Receptive fields

Prfpy contains a set of simple functions that define receptive field structures. These are used by Gridder objects to calculate the receptive fields.

prfpy.rf.gauss1D_cart(x, mu=0.0, sigma=1.0)

gauss1D_cart takes a 1D array x, a mean and standard deviation, and produces a gaussian with given parameters, with a peak of height 1.

xnumpy.ndarray (1D)

space on which to calculate the gauss

mufloat, optional

mean/mode of gaussian (the default is 0.0)

sigmafloat, optional

standard deviation of gaussian (the default is 1.0)

numpy.ndarray

gaussian values at x

prfpy.rf.gauss1D_log(x, mu=0.0, sigma=1.0)

gauss1D_log takes a 1D array x, a mean and standard deviation, and produces a pRF with given parameters with the distance between mean and x log-scaled

xnumpy.ndarray (1D)

space on which to calculate the gauss

mufloat, optional

mean/mode of gaussian (the default is 0.0)

sigmafloat, optional

standard deviation of gaussian (the default is 1.0)

numpy.ndarray

gaussian values at log(x)

prfpy.rf.gauss2D_iso_cart(x, y, mu=(0.0, 0.0), sigma=1.0, normalize_RFs=False)

gauss2D_iso_cart takes two-dimensional arrays x and y, containing the x and y coordinates at which to evaluate the 2D isotropic gaussian function, with a given sigma, and returns a 2D array of Z values.

xnumpy.ndarray, 2D or flattened by masking

2D, containing x coordinates

ynumpy.ndarray, 2D or flattened by masking

2D, containing y coordinates

mutuple, optional

mean, 2D coordinates of mean/mode of gauss (the default is (0.0,0.0))

sigmafloat, optional

standard deviation of gauss (the default is 1.0)

numpy.ndarray, 2D or flattened by masking

gaussian values evaluated at (x,y)

prfpy.rf.gauss2D_logpolar(ecc, polar, mu=(1.0, 0.0), sigma=1.0, kappa=1.0)

gauss2D_logpolar takes two-dimensional arrays ecc and polar, containing the eccentricity and polar angle coordinates at which to evaluate the 2D gaussian, which in this case is a von Mises in the polar angle direction, and a log gauss in the eccentricity dimension, and returns a 2D array of Z values. We recommend entering the ecc and polar angles ordered as x and y for easy visualization.

eccnumpy.ndarray, 2D or flattened by masking

2D, containing eccentricity

polarnumpy.ndarray, 2D or flattened by masking

2D, containing polar angle coordinates (0, 2*np.pi)

mutuple, optional

mean, 2D coordinates of mean/mode of gauss (ecc) and von Mises (polar) (the default is (0.0,0.0))

sigmafloat, optional

standard deviation of gauss (the default is 1.0)

kappafloat, optional

dispersion coefficient of the von Mises, akin to inverse of standard deviation of gaussian (the default is 1.0)

numpy.ndarray, 2D or flattened by masking

values evaluated at (ecc, polar), peak has y-value of 1.

prfpy.rf.gauss2D_rot_cart(x, y, mu=(0.0, 0.0), sigma=1.0, theta=0.0, ar=1.0)

gauss2D_rot_cart takes two-dimensional arrays x and y, containing the x and y coordinates at which to evaluate the 2D non-isotropic gaussian function, with a given sigma, angle of rotation theta, and aspect ratio ar. it returns a 2D array of Z values. Default is an isotropic gauss.

xnumpy.ndarray, 2D

2D, containing x coordinates or flattened by masking

ynumpy.ndarray, 2D

2D, containing y coordinates or flattened by masking

mutuple, optional

mean, 2D coordinates of mean/mode of gauss (the default is (0.0,0.0))

sigmafloat, optional

standard deviation of gauss (the default is 1.0)

thetafloat, optional

angle of rotation of gauss (the default is 0.0)

arfloat, optional

aspect ratio of gauss, multiplies the rotated y parameters (the default is 1.0)

numpy.ndarray, 2D or flattened by masking

gaussian values evaluated at (x,y)

prfpy.rf.vonMises1D(x, mu=0.0, kappa=1.0)

vonMises1D takes a 1D array x, a mean and kappa (inverse of standard deviation), and produces a von Mises pRF with given parameters. This shape can be thought of as a circular gaussian shape. Used for orientation or motion direction pRFs, for instance.

xnumpy.ndarray (1D)

space on which to calculate the von Mises. Assumed to be in the range (0, 2*np.pi)

mufloat, optional

mean/mode of von Mises (the default is 0.0)

kappafloat, optional

dispersion coefficient of the von Mises, akin to invers of standard deviation of gaussian (the default is 1.0)

numpy.ndarray

von Mises values at x, peak has y-value of 1