguishing cues. Results from our computer simulations are shown next to the measurements results of real-life animals as reported in Knierim et al. (1995). In both cases the spatial representation can be seen to follow the rotation of the cue card. Directional consistency values for fixed-direction activity over the arena are clustered closely around 1.0, while all cells but one feature a single distinct field of activity (Figure 7). Since the model is based solely on visual information, this result is not unexpected but does confirm a basic ability of the model to generalize: the cue card has to be recognized via edge and color information rather than by memorizing a distinct pixel pattern. Such patterns depend on location and perspective and change when the cue card is rotated.


Modeling place field activity with hierarchical slow feature analysis.

Schönfeld F, Wiskott L - Front Comput Neurosci (2015)

Left: directional consistency of the artificial cell population during cue rotation. Right: spatial consistency of the population.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4441153&req=5

Figure 7: Left: directional consistency of the artificial cell population during cue rotation. Right: spatial consistency of the population.
Mentions: Figure 6 presents the activity of two representative cells before and after the rotation of a white cue card in an environment that lacks any other distinguishing cues. Results from our computer simulations are shown next to the measurements results of real-life animals as reported in Knierim et al. (1995). In both cases the spatial representation can be seen to follow the rotation of the cue card. Directional consistency values for fixed-direction activity over the arena are clustered closely around 1.0, while all cells but one feature a single distinct field of activity (Figure 7). Since the model is based solely on visual information, this result is not unexpected but does confirm a basic ability of the model to generalize: the cue card has to be recognized via edge and color information rather than by memorizing a distinct pixel pattern. Such patterns depend on location and perspective and change when the cue card is rotated.

Bottom Line: Simulations are based on a hierarchical Slow Feature Analysis (SFA) network topped by a principal component analysis (ICA) output layer.The slowness principle is shown to account for the main findings of the presented experimental studies.Future iterations of the model will thus have to incorporate additional information, such as path integration and grid cell activity, in order to be able to also replicate studies that take place during darkness.

View Article: PubMed Central - PubMed

Affiliation: Theory of Neural Systems Group, Institut für Neuroinformatik, Ruhr Universität Bochum Bochum, Germany.

ABSTRACT
What are the computational laws of hippocampal activity? In this paper we argue for the slowness principle as a fundamental processing paradigm behind hippocampal place cell firing. We present six different studies from the experimental literature, performed with real-life rats, that we replicated in computer simulations. Each of the chosen studies allows rodents to develop stable place fields and then examines a distinct property of the established spatial encoding: adaptation to cue relocation and removal; directional dependent firing in the linear track and open field; and morphing and scaling the environment itself. Simulations are based on a hierarchical Slow Feature Analysis (SFA) network topped by a principal component analysis (ICA) output layer. The slowness principle is shown to account for the main findings of the presented experimental studies. The SFA network generates its responses using raw visual input only, which adds to its biological plausibility but requires experiments performed in light conditions. Future iterations of the model will thus have to incorporate additional information, such as path integration and grid cell activity, in order to be able to also replicate studies that take place during darkness.

No MeSH data available.


Related in: MedlinePlus