Project code associated with Hu_etal '17 paper on contour integration and border-ownership assignment
The program is written in MATLAB (Mathworks). The code is known to run on R2014a, but should also be compatible with other versions. The main program function is demo.m. Running this program will simulate the model network's response to different input stimuli, corresponding to the contour integration experiments (Chen et al., 2014) and the border-ownership experiments (Qiu et al., 2007). Please be patient as these simulations may take some time. Please also note that our final results are based on averages over multiple simulations, while the demo here just shows results from a single simulation. To reproduce a subset of the figures shown in the paper using the actual data from our simulations, run plot_figs.m. For more details about the model and/or experiments, please see the following references:
@Article{Hu_Niebur17,
Title = {A recurrent neural model of proto-object based contour integration and figure-ground segregation},
Author = {Hu, Brian and Niebur, Ernst},
Journal = {Journal of Computational Neuroscience},
Year = {2017},
Volume = {43},
Number = {3},
Pages = {227--242},
Doi = {10.1007/s10827-017-0659-3},
}
@Article{Chen_etal14,
Title = {Incremental integration of global contours through interplay between visual cortical areas},
Author = {Chen, Minggui and Yan, Yin and Gong, Xiajing and Gilbert, Charles D and Liang, Hualou and Li, Wu},
Journal = {Neuron},
Year = {2014},
Volume = {82},
Number = {3},
Pages = {682--694},
Publisher = {Elsevier},
}
@Article{Qiu_etal07,
Title = {Figure-ground mechanisms provide structure for selective attention},
Author = {F. T. Qiu and T. Sugihara and R. von der Heydt},
Journal = {Nat. Neurosci.},
Year = {2007},
Month = {October},
Number = {11},
Pages = {1492-9},
Volume = {10}
}
The data from our final simulation results can be found in the Results directory. Our paper detailing the recurrent neural model can be found in the resources directory.
Special thanks to Danny Jeck (@dannyjeck) for contributing portions of the code in his re-write of an earlier model that the current work was based upon. If you have any questions, please feel free to contact me at bhu6 (AT) jhmi (DOT) edu.